diff --git a/daemons/execd/remoted_pidone.c b/daemons/execd/remoted_pidone.c index 0a6c251066..212eb4b64a 100644 --- a/daemons/execd/remoted_pidone.c +++ b/daemons/execd/remoted_pidone.c @@ -1,302 +1,303 @@ /* - * Copyright 2017-2024 the Pacemaker project contributors + * Copyright 2017-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include "pacemaker-execd.h" static pid_t main_pid = 0; static void sigdone(void) { crm_exit(CRM_EX_OK); } static void sigreap(void) { pid_t pid = 0; int status; do { /* * Opinions seem to differ as to what to put here: * -1, any child process * 0, any child process whose process group ID is equal to that of the calling process */ pid = waitpid(-1, &status, WNOHANG); if (pid == main_pid) { /* Exit when pacemaker-remote exits and use the same return code */ if (WIFEXITED(status)) { crm_exit(WEXITSTATUS(status)); } crm_exit(CRM_EX_ERROR); } } while (pid > 0); } static struct { int sig; void (*handler)(void); } sigmap[] = { { SIGCHLD, sigreap }, { SIGINT, sigdone }, }; /*! * \internal * \brief Check a line of text for a valid environment variable name * * \param[in] line Text to check * \param[out] first First character of valid name if found, NULL otherwise * \param[out] last Last character of valid name if found, NULL otherwise * * \return TRUE if valid name found, FALSE otherwise * \note It's reasonable to impose limitations on environment variable names * beyond what C or setenv() does: We only allow names that contain only * [a-zA-Z0-9_] characters and do not start with a digit. */ static bool find_env_var_name(char *line, char **first, char **last) { // Skip leading whitespace *first = line; while (isspace(**first)) { ++*first; } if (isalpha(**first) || (**first == '_')) { // Valid first character *last = *first; while (isalnum(*(*last + 1)) || (*(*last + 1) == '_')) { ++*last; } return TRUE; } *first = *last = NULL; return FALSE; } static void load_env_vars(const char *filename) { /* We haven't forked or initialized logging yet, so don't leave any file * descriptors open, and don't log -- silently ignore errors. */ FILE *fp = fopen(filename, "r"); if (fp != NULL) { char line[LINE_MAX] = { '\0', }; while (fgets(line, LINE_MAX, fp) != NULL) { char *name = NULL; char *end = NULL; char *value = NULL; char *quote = NULL; // Look for valid name immediately followed by equals sign if (find_env_var_name(line, &name, &end) && (*++end == '=')) { // Null-terminate name, and advance beyond equals sign *end++ = '\0'; // Check whether value is quoted if ((*end == '\'') || (*end == '"')) { quote = end++; } value = end; if (quote) { /* Value is remaining characters up to next non-backslashed * matching quote character. */ while (((*end != *quote) || (*(end - 1) == '\\')) && (*end != '\0')) { end++; } if (*end == *quote) { // Null-terminate value, and advance beyond close quote *end++ = '\0'; } else { // Matching closing quote wasn't found value = NULL; } } else { /* Value is remaining characters up to next non-backslashed * whitespace. */ while ((!isspace(*end) || (*(end - 1) == '\\')) && (*end != '\0')) { ++end; } if (end == (line + LINE_MAX - 1)) { // Line was too long value = NULL; } // Do NOT null-terminate value (yet) } /* We have a valid name and value, and end is now the character * after the closing quote or the first whitespace after the * unquoted value. Make sure the rest of the line is just * whitespace or a comment. */ if (value) { char *value_end = end; while (isspace(*end) && (*end != '\n')) { ++end; } if ((*end == '\n') || (*end == '#')) { if (quote == NULL) { // Now we can null-terminate an unquoted value *value_end = '\0'; } // Don't overwrite (bundle options take precedence) + // coverity[tainted_string] This can't easily be changed right now setenv(name, value, 0); } else { value = NULL; } } } if ((value == NULL) && (strchr(line, '\n') == NULL)) { // Eat remainder of line beyond LINE_MAX if (fscanf(fp, "%*[^\n]\n") == EOF) { value = NULL; // Don't care, make compiler happy } } } fclose(fp); } } void remoted_spawn_pidone(int argc, char **argv, char **envp) { sigset_t set; /* This environment variable exists for two purposes: * - For testing, setting it to "full" enables full PID 1 behavior even * when PID is not 1 * - Setting to "vars" enables just the loading of environment variables * from /etc/pacemaker/pcmk-init.env, which could be useful for testing or * containers with a custom PID 1 script that launches pacemaker-remoted. */ const char *pid1 = PCMK_VALUE_DEFAULT; if (getpid() != 1) { pid1 = pcmk__env_option(PCMK__ENV_REMOTE_PID1); if (!pcmk__str_any_of(pid1, "full", "vars", NULL)) { // Default, unset, or invalid return; } } /* When a container is launched, it may be given specific environment * variables, which for Pacemaker bundles are given in the bundle * configuration. However, that does not allow for host-specific values. * To allow for that, look for a special file containing a shell-like syntax * of name/value pairs, and export those into the environment. */ load_env_vars("/etc/pacemaker/pcmk-init.env"); if (strcmp(pid1, "vars") == 0) { return; } /* Containers can be expected to have /var/log, but they may not have * /var/log/pacemaker, so use a different default if no value has been * explicitly configured in the container's environment. */ if (pcmk__env_option(PCMK__ENV_LOGFILE) == NULL) { pcmk__set_env_option(PCMK__ENV_LOGFILE, "/var/log/pcmk-init.log", true); } sigfillset(&set); sigprocmask(SIG_BLOCK, &set, 0); main_pid = fork(); switch (main_pid) { case 0: sigprocmask(SIG_UNBLOCK, &set, NULL); setsid(); setpgid(0, 0); // Child remains as pacemaker-remoted return; case -1: crm_err("fork failed: %s", pcmk_rc_str(errno)); } /* Parent becomes the reaper of zombie processes */ /* Safe to initialize logging now if needed */ # ifdef HAVE_PROGNAME /* Differentiate ourselves in the 'ps' output */ { char *p; int i, maxlen; char *LastArgv = NULL; const char *name = "pcmk-init"; for (i = 0; i < argc; i++) { if (!i || (LastArgv + 1 == argv[i])) LastArgv = argv[i] + strlen(argv[i]); } for (i = 0; envp[i] != NULL; i++) { if ((LastArgv + 1) == envp[i]) { LastArgv = envp[i] + strlen(envp[i]); } } maxlen = (LastArgv - argv[0]) - 2; i = strlen(name); /* We can overwrite individual argv[] arguments */ snprintf(argv[0], maxlen, "%s", name); /* Now zero out everything else */ p = &argv[0][i]; while (p < LastArgv) { *p++ = '\0'; } argv[1] = NULL; } # endif // HAVE_PROGNAME while (1) { int sig; size_t i; sigwait(&set, &sig); for (i = 0; i < PCMK__NELEM(sigmap); i++) { if (sigmap[i].sig == sig) { sigmap[i].handler(); break; } } } } diff --git a/daemons/execd/remoted_schemas.c b/daemons/execd/remoted_schemas.c index d8ee8f405e..843a85bf02 100644 --- a/daemons/execd/remoted_schemas.c +++ b/daemons/execd/remoted_schemas.c @@ -1,290 +1,291 @@ /* - * Copyright 2023-2024 the Pacemaker project contributors + * Copyright 2023-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include #include "pacemaker-execd.h" static pid_t schema_fetch_pid = 0; static int rm_files(const char *pathname, const struct stat *sbuf, int type, struct FTW *ftwb) { /* Don't delete PCMK__REMOTE_SCHEMA_DIR . */ if (ftwb->level == 0) { return 0; } if (remove(pathname) != 0) { int rc = errno; crm_err("Could not remove %s: %s", pathname, pcmk_rc_str(rc)); return -1; } return 0; } static void clean_up_extra_schema_files(void) { const char *remote_schema_dir = pcmk__remote_schema_dir(); - struct stat sb; int rc; - rc = stat(remote_schema_dir, &sb); + /* Try to create the remote schema directory first. */ + rc = mkdir(remote_schema_dir, 0755); - if (rc == -1) { - if (errno == ENOENT) { - /* If the directory doesn't exist, try to make it first. */ - if (mkdir(remote_schema_dir, 0755) != 0) { - rc = errno; - crm_err("Could not create directory for schemas: %s", - pcmk_rc_str(rc)); - } + if (rc == 0) { + /* Success. */ + return; + } + + if (errno == EEXIST) { + /* The path already exists. Assume it's a directory and try to clear + * it out so we can download new schema files. If it's not a directory, + * nftw will fail and set errno. + */ + rc = nftw(remote_schema_dir, rm_files, 10, FTW_DEPTH|FTW_MOUNT|FTW_PHYS); + if (rc == 0) { + /* Success. */ + return; + } + + if (errno == ENOTDIR) { + /* Something other than a directory already has that name. */ + crm_err("%s already exists but is not a directory", + remote_schema_dir); } else { rc = errno; - crm_err("Could not create directory for schemas: %s", + crm_err("Could not clear directory %s: %s", remote_schema_dir, pcmk_rc_str(rc)); } - } else if (!S_ISDIR(sb.st_mode)) { - /* If something exists with the same name that's not a directory, that's - * an error. - */ - crm_err("%s already exists but is not a directory", remote_schema_dir); - } else { - /* It's a directory - clear it out so we can download potentially new - * schema files. - */ - rc = nftw(remote_schema_dir, rm_files, 10, FTW_DEPTH|FTW_MOUNT|FTW_PHYS); - - if (rc != 0) { - crm_err("Could not remove %s: %s", remote_schema_dir, pcmk_rc_str(rc)); - } + rc = errno; + crm_err("Could not create directory for schemas: %s", + pcmk_rc_str(rc)); } } static void write_extra_schema_file(xmlNode *xml, void *user_data) { const char *remote_schema_dir = pcmk__remote_schema_dir(); const char *file = NULL; char *path = NULL; int rc; file = crm_element_value(xml, PCMK_XA_PATH); if (file == NULL) { crm_warn("No destination path given in schema request"); return; } path = crm_strdup_printf("%s/%s", remote_schema_dir, file); /* The schema is a CDATA node, which is a child of the node. Traverse * all children and look for the first CDATA child. There can't be more than * one because we only have one file attribute on the parent. */ for (xmlNode *child = xml->children; child != NULL; child = child->next) { FILE *stream = NULL; if (child->type != XML_CDATA_SECTION_NODE) { continue; } stream = fopen(path, "w+"); if (stream == NULL) { crm_warn("Could not write schema file %s: %s", path, strerror(errno)); } else { rc = fprintf(stream, "%s", child->content); if (rc < 0) { crm_warn("Could not write schema file %s: %s", path, strerror(errno)); } fclose(stream); } break; } free(path); } static void get_schema_files(void) { int rc = pcmk_rc_ok; cib_t *cib = NULL; xmlNode *reply; cib = cib_new(); if (cib == NULL) { pcmk_common_cleanup(); _exit(CRM_EX_OSERR); } rc = cib->cmds->signon(cib, crm_system_name, cib_query); rc = pcmk_legacy2rc(rc); if (rc != pcmk_rc_ok) { crm_err("Could not connect to the CIB manager: %s", pcmk_rc_str(rc)); pcmk_common_cleanup(); _exit(pcmk_rc2exitc(rc)); } rc = cib->cmds->fetch_schemas(cib, &reply, pcmk__highest_schema_name(), cib_sync_call); if (rc != pcmk_ok) { crm_err("Could not get schema files: %s", pcmk_strerror(rc)); rc = pcmk_legacy2rc(rc); } else if (reply->children != NULL) { /* The returned document looks something like this: * * * * * * * * * ... * * * * ... * * * * * All the and tags are really just there for organizing * the XML a little better. What we really care about are the nodes, * and specifically the path attributes and the CDATA children (not shown) * of each. We can use an xpath query to reach down and get all the * nodes at once. * * If we already have the latest schema version, or we asked for one later * than what the cluster supports, we'll get back an empty node, * so all this will continue to work. It just won't do anything. */ crm_foreach_xpath_result(reply, "//" PCMK_XA_FILE, write_extra_schema_file, NULL); } free_xml(reply); cib__clean_up_connection(&cib); pcmk_common_cleanup(); _exit(pcmk_rc2exitc(rc)); } /* Load any additional schema files when the child is finished fetching and * saving them to disk. */ static void get_schema_files_complete(mainloop_child_t *p, pid_t pid, int core, int signo, int exitcode) { const char *errmsg = "Could not load additional schema files"; if ((signo == 0) && (exitcode == 0)) { const char *remote_schema_dir = pcmk__remote_schema_dir(); /* Don't just crm_schema_init here because that will load the base * schemas again too. Instead just load the things we fetched. */ pcmk__load_schemas_from_dir(remote_schema_dir); pcmk__sort_schemas(); crm_info("Fetching extra schema files completed successfully"); } else { if (signo == 0) { crm_err("%s: process %d exited %d", errmsg, (int) pid, exitcode); } else { crm_err("%s: process %d terminated with signal %d (%s)%s", errmsg, (int) pid, signo, strsignal(signo), (core? " and dumped core" : "")); } /* Clean up any incomplete schema data we might have been downloading when * the process timed out or crashed. We don't need to do any extra cleanup * because we never loaded the extra schemas, and we don't need to call * crm_schema_init because that was called in remoted_request_cib_schema_files * before this function. */ clean_up_extra_schema_files(); } } void remoted_request_cib_schema_files(void) { pid_t pid; int rc; /* If a previous schema-fetch process is still running when we're called * again, it's hung. Attempt to kill it before cleaning up the extra * directory. */ if (schema_fetch_pid != 0) { if (mainloop_child_kill(schema_fetch_pid) == FALSE) { crm_warn("Unable to kill pre-existing schema-fetch process"); return; } schema_fetch_pid = 0; } /* Clean up any extra schema files we downloaded from a previous cluster * connection. After the files are gone, we need to wipe them from * known_schemas, but there's no opposite operation for add_schema(). * * Instead, unload all the schemas. This means we'll also forget about all * installed schemas as well, which means that pcmk__highest_schema_name() * would fail. So we need to load the base schemas right now. */ clean_up_extra_schema_files(); crm_schema_cleanup(); crm_schema_init(); crm_info("Fetching extra schema files from cluster"); pid = fork(); switch (pid) { case -1: { rc = errno; crm_warn("Could not spawn process to get schema files: %s", pcmk_rc_str(rc)); break; } case 0: /* child */ get_schema_files(); break; default: /* parent */ schema_fetch_pid = pid; mainloop_child_add_with_flags(pid, 5 * 60 * 1000, "schema-fetch", NULL, mainloop_leave_pid_group, get_schema_files_complete); break; } } diff --git a/daemons/fenced/fenced_commands.c b/daemons/fenced/fenced_commands.c index 2501c63cb7..758420d4ef 100644 --- a/daemons/fenced/fenced_commands.c +++ b/daemons/fenced/fenced_commands.c @@ -1,3727 +1,3727 @@ /* - * Copyright 2009-2024 the Pacemaker project contributors + * Copyright 2009-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU General Public License version 2 * or later (GPLv2+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include GHashTable *device_list = NULL; GHashTable *topology = NULL; static GList *cmd_list = NULL; static GHashTable *fenced_handlers = NULL; struct device_search_s { /* target of fence action */ char *host; /* requested fence action */ char *action; /* timeout to use if a device is queried dynamically for possible targets */ int per_device_timeout; /* number of registered fencing devices at time of request */ int replies_needed; /* number of device replies received so far */ int replies_received; /* whether the target is eligible to perform requested action (or off) */ bool allow_self; /* private data to pass to search callback function */ void *user_data; /* function to call when all replies have been received */ void (*callback) (GList * devices, void *user_data); /* devices capable of performing requested action (or off if remapping) */ GList *capable; /* Whether to perform searches that support the action */ uint32_t support_action_only; }; static gboolean stonith_device_dispatch(gpointer user_data); static void st_child_done(int pid, const pcmk__action_result_t *result, void *user_data); static void search_devices_record_result(struct device_search_s *search, const char *device, gboolean can_fence); static int get_agent_metadata(const char *agent, xmlNode **metadata); static void read_action_metadata(stonith_device_t *device); static enum fenced_target_by unpack_level_kind(const xmlNode *level); typedef struct async_command_s { int id; int pid; int fd_stdout; uint32_t options; int default_timeout; /* seconds */ int timeout; /* seconds */ int start_delay; // seconds (-1 means disable static/random fencing delays) int delay_id; char *op; char *origin; char *client; char *client_name; char *remote_op_id; char *target; uint32_t target_nodeid; char *action; char *device; GList *device_list; GList *next_device_iter; // device_list entry for next device to execute void *internal_user_data; void (*done_cb) (int pid, const pcmk__action_result_t *result, void *user_data); guint timer_sigterm; guint timer_sigkill; /*! If the operation timed out, this is the last signal * we sent to the process to get it to terminate */ int last_timeout_signo; stonith_device_t *active_on; stonith_device_t *activating_on; } async_command_t; static xmlNode *construct_async_reply(const async_command_t *cmd, const pcmk__action_result_t *result); static gboolean is_action_required(const char *action, const stonith_device_t *device) { return (device != NULL) && device->automatic_unfencing && pcmk__str_eq(action, PCMK_ACTION_ON, pcmk__str_none); } static int get_action_delay_max(const stonith_device_t *device, const char *action) { const char *value = NULL; guint delay_max = 0U; if (!pcmk__is_fencing_action(action)) { return 0; } value = g_hash_table_lookup(device->params, PCMK_STONITH_DELAY_MAX); if (value) { pcmk_parse_interval_spec(value, &delay_max); delay_max /= 1000; } return (int) delay_max; } static int get_action_delay_base(const stonith_device_t *device, const char *action, const char *target) { char *hash_value = NULL; guint delay_base = 0U; if (!pcmk__is_fencing_action(action)) { return 0; } hash_value = g_hash_table_lookup(device->params, PCMK_STONITH_DELAY_BASE); if (hash_value) { char *value = pcmk__str_copy(hash_value); char *valptr = value; if (target != NULL) { for (char *val = strtok(value, "; \t"); val != NULL; val = strtok(NULL, "; \t")) { char *mapval = strchr(val, ':'); if (mapval == NULL || mapval[1] == 0) { crm_err("pcmk_delay_base: empty value in mapping", val); continue; } if (mapval != val && strncasecmp(target, val, (size_t)(mapval - val)) == 0) { value = mapval + 1; crm_debug("pcmk_delay_base mapped to %s for %s", value, target); break; } } } if (strchr(value, ':') == 0) { pcmk_parse_interval_spec(value, &delay_base); delay_base /= 1000; } free(valptr); } return (int) delay_base; } /*! * \internal * \brief Override STONITH timeout with pcmk_*_timeout if available * * \param[in] device STONITH device to use * \param[in] action STONITH action name * \param[in] default_timeout Timeout to use if device does not have * a pcmk_*_timeout parameter for action * * \return Value of pcmk_(action)_timeout if available, otherwise default_timeout * \note For consistency, it would be nice if reboot/off/on timeouts could be * set the same way as start/stop/monitor timeouts, i.e. with an * entry in the fencing resource configuration. However that * is insufficient because fencing devices may be registered directly via * the fencer's register_device() API instead of going through the CIB * (e.g. stonith_admin uses it for its -R option, and the executor uses it * to ensure a device is registered when a command is issued). As device * properties, pcmk_*_timeout parameters can be grabbed by the fencer when * the device is registered, whether by CIB change or API call. */ static int get_action_timeout(const stonith_device_t *device, const char *action, int default_timeout) { if (action && device && device->params) { char buffer[64] = { 0, }; const char *value = NULL; /* If "reboot" was requested but the device does not support it, * we will remap to "off", so check timeout for "off" instead */ if (pcmk__str_eq(action, PCMK_ACTION_REBOOT, pcmk__str_none) && !pcmk_is_set(device->flags, st_device_supports_reboot)) { crm_trace("%s doesn't support reboot, using timeout for off instead", device->id); action = PCMK_ACTION_OFF; } /* If the device config specified an action-specific timeout, use it */ snprintf(buffer, sizeof(buffer), "pcmk_%s_timeout", action); value = g_hash_table_lookup(device->params, buffer); if (value) { long long timeout_ms = crm_get_msec(value); return (int) QB_MIN(timeout_ms / 1000, INT_MAX); } } return default_timeout; } /*! * \internal * \brief Get the currently executing device for a fencing operation * * \param[in] cmd Fencing operation to check * * \return Currently executing device for \p cmd if any, otherwise NULL */ static stonith_device_t * cmd_device(const async_command_t *cmd) { if ((cmd == NULL) || (cmd->device == NULL) || (device_list == NULL)) { return NULL; } return g_hash_table_lookup(device_list, cmd->device); } /*! * \internal * \brief Return the configured reboot action for a given device * * \param[in] device_id Device ID * * \return Configured reboot action for \p device_id */ const char * fenced_device_reboot_action(const char *device_id) { const char *action = NULL; if ((device_list != NULL) && (device_id != NULL)) { stonith_device_t *device = g_hash_table_lookup(device_list, device_id); if ((device != NULL) && (device->params != NULL)) { action = g_hash_table_lookup(device->params, "pcmk_reboot_action"); } } return pcmk__s(action, PCMK_ACTION_REBOOT); } /*! * \internal * \brief Check whether a given device supports the "on" action * * \param[in] device_id Device ID * * \return true if \p device_id supports "on", otherwise false */ bool fenced_device_supports_on(const char *device_id) { if ((device_list != NULL) && (device_id != NULL)) { stonith_device_t *device = g_hash_table_lookup(device_list, device_id); if (device != NULL) { return pcmk_is_set(device->flags, st_device_supports_on); } } return false; } static void free_async_command(async_command_t * cmd) { if (!cmd) { return; } if (cmd->delay_id) { g_source_remove(cmd->delay_id); } cmd_list = g_list_remove(cmd_list, cmd); g_list_free_full(cmd->device_list, free); free(cmd->device); free(cmd->action); free(cmd->target); free(cmd->remote_op_id); free(cmd->client); free(cmd->client_name); free(cmd->origin); free(cmd->op); free(cmd); } /*! * \internal * \brief Create a new asynchronous fencing operation from request XML * * \param[in] msg Fencing request XML (from IPC or CPG) * * \return Newly allocated fencing operation on success, otherwise NULL * * \note This asserts on memory errors, so a NULL return indicates an * unparseable message. */ static async_command_t * create_async_command(xmlNode *msg) { xmlNode *op = NULL; async_command_t *cmd = NULL; int rc = pcmk_rc_ok; if (msg == NULL) { return NULL; } op = get_xpath_object("//@" PCMK__XE_ST_DEVICE_ACTION, msg, LOG_ERR); if (op == NULL) { return NULL; } cmd = pcmk__assert_alloc(1, sizeof(async_command_t)); // All messages must include these cmd->action = crm_element_value_copy(op, PCMK__XA_ST_DEVICE_ACTION); cmd->op = crm_element_value_copy(msg, PCMK__XA_ST_OP); cmd->client = crm_element_value_copy(msg, PCMK__XA_ST_CLIENTID); if ((cmd->action == NULL) || (cmd->op == NULL) || (cmd->client == NULL)) { free_async_command(cmd); return NULL; } crm_element_value_int(msg, PCMK__XA_ST_CALLID, &(cmd->id)); crm_element_value_int(msg, PCMK__XA_ST_DELAY, &(cmd->start_delay)); crm_element_value_int(msg, PCMK__XA_ST_TIMEOUT, &(cmd->default_timeout)); cmd->timeout = cmd->default_timeout; rc = pcmk__xe_get_flags(msg, PCMK__XA_ST_CALLOPT, &(cmd->options), st_opt_none); if (rc != pcmk_rc_ok) { crm_warn("Couldn't parse options from request: %s", pcmk_rc_str(rc)); } cmd->origin = crm_element_value_copy(msg, PCMK__XA_SRC); cmd->remote_op_id = crm_element_value_copy(msg, PCMK__XA_ST_REMOTE_OP); cmd->client_name = crm_element_value_copy(msg, PCMK__XA_ST_CLIENTNAME); cmd->target = crm_element_value_copy(op, PCMK__XA_ST_TARGET); cmd->device = crm_element_value_copy(op, PCMK__XA_ST_DEVICE_ID); cmd->done_cb = st_child_done; // Track in global command list cmd_list = g_list_append(cmd_list, cmd); return cmd; } static int get_action_limit(stonith_device_t * device) { const char *value = NULL; int action_limit = 1; value = g_hash_table_lookup(device->params, PCMK_STONITH_ACTION_LIMIT); if ((value == NULL) || (pcmk__scan_min_int(value, &action_limit, INT_MIN) != pcmk_rc_ok) || (action_limit == 0)) { action_limit = 1; } return action_limit; } static int get_active_cmds(stonith_device_t * device) { int counter = 0; GList *gIter = NULL; GList *gIterNext = NULL; CRM_CHECK(device != NULL, return 0); for (gIter = cmd_list; gIter != NULL; gIter = gIterNext) { async_command_t *cmd = gIter->data; gIterNext = gIter->next; if (cmd->active_on == device) { counter++; } } return counter; } static void fork_cb(int pid, void *user_data) { async_command_t *cmd = (async_command_t *) user_data; stonith_device_t * device = /* in case of a retry we've done the move from activating_on to active_on already */ cmd->activating_on?cmd->activating_on:cmd->active_on; pcmk__assert(device != NULL); crm_debug("Operation '%s' [%d]%s%s using %s now running with %ds timeout", cmd->action, pid, ((cmd->target == NULL)? "" : " targeting "), pcmk__s(cmd->target, ""), device->id, cmd->timeout); cmd->active_on = device; cmd->activating_on = NULL; } static int get_agent_metadata_cb(gpointer data) { stonith_device_t *device = data; guint period_ms; switch (get_agent_metadata(device->agent, &device->agent_metadata)) { case pcmk_rc_ok: if (device->agent_metadata) { read_action_metadata(device); stonith__device_parameter_flags(&(device->flags), device->id, device->agent_metadata); } return G_SOURCE_REMOVE; case EAGAIN: period_ms = pcmk__mainloop_timer_get_period(device->timer); if (period_ms < 160 * 1000) { mainloop_timer_set_period(device->timer, 2 * period_ms); } return G_SOURCE_CONTINUE; default: return G_SOURCE_REMOVE; } } /*! * \internal * \brief Call a command's action callback for an internal (not library) result * * \param[in,out] cmd Command to report result for * \param[in] execution_status Execution status to use for result * \param[in] exit_status Exit status to use for result * \param[in] exit_reason Exit reason to use for result */ static void report_internal_result(async_command_t *cmd, int exit_status, int execution_status, const char *exit_reason) { pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; pcmk__set_result(&result, exit_status, execution_status, exit_reason); cmd->done_cb(0, &result, cmd); pcmk__reset_result(&result); } static gboolean stonith_device_execute(stonith_device_t * device) { int exec_rc = 0; const char *action_str = NULL; const char *host_arg = NULL; async_command_t *cmd = NULL; stonith_action_t *action = NULL; int active_cmds = 0; int action_limit = 0; GList *gIter = NULL; GList *gIterNext = NULL; CRM_CHECK(device != NULL, return FALSE); active_cmds = get_active_cmds(device); action_limit = get_action_limit(device); if (action_limit > -1 && active_cmds >= action_limit) { crm_trace("%s is over its action limit of %d (%u active action%s)", device->id, action_limit, active_cmds, pcmk__plural_s(active_cmds)); return TRUE; } for (gIter = device->pending_ops; gIter != NULL; gIter = gIterNext) { async_command_t *pending_op = gIter->data; gIterNext = gIter->next; if (pending_op && pending_op->delay_id) { crm_trace("Operation '%s'%s%s using %s was asked to run too early, " "waiting for start delay of %ds", pending_op->action, ((pending_op->target == NULL)? "" : " targeting "), pcmk__s(pending_op->target, ""), device->id, pending_op->start_delay); continue; } device->pending_ops = g_list_remove_link(device->pending_ops, gIter); g_list_free_1(gIter); cmd = pending_op; break; } if (cmd == NULL) { crm_trace("No actions using %s are needed", device->id); return TRUE; } if (pcmk__str_any_of(device->agent, STONITH_WATCHDOG_AGENT, STONITH_WATCHDOG_AGENT_INTERNAL, NULL)) { if (pcmk__is_fencing_action(cmd->action)) { if (node_does_watchdog_fencing(stonith_our_uname)) { pcmk__panic(__func__); goto done; } } else { crm_info("Faking success for %s watchdog operation", cmd->action); report_internal_result(cmd, CRM_EX_OK, PCMK_EXEC_DONE, NULL); goto done; } } #if SUPPORT_CIBSECRETS exec_rc = pcmk__substitute_secrets(device->id, device->params); if (exec_rc != pcmk_rc_ok) { if (pcmk__str_eq(cmd->action, PCMK_ACTION_STOP, pcmk__str_none)) { crm_info("Proceeding with stop operation for %s " "despite being unable to load CIB secrets (%s)", device->id, pcmk_rc_str(exec_rc)); } else { crm_err("Considering %s unconfigured " "because unable to load CIB secrets: %s", device->id, pcmk_rc_str(exec_rc)); report_internal_result(cmd, CRM_EX_ERROR, PCMK_EXEC_NO_SECRETS, "Failed to get CIB secrets"); goto done; } } #endif action_str = cmd->action; if (pcmk__str_eq(cmd->action, PCMK_ACTION_REBOOT, pcmk__str_none) && !pcmk_is_set(device->flags, st_device_supports_reboot)) { crm_notice("Remapping 'reboot' action%s%s using %s to 'off' " "because agent '%s' does not support reboot", ((cmd->target == NULL)? "" : " targeting "), pcmk__s(cmd->target, ""), device->id, device->agent); action_str = PCMK_ACTION_OFF; } if (pcmk_is_set(device->flags, st_device_supports_parameter_port)) { host_arg = "port"; } else if (pcmk_is_set(device->flags, st_device_supports_parameter_plug)) { host_arg = "plug"; } action = stonith__action_create(device->agent, action_str, cmd->target, cmd->target_nodeid, cmd->timeout, device->params, device->aliases, host_arg); /* for async exec, exec_rc is negative for early error exit otherwise handling of success/errors is done via callbacks */ cmd->activating_on = device; exec_rc = stonith__execute_async(action, (void *)cmd, cmd->done_cb, fork_cb); if (exec_rc < 0) { cmd->activating_on = NULL; cmd->done_cb(0, stonith__action_result(action), cmd); stonith__destroy_action(action); } done: /* Device might get triggered to work by multiple fencing commands * simultaneously. Trigger the device again to make sure any * remaining concurrent commands get executed. */ if (device->pending_ops) { mainloop_set_trigger(device->work); } return TRUE; } static gboolean stonith_device_dispatch(gpointer user_data) { return stonith_device_execute(user_data); } static gboolean start_delay_helper(gpointer data) { async_command_t *cmd = data; stonith_device_t *device = cmd_device(cmd); cmd->delay_id = 0; if (device) { mainloop_set_trigger(device->work); } return FALSE; } static void schedule_stonith_command(async_command_t * cmd, stonith_device_t * device) { int delay_max = 0; int delay_base = 0; int requested_delay = cmd->start_delay; CRM_CHECK(cmd != NULL, return); CRM_CHECK(device != NULL, return); if (cmd->device) { free(cmd->device); } if (device->include_nodeid && (cmd->target != NULL)) { crm_node_t *node = pcmk__get_node(0, cmd->target, NULL, pcmk__node_search_cluster_member); cmd->target_nodeid = node->id; } cmd->device = pcmk__str_copy(device->id); cmd->timeout = get_action_timeout(device, cmd->action, cmd->default_timeout); if (cmd->remote_op_id) { crm_debug("Scheduling '%s' action%s%s using %s for remote peer %s " "with op id %.8s and timeout %ds", cmd->action, (cmd->target == NULL)? "" : " targeting ", pcmk__s(cmd->target, ""), device->id, cmd->origin, cmd->remote_op_id, cmd->timeout); } else { crm_debug("Scheduling '%s' action%s%s using %s for %s with timeout %ds", cmd->action, (cmd->target == NULL)? "" : " targeting ", pcmk__s(cmd->target, ""), device->id, cmd->client, cmd->timeout); } device->pending_ops = g_list_append(device->pending_ops, cmd); mainloop_set_trigger(device->work); // Value -1 means disable any static/random fencing delays if (requested_delay < 0) { return; } delay_max = get_action_delay_max(device, cmd->action); delay_base = get_action_delay_base(device, cmd->action, cmd->target); if (delay_max == 0) { delay_max = delay_base; } if (delay_max < delay_base) { crm_warn(PCMK_STONITH_DELAY_BASE " (%ds) is larger than " PCMK_STONITH_DELAY_MAX " (%ds) for %s using %s " "(limiting to maximum delay)", delay_base, delay_max, cmd->action, device->id); delay_base = delay_max; } if (delay_max > 0) { - // coverity[dontcall] It doesn't matter here if rand() is predictable cmd->start_delay += + // coverity[dont_call] It doesn't matter here if rand() is predictable ((delay_max != delay_base)?(rand() % (delay_max - delay_base)):0) + delay_base; } if (cmd->start_delay > 0) { crm_notice("Delaying '%s' action%s%s using %s for %ds " CRM_XS " timeout=%ds requested_delay=%ds base=%ds max=%ds", cmd->action, (cmd->target == NULL)? "" : " targeting ", pcmk__s(cmd->target, ""), device->id, cmd->start_delay, cmd->timeout, requested_delay, delay_base, delay_max); cmd->delay_id = g_timeout_add_seconds(cmd->start_delay, start_delay_helper, cmd); } } static void free_device(gpointer data) { GList *gIter = NULL; stonith_device_t *device = data; g_hash_table_destroy(device->params); g_hash_table_destroy(device->aliases); for (gIter = device->pending_ops; gIter != NULL; gIter = gIter->next) { async_command_t *cmd = gIter->data; crm_warn("Removal of device '%s' purged operation '%s'", device->id, cmd->action); report_internal_result(cmd, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, "Device was removed before action could be executed"); } g_list_free(device->pending_ops); g_list_free_full(device->targets, free); if (device->timer) { mainloop_timer_stop(device->timer); mainloop_timer_del(device->timer); } mainloop_destroy_trigger(device->work); free_xml(device->agent_metadata); free(device->namespace); if (device->on_target_actions != NULL) { g_string_free(device->on_target_actions, TRUE); } free(device->agent); free(device->id); free(device); } void free_device_list(void) { if (device_list != NULL) { g_hash_table_destroy(device_list); device_list = NULL; } } void init_device_list(void) { if (device_list == NULL) { device_list = pcmk__strkey_table(NULL, free_device); } } static GHashTable * build_port_aliases(const char *hostmap, GList ** targets) { char *name = NULL; int last = 0, lpc = 0, max = 0, added = 0; GHashTable *aliases = pcmk__strikey_table(free, free); if (hostmap == NULL) { return aliases; } max = strlen(hostmap); for (; lpc <= max; lpc++) { switch (hostmap[lpc]) { /* Skip escaped chars */ case '\\': lpc++; break; /* Assignment chars */ case '=': case ':': if (lpc > last) { free(name); name = pcmk__assert_alloc(1, 1 + lpc - last); memcpy(name, hostmap + last, lpc - last); } last = lpc + 1; break; /* Delimeter chars */ /* case ',': Potentially used to specify multiple ports */ case 0: case ';': case ' ': case '\t': if (name) { char *value = NULL; int k = 0; value = pcmk__assert_alloc(1, 1 + lpc - last); memcpy(value, hostmap + last, lpc - last); for (int i = 0; value[i] != '\0'; i++) { if (value[i] != '\\') { value[k++] = value[i]; } } value[k] = '\0'; crm_debug("Adding alias '%s'='%s'", name, value); g_hash_table_replace(aliases, name, value); if (targets) { *targets = g_list_append(*targets, pcmk__str_copy(value)); } value = NULL; name = NULL; added++; } else if (lpc > last) { crm_debug("Parse error at offset %d near '%s'", lpc - last, hostmap + last); } last = lpc + 1; break; } if (hostmap[lpc] == 0) { break; } } if (added == 0) { crm_info("No host mappings detected in '%s'", hostmap); } free(name); return aliases; } GHashTable *metadata_cache = NULL; void free_metadata_cache(void) { if (metadata_cache != NULL) { g_hash_table_destroy(metadata_cache); metadata_cache = NULL; } } static void init_metadata_cache(void) { if (metadata_cache == NULL) { metadata_cache = pcmk__strkey_table(free, free); } } int get_agent_metadata(const char *agent, xmlNode ** metadata) { char *buffer = NULL; if (metadata == NULL) { return EINVAL; } *metadata = NULL; if (pcmk__str_eq(agent, STONITH_WATCHDOG_AGENT_INTERNAL, pcmk__str_none)) { return pcmk_rc_ok; } init_metadata_cache(); buffer = g_hash_table_lookup(metadata_cache, agent); if (buffer == NULL) { stonith_t *st = stonith_api_new(); int rc; if (st == NULL) { crm_warn("Could not get agent meta-data: " "API memory allocation failed"); return EAGAIN; } rc = st->cmds->metadata(st, st_opt_sync_call, agent, NULL, &buffer, 10); stonith_api_delete(st); if (rc || !buffer) { crm_err("Could not retrieve metadata for fencing agent %s", agent); return EAGAIN; } g_hash_table_replace(metadata_cache, pcmk__str_copy(agent), buffer); } *metadata = pcmk__xml_parse(buffer); return pcmk_rc_ok; } static gboolean is_nodeid_required(xmlNode * xml) { xmlXPathObjectPtr xpath = NULL; if (stand_alone) { return FALSE; } if (!xml) { return FALSE; } xpath = xpath_search(xml, "//" PCMK_XE_PARAMETER "[@" PCMK_XA_NAME "='nodeid']"); if (numXpathResults(xpath) <= 0) { freeXpathObject(xpath); return FALSE; } freeXpathObject(xpath); return TRUE; } static void read_action_metadata(stonith_device_t *device) { xmlXPathObjectPtr xpath = NULL; int max = 0; int lpc = 0; if (device->agent_metadata == NULL) { return; } xpath = xpath_search(device->agent_metadata, "//action"); max = numXpathResults(xpath); if (max <= 0) { freeXpathObject(xpath); return; } for (lpc = 0; lpc < max; lpc++) { const char *action = NULL; xmlNode *match = getXpathResult(xpath, lpc); CRM_LOG_ASSERT(match != NULL); if(match == NULL) { continue; }; action = crm_element_value(match, PCMK_XA_NAME); if (pcmk__str_eq(action, PCMK_ACTION_LIST, pcmk__str_none)) { stonith__set_device_flags(device->flags, device->id, st_device_supports_list); } else if (pcmk__str_eq(action, PCMK_ACTION_STATUS, pcmk__str_none)) { stonith__set_device_flags(device->flags, device->id, st_device_supports_status); } else if (pcmk__str_eq(action, PCMK_ACTION_REBOOT, pcmk__str_none)) { stonith__set_device_flags(device->flags, device->id, st_device_supports_reboot); } else if (pcmk__str_eq(action, PCMK_ACTION_ON, pcmk__str_none)) { /* PCMK_XA_AUTOMATIC means the cluster will unfence a node when it * joins. * * @COMPAT PCMK__XA_REQUIRED is a deprecated synonym for * PCMK_XA_AUTOMATIC. */ if (pcmk__xe_attr_is_true(match, PCMK_XA_AUTOMATIC) || pcmk__xe_attr_is_true(match, PCMK__XA_REQUIRED)) { device->automatic_unfencing = TRUE; } stonith__set_device_flags(device->flags, device->id, st_device_supports_on); } if ((action != NULL) && pcmk__xe_attr_is_true(match, PCMK_XA_ON_TARGET)) { pcmk__add_word(&(device->on_target_actions), 64, action); } } freeXpathObject(xpath); } /*! * \internal * \brief Set a pcmk_*_action parameter if not already set * * \param[in,out] params Device parameters * \param[in] action Name of action * \param[in] value Value to use if action is not already set */ static void map_action(GHashTable *params, const char *action, const char *value) { char *key = crm_strdup_printf("pcmk_%s_action", action); if (g_hash_table_lookup(params, key)) { crm_warn("Ignoring %s='%s', see %s instead", STONITH_ATTR_ACTION_OP, value, key); free(key); } else { crm_warn("Mapping %s='%s' to %s='%s'", STONITH_ATTR_ACTION_OP, value, key, value); g_hash_table_insert(params, key, pcmk__str_copy(value)); } } /*! * \internal * \brief Create device parameter table from XML * * \param[in] name Device name (used for logging only) * \param[in] dev XML containing device parameters */ static GHashTable * xml2device_params(const char *name, const xmlNode *dev) { GHashTable *params = xml2list(dev); const char *value; /* Action should never be specified in the device configuration, * but we support it for users who are familiar with other software * that worked that way. */ value = g_hash_table_lookup(params, STONITH_ATTR_ACTION_OP); if (value != NULL) { crm_warn("%s has '%s' parameter, which should never be specified in configuration", name, STONITH_ATTR_ACTION_OP); if (*value == '\0') { crm_warn("Ignoring empty '%s' parameter", STONITH_ATTR_ACTION_OP); } else if (strcmp(value, PCMK_ACTION_REBOOT) == 0) { crm_warn("Ignoring %s='reboot' (see " PCMK_OPT_STONITH_ACTION " cluster property instead)", STONITH_ATTR_ACTION_OP); } else if (strcmp(value, PCMK_ACTION_OFF) == 0) { map_action(params, PCMK_ACTION_REBOOT, value); } else { map_action(params, PCMK_ACTION_OFF, value); map_action(params, PCMK_ACTION_REBOOT, value); } g_hash_table_remove(params, STONITH_ATTR_ACTION_OP); } return params; } static const char * target_list_type(stonith_device_t * dev) { const char *check_type = NULL; check_type = g_hash_table_lookup(dev->params, PCMK_STONITH_HOST_CHECK); if (check_type == NULL) { if (g_hash_table_lookup(dev->params, PCMK_STONITH_HOST_LIST)) { check_type = PCMK_VALUE_STATIC_LIST; } else if (g_hash_table_lookup(dev->params, PCMK_STONITH_HOST_MAP)) { check_type = PCMK_VALUE_STATIC_LIST; } else if (pcmk_is_set(dev->flags, st_device_supports_list)) { check_type = PCMK_VALUE_DYNAMIC_LIST; } else if (pcmk_is_set(dev->flags, st_device_supports_status)) { check_type = PCMK_VALUE_STATUS; } else { check_type = PCMK_VALUE_NONE; } } return check_type; } static stonith_device_t * build_device_from_xml(xmlNode *dev) { const char *value; stonith_device_t *device = NULL; char *agent = crm_element_value_copy(dev, PCMK_XA_AGENT); CRM_CHECK(agent != NULL, return device); device = pcmk__assert_alloc(1, sizeof(stonith_device_t)); device->id = crm_element_value_copy(dev, PCMK_XA_ID); device->agent = agent; device->namespace = crm_element_value_copy(dev, PCMK__XA_NAMESPACE); device->params = xml2device_params(device->id, dev); value = g_hash_table_lookup(device->params, PCMK_STONITH_HOST_LIST); if (value) { device->targets = stonith__parse_targets(value); } value = g_hash_table_lookup(device->params, PCMK_STONITH_HOST_MAP); device->aliases = build_port_aliases(value, &(device->targets)); value = target_list_type(device); if (!pcmk__str_eq(value, PCMK_VALUE_STATIC_LIST, pcmk__str_casei) && (device->targets != NULL)) { // device->targets is necessary only with PCMK_VALUE_STATIC_LIST g_list_free_full(device->targets, free); device->targets = NULL; } switch (get_agent_metadata(device->agent, &device->agent_metadata)) { case pcmk_rc_ok: if (device->agent_metadata) { read_action_metadata(device); stonith__device_parameter_flags(&(device->flags), device->id, device->agent_metadata); } break; case EAGAIN: if (device->timer == NULL) { device->timer = mainloop_timer_add("get_agent_metadata", 10 * 1000, TRUE, get_agent_metadata_cb, device); } if (!mainloop_timer_running(device->timer)) { mainloop_timer_start(device->timer); } break; default: break; } value = g_hash_table_lookup(device->params, "nodeid"); if (!value) { device->include_nodeid = is_nodeid_required(device->agent_metadata); } value = crm_element_value(dev, PCMK__XA_RSC_PROVIDES); if (pcmk__str_eq(value, PCMK_VALUE_UNFENCING, pcmk__str_casei)) { device->automatic_unfencing = TRUE; } if (is_action_required(PCMK_ACTION_ON, device)) { crm_info("Fencing device '%s' requires unfencing", device->id); } if (device->on_target_actions != NULL) { crm_info("Fencing device '%s' requires actions (%s) to be executed " "on target", device->id, (const char *) device->on_target_actions->str); } device->work = mainloop_add_trigger(G_PRIORITY_HIGH, stonith_device_dispatch, device); /* TODO: Hook up priority */ return device; } static void schedule_internal_command(const char *origin, stonith_device_t * device, const char *action, const char *target, int timeout, void *internal_user_data, void (*done_cb) (int pid, const pcmk__action_result_t *result, void *user_data)) { async_command_t *cmd = NULL; cmd = pcmk__assert_alloc(1, sizeof(async_command_t)); cmd->id = -1; cmd->default_timeout = timeout ? timeout : 60; cmd->timeout = cmd->default_timeout; cmd->action = pcmk__str_copy(action); cmd->target = pcmk__str_copy(target); cmd->device = pcmk__str_copy(device->id); cmd->origin = pcmk__str_copy(origin); cmd->client = pcmk__str_copy(crm_system_name); cmd->client_name = pcmk__str_copy(crm_system_name); cmd->internal_user_data = internal_user_data; cmd->done_cb = done_cb; /* cmd, not internal_user_data, is passed to 'done_cb' as the userdata */ schedule_stonith_command(cmd, device); } // Fence agent status commands use custom exit status codes enum fence_status_code { fence_status_invalid = -1, fence_status_active = 0, fence_status_unknown = 1, fence_status_inactive = 2, }; static void status_search_cb(int pid, const pcmk__action_result_t *result, void *user_data) { async_command_t *cmd = user_data; struct device_search_s *search = cmd->internal_user_data; stonith_device_t *dev = cmd_device(cmd); gboolean can = FALSE; free_async_command(cmd); if (!dev) { search_devices_record_result(search, NULL, FALSE); return; } mainloop_set_trigger(dev->work); if (result->execution_status != PCMK_EXEC_DONE) { crm_warn("Assuming %s cannot fence %s " "because status could not be executed: %s%s%s%s", dev->id, search->host, pcmk_exec_status_str(result->execution_status), ((result->exit_reason == NULL)? "" : " ("), ((result->exit_reason == NULL)? "" : result->exit_reason), ((result->exit_reason == NULL)? "" : ")")); search_devices_record_result(search, dev->id, FALSE); return; } switch (result->exit_status) { case fence_status_unknown: crm_trace("%s reported it cannot fence %s", dev->id, search->host); break; case fence_status_active: case fence_status_inactive: crm_trace("%s reported it can fence %s", dev->id, search->host); can = TRUE; break; default: crm_warn("Assuming %s cannot fence %s " "(status returned unknown code %d)", dev->id, search->host, result->exit_status); break; } search_devices_record_result(search, dev->id, can); } static void dynamic_list_search_cb(int pid, const pcmk__action_result_t *result, void *user_data) { async_command_t *cmd = user_data; struct device_search_s *search = cmd->internal_user_data; stonith_device_t *dev = cmd_device(cmd); gboolean can_fence = FALSE; free_async_command(cmd); /* Host/alias must be in the list output to be eligible to be fenced * * Will cause problems if down'd nodes aren't listed or (for virtual nodes) * if the guest is still listed despite being moved to another machine */ if (!dev) { search_devices_record_result(search, NULL, FALSE); return; } mainloop_set_trigger(dev->work); if (pcmk__result_ok(result)) { crm_info("Refreshing target list for %s", dev->id); g_list_free_full(dev->targets, free); dev->targets = stonith__parse_targets(result->action_stdout); dev->targets_age = time(NULL); } else if (dev->targets != NULL) { if (result->execution_status == PCMK_EXEC_DONE) { crm_info("Reusing most recent target list for %s " "because list returned error code %d", dev->id, result->exit_status); } else { crm_info("Reusing most recent target list for %s " "because list could not be executed: %s%s%s%s", dev->id, pcmk_exec_status_str(result->execution_status), ((result->exit_reason == NULL)? "" : " ("), ((result->exit_reason == NULL)? "" : result->exit_reason), ((result->exit_reason == NULL)? "" : ")")); } } else { // We have never successfully executed list if (result->execution_status == PCMK_EXEC_DONE) { crm_warn("Assuming %s cannot fence %s " "because list returned error code %d", dev->id, search->host, result->exit_status); } else { crm_warn("Assuming %s cannot fence %s " "because list could not be executed: %s%s%s%s", dev->id, search->host, pcmk_exec_status_str(result->execution_status), ((result->exit_reason == NULL)? "" : " ("), ((result->exit_reason == NULL)? "" : result->exit_reason), ((result->exit_reason == NULL)? "" : ")")); } /* Fall back to pcmk_host_check=PCMK_VALUE_STATUS if the user didn't * explicitly specify PCMK_VALUE_DYNAMIC_LIST */ if (g_hash_table_lookup(dev->params, PCMK_STONITH_HOST_CHECK) == NULL) { crm_notice("Switching to pcmk_host_check='status' for %s", dev->id); pcmk__insert_dup(dev->params, PCMK_STONITH_HOST_CHECK, PCMK_VALUE_STATUS); } } if (dev->targets) { const char *alias = g_hash_table_lookup(dev->aliases, search->host); if (!alias) { alias = search->host; } if (pcmk__str_in_list(alias, dev->targets, pcmk__str_casei)) { can_fence = TRUE; } } search_devices_record_result(search, dev->id, can_fence); } /*! * \internal * \brief Returns true if any key in first is not in second or second has a different value for key */ static int device_params_diff(GHashTable *first, GHashTable *second) { char *key = NULL; char *value = NULL; GHashTableIter gIter; g_hash_table_iter_init(&gIter, first); while (g_hash_table_iter_next(&gIter, (void **)&key, (void **)&value)) { if(strstr(key, "CRM_meta") == key) { continue; } else if (strcmp(key, PCMK_XA_CRM_FEATURE_SET) == 0) { continue; } else { char *other_value = g_hash_table_lookup(second, key); if (!other_value || !pcmk__str_eq(other_value, value, pcmk__str_casei)) { crm_trace("Different value for %s: %s != %s", key, other_value, value); return 1; } } } return 0; } /*! * \internal * \brief Checks to see if an identical device already exists in the device_list */ static stonith_device_t * device_has_duplicate(const stonith_device_t *device) { stonith_device_t *dup = g_hash_table_lookup(device_list, device->id); if (!dup) { crm_trace("No match for %s", device->id); return NULL; } else if (!pcmk__str_eq(dup->agent, device->agent, pcmk__str_casei)) { crm_trace("Different agent: %s != %s", dup->agent, device->agent); return NULL; } /* Use calculate_operation_digest() here? */ if (device_params_diff(device->params, dup->params) || device_params_diff(dup->params, device->params)) { return NULL; } crm_trace("Match"); return dup; } int stonith_device_register(xmlNode *dev, gboolean from_cib) { stonith_device_t *dup = NULL; stonith_device_t *device = build_device_from_xml(dev); guint ndevices = 0; int rv = pcmk_ok; CRM_CHECK(device != NULL, return -ENOMEM); /* do we have a watchdog-device? */ if (pcmk__str_eq(device->id, STONITH_WATCHDOG_ID, pcmk__str_none) || pcmk__str_any_of(device->agent, STONITH_WATCHDOG_AGENT, STONITH_WATCHDOG_AGENT_INTERNAL, NULL)) do { if (stonith_watchdog_timeout_ms <= 0) { crm_err("Ignoring watchdog fence device without " PCMK_OPT_STONITH_WATCHDOG_TIMEOUT " set."); rv = -ENODEV; /* fall through to cleanup & return */ } else if (!pcmk__str_any_of(device->agent, STONITH_WATCHDOG_AGENT, STONITH_WATCHDOG_AGENT_INTERNAL, NULL)) { crm_err("Ignoring watchdog fence device with unknown " "agent '%s' unequal '" STONITH_WATCHDOG_AGENT "'.", device->agent?device->agent:""); rv = -ENODEV; /* fall through to cleanup & return */ } else if (!pcmk__str_eq(device->id, STONITH_WATCHDOG_ID, pcmk__str_none)) { crm_err("Ignoring watchdog fence device " "named %s !='"STONITH_WATCHDOG_ID"'.", device->id?device->id:""); rv = -ENODEV; /* fall through to cleanup & return */ } else { if (pcmk__str_eq(device->agent, STONITH_WATCHDOG_AGENT, pcmk__str_none)) { /* this either has an empty list or the targets configured for watchdog-fencing */ g_list_free_full(stonith_watchdog_targets, free); stonith_watchdog_targets = device->targets; device->targets = NULL; } if (node_does_watchdog_fencing(stonith_our_uname)) { g_list_free_full(device->targets, free); device->targets = stonith__parse_targets(stonith_our_uname); pcmk__insert_dup(device->params, PCMK_STONITH_HOST_LIST, stonith_our_uname); /* proceed as with any other stonith-device */ break; } crm_debug("Skip registration of watchdog fence device on node not in host-list."); /* cleanup and fall through to more cleanup and return */ device->targets = NULL; stonith_device_remove(device->id, from_cib); } free_device(device); return rv; } while (0); dup = device_has_duplicate(device); if (dup) { ndevices = g_hash_table_size(device_list); crm_debug("Device '%s' already in device list (%d active device%s)", device->id, ndevices, pcmk__plural_s(ndevices)); free_device(device); device = dup; dup = g_hash_table_lookup(device_list, device->id); dup->dirty = FALSE; } else { stonith_device_t *old = g_hash_table_lookup(device_list, device->id); if (from_cib && old && old->api_registered) { /* If the cib is writing over an entry that is shared with a stonith client, * copy any pending ops that currently exist on the old entry to the new one. * Otherwise the pending ops will be reported as failures */ crm_info("Overwriting existing entry for %s from CIB", device->id); device->pending_ops = old->pending_ops; device->api_registered = TRUE; old->pending_ops = NULL; if (device->pending_ops) { mainloop_set_trigger(device->work); } } g_hash_table_replace(device_list, device->id, device); ndevices = g_hash_table_size(device_list); crm_notice("Added '%s' to device list (%d active device%s)", device->id, ndevices, pcmk__plural_s(ndevices)); } if (from_cib) { device->cib_registered = TRUE; } else { device->api_registered = TRUE; } return pcmk_ok; } void stonith_device_remove(const char *id, bool from_cib) { stonith_device_t *device = g_hash_table_lookup(device_list, id); guint ndevices = 0; if (!device) { ndevices = g_hash_table_size(device_list); crm_info("Device '%s' not found (%d active device%s)", id, ndevices, pcmk__plural_s(ndevices)); return; } if (from_cib) { device->cib_registered = FALSE; } else { device->verified = FALSE; device->api_registered = FALSE; } if (!device->cib_registered && !device->api_registered) { g_hash_table_remove(device_list, id); ndevices = g_hash_table_size(device_list); crm_info("Removed '%s' from device list (%d active device%s)", id, ndevices, pcmk__plural_s(ndevices)); } else { crm_trace("Not removing '%s' from device list (%d active) because " "still registered via:%s%s", id, g_hash_table_size(device_list), (device->cib_registered? " cib" : ""), (device->api_registered? " api" : "")); } } /*! * \internal * \brief Return the number of stonith levels registered for a node * * \param[in] tp Node's topology table entry * * \return Number of non-NULL levels in topology entry * \note This function is used only for log messages. */ static int count_active_levels(const stonith_topology_t *tp) { int lpc = 0; int count = 0; for (lpc = 0; lpc < ST__LEVEL_COUNT; lpc++) { if (tp->levels[lpc] != NULL) { count++; } } return count; } static void free_topology_entry(gpointer data) { stonith_topology_t *tp = data; int lpc = 0; for (lpc = 0; lpc < ST__LEVEL_COUNT; lpc++) { if (tp->levels[lpc] != NULL) { g_list_free_full(tp->levels[lpc], free); } } free(tp->target); free(tp->target_value); free(tp->target_pattern); free(tp->target_attribute); free(tp); } void free_topology_list(void) { if (topology != NULL) { g_hash_table_destroy(topology); topology = NULL; } } void init_topology_list(void) { if (topology == NULL) { topology = pcmk__strkey_table(NULL, free_topology_entry); } } char * stonith_level_key(const xmlNode *level, enum fenced_target_by mode) { if (mode == fenced_target_by_unknown) { mode = unpack_level_kind(level); } switch (mode) { case fenced_target_by_name: return crm_element_value_copy(level, PCMK_XA_TARGET); case fenced_target_by_pattern: return crm_element_value_copy(level, PCMK_XA_TARGET_PATTERN); case fenced_target_by_attribute: return crm_strdup_printf("%s=%s", crm_element_value(level, PCMK_XA_TARGET_ATTRIBUTE), crm_element_value(level, PCMK_XA_TARGET_VALUE)); default: return crm_strdup_printf("unknown-%s", pcmk__xe_id(level)); } } /*! * \internal * \brief Parse target identification from topology level XML * * \param[in] level Topology level XML to parse * * \return How to identify target of \p level */ static enum fenced_target_by unpack_level_kind(const xmlNode *level) { if (crm_element_value(level, PCMK_XA_TARGET) != NULL) { return fenced_target_by_name; } if (crm_element_value(level, PCMK_XA_TARGET_PATTERN) != NULL) { return fenced_target_by_pattern; } if (!stand_alone /* if standalone, there's no attribute manager */ && (crm_element_value(level, PCMK_XA_TARGET_ATTRIBUTE) != NULL) && (crm_element_value(level, PCMK_XA_TARGET_VALUE) != NULL)) { return fenced_target_by_attribute; } return fenced_target_by_unknown; } static stonith_key_value_t * parse_device_list(const char *devices) { int lpc = 0; int max = 0; int last = 0; stonith_key_value_t *output = NULL; if (devices == NULL) { return output; } max = strlen(devices); for (lpc = 0; lpc <= max; lpc++) { if (devices[lpc] == ',' || devices[lpc] == 0) { char *line = strndup(devices + last, lpc - last); output = stonith_key_value_add(output, NULL, line); free(line); last = lpc + 1; } } return output; } /*! * \internal * \brief Unpack essential information from topology request XML * * \param[in] xml Request XML to search * \param[out] mode If not NULL, where to store level kind * \param[out] target If not NULL, where to store representation of target * \param[out] id If not NULL, where to store level number * \param[out] desc If not NULL, where to store log-friendly level description * * \return Topology level XML from within \p xml, or NULL if not found * \note The caller is responsible for freeing \p *target and \p *desc if set. */ static xmlNode * unpack_level_request(xmlNode *xml, enum fenced_target_by *mode, char **target, int *id, char **desc) { enum fenced_target_by local_mode = fenced_target_by_unknown; char *local_target = NULL; int local_id = 0; /* The level element can be the top element or lower. If top level, don't * search by xpath, because it might give multiple hits if the XML is the * CIB. */ if ((xml != NULL) && !pcmk__xe_is(xml, PCMK_XE_FENCING_LEVEL)) { xml = get_xpath_object("//" PCMK_XE_FENCING_LEVEL, xml, LOG_WARNING); } if (xml == NULL) { if (desc != NULL) { *desc = crm_strdup_printf("missing"); } } else { local_mode = unpack_level_kind(xml); local_target = stonith_level_key(xml, local_mode); crm_element_value_int(xml, PCMK_XA_INDEX, &local_id); if (desc != NULL) { *desc = crm_strdup_printf("%s[%d]", local_target, local_id); } } if (mode != NULL) { *mode = local_mode; } if (id != NULL) { *id = local_id; } if (target != NULL) { *target = local_target; } else { free(local_target); } return xml; } /*! * \internal * \brief Register a fencing topology level for a target * * Given an XML request specifying the target name, level index, and device IDs * for the level, this will create an entry for the target in the global topology * table if one does not already exist, then append the specified device IDs to * the entry's device list for the specified level. * * \param[in] msg XML request for STONITH level registration * \param[out] desc If not NULL, set to string representation "TARGET[LEVEL]" * \param[out] result Where to set result of registration */ void fenced_register_level(xmlNode *msg, char **desc, pcmk__action_result_t *result) { int id = 0; xmlNode *level; enum fenced_target_by mode; char *target; stonith_topology_t *tp; stonith_key_value_t *dIter = NULL; stonith_key_value_t *devices = NULL; CRM_CHECK((msg != NULL) && (result != NULL), return); level = unpack_level_request(msg, &mode, &target, &id, desc); if (level == NULL) { fenced_set_protocol_error(result); return; } // Ensure an ID was given (even the client API adds an ID) if (pcmk__str_empty(pcmk__xe_id(level))) { crm_warn("Ignoring registration for topology level without ID"); free(target); crm_log_xml_trace(level, "Bad level"); pcmk__format_result(result, CRM_EX_INVALID_PARAM, PCMK_EXEC_INVALID, "Topology level is invalid without ID"); return; } // Ensure a valid target was specified if (mode == fenced_target_by_unknown) { crm_warn("Ignoring registration for topology level '%s' " "without valid target", pcmk__xe_id(level)); free(target); crm_log_xml_trace(level, "Bad level"); pcmk__format_result(result, CRM_EX_INVALID_PARAM, PCMK_EXEC_INVALID, "Invalid target for topology level '%s'", pcmk__xe_id(level)); return; } // Ensure level ID is in allowed range if ((id < ST__LEVEL_MIN) || (id > ST__LEVEL_MAX)) { crm_warn("Ignoring topology registration for %s with invalid level %d", target, id); free(target); crm_log_xml_trace(level, "Bad level"); pcmk__format_result(result, CRM_EX_INVALID_PARAM, PCMK_EXEC_INVALID, "Invalid level number '%s' for topology level '%s'", pcmk__s(crm_element_value(level, PCMK_XA_INDEX), ""), pcmk__xe_id(level)); return; } /* Find or create topology table entry */ tp = g_hash_table_lookup(topology, target); if (tp == NULL) { tp = pcmk__assert_alloc(1, sizeof(stonith_topology_t)); tp->kind = mode; tp->target = target; tp->target_value = crm_element_value_copy(level, PCMK_XA_TARGET_VALUE); tp->target_pattern = crm_element_value_copy(level, PCMK_XA_TARGET_PATTERN); tp->target_attribute = crm_element_value_copy(level, PCMK_XA_TARGET_ATTRIBUTE); g_hash_table_replace(topology, tp->target, tp); crm_trace("Added %s (%d) to the topology (%d active entries)", target, (int) mode, g_hash_table_size(topology)); } else { free(target); } if (tp->levels[id] != NULL) { crm_info("Adding to the existing %s[%d] topology entry", tp->target, id); } devices = parse_device_list(crm_element_value(level, PCMK_XA_DEVICES)); for (dIter = devices; dIter; dIter = dIter->next) { const char *device = dIter->value; crm_trace("Adding device '%s' for %s[%d]", device, tp->target, id); tp->levels[id] = g_list_append(tp->levels[id], pcmk__str_copy(device)); } stonith_key_value_freeall(devices, 1, 1); { int nlevels = count_active_levels(tp); crm_info("Target %s has %d active fencing level%s", tp->target, nlevels, pcmk__plural_s(nlevels)); } pcmk__set_result(result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); } /*! * \internal * \brief Unregister a fencing topology level for a target * * Given an XML request specifying the target name and level index (or 0 for all * levels), this will remove any corresponding entry for the target from the * global topology table. * * \param[in] msg XML request for STONITH level registration * \param[out] desc If not NULL, set to string representation "TARGET[LEVEL]" * \param[out] result Where to set result of unregistration */ void fenced_unregister_level(xmlNode *msg, char **desc, pcmk__action_result_t *result) { int id = -1; stonith_topology_t *tp; char *target; xmlNode *level = NULL; CRM_CHECK(result != NULL, return); level = unpack_level_request(msg, NULL, &target, &id, desc); if (level == NULL) { fenced_set_protocol_error(result); return; } // Ensure level ID is in allowed range if ((id < 0) || (id >= ST__LEVEL_COUNT)) { crm_warn("Ignoring topology unregistration for %s with invalid level %d", target, id); free(target); crm_log_xml_trace(level, "Bad level"); pcmk__format_result(result, CRM_EX_INVALID_PARAM, PCMK_EXEC_INVALID, "Invalid level number '%s' for topology level %s", pcmk__s(crm_element_value(level, PCMK_XA_INDEX), ""), // Client API doesn't add ID to unregistration XML pcmk__s(pcmk__xe_id(level), "")); return; } tp = g_hash_table_lookup(topology, target); if (tp == NULL) { guint nentries = g_hash_table_size(topology); crm_info("No fencing topology found for %s (%d active %s)", target, nentries, pcmk__plural_alt(nentries, "entry", "entries")); } else if (id == 0 && g_hash_table_remove(topology, target)) { guint nentries = g_hash_table_size(topology); crm_info("Removed all fencing topology entries related to %s " "(%d active %s remaining)", target, nentries, pcmk__plural_alt(nentries, "entry", "entries")); } else if (tp->levels[id] != NULL) { guint nlevels; g_list_free_full(tp->levels[id], free); tp->levels[id] = NULL; nlevels = count_active_levels(tp); crm_info("Removed level %d from fencing topology for %s " "(%d active level%s remaining)", id, target, nlevels, pcmk__plural_s(nlevels)); } free(target); pcmk__set_result(result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); } static char * list_to_string(GList *list, const char *delim, gboolean terminate_with_delim) { int max = g_list_length(list); size_t delim_len = delim?strlen(delim):0; size_t alloc_size = 1 + (max?((max-1+(terminate_with_delim?1:0))*delim_len):0); char *rv; GList *gIter; char *pos = NULL; const char *lead_delim = ""; for (gIter = list; gIter != NULL; gIter = gIter->next) { const char *value = (const char *) gIter->data; alloc_size += strlen(value); } rv = pcmk__assert_alloc(alloc_size, sizeof(char)); pos = rv; for (gIter = list; gIter != NULL; gIter = gIter->next) { const char *value = (const char *) gIter->data; pos = &pos[sprintf(pos, "%s%s", lead_delim, value)]; lead_delim = delim; } if (max && terminate_with_delim) { sprintf(pos, "%s", delim); } return rv; } /*! * \internal * \brief Execute a fence agent action directly (and asynchronously) * * Handle a STONITH_OP_EXEC API message by scheduling a requested agent action * directly on a specified device. Only list, monitor, and status actions are * expected to use this call, though it should work with any agent command. * * \param[in] msg Request XML specifying action * \param[out] result Where to store result of action * * \note If the action is monitor, the device must be registered via the API * (CIB registration is not sufficient), because monitor should not be * possible unless the device is "started" (API registered). */ static void execute_agent_action(xmlNode *msg, pcmk__action_result_t *result) { xmlNode *dev = get_xpath_object("//" PCMK__XE_ST_DEVICE_ID, msg, LOG_ERR); xmlNode *op = get_xpath_object("//@" PCMK__XE_ST_DEVICE_ACTION, msg, LOG_ERR); const char *id = crm_element_value(dev, PCMK__XA_ST_DEVICE_ID); const char *action = crm_element_value(op, PCMK__XA_ST_DEVICE_ACTION); async_command_t *cmd = NULL; stonith_device_t *device = NULL; if ((id == NULL) || (action == NULL)) { crm_info("Malformed API action request: device %s, action %s", (id? id : "not specified"), (action? action : "not specified")); fenced_set_protocol_error(result); return; } if (pcmk__str_eq(id, STONITH_WATCHDOG_ID, pcmk__str_none)) { // Watchdog agent actions are implemented internally if (stonith_watchdog_timeout_ms <= 0) { pcmk__set_result(result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, "Watchdog fence device not configured"); return; } else if (pcmk__str_eq(action, PCMK_ACTION_LIST, pcmk__str_none)) { pcmk__set_result(result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); pcmk__set_result_output(result, list_to_string(stonith_watchdog_targets, "\n", TRUE), NULL); return; } else if (pcmk__str_eq(action, PCMK_ACTION_MONITOR, pcmk__str_none)) { pcmk__set_result(result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); return; } } device = g_hash_table_lookup(device_list, id); if (device == NULL) { crm_info("Ignoring API '%s' action request because device %s not found", action, id); pcmk__format_result(result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, "'%s' not found", id); return; } else if (!device->api_registered && (strcmp(action, PCMK_ACTION_MONITOR) == 0)) { // Monitors may run only on "started" (API-registered) devices crm_info("Ignoring API '%s' action request because device %s not active", action, id); pcmk__format_result(result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, "'%s' not active", id); return; } cmd = create_async_command(msg); if (cmd == NULL) { crm_log_xml_warn(msg, "invalid"); fenced_set_protocol_error(result); return; } schedule_stonith_command(cmd, device); pcmk__set_result(result, CRM_EX_OK, PCMK_EXEC_PENDING, NULL); } static void search_devices_record_result(struct device_search_s *search, const char *device, gboolean can_fence) { search->replies_received++; if (can_fence && device) { if (search->support_action_only != st_device_supports_none) { stonith_device_t *dev = g_hash_table_lookup(device_list, device); if (dev && !pcmk_is_set(dev->flags, search->support_action_only)) { return; } } search->capable = g_list_append(search->capable, pcmk__str_copy(device)); } if (search->replies_needed == search->replies_received) { guint ndevices = g_list_length(search->capable); crm_debug("Search found %d device%s that can perform '%s' targeting %s", ndevices, pcmk__plural_s(ndevices), (search->action? search->action : "unknown action"), (search->host? search->host : "any node")); search->callback(search->capable, search->user_data); free(search->host); free(search->action); free(search); } } /*! * \internal * \brief Check whether the local host is allowed to execute a fencing action * * \param[in] device Fence device to check * \param[in] action Fence action to check * \param[in] target Hostname of fence target * \param[in] allow_self Whether self-fencing is allowed for this operation * * \return TRUE if local host is allowed to execute action, FALSE otherwise */ static gboolean localhost_is_eligible(const stonith_device_t *device, const char *action, const char *target, gboolean allow_self) { gboolean localhost_is_target = pcmk__str_eq(target, stonith_our_uname, pcmk__str_casei); if ((device != NULL) && (action != NULL) && (device->on_target_actions != NULL) && (strstr((const char*) device->on_target_actions->str, action) != NULL)) { if (!localhost_is_target) { crm_trace("Operation '%s' using %s can only be executed for local " "host, not %s", action, device->id, target); return FALSE; } } else if (localhost_is_target && !allow_self) { crm_trace("'%s' operation does not support self-fencing", action); return FALSE; } return TRUE; } /*! * \internal * \brief Check if local node is allowed to execute (possibly remapped) action * * \param[in] device Fence device to check * \param[in] action Fence action to check * \param[in] target Node name of fence target * \param[in] allow_self Whether self-fencing is allowed for this operation * * \return true if local node is allowed to execute \p action or any actions it * might be remapped to, otherwise false */ static bool localhost_is_eligible_with_remap(const stonith_device_t *device, const char *action, const char *target, gboolean allow_self) { // Check exact action if (localhost_is_eligible(device, action, target, allow_self)) { return true; } // Check potential remaps if (pcmk__str_eq(action, PCMK_ACTION_REBOOT, pcmk__str_none)) { /* "reboot" might get remapped to "off" then "on", so even if reboot is * disallowed, return true if either of those is allowed. We'll report * the disallowed actions with the results. We never allow self-fencing * for remapped "on" actions because the target is off at that point. */ if (localhost_is_eligible(device, PCMK_ACTION_OFF, target, allow_self) || localhost_is_eligible(device, PCMK_ACTION_ON, target, FALSE)) { return true; } } return false; } static void can_fence_host_with_device(stonith_device_t *dev, struct device_search_s *search) { gboolean can = FALSE; const char *check_type = "Internal bug"; const char *target = NULL; const char *alias = NULL; const char *dev_id = "Unspecified device"; const char *action = (search == NULL)? NULL : search->action; CRM_CHECK((dev != NULL) && (action != NULL), goto search_report_results); if (dev->id != NULL) { dev_id = dev->id; } target = search->host; if (target == NULL) { can = TRUE; check_type = "No target"; goto search_report_results; } /* Answer immediately if the device does not support the action * or the local node is not allowed to perform it */ if (pcmk__str_eq(action, PCMK_ACTION_ON, pcmk__str_none) && !pcmk_is_set(dev->flags, st_device_supports_on)) { check_type = "Agent does not support 'on'"; goto search_report_results; } else if (!localhost_is_eligible_with_remap(dev, action, target, search->allow_self)) { check_type = "This node is not allowed to execute action"; goto search_report_results; } // Check eligibility as specified by pcmk_host_check check_type = target_list_type(dev); alias = g_hash_table_lookup(dev->aliases, target); if (pcmk__str_eq(check_type, PCMK_VALUE_NONE, pcmk__str_casei)) { can = TRUE; } else if (pcmk__str_eq(check_type, PCMK_VALUE_STATIC_LIST, pcmk__str_casei)) { if (pcmk__str_in_list(target, dev->targets, pcmk__str_casei)) { can = TRUE; } else if (g_hash_table_lookup(dev->params, PCMK_STONITH_HOST_MAP) && g_hash_table_lookup(dev->aliases, target)) { can = TRUE; } } else if (pcmk__str_eq(check_type, PCMK_VALUE_DYNAMIC_LIST, pcmk__str_casei)) { time_t now = time(NULL); if (dev->targets == NULL || dev->targets_age + 60 < now) { int device_timeout = get_action_timeout(dev, PCMK_ACTION_LIST, search->per_device_timeout); if (device_timeout > search->per_device_timeout) { crm_notice("Since the pcmk_list_timeout (%ds) parameter of %s " "is larger than " PCMK_OPT_STONITH_TIMEOUT " (%ds), timeout may occur", device_timeout, dev_id, search->per_device_timeout); } crm_trace("Running '%s' to check whether %s is eligible to fence %s (%s)", check_type, dev_id, target, action); schedule_internal_command(__func__, dev, PCMK_ACTION_LIST, NULL, search->per_device_timeout, search, dynamic_list_search_cb); /* we'll respond to this search request async in the cb */ return; } if (pcmk__str_in_list(((alias == NULL)? target : alias), dev->targets, pcmk__str_casei)) { can = TRUE; } } else if (pcmk__str_eq(check_type, PCMK_VALUE_STATUS, pcmk__str_casei)) { int device_timeout = get_action_timeout(dev, check_type, search->per_device_timeout); if (device_timeout > search->per_device_timeout) { crm_notice("Since the pcmk_status_timeout (%ds) parameter of %s is " "larger than " PCMK_OPT_STONITH_TIMEOUT " (%ds), " "timeout may occur", device_timeout, dev_id, search->per_device_timeout); } crm_trace("Running '%s' to check whether %s is eligible to fence %s (%s)", check_type, dev_id, target, action); schedule_internal_command(__func__, dev, PCMK_ACTION_STATUS, target, search->per_device_timeout, search, status_search_cb); /* we'll respond to this search request async in the cb */ return; } else { crm_err("Invalid value for " PCMK_STONITH_HOST_CHECK ": %s", check_type); check_type = "Invalid " PCMK_STONITH_HOST_CHECK; } search_report_results: crm_info("%s is%s eligible to fence (%s) %s%s%s%s: %s", dev_id, (can? "" : " not"), pcmk__s(action, "unspecified action"), pcmk__s(target, "unspecified target"), (alias == NULL)? "" : " (as '", pcmk__s(alias, ""), (alias == NULL)? "" : "')", check_type); search_devices_record_result(search, ((dev == NULL)? NULL : dev_id), can); } static void search_devices(gpointer key, gpointer value, gpointer user_data) { stonith_device_t *dev = value; struct device_search_s *search = user_data; can_fence_host_with_device(dev, search); } #define DEFAULT_QUERY_TIMEOUT 20 static void get_capable_devices(const char *host, const char *action, int timeout, bool allow_self, void *user_data, void (*callback) (GList * devices, void *user_data), uint32_t support_action_only) { struct device_search_s *search; guint ndevices = g_hash_table_size(device_list); if (ndevices == 0) { callback(NULL, user_data); return; } search = pcmk__assert_alloc(1, sizeof(struct device_search_s)); search->host = pcmk__str_copy(host); search->action = pcmk__str_copy(action); search->per_device_timeout = timeout; search->allow_self = allow_self; search->callback = callback; search->user_data = user_data; search->support_action_only = support_action_only; /* We are guaranteed this many replies, even if a device is * unregistered while the search is in progress. */ search->replies_needed = ndevices; crm_debug("Searching %d device%s to see which can execute '%s' targeting %s", ndevices, pcmk__plural_s(ndevices), (search->action? search->action : "unknown action"), (search->host? search->host : "any node")); g_hash_table_foreach(device_list, search_devices, search); } struct st_query_data { xmlNode *reply; char *remote_peer; char *client_id; char *target; char *action; int call_options; }; /*! * \internal * \brief Add action-specific attributes to query reply XML * * \param[in,out] xml XML to add attributes to * \param[in] action Fence action * \param[in] device Fence device * \param[in] target Fence target */ static void add_action_specific_attributes(xmlNode *xml, const char *action, const stonith_device_t *device, const char *target) { int action_specific_timeout; int delay_max; int delay_base; CRM_CHECK(xml && action && device, return); // PCMK__XA_ST_REQUIRED is currently used only for unfencing if (is_action_required(action, device)) { crm_trace("Action '%s' is required using %s", action, device->id); crm_xml_add_int(xml, PCMK__XA_ST_REQUIRED, 1); } // pcmk__timeout if configured action_specific_timeout = get_action_timeout(device, action, 0); if (action_specific_timeout) { crm_trace("Action '%s' has timeout %ds using %s", action, action_specific_timeout, device->id); crm_xml_add_int(xml, PCMK__XA_ST_ACTION_TIMEOUT, action_specific_timeout); } delay_max = get_action_delay_max(device, action); if (delay_max > 0) { crm_trace("Action '%s' has maximum random delay %ds using %s", action, delay_max, device->id); crm_xml_add_int(xml, PCMK__XA_ST_DELAY_MAX, delay_max); } delay_base = get_action_delay_base(device, action, target); if (delay_base > 0) { crm_xml_add_int(xml, PCMK__XA_ST_DELAY_BASE, delay_base); } if ((delay_max > 0) && (delay_base == 0)) { crm_trace("Action '%s' has maximum random delay %ds using %s", action, delay_max, device->id); } else if ((delay_max == 0) && (delay_base > 0)) { crm_trace("Action '%s' has a static delay of %ds using %s", action, delay_base, device->id); } else if ((delay_max > 0) && (delay_base > 0)) { crm_trace("Action '%s' has a minimum delay of %ds and a randomly chosen " "maximum delay of %ds using %s", action, delay_base, delay_max, device->id); } } /*! * \internal * \brief Add "disallowed" attribute to query reply XML if appropriate * * \param[in,out] xml XML to add attribute to * \param[in] action Fence action * \param[in] device Fence device * \param[in] target Fence target * \param[in] allow_self Whether self-fencing is allowed */ static void add_disallowed(xmlNode *xml, const char *action, const stonith_device_t *device, const char *target, gboolean allow_self) { if (!localhost_is_eligible(device, action, target, allow_self)) { crm_trace("Action '%s' using %s is disallowed for local host", action, device->id); pcmk__xe_set_bool_attr(xml, PCMK__XA_ST_ACTION_DISALLOWED, true); } } /*! * \internal * \brief Add child element with action-specific values to query reply XML * * \param[in,out] xml XML to add attribute to * \param[in] action Fence action * \param[in] device Fence device * \param[in] target Fence target * \param[in] allow_self Whether self-fencing is allowed */ static void add_action_reply(xmlNode *xml, const char *action, const stonith_device_t *device, const char *target, gboolean allow_self) { xmlNode *child = pcmk__xe_create(xml, PCMK__XE_ST_DEVICE_ACTION); crm_xml_add(child, PCMK_XA_ID, action); add_action_specific_attributes(child, action, device, target); add_disallowed(child, action, device, target, allow_self); } /*! * \internal * \brief Send a reply to a CPG peer or IPC client * * \param[in] reply XML reply to send * \param[in] call_options Send synchronously if st_opt_sync_call is set * \param[in] remote_peer If not NULL, name of peer node to send CPG reply * \param[in,out] client If not NULL, client to send IPC reply */ static void stonith_send_reply(const xmlNode *reply, int call_options, const char *remote_peer, pcmk__client_t *client) { CRM_CHECK((reply != NULL) && ((remote_peer != NULL) || (client != NULL)), return); if (remote_peer == NULL) { do_local_reply(reply, client, call_options); } else { const crm_node_t *node = pcmk__get_node(0, remote_peer, NULL, pcmk__node_search_cluster_member); pcmk__cluster_send_message(node, crm_msg_stonith_ng, reply); } } static void stonith_query_capable_device_cb(GList * devices, void *user_data) { struct st_query_data *query = user_data; int available_devices = 0; xmlNode *wrapper = NULL; xmlNode *list = NULL; GList *lpc = NULL; pcmk__client_t *client = NULL; if (query->client_id != NULL) { client = pcmk__find_client_by_id(query->client_id); if ((client == NULL) && (query->remote_peer == NULL)) { crm_trace("Skipping reply to %s: no longer a client", query->client_id); goto done; } } // Pack the results into XML wrapper = pcmk__xe_create(query->reply, PCMK__XE_ST_CALLDATA); list = pcmk__xe_create(wrapper, __func__); crm_xml_add(list, PCMK__XA_ST_TARGET, query->target); for (lpc = devices; lpc != NULL; lpc = lpc->next) { stonith_device_t *device = g_hash_table_lookup(device_list, lpc->data); const char *action = query->action; xmlNode *dev = NULL; if (!device) { /* It is possible the device got unregistered while * determining who can fence the target */ continue; } available_devices++; dev = pcmk__xe_create(list, PCMK__XE_ST_DEVICE_ID); crm_xml_add(dev, PCMK_XA_ID, device->id); crm_xml_add(dev, PCMK__XA_NAMESPACE, device->namespace); crm_xml_add(dev, PCMK_XA_AGENT, device->agent); // Has had successful monitor, list, or status on this node crm_xml_add_int(dev, PCMK__XA_ST_MONITOR_VERIFIED, device->verified); crm_xml_add_int(dev, PCMK__XA_ST_DEVICE_SUPPORT_FLAGS, device->flags); /* If the originating fencer wants to reboot the node, and we have a * capable device that doesn't support "reboot", remap to "off" instead. */ if (!pcmk_is_set(device->flags, st_device_supports_reboot) && pcmk__str_eq(query->action, PCMK_ACTION_REBOOT, pcmk__str_none)) { crm_trace("%s doesn't support reboot, using values for off instead", device->id); action = PCMK_ACTION_OFF; } /* Add action-specific values if available */ add_action_specific_attributes(dev, action, device, query->target); if (pcmk__str_eq(query->action, PCMK_ACTION_REBOOT, pcmk__str_none)) { /* A "reboot" *might* get remapped to "off" then "on", so after * sending the "reboot"-specific values in the main element, we add * sub-elements for "off" and "on" values. * * We short-circuited earlier if "reboot", "off" and "on" are all * disallowed for the local host. However if only one or two are * disallowed, we send back the results and mark which ones are * disallowed. If "reboot" is disallowed, this might cause problems * with older fencer versions, which won't check for it. Older * versions will ignore "off" and "on", so they are not a problem. */ add_disallowed(dev, action, device, query->target, pcmk_is_set(query->call_options, st_opt_allow_self_fencing)); add_action_reply(dev, PCMK_ACTION_OFF, device, query->target, pcmk_is_set(query->call_options, st_opt_allow_self_fencing)); add_action_reply(dev, PCMK_ACTION_ON, device, query->target, FALSE); } /* A query without a target wants device parameters */ if (query->target == NULL) { xmlNode *attrs = pcmk__xe_create(dev, PCMK__XE_ATTRIBUTES); g_hash_table_foreach(device->params, hash2field, attrs); } } crm_xml_add_int(list, PCMK__XA_ST_AVAILABLE_DEVICES, available_devices); if (query->target) { crm_debug("Found %d matching device%s for target '%s'", available_devices, pcmk__plural_s(available_devices), query->target); } else { crm_debug("%d device%s installed", available_devices, pcmk__plural_s(available_devices)); } crm_log_xml_trace(list, "query-result"); stonith_send_reply(query->reply, query->call_options, query->remote_peer, client); done: free_xml(query->reply); free(query->remote_peer); free(query->client_id); free(query->target); free(query->action); free(query); g_list_free_full(devices, free); } /*! * \internal * \brief Log the result of an asynchronous command * * \param[in] cmd Command the result is for * \param[in] result Result of command * \param[in] pid Process ID of command, if available * \param[in] next Alternate device that will be tried if command failed * \param[in] op_merged Whether this command was merged with an earlier one */ static void log_async_result(const async_command_t *cmd, const pcmk__action_result_t *result, int pid, const char *next, bool op_merged) { int log_level = LOG_ERR; int output_log_level = LOG_NEVER; guint devices_remaining = g_list_length(cmd->next_device_iter); GString *msg = g_string_sized_new(80); // Reasonable starting size // Choose log levels appropriately if we have a result if (pcmk__result_ok(result)) { log_level = (cmd->target == NULL)? LOG_DEBUG : LOG_NOTICE; if ((result->action_stdout != NULL) && !pcmk__str_eq(cmd->action, PCMK_ACTION_METADATA, pcmk__str_none)) { output_log_level = LOG_DEBUG; } next = NULL; } else { log_level = (cmd->target == NULL)? LOG_NOTICE : LOG_ERR; if ((result->action_stdout != NULL) && !pcmk__str_eq(cmd->action, PCMK_ACTION_METADATA, pcmk__str_none)) { output_log_level = LOG_WARNING; } } // Build the log message piece by piece pcmk__g_strcat(msg, "Operation '", cmd->action, "' ", NULL); if (pid != 0) { g_string_append_printf(msg, "[%d] ", pid); } if (cmd->target != NULL) { pcmk__g_strcat(msg, "targeting ", cmd->target, " ", NULL); } if (cmd->device != NULL) { pcmk__g_strcat(msg, "using ", cmd->device, " ", NULL); } // Add exit status or execution status as appropriate if (result->execution_status == PCMK_EXEC_DONE) { g_string_append_printf(msg, "returned %d", result->exit_status); } else { pcmk__g_strcat(msg, "could not be executed: ", pcmk_exec_status_str(result->execution_status), NULL); } // Add exit reason and next device if appropriate if (result->exit_reason != NULL) { pcmk__g_strcat(msg, " (", result->exit_reason, ")", NULL); } if (next != NULL) { pcmk__g_strcat(msg, ", retrying with ", next, NULL); } if (devices_remaining > 0) { g_string_append_printf(msg, " (%u device%s remaining)", (unsigned int) devices_remaining, pcmk__plural_s(devices_remaining)); } g_string_append_printf(msg, " " CRM_XS " %scall %d from %s", (op_merged? "merged " : ""), cmd->id, cmd->client_name); // Log the result do_crm_log(log_level, "%s", msg->str); g_string_free(msg, TRUE); // Log the output (which may have multiple lines), if appropriate if (output_log_level != LOG_NEVER) { char *prefix = crm_strdup_printf("%s[%d]", cmd->device, pid); crm_log_output(output_log_level, prefix, result->action_stdout); free(prefix); } } /*! * \internal * \brief Reply to requester after asynchronous command completion * * \param[in] cmd Command that completed * \param[in] result Result of command * \param[in] pid Process ID of command, if available * \param[in] merged If true, command was merged with another, not executed */ static void send_async_reply(const async_command_t *cmd, const pcmk__action_result_t *result, int pid, bool merged) { xmlNode *reply = NULL; pcmk__client_t *client = NULL; CRM_CHECK((cmd != NULL) && (result != NULL), return); log_async_result(cmd, result, pid, NULL, merged); if (cmd->client != NULL) { client = pcmk__find_client_by_id(cmd->client); if ((client == NULL) && (cmd->origin == NULL)) { crm_trace("Skipping reply to %s: no longer a client", cmd->client); return; } } reply = construct_async_reply(cmd, result); if (merged) { pcmk__xe_set_bool_attr(reply, PCMK__XA_ST_OP_MERGED, true); } if (!stand_alone && pcmk__is_fencing_action(cmd->action) && pcmk__str_eq(cmd->origin, cmd->target, pcmk__str_casei)) { /* The target was also the originator, so broadcast the result on its * behalf (since it will be unable to). */ crm_trace("Broadcast '%s' result for %s (target was also originator)", cmd->action, cmd->target); crm_xml_add(reply, PCMK__XA_SUBT, PCMK__VALUE_BROADCAST); crm_xml_add(reply, PCMK__XA_ST_OP, STONITH_OP_NOTIFY); pcmk__cluster_send_message(NULL, crm_msg_stonith_ng, reply); } else { // Reply only to the originator stonith_send_reply(reply, cmd->options, cmd->origin, client); } crm_log_xml_trace(reply, "Reply"); free_xml(reply); if (stand_alone) { /* Do notification with a clean data object */ xmlNode *notify_data = pcmk__xe_create(NULL, PCMK__XE_ST_NOTIFY_FENCE); stonith__xe_set_result(notify_data, result); crm_xml_add(notify_data, PCMK__XA_ST_TARGET, cmd->target); crm_xml_add(notify_data, PCMK__XA_ST_OP, cmd->op); crm_xml_add(notify_data, PCMK__XA_ST_DELEGATE, "localhost"); crm_xml_add(notify_data, PCMK__XA_ST_DEVICE_ID, cmd->device); crm_xml_add(notify_data, PCMK__XA_ST_REMOTE_OP, cmd->remote_op_id); crm_xml_add(notify_data, PCMK__XA_ST_ORIGIN, cmd->client); fenced_send_notification(PCMK__VALUE_ST_NOTIFY_FENCE, result, notify_data); fenced_send_notification(PCMK__VALUE_ST_NOTIFY_HISTORY, NULL, NULL); } } static void cancel_stonith_command(async_command_t * cmd) { stonith_device_t *device = cmd_device(cmd); if (device) { crm_trace("Cancel scheduled '%s' action using %s", cmd->action, device->id); device->pending_ops = g_list_remove(device->pending_ops, cmd); } } /*! * \internal * \brief Cancel and reply to any duplicates of a just-completed operation * * Check whether any fencing operations are scheduled to do the same thing as * one that just succeeded. If so, rather than performing the same operation * twice, return the result of this operation for all matching pending commands. * * \param[in,out] cmd Fencing operation that just succeeded * \param[in] result Result of \p cmd * \param[in] pid If nonzero, process ID of agent invocation (for logs) * * \note Duplicate merging will do the right thing for either type of remapped * reboot. If the executing fencer remapped an unsupported reboot to off, * then cmd->action will be "reboot" and will be merged with any other * reboot requests. If the originating fencer remapped a topology reboot * to off then on, we will get here once with cmd->action "off" and once * with "on", and they will be merged separately with similar requests. */ static void reply_to_duplicates(async_command_t *cmd, const pcmk__action_result_t *result, int pid) { GList *next = NULL; for (GList *iter = cmd_list; iter != NULL; iter = next) { async_command_t *cmd_other = iter->data; next = iter->next; // We might delete this entry, so grab next now if (cmd == cmd_other) { continue; } /* A pending operation matches if: * 1. The client connections are different. * 2. The target is the same. * 3. The fencing action is the same. * 4. The device scheduled to execute the action is the same. */ if (pcmk__str_eq(cmd->client, cmd_other->client, pcmk__str_casei) || !pcmk__str_eq(cmd->target, cmd_other->target, pcmk__str_casei) || !pcmk__str_eq(cmd->action, cmd_other->action, pcmk__str_none) || !pcmk__str_eq(cmd->device, cmd_other->device, pcmk__str_casei)) { continue; } crm_notice("Merging fencing action '%s'%s%s originating from " "client %s with identical fencing request from client %s", cmd_other->action, (cmd_other->target == NULL)? "" : " targeting ", pcmk__s(cmd_other->target, ""), cmd_other->client_name, cmd->client_name); // Stop tracking the duplicate, send its result, and cancel it cmd_list = g_list_remove_link(cmd_list, iter); send_async_reply(cmd_other, result, pid, true); cancel_stonith_command(cmd_other); free_async_command(cmd_other); g_list_free_1(iter); } } /*! * \internal * \brief Return the next required device (if any) for an operation * * \param[in,out] cmd Fencing operation that just succeeded * * \return Next device required for action if any, otherwise NULL */ static stonith_device_t * next_required_device(async_command_t *cmd) { for (GList *iter = cmd->next_device_iter; iter != NULL; iter = iter->next) { stonith_device_t *next_device = g_hash_table_lookup(device_list, iter->data); if (is_action_required(cmd->action, next_device)) { /* This is only called for successful actions, so it's OK to skip * non-required devices. */ cmd->next_device_iter = iter->next; return next_device; } } return NULL; } static void st_child_done(int pid, const pcmk__action_result_t *result, void *user_data) { async_command_t *cmd = user_data; stonith_device_t *device = NULL; stonith_device_t *next_device = NULL; CRM_CHECK(cmd != NULL, return); device = cmd_device(cmd); cmd->active_on = NULL; /* The device is ready to do something else now */ if (device) { if (!device->verified && pcmk__result_ok(result) && pcmk__strcase_any_of(cmd->action, PCMK_ACTION_LIST, PCMK_ACTION_MONITOR, PCMK_ACTION_STATUS, NULL)) { device->verified = TRUE; } mainloop_set_trigger(device->work); } if (pcmk__result_ok(result)) { next_device = next_required_device(cmd); } else if ((cmd->next_device_iter != NULL) && !is_action_required(cmd->action, device)) { /* if this device didn't work out, see if there are any others we can try. * if the failed device was 'required', we can't pick another device. */ next_device = g_hash_table_lookup(device_list, cmd->next_device_iter->data); cmd->next_device_iter = cmd->next_device_iter->next; } if (next_device == NULL) { send_async_reply(cmd, result, pid, false); if (pcmk__result_ok(result)) { reply_to_duplicates(cmd, result, pid); } free_async_command(cmd); } else { // This operation requires more fencing log_async_result(cmd, result, pid, next_device->id, false); schedule_stonith_command(cmd, next_device); } } static gint sort_device_priority(gconstpointer a, gconstpointer b) { const stonith_device_t *dev_a = a; const stonith_device_t *dev_b = b; if (dev_a->priority > dev_b->priority) { return -1; } else if (dev_a->priority < dev_b->priority) { return 1; } return 0; } static void stonith_fence_get_devices_cb(GList * devices, void *user_data) { async_command_t *cmd = user_data; stonith_device_t *device = NULL; guint ndevices = g_list_length(devices); crm_info("Found %d matching device%s for target '%s'", ndevices, pcmk__plural_s(ndevices), cmd->target); if (devices != NULL) { /* Order based on priority */ devices = g_list_sort(devices, sort_device_priority); device = g_hash_table_lookup(device_list, devices->data); } if (device == NULL) { // No device found pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; pcmk__format_result(&result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, "No device configured for target '%s'", cmd->target); send_async_reply(cmd, &result, 0, false); pcmk__reset_result(&result); free_async_command(cmd); g_list_free_full(devices, free); } else { // Device found, schedule it for fencing cmd->device_list = devices; cmd->next_device_iter = devices->next; schedule_stonith_command(cmd, device); } } /*! * \internal * \brief Execute a fence action via the local node * * \param[in] msg Fencing request * \param[out] result Where to store result of fence action */ static void fence_locally(xmlNode *msg, pcmk__action_result_t *result) { const char *device_id = NULL; stonith_device_t *device = NULL; async_command_t *cmd = NULL; xmlNode *dev = NULL; CRM_CHECK((msg != NULL) && (result != NULL), return); dev = get_xpath_object("//@" PCMK__XA_ST_TARGET, msg, LOG_ERR); cmd = create_async_command(msg); if (cmd == NULL) { crm_log_xml_warn(msg, "invalid"); fenced_set_protocol_error(result); return; } device_id = crm_element_value(dev, PCMK__XA_ST_DEVICE_ID); if (device_id != NULL) { device = g_hash_table_lookup(device_list, device_id); if (device == NULL) { crm_err("Requested device '%s' is not available", device_id); pcmk__format_result(result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, "Requested device '%s' not found", device_id); return; } schedule_stonith_command(cmd, device); } else { const char *host = crm_element_value(dev, PCMK__XA_ST_TARGET); if (pcmk_is_set(cmd->options, st_opt_cs_nodeid)) { int nodeid = 0; crm_node_t *node = NULL; pcmk__scan_min_int(host, &nodeid, 0); node = pcmk__search_node_caches(nodeid, NULL, NULL, pcmk__node_search_any |pcmk__node_search_cluster_cib); if (node != NULL) { host = node->uname; } } /* If we get to here, then self-fencing is implicitly allowed */ get_capable_devices(host, cmd->action, cmd->default_timeout, TRUE, cmd, stonith_fence_get_devices_cb, fenced_support_flag(cmd->action)); } pcmk__set_result(result, CRM_EX_OK, PCMK_EXEC_PENDING, NULL); } /*! * \internal * \brief Build an XML reply for a fencing operation * * \param[in] request Request that reply is for * \param[in] data If not NULL, add to reply as call data * \param[in] result Full result of fencing operation * * \return Newly created XML reply * \note The caller is responsible for freeing the result. * \note This has some overlap with construct_async_reply(), but that copies * values from an async_command_t, whereas this one copies them from the * request. */ xmlNode * fenced_construct_reply(const xmlNode *request, xmlNode *data, const pcmk__action_result_t *result) { xmlNode *reply = NULL; reply = pcmk__xe_create(NULL, PCMK__XE_ST_REPLY); crm_xml_add(reply, PCMK__XA_ST_ORIGIN, __func__); crm_xml_add(reply, PCMK__XA_T, PCMK__VALUE_STONITH_NG); stonith__xe_set_result(reply, result); if (request == NULL) { /* Most likely, this is the result of a stonith operation that was * initiated before we came up. Unfortunately that means we lack enough * information to provide clients with a full result. * * @TODO Maybe synchronize this information at start-up? */ crm_warn("Missing request information for client notifications for " "operation with result '%s' (initiated before we came up?)", pcmk_exec_status_str(result->execution_status)); } else { const char *name = NULL; const char *value = NULL; // Attributes to copy from request to reply const char *names[] = { PCMK__XA_ST_OP, PCMK__XA_ST_CALLID, PCMK__XA_ST_CLIENTID, PCMK__XA_ST_CLIENTNAME, PCMK__XA_ST_REMOTE_OP, PCMK__XA_ST_CALLOPT, }; for (int lpc = 0; lpc < PCMK__NELEM(names); lpc++) { name = names[lpc]; value = crm_element_value(request, name); crm_xml_add(reply, name, value); } if (data != NULL) { xmlNode *wrapper = pcmk__xe_create(reply, PCMK__XE_ST_CALLDATA); pcmk__xml_copy(wrapper, data); } } return reply; } /*! * \internal * \brief Build an XML reply to an asynchronous fencing command * * \param[in] cmd Fencing command that reply is for * \param[in] result Command result */ static xmlNode * construct_async_reply(const async_command_t *cmd, const pcmk__action_result_t *result) { xmlNode *reply = pcmk__xe_create(NULL, PCMK__XE_ST_REPLY); crm_xml_add(reply, PCMK__XA_ST_ORIGIN, __func__); crm_xml_add(reply, PCMK__XA_T, PCMK__VALUE_STONITH_NG); crm_xml_add(reply, PCMK__XA_ST_OP, cmd->op); crm_xml_add(reply, PCMK__XA_ST_DEVICE_ID, cmd->device); crm_xml_add(reply, PCMK__XA_ST_REMOTE_OP, cmd->remote_op_id); crm_xml_add(reply, PCMK__XA_ST_CLIENTID, cmd->client); crm_xml_add(reply, PCMK__XA_ST_CLIENTNAME, cmd->client_name); crm_xml_add(reply, PCMK__XA_ST_TARGET, cmd->target); crm_xml_add(reply, PCMK__XA_ST_DEVICE_ACTION, cmd->op); crm_xml_add(reply, PCMK__XA_ST_ORIGIN, cmd->origin); crm_xml_add_int(reply, PCMK__XA_ST_CALLID, cmd->id); crm_xml_add_int(reply, PCMK__XA_ST_CALLOPT, cmd->options); stonith__xe_set_result(reply, result); return reply; } bool fencing_peer_active(crm_node_t *peer) { if (peer == NULL) { return FALSE; } else if (peer->uname == NULL) { return FALSE; } else if (pcmk_is_set(peer->processes, crm_get_cluster_proc())) { return TRUE; } return FALSE; } void set_fencing_completed(remote_fencing_op_t *op) { struct timespec tv; qb_util_timespec_from_epoch_get(&tv); op->completed = tv.tv_sec; op->completed_nsec = tv.tv_nsec; } /*! * \internal * \brief Look for alternate node needed if local node shouldn't fence target * * \param[in] target Node that must be fenced * * \return Name of an alternate node that should fence \p target if any, * or NULL otherwise */ static const char * check_alternate_host(const char *target) { if (pcmk__str_eq(target, stonith_our_uname, pcmk__str_casei)) { GHashTableIter gIter; crm_node_t *entry = NULL; g_hash_table_iter_init(&gIter, crm_peer_cache); while (g_hash_table_iter_next(&gIter, NULL, (void **)&entry)) { if (fencing_peer_active(entry) && !pcmk__str_eq(entry->uname, target, pcmk__str_casei)) { crm_notice("Forwarding self-fencing request to %s", entry->uname); return entry->uname; } } crm_warn("Will handle own fencing because no peer can"); } return NULL; } static void remove_relay_op(xmlNode * request) { xmlNode *dev = get_xpath_object("//@" PCMK__XE_ST_DEVICE_ACTION, request, LOG_TRACE); const char *relay_op_id = NULL; const char *op_id = NULL; const char *client_name = NULL; const char *target = NULL; remote_fencing_op_t *relay_op = NULL; if (dev) { target = crm_element_value(dev, PCMK__XA_ST_TARGET); } relay_op_id = crm_element_value(request, PCMK__XA_ST_REMOTE_OP_RELAY); op_id = crm_element_value(request, PCMK__XA_ST_REMOTE_OP); client_name = crm_element_value(request, PCMK__XA_ST_CLIENTNAME); /* Delete RELAY operation. */ if (relay_op_id && target && pcmk__str_eq(target, stonith_our_uname, pcmk__str_casei)) { relay_op = g_hash_table_lookup(stonith_remote_op_list, relay_op_id); if (relay_op) { GHashTableIter iter; remote_fencing_op_t *list_op = NULL; g_hash_table_iter_init(&iter, stonith_remote_op_list); /* If the operation to be deleted is registered as a duplicate, delete the registration. */ while (g_hash_table_iter_next(&iter, NULL, (void **)&list_op)) { GList *dup_iter = NULL; if (list_op != relay_op) { for (dup_iter = list_op->duplicates; dup_iter != NULL; dup_iter = dup_iter->next) { remote_fencing_op_t *other = dup_iter->data; if (other == relay_op) { other->duplicates = g_list_remove(other->duplicates, relay_op); break; } } } } crm_debug("Deleting relay op %s ('%s'%s%s for %s), " "replaced by op %s ('%s'%s%s for %s)", relay_op->id, relay_op->action, (relay_op->target == NULL)? "" : " targeting ", pcmk__s(relay_op->target, ""), relay_op->client_name, op_id, relay_op->action, (target == NULL)? "" : " targeting ", pcmk__s(target, ""), client_name); g_hash_table_remove(stonith_remote_op_list, relay_op_id); } } } /*! * \internal * \brief Check whether an API request was sent by a privileged user * * API commands related to fencing configuration may be done only by privileged * IPC users (i.e. root or hacluster), because all other users should go through * the CIB to have ACLs applied. If no client was given, this is a peer request, * which is always allowed. * * \param[in] c IPC client that sent request (or NULL if sent by CPG peer) * \param[in] op Requested API operation (for logging only) * * \return true if sender is peer or privileged client, otherwise false */ static inline bool is_privileged(const pcmk__client_t *c, const char *op) { if ((c == NULL) || pcmk_is_set(c->flags, pcmk__client_privileged)) { return true; } else { crm_warn("Rejecting IPC request '%s' from unprivileged client %s", pcmk__s(op, ""), pcmk__client_name(c)); return false; } } // CRM_OP_REGISTER static xmlNode * handle_register_request(pcmk__request_t *request) { xmlNode *reply = pcmk__xe_create(NULL, "reply"); pcmk__assert(request->ipc_client != NULL); crm_xml_add(reply, PCMK__XA_ST_OP, CRM_OP_REGISTER); crm_xml_add(reply, PCMK__XA_ST_CLIENTID, request->ipc_client->id); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); pcmk__set_request_flags(request, pcmk__request_reuse_options); return reply; } // STONITH_OP_EXEC static xmlNode * handle_agent_request(pcmk__request_t *request) { execute_agent_action(request->xml, &request->result); if (request->result.execution_status == PCMK_EXEC_PENDING) { return NULL; } return fenced_construct_reply(request->xml, NULL, &request->result); } // STONITH_OP_TIMEOUT_UPDATE static xmlNode * handle_update_timeout_request(pcmk__request_t *request) { const char *call_id = crm_element_value(request->xml, PCMK__XA_ST_CALLID); const char *client_id = crm_element_value(request->xml, PCMK__XA_ST_CLIENTID); int op_timeout = 0; crm_element_value_int(request->xml, PCMK__XA_ST_TIMEOUT, &op_timeout); do_stonith_async_timeout_update(client_id, call_id, op_timeout); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); return NULL; } // STONITH_OP_QUERY static xmlNode * handle_query_request(pcmk__request_t *request) { int timeout = 0; xmlNode *dev = NULL; const char *action = NULL; const char *target = NULL; const char *client_id = crm_element_value(request->xml, PCMK__XA_ST_CLIENTID); struct st_query_data *query = NULL; if (request->peer != NULL) { // Record it for the future notification create_remote_stonith_op(client_id, request->xml, TRUE); } /* Delete the DC node RELAY operation. */ remove_relay_op(request->xml); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); dev = get_xpath_object("//@" PCMK__XE_ST_DEVICE_ACTION, request->xml, LOG_NEVER); if (dev != NULL) { const char *device = crm_element_value(dev, PCMK__XA_ST_DEVICE_ID); if (pcmk__str_eq(device, "manual_ack", pcmk__str_casei)) { return NULL; // No query or reply necessary } target = crm_element_value(dev, PCMK__XA_ST_TARGET); action = crm_element_value(dev, PCMK__XA_ST_DEVICE_ACTION); } crm_log_xml_trace(request->xml, "Query"); query = pcmk__assert_alloc(1, sizeof(struct st_query_data)); query->reply = fenced_construct_reply(request->xml, NULL, &request->result); query->remote_peer = pcmk__str_copy(request->peer); query->client_id = pcmk__str_copy(client_id); query->target = pcmk__str_copy(target); query->action = pcmk__str_copy(action); query->call_options = request->call_options; crm_element_value_int(request->xml, PCMK__XA_ST_TIMEOUT, &timeout); get_capable_devices(target, action, timeout, pcmk_is_set(query->call_options, st_opt_allow_self_fencing), query, stonith_query_capable_device_cb, st_device_supports_none); return NULL; } // STONITH_OP_NOTIFY static xmlNode * handle_notify_request(pcmk__request_t *request) { const char *flag_name = NULL; pcmk__assert(request->ipc_client != NULL); flag_name = crm_element_value(request->xml, PCMK__XA_ST_NOTIFY_ACTIVATE); if (flag_name != NULL) { crm_debug("Enabling %s callbacks for client %s", flag_name, pcmk__request_origin(request)); pcmk__set_client_flags(request->ipc_client, get_stonith_flag(flag_name)); } flag_name = crm_element_value(request->xml, PCMK__XA_ST_NOTIFY_DEACTIVATE); if (flag_name != NULL) { crm_debug("Disabling %s callbacks for client %s", flag_name, pcmk__request_origin(request)); pcmk__clear_client_flags(request->ipc_client, get_stonith_flag(flag_name)); } pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); pcmk__set_request_flags(request, pcmk__request_reuse_options); return pcmk__ipc_create_ack(request->ipc_flags, PCMK__XE_ACK, NULL, CRM_EX_OK); } // STONITH_OP_RELAY static xmlNode * handle_relay_request(pcmk__request_t *request) { xmlNode *dev = get_xpath_object("//@" PCMK__XA_ST_TARGET, request->xml, LOG_TRACE); crm_notice("Received forwarded fencing request from " "%s %s to fence (%s) peer %s", pcmk__request_origin_type(request), pcmk__request_origin(request), crm_element_value(dev, PCMK__XA_ST_DEVICE_ACTION), crm_element_value(dev, PCMK__XA_ST_TARGET)); if (initiate_remote_stonith_op(NULL, request->xml, FALSE) == NULL) { fenced_set_protocol_error(&request->result); return fenced_construct_reply(request->xml, NULL, &request->result); } pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_PENDING, NULL); return NULL; } // STONITH_OP_FENCE static xmlNode * handle_fence_request(pcmk__request_t *request) { if ((request->peer != NULL) || stand_alone) { fence_locally(request->xml, &request->result); } else if (pcmk_is_set(request->call_options, st_opt_manual_ack)) { switch (fenced_handle_manual_confirmation(request->ipc_client, request->xml)) { case pcmk_rc_ok: pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); break; case EINPROGRESS: pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_PENDING, NULL); break; default: fenced_set_protocol_error(&request->result); break; } } else { const char *alternate_host = NULL; xmlNode *dev = get_xpath_object("//@" PCMK__XA_ST_TARGET, request->xml, LOG_TRACE); const char *target = crm_element_value(dev, PCMK__XA_ST_TARGET); const char *action = crm_element_value(dev, PCMK__XA_ST_DEVICE_ACTION); const char *device = crm_element_value(dev, PCMK__XA_ST_DEVICE_ID); if (request->ipc_client != NULL) { int tolerance = 0; crm_notice("Client %s wants to fence (%s) %s using %s", pcmk__request_origin(request), action, target, (device? device : "any device")); crm_element_value_int(dev, PCMK__XA_ST_TOLERANCE, &tolerance); if (stonith_check_fence_tolerance(tolerance, target, action)) { pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); return fenced_construct_reply(request->xml, NULL, &request->result); } alternate_host = check_alternate_host(target); } else { crm_notice("Peer %s wants to fence (%s) '%s' with device '%s'", request->peer, action, target, (device == NULL)? "(any)" : device); } if (alternate_host != NULL) { const char *client_id = NULL; remote_fencing_op_t *op = NULL; crm_node_t *node = pcmk__get_node(0, alternate_host, NULL, pcmk__node_search_cluster_member); if (request->ipc_client->id == 0) { client_id = crm_element_value(request->xml, PCMK__XA_ST_CLIENTID); } else { client_id = request->ipc_client->id; } /* Create a duplicate fencing operation to relay with the client ID. * When a query response is received, this operation should be * deleted to avoid keeping the duplicate around. */ op = create_remote_stonith_op(client_id, request->xml, FALSE); crm_xml_add(request->xml, PCMK__XA_ST_OP, STONITH_OP_RELAY); crm_xml_add(request->xml, PCMK__XA_ST_CLIENTID, request->ipc_client->id); crm_xml_add(request->xml, PCMK__XA_ST_REMOTE_OP, op->id); pcmk__cluster_send_message(node, crm_msg_stonith_ng, request->xml); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_PENDING, NULL); } else if (initiate_remote_stonith_op(request->ipc_client, request->xml, FALSE) == NULL) { fenced_set_protocol_error(&request->result); } else { pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_PENDING, NULL); } } if (request->result.execution_status == PCMK_EXEC_PENDING) { return NULL; } return fenced_construct_reply(request->xml, NULL, &request->result); } // STONITH_OP_FENCE_HISTORY static xmlNode * handle_history_request(pcmk__request_t *request) { xmlNode *reply = NULL; xmlNode *data = NULL; stonith_fence_history(request->xml, &data, request->peer, request->call_options); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); if (!pcmk_is_set(request->call_options, st_opt_discard_reply)) { /* When the local node broadcasts its history, it sets * st_opt_discard_reply and doesn't need a reply. */ reply = fenced_construct_reply(request->xml, data, &request->result); } free_xml(data); return reply; } // STONITH_OP_DEVICE_ADD static xmlNode * handle_device_add_request(pcmk__request_t *request) { const char *op = crm_element_value(request->xml, PCMK__XA_ST_OP); xmlNode *dev = get_xpath_object("//" PCMK__XE_ST_DEVICE_ID, request->xml, LOG_ERR); if (is_privileged(request->ipc_client, op)) { int rc = stonith_device_register(dev, FALSE); pcmk__set_result(&request->result, ((rc == pcmk_ok)? CRM_EX_OK : CRM_EX_ERROR), stonith__legacy2status(rc), ((rc == pcmk_ok)? NULL : pcmk_strerror(rc))); } else { pcmk__set_result(&request->result, CRM_EX_INSUFFICIENT_PRIV, PCMK_EXEC_INVALID, "Unprivileged users must register device via CIB"); } fenced_send_config_notification(op, &request->result, (dev == NULL)? NULL : pcmk__xe_id(dev)); return fenced_construct_reply(request->xml, NULL, &request->result); } // STONITH_OP_DEVICE_DEL static xmlNode * handle_device_delete_request(pcmk__request_t *request) { xmlNode *dev = get_xpath_object("//" PCMK__XE_ST_DEVICE_ID, request->xml, LOG_ERR); const char *device_id = crm_element_value(dev, PCMK_XA_ID); const char *op = crm_element_value(request->xml, PCMK__XA_ST_OP); if (is_privileged(request->ipc_client, op)) { stonith_device_remove(device_id, false); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); } else { pcmk__set_result(&request->result, CRM_EX_INSUFFICIENT_PRIV, PCMK_EXEC_INVALID, "Unprivileged users must delete device via CIB"); } fenced_send_config_notification(op, &request->result, device_id); return fenced_construct_reply(request->xml, NULL, &request->result); } // STONITH_OP_LEVEL_ADD static xmlNode * handle_level_add_request(pcmk__request_t *request) { char *desc = NULL; const char *op = crm_element_value(request->xml, PCMK__XA_ST_OP); if (is_privileged(request->ipc_client, op)) { fenced_register_level(request->xml, &desc, &request->result); } else { unpack_level_request(request->xml, NULL, NULL, NULL, &desc); pcmk__set_result(&request->result, CRM_EX_INSUFFICIENT_PRIV, PCMK_EXEC_INVALID, "Unprivileged users must add level via CIB"); } fenced_send_config_notification(op, &request->result, desc); free(desc); return fenced_construct_reply(request->xml, NULL, &request->result); } // STONITH_OP_LEVEL_DEL static xmlNode * handle_level_delete_request(pcmk__request_t *request) { char *desc = NULL; const char *op = crm_element_value(request->xml, PCMK__XA_ST_OP); if (is_privileged(request->ipc_client, op)) { fenced_unregister_level(request->xml, &desc, &request->result); } else { unpack_level_request(request->xml, NULL, NULL, NULL, &desc); pcmk__set_result(&request->result, CRM_EX_INSUFFICIENT_PRIV, PCMK_EXEC_INVALID, "Unprivileged users must delete level via CIB"); } fenced_send_config_notification(op, &request->result, desc); free(desc); return fenced_construct_reply(request->xml, NULL, &request->result); } // CRM_OP_RM_NODE_CACHE static xmlNode * handle_cache_request(pcmk__request_t *request) { int node_id = 0; const char *name = NULL; crm_element_value_int(request->xml, PCMK_XA_ID, &node_id); name = crm_element_value(request->xml, PCMK_XA_UNAME); pcmk__cluster_forget_cluster_node(node_id, name); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); return NULL; } static xmlNode * handle_unknown_request(pcmk__request_t *request) { crm_err("Unknown IPC request %s from %s %s", request->op, pcmk__request_origin_type(request), pcmk__request_origin(request)); pcmk__format_result(&request->result, CRM_EX_PROTOCOL, PCMK_EXEC_INVALID, "Unknown IPC request type '%s' (bug?)", request->op); return fenced_construct_reply(request->xml, NULL, &request->result); } static void fenced_register_handlers(void) { pcmk__server_command_t handlers[] = { { CRM_OP_REGISTER, handle_register_request }, { STONITH_OP_EXEC, handle_agent_request }, { STONITH_OP_TIMEOUT_UPDATE, handle_update_timeout_request }, { STONITH_OP_QUERY, handle_query_request }, { STONITH_OP_NOTIFY, handle_notify_request }, { STONITH_OP_RELAY, handle_relay_request }, { STONITH_OP_FENCE, handle_fence_request }, { STONITH_OP_FENCE_HISTORY, handle_history_request }, { STONITH_OP_DEVICE_ADD, handle_device_add_request }, { STONITH_OP_DEVICE_DEL, handle_device_delete_request }, { STONITH_OP_LEVEL_ADD, handle_level_add_request }, { STONITH_OP_LEVEL_DEL, handle_level_delete_request }, { CRM_OP_RM_NODE_CACHE, handle_cache_request }, { NULL, handle_unknown_request }, }; fenced_handlers = pcmk__register_handlers(handlers); } void fenced_unregister_handlers(void) { if (fenced_handlers != NULL) { g_hash_table_destroy(fenced_handlers); fenced_handlers = NULL; } } static void handle_request(pcmk__request_t *request) { xmlNode *reply = NULL; const char *reason = NULL; if (fenced_handlers == NULL) { fenced_register_handlers(); } reply = pcmk__process_request(request, fenced_handlers); if (reply != NULL) { if (pcmk_is_set(request->flags, pcmk__request_reuse_options) && (request->ipc_client != NULL)) { /* Certain IPC-only commands must reuse the call options from the * original request rather than the ones set by stonith_send_reply() * -> do_local_reply(). */ pcmk__ipc_send_xml(request->ipc_client, request->ipc_id, reply, request->ipc_flags); request->ipc_client->request_id = 0; } else { stonith_send_reply(reply, request->call_options, request->peer, request->ipc_client); } free_xml(reply); } reason = request->result.exit_reason; crm_debug("Processed %s request from %s %s: %s%s%s%s", request->op, pcmk__request_origin_type(request), pcmk__request_origin(request), pcmk_exec_status_str(request->result.execution_status), (reason == NULL)? "" : " (", (reason == NULL)? "" : reason, (reason == NULL)? "" : ")"); } static void handle_reply(pcmk__client_t *client, xmlNode *request, const char *remote_peer) { // Copy, because request might be freed before we want to log this char *op = crm_element_value_copy(request, PCMK__XA_ST_OP); if (pcmk__str_eq(op, STONITH_OP_QUERY, pcmk__str_none)) { process_remote_stonith_query(request); } else if (pcmk__str_any_of(op, STONITH_OP_NOTIFY, STONITH_OP_FENCE, NULL)) { fenced_process_fencing_reply(request); } else { crm_err("Ignoring unknown %s reply from %s %s", pcmk__s(op, "untyped"), ((client == NULL)? "peer" : "client"), ((client == NULL)? remote_peer : pcmk__client_name(client))); crm_log_xml_warn(request, "UnknownOp"); free(op); return; } crm_debug("Processed %s reply from %s %s", op, ((client == NULL)? "peer" : "client"), ((client == NULL)? remote_peer : pcmk__client_name(client))); free(op); } /*! * \internal * \brief Handle a message from an IPC client or CPG peer * * \param[in,out] client If not NULL, IPC client that sent message * \param[in] id If from IPC client, IPC message ID * \param[in] flags Message flags * \param[in,out] message Message XML * \param[in] remote_peer If not NULL, CPG peer that sent message */ void stonith_command(pcmk__client_t *client, uint32_t id, uint32_t flags, xmlNode *message, const char *remote_peer) { uint32_t call_options = st_opt_none; int rc = pcmk_rc_ok; bool is_reply = false; CRM_CHECK(message != NULL, return); if (get_xpath_object("//" PCMK__XE_ST_REPLY, message, LOG_NEVER) != NULL) { is_reply = true; } rc = pcmk__xe_get_flags(message, PCMK__XA_ST_CALLOPT, &call_options, st_opt_none); if (rc != pcmk_rc_ok) { crm_warn("Couldn't parse options from message: %s", pcmk_rc_str(rc)); } crm_debug("Processing %ssynchronous %s %s %u from %s %s", pcmk_is_set(call_options, st_opt_sync_call)? "" : "a", crm_element_value(message, PCMK__XA_ST_OP), (is_reply? "reply" : "request"), id, ((client == NULL)? "peer" : "client"), ((client == NULL)? remote_peer : pcmk__client_name(client))); if (pcmk_is_set(call_options, st_opt_sync_call)) { pcmk__assert((client == NULL) || (client->request_id == id)); } if (is_reply) { handle_reply(client, message, remote_peer); } else { pcmk__request_t request = { .ipc_client = client, .ipc_id = id, .ipc_flags = flags, .peer = remote_peer, .xml = message, .call_options = call_options, .result = PCMK__UNKNOWN_RESULT, }; request.op = crm_element_value_copy(request.xml, PCMK__XA_ST_OP); CRM_CHECK(request.op != NULL, return); if (pcmk_is_set(request.call_options, st_opt_sync_call)) { pcmk__set_request_flags(&request, pcmk__request_sync); } handle_request(&request); pcmk__reset_request(&request); } } diff --git a/daemons/fenced/fenced_remote.c b/daemons/fenced/fenced_remote.c index d04ae85cf9..8ff2136ec7 100644 --- a/daemons/fenced/fenced_remote.c +++ b/daemons/fenced/fenced_remote.c @@ -1,2608 +1,2606 @@ /* - * Copyright 2009-2024 the Pacemaker project contributors + * Copyright 2009-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU General Public License version 2 * or later (GPLv2+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #define TIMEOUT_MULTIPLY_FACTOR 1.2 /* When one fencer queries its peers for devices able to handle a fencing * request, each peer will reply with a list of such devices available to it. * Each reply will be parsed into a peer_device_info_t, with each device's * information kept in a device_properties_t. */ typedef struct device_properties_s { /* Whether access to this device has been verified */ gboolean verified; /* The remaining members are indexed by the operation's "phase" */ /* Whether this device has been executed in each phase */ gboolean executed[st_phase_max]; /* Whether this device is disallowed from executing in each phase */ gboolean disallowed[st_phase_max]; /* Action-specific timeout for each phase */ int custom_action_timeout[st_phase_max]; /* Action-specific maximum random delay for each phase */ int delay_max[st_phase_max]; /* Action-specific base delay for each phase */ int delay_base[st_phase_max]; /* Group of enum st_device_flags */ uint32_t device_support_flags; } device_properties_t; typedef struct { /* Name of peer that sent this result */ char *host; /* Only try peers for non-topology based operations once */ gboolean tried; /* Number of entries in the devices table */ int ndevices; /* Devices available to this host that are capable of fencing the target */ GHashTable *devices; } peer_device_info_t; GHashTable *stonith_remote_op_list = NULL; extern xmlNode *stonith_create_op(int call_id, const char *token, const char *op, xmlNode * data, int call_options); static void request_peer_fencing(remote_fencing_op_t *op, peer_device_info_t *peer); static void finalize_op(remote_fencing_op_t *op, xmlNode *data, bool dup); static void report_timeout_period(remote_fencing_op_t * op, int op_timeout); static int get_op_total_timeout(const remote_fencing_op_t *op, const peer_device_info_t *chosen_peer); static gint sort_strings(gconstpointer a, gconstpointer b) { return strcmp(a, b); } static void free_remote_query(gpointer data) { if (data != NULL) { peer_device_info_t *peer = data; g_hash_table_destroy(peer->devices); free(peer->host); free(peer); } } void free_stonith_remote_op_list(void) { if (stonith_remote_op_list != NULL) { g_hash_table_destroy(stonith_remote_op_list); stonith_remote_op_list = NULL; } } struct peer_count_data { const remote_fencing_op_t *op; gboolean verified_only; uint32_t support_action_only; int count; }; /*! * \internal * \brief Increment a counter if a device has not been executed yet * * \param[in] key Device ID (ignored) * \param[in] value Device properties * \param[in,out] user_data Peer count data */ static void count_peer_device(gpointer key, gpointer value, gpointer user_data) { device_properties_t *props = (device_properties_t*)value; struct peer_count_data *data = user_data; if (!props->executed[data->op->phase] && (!data->verified_only || props->verified) && ((data->support_action_only == st_device_supports_none) || pcmk_is_set(props->device_support_flags, data->support_action_only))) { ++(data->count); } } /*! * \internal * \brief Check the number of available devices in a peer's query results * * \param[in] op Operation that results are for * \param[in] peer Peer to count * \param[in] verified_only Whether to count only verified devices * \param[in] support_action_only Whether to count only devices that support action * * \return Number of devices available to peer that were not already executed */ static int count_peer_devices(const remote_fencing_op_t *op, const peer_device_info_t *peer, gboolean verified_only, uint32_t support_on_action_only) { struct peer_count_data data; data.op = op; data.verified_only = verified_only; data.support_action_only = support_on_action_only; data.count = 0; if (peer) { g_hash_table_foreach(peer->devices, count_peer_device, &data); } return data.count; } /*! * \internal * \brief Search for a device in a query result * * \param[in] op Operation that result is for * \param[in] peer Query result for a peer * \param[in] device Device ID to search for * * \return Device properties if found, NULL otherwise */ static device_properties_t * find_peer_device(const remote_fencing_op_t *op, const peer_device_info_t *peer, const char *device, uint32_t support_action_only) { device_properties_t *props = g_hash_table_lookup(peer->devices, device); if (props && support_action_only != st_device_supports_none && !pcmk_is_set(props->device_support_flags, support_action_only)) { return NULL; } return (props && !props->executed[op->phase] && !props->disallowed[op->phase])? props : NULL; } /*! * \internal * \brief Find a device in a peer's device list and mark it as executed * * \param[in] op Operation that peer result is for * \param[in,out] peer Peer with results to search * \param[in] device ID of device to mark as done * \param[in] verified_devices_only Only consider verified devices * * \return TRUE if device was found and marked, FALSE otherwise */ static gboolean grab_peer_device(const remote_fencing_op_t *op, peer_device_info_t *peer, const char *device, gboolean verified_devices_only) { device_properties_t *props = find_peer_device(op, peer, device, fenced_support_flag(op->action)); if ((props == NULL) || (verified_devices_only && !props->verified)) { return FALSE; } crm_trace("Removing %s from %s (%d remaining)", device, peer->host, count_peer_devices(op, peer, FALSE, st_device_supports_none)); props->executed[op->phase] = TRUE; return TRUE; } static void clear_remote_op_timers(remote_fencing_op_t * op) { if (op->query_timer) { g_source_remove(op->query_timer); op->query_timer = 0; } if (op->op_timer_total) { g_source_remove(op->op_timer_total); op->op_timer_total = 0; } if (op->op_timer_one) { g_source_remove(op->op_timer_one); op->op_timer_one = 0; } } static void free_remote_op(gpointer data) { remote_fencing_op_t *op = data; crm_log_xml_debug(op->request, "Destroying"); clear_remote_op_timers(op); free(op->id); free(op->action); free(op->delegate); free(op->target); free(op->client_id); free(op->client_name); free(op->originator); if (op->query_results) { g_list_free_full(op->query_results, free_remote_query); } if (op->request) { free_xml(op->request); op->request = NULL; } if (op->devices_list) { g_list_free_full(op->devices_list, free); op->devices_list = NULL; } g_list_free_full(op->automatic_list, free); g_list_free(op->duplicates); pcmk__reset_result(&op->result); free(op); } void init_stonith_remote_op_hash_table(GHashTable **table) { if (*table == NULL) { *table = pcmk__strkey_table(NULL, free_remote_op); } } /*! * \internal * \brief Return an operation's originally requested action (before any remap) * * \param[in] op Operation to check * * \return Operation's original action */ static const char * op_requested_action(const remote_fencing_op_t *op) { return ((op->phase > st_phase_requested)? PCMK_ACTION_REBOOT : op->action); } /*! * \internal * \brief Remap a "reboot" operation to the "off" phase * * \param[in,out] op Operation to remap */ static void op_phase_off(remote_fencing_op_t *op) { crm_info("Remapping multiple-device reboot targeting %s to 'off' " CRM_XS " id=%.8s", op->target, op->id); op->phase = st_phase_off; /* Happily, "off" and "on" are shorter than "reboot", so we can reuse the * memory allocation at each phase. */ strcpy(op->action, PCMK_ACTION_OFF); } /*! * \internal * \brief Advance a remapped reboot operation to the "on" phase * * \param[in,out] op Operation to remap */ static void op_phase_on(remote_fencing_op_t *op) { GList *iter = NULL; crm_info("Remapped 'off' targeting %s complete, " "remapping to 'on' for %s " CRM_XS " id=%.8s", op->target, op->client_name, op->id); op->phase = st_phase_on; strcpy(op->action, PCMK_ACTION_ON); /* Skip devices with automatic unfencing, because the cluster will handle it * when the node rejoins. */ for (iter = op->automatic_list; iter != NULL; iter = iter->next) { GList *match = g_list_find_custom(op->devices_list, iter->data, sort_strings); if (match) { op->devices_list = g_list_remove(op->devices_list, match->data); } } g_list_free_full(op->automatic_list, free); op->automatic_list = NULL; /* Rewind device list pointer */ op->devices = op->devices_list; } /*! * \internal * \brief Reset a remapped reboot operation * * \param[in,out] op Operation to reset */ static void undo_op_remap(remote_fencing_op_t *op) { if (op->phase > 0) { crm_info("Undoing remap of reboot targeting %s for %s " CRM_XS " id=%.8s", op->target, op->client_name, op->id); op->phase = st_phase_requested; strcpy(op->action, PCMK_ACTION_REBOOT); } } /*! * \internal * \brief Create notification data XML for a fencing operation result * * \param[in,out] parent Parent XML element for newly created element * \param[in] op Fencer operation that completed * * \return Newly created XML to add as notification data * \note The caller is responsible for freeing the result. */ static xmlNode * fencing_result2xml(xmlNode *parent, const remote_fencing_op_t *op) { xmlNode *notify_data = pcmk__xe_create(parent, PCMK__XE_ST_NOTIFY_FENCE); crm_xml_add_int(notify_data, PCMK_XA_STATE, op->state); crm_xml_add(notify_data, PCMK__XA_ST_TARGET, op->target); crm_xml_add(notify_data, PCMK__XA_ST_DEVICE_ACTION, op->action); crm_xml_add(notify_data, PCMK__XA_ST_DELEGATE, op->delegate); crm_xml_add(notify_data, PCMK__XA_ST_REMOTE_OP, op->id); crm_xml_add(notify_data, PCMK__XA_ST_ORIGIN, op->originator); crm_xml_add(notify_data, PCMK__XA_ST_CLIENTID, op->client_id); crm_xml_add(notify_data, PCMK__XA_ST_CLIENTNAME, op->client_name); return notify_data; } /*! * \internal * \brief Broadcast a fence result notification to all CPG peers * * \param[in] op Fencer operation that completed * \param[in] op_merged Whether this operation is a duplicate of another */ void fenced_broadcast_op_result(const remote_fencing_op_t *op, bool op_merged) { static int count = 0; xmlNode *bcast = pcmk__xe_create(NULL, PCMK__XE_ST_REPLY); xmlNode *wrapper = NULL; xmlNode *notify_data = NULL; count++; crm_trace("Broadcasting result to peers"); crm_xml_add(bcast, PCMK__XA_T, PCMK__VALUE_ST_NOTIFY); crm_xml_add(bcast, PCMK__XA_SUBT, PCMK__VALUE_BROADCAST); crm_xml_add(bcast, PCMK__XA_ST_OP, STONITH_OP_NOTIFY); crm_xml_add_int(bcast, PCMK_XA_COUNT, count); if (op_merged) { pcmk__xe_set_bool_attr(bcast, PCMK__XA_ST_OP_MERGED, true); } wrapper = pcmk__xe_create(bcast, PCMK__XE_ST_CALLDATA); notify_data = fencing_result2xml(wrapper, op); stonith__xe_set_result(notify_data, &op->result); pcmk__cluster_send_message(NULL, crm_msg_stonith_ng, bcast); free_xml(bcast); return; } /*! * \internal * \brief Reply to a local request originator and notify all subscribed clients * * \param[in,out] op Fencer operation that completed * \param[in,out] data Top-level XML to add notification to */ static void handle_local_reply_and_notify(remote_fencing_op_t *op, xmlNode *data) { xmlNode *notify_data = NULL; xmlNode *reply = NULL; pcmk__client_t *client = NULL; if (op->notify_sent == TRUE) { /* nothing to do */ return; } /* Do notification with a clean data object */ crm_xml_add_int(data, PCMK_XA_STATE, op->state); crm_xml_add(data, PCMK__XA_ST_TARGET, op->target); crm_xml_add(data, PCMK__XA_ST_OP, op->action); reply = fenced_construct_reply(op->request, data, &op->result); crm_xml_add(reply, PCMK__XA_ST_DELEGATE, op->delegate); /* Send fencing OP reply to local client that initiated fencing */ client = pcmk__find_client_by_id(op->client_id); if (client == NULL) { crm_trace("Skipping reply to %s: no longer a client", op->client_id); } else { do_local_reply(reply, client, op->call_options); } /* bcast to all local clients that the fencing operation happend */ notify_data = fencing_result2xml(NULL, op); fenced_send_notification(PCMK__VALUE_ST_NOTIFY_FENCE, &op->result, notify_data); free_xml(notify_data); fenced_send_notification(PCMK__VALUE_ST_NOTIFY_HISTORY, NULL, NULL); /* mark this op as having notify's already sent */ op->notify_sent = TRUE; free_xml(reply); } /*! * \internal * \brief Finalize all duplicates of a given fencer operation * * \param[in,out] op Fencer operation that completed * \param[in,out] data Top-level XML to add notification to */ static void finalize_op_duplicates(remote_fencing_op_t *op, xmlNode *data) { for (GList *iter = op->duplicates; iter != NULL; iter = iter->next) { remote_fencing_op_t *other = iter->data; if (other->state == st_duplicate) { other->state = op->state; crm_debug("Performing duplicate notification for %s@%s: %s " CRM_XS " id=%.8s", other->client_name, other->originator, pcmk_exec_status_str(op->result.execution_status), other->id); pcmk__copy_result(&op->result, &other->result); finalize_op(other, data, true); } else { // Possible if (for example) it timed out already crm_err("Skipping duplicate notification for %s@%s " CRM_XS " state=%s id=%.8s", other->client_name, other->originator, stonith_op_state_str(other->state), other->id); } } } static char * delegate_from_xml(xmlNode *xml) { xmlNode *match = get_xpath_object("//@" PCMK__XA_ST_DELEGATE, xml, LOG_NEVER); if (match == NULL) { return crm_element_value_copy(xml, PCMK__XA_SRC); } else { return crm_element_value_copy(match, PCMK__XA_ST_DELEGATE); } } /*! * \internal * \brief Finalize a peer fencing operation * * Clean up after a fencing operation completes. This function has two code * paths: the executioner uses it to broadcast the result to CPG peers, and then * each peer (including the executioner) uses it to process that broadcast and * notify its IPC clients of the result. * * \param[in,out] op Fencer operation that completed * \param[in,out] data If not NULL, XML reply of last delegated operation * \param[in] dup Whether this operation is a duplicate of another * (in which case, do not broadcast the result) * * \note The operation result should be set before calling this function. */ static void finalize_op(remote_fencing_op_t *op, xmlNode *data, bool dup) { int level = LOG_ERR; const char *subt = NULL; xmlNode *local_data = NULL; gboolean op_merged = FALSE; CRM_CHECK((op != NULL), return); // This is a no-op if timers have already been cleared clear_remote_op_timers(op); if (op->notify_sent) { // Most likely, this is a timed-out action that eventually completed crm_notice("Operation '%s'%s%s by %s for %s@%s%s: " "Result arrived too late " CRM_XS " id=%.8s", op->action, (op->target? " targeting " : ""), (op->target? op->target : ""), (op->delegate? op->delegate : "unknown node"), op->client_name, op->originator, (op_merged? " (merged)" : ""), op->id); return; } set_fencing_completed(op); undo_op_remap(op); if (data == NULL) { data = pcmk__xe_create(NULL, "remote-op"); local_data = data; } else if (op->delegate == NULL) { switch (op->result.execution_status) { case PCMK_EXEC_NO_FENCE_DEVICE: break; case PCMK_EXEC_INVALID: if (op->result.exit_status != CRM_EX_EXPIRED) { op->delegate = delegate_from_xml(data); } break; default: op->delegate = delegate_from_xml(data); break; } } if (dup || (crm_element_value(data, PCMK__XA_ST_OP_MERGED) != NULL)) { op_merged = true; } /* Tell everyone the operation is done, we will continue * with doing the local notifications once we receive * the broadcast back. */ subt = crm_element_value(data, PCMK__XA_SUBT); if (!dup && !pcmk__str_eq(subt, PCMK__VALUE_BROADCAST, pcmk__str_none)) { /* Defer notification until the bcast message arrives */ fenced_broadcast_op_result(op, op_merged); free_xml(local_data); return; } if (pcmk__result_ok(&op->result) || dup || !pcmk__str_eq(op->originator, stonith_our_uname, pcmk__str_casei)) { level = LOG_NOTICE; } do_crm_log(level, "Operation '%s'%s%s by %s for %s@%s%s: %s (%s%s%s) " CRM_XS " id=%.8s", op->action, (op->target? " targeting " : ""), (op->target? op->target : ""), (op->delegate? op->delegate : "unknown node"), op->client_name, op->originator, (op_merged? " (merged)" : ""), crm_exit_str(op->result.exit_status), pcmk_exec_status_str(op->result.execution_status), ((op->result.exit_reason == NULL)? "" : ": "), ((op->result.exit_reason == NULL)? "" : op->result.exit_reason), op->id); handle_local_reply_and_notify(op, data); if (!dup) { finalize_op_duplicates(op, data); } /* Free non-essential parts of the record * Keep the record around so we can query the history */ if (op->query_results) { g_list_free_full(op->query_results, free_remote_query); op->query_results = NULL; } if (op->request) { free_xml(op->request); op->request = NULL; } free_xml(local_data); } /*! * \internal * \brief Finalize a watchdog fencer op after the waiting time expires * * \param[in,out] userdata Fencer operation that completed * * \return G_SOURCE_REMOVE (which tells glib not to restart timer) */ static gboolean remote_op_watchdog_done(gpointer userdata) { remote_fencing_op_t *op = userdata; op->op_timer_one = 0; crm_notice("Self-fencing (%s) by %s for %s assumed complete " CRM_XS " id=%.8s", op->action, op->target, op->client_name, op->id); op->state = st_done; pcmk__set_result(&op->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); finalize_op(op, NULL, false); return G_SOURCE_REMOVE; } static gboolean remote_op_timeout_one(gpointer userdata) { remote_fencing_op_t *op = userdata; op->op_timer_one = 0; crm_notice("Peer's '%s' action targeting %s for client %s timed out " CRM_XS " id=%.8s", op->action, op->target, op->client_name, op->id); pcmk__set_result(&op->result, CRM_EX_ERROR, PCMK_EXEC_TIMEOUT, "Peer did not return fence result within timeout"); // The requested delay has been applied for the first device if (op->client_delay > 0) { op->client_delay = 0; crm_trace("Try another device for '%s' action targeting %s " "for client %s without delay " CRM_XS " id=%.8s", op->action, op->target, op->client_name, op->id); } // Try another device, if appropriate request_peer_fencing(op, NULL); return G_SOURCE_REMOVE; } /*! * \internal * \brief Finalize a remote fencer operation that timed out * * \param[in,out] op Fencer operation that timed out * \param[in] reason Readable description of what step timed out */ static void finalize_timed_out_op(remote_fencing_op_t *op, const char *reason) { crm_debug("Action '%s' targeting %s for client %s timed out " CRM_XS " id=%.8s", op->action, op->target, op->client_name, op->id); if (op->phase == st_phase_on) { /* A remapped reboot operation timed out in the "on" phase, but the * "off" phase completed successfully, so quit trying any further * devices, and return success. */ op->state = st_done; pcmk__set_result(&op->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); } else { op->state = st_failed; pcmk__set_result(&op->result, CRM_EX_ERROR, PCMK_EXEC_TIMEOUT, reason); } finalize_op(op, NULL, false); } /*! * \internal * \brief Finalize a remote fencer operation that timed out * * \param[in,out] userdata Fencer operation that timed out * * \return G_SOURCE_REMOVE (which tells glib not to restart timer) */ static gboolean remote_op_timeout(gpointer userdata) { remote_fencing_op_t *op = userdata; op->op_timer_total = 0; if (op->state == st_done) { crm_debug("Action '%s' targeting %s for client %s already completed " CRM_XS " id=%.8s", op->action, op->target, op->client_name, op->id); } else { finalize_timed_out_op(userdata, "Fencing did not complete within a " "total timeout based on the " "configured timeout and retries for " "any devices attempted"); } return G_SOURCE_REMOVE; } static gboolean remote_op_query_timeout(gpointer data) { remote_fencing_op_t *op = data; op->query_timer = 0; if (op->state == st_done) { crm_debug("Operation %.8s targeting %s already completed", op->id, op->target); } else if (op->state == st_exec) { crm_debug("Operation %.8s targeting %s already in progress", op->id, op->target); } else if (op->query_results) { // Query succeeded, so attempt the actual fencing crm_debug("Query %.8s targeting %s complete (state=%s)", op->id, op->target, stonith_op_state_str(op->state)); request_peer_fencing(op, NULL); } else { crm_debug("Query %.8s targeting %s timed out (state=%s)", op->id, op->target, stonith_op_state_str(op->state)); finalize_timed_out_op(op, "No capable peers replied to device query " "within timeout"); } return G_SOURCE_REMOVE; } static gboolean topology_is_empty(stonith_topology_t *tp) { int i; if (tp == NULL) { return TRUE; } for (i = 0; i < ST__LEVEL_COUNT; i++) { if (tp->levels[i] != NULL) { return FALSE; } } return TRUE; } /*! * \internal * \brief Add a device to an operation's automatic unfencing list * * \param[in,out] op Operation to modify * \param[in] device Device ID to add */ static void add_required_device(remote_fencing_op_t *op, const char *device) { GList *match = g_list_find_custom(op->automatic_list, device, sort_strings); if (!match) { op->automatic_list = g_list_prepend(op->automatic_list, pcmk__str_copy(device)); } } /*! * \internal * \brief Remove a device from the automatic unfencing list * * \param[in,out] op Operation to modify * \param[in] device Device ID to remove */ static void remove_required_device(remote_fencing_op_t *op, const char *device) { GList *match = g_list_find_custom(op->automatic_list, device, sort_strings); if (match) { op->automatic_list = g_list_remove(op->automatic_list, match->data); } } /* deep copy the device list */ static void set_op_device_list(remote_fencing_op_t * op, GList *devices) { GList *lpc = NULL; if (op->devices_list) { g_list_free_full(op->devices_list, free); op->devices_list = NULL; } for (lpc = devices; lpc != NULL; lpc = lpc->next) { const char *device = lpc->data; op->devices_list = g_list_append(op->devices_list, pcmk__str_copy(device)); } op->devices = op->devices_list; } /*! * \internal * \brief Check whether a node matches a topology target * * \param[in] tp Topology table entry to check * \param[in] node Name of node to check * * \return TRUE if node matches topology target */ static gboolean topology_matches(const stonith_topology_t *tp, const char *node) { regex_t r_patt; CRM_CHECK(node && tp && tp->target, return FALSE); switch (tp->kind) { case fenced_target_by_attribute: /* This level targets by attribute, so tp->target is a NAME=VALUE pair * of a permanent attribute applied to targeted nodes. The test below * relies on the locally cached copy of the CIB, so if fencing needs to * be done before the initial CIB is received or after a malformed CIB * is received, then the topology will be unable to be used. */ if (node_has_attr(node, tp->target_attribute, tp->target_value)) { crm_notice("Matched %s with %s by attribute", node, tp->target); return TRUE; } break; case fenced_target_by_pattern: /* This level targets node names matching a pattern, so tp->target * (and tp->target_pattern) is a regular expression. */ if (regcomp(&r_patt, tp->target_pattern, REG_EXTENDED|REG_NOSUB)) { crm_info("Bad regex '%s' for fencing level", tp->target); } else { int status = regexec(&r_patt, node, 0, NULL, 0); regfree(&r_patt); if (status == 0) { crm_notice("Matched %s with %s by name", node, tp->target); return TRUE; } } break; case fenced_target_by_name: crm_trace("Testing %s against %s", node, tp->target); return pcmk__str_eq(tp->target, node, pcmk__str_casei); default: break; } crm_trace("No match for %s with %s", node, tp->target); return FALSE; } stonith_topology_t * find_topology_for_host(const char *host) { GHashTableIter tIter; stonith_topology_t *tp = g_hash_table_lookup(topology, host); if(tp != NULL) { crm_trace("Found %s for %s in %d entries", tp->target, host, g_hash_table_size(topology)); return tp; } g_hash_table_iter_init(&tIter, topology); while (g_hash_table_iter_next(&tIter, NULL, (gpointer *) & tp)) { if (topology_matches(tp, host)) { crm_trace("Found %s for %s in %d entries", tp->target, host, g_hash_table_size(topology)); return tp; } } crm_trace("No matches for %s in %d topology entries", host, g_hash_table_size(topology)); return NULL; } /*! * \internal * \brief Set fencing operation's device list to target's next topology level * * \param[in,out] op Remote fencing operation to modify * \param[in] empty_ok If true, an operation without a target (i.e. * queries) or a target without a topology will get a * pcmk_rc_ok return value instead of ENODEV * * \return Standard Pacemaker return value */ static int advance_topology_level(remote_fencing_op_t *op, bool empty_ok) { stonith_topology_t *tp = NULL; if (op->target) { tp = find_topology_for_host(op->target); } if (topology_is_empty(tp)) { return empty_ok? pcmk_rc_ok : ENODEV; } pcmk__assert(tp->levels != NULL); stonith__set_call_options(op->call_options, op->id, st_opt_topology); /* This is a new level, so undo any remapping left over from previous */ undo_op_remap(op); do { op->level++; } while (op->level < ST__LEVEL_COUNT && tp->levels[op->level] == NULL); if (op->level < ST__LEVEL_COUNT) { crm_trace("Attempting fencing level %d targeting %s (%d devices) " "for client %s@%s (id=%.8s)", op->level, op->target, g_list_length(tp->levels[op->level]), op->client_name, op->originator, op->id); set_op_device_list(op, tp->levels[op->level]); // The requested delay has been applied for the first fencing level if ((op->level > 1) && (op->client_delay > 0)) { op->client_delay = 0; } if ((g_list_next(op->devices_list) != NULL) && pcmk__str_eq(op->action, PCMK_ACTION_REBOOT, pcmk__str_none)) { /* A reboot has been requested for a topology level with multiple * devices. Instead of rebooting the devices sequentially, we will * turn them all off, then turn them all on again. (Think about * switched power outlets for redundant power supplies.) */ op_phase_off(op); } return pcmk_rc_ok; } crm_info("All %sfencing options targeting %s for client %s@%s failed " CRM_XS " id=%.8s", (stonith_watchdog_timeout_ms > 0)?"non-watchdog ":"", op->target, op->client_name, op->originator, op->id); return ENODEV; } /*! * \internal * \brief If fencing operation is a duplicate, merge it into the other one * * \param[in,out] op Fencing operation to check */ static void merge_duplicates(remote_fencing_op_t *op) { GHashTableIter iter; remote_fencing_op_t *other = NULL; time_t now = time(NULL); g_hash_table_iter_init(&iter, stonith_remote_op_list); while (g_hash_table_iter_next(&iter, NULL, (void **)&other)) { const char *other_action = op_requested_action(other); crm_node_t *node = NULL; if (!strcmp(op->id, other->id)) { continue; // Don't compare against self } if (other->state > st_exec) { crm_trace("%.8s not duplicate of %.8s: not in progress", op->id, other->id); continue; } if (!pcmk__str_eq(op->target, other->target, pcmk__str_casei)) { crm_trace("%.8s not duplicate of %.8s: node %s vs. %s", op->id, other->id, op->target, other->target); continue; } if (!pcmk__str_eq(op->action, other_action, pcmk__str_none)) { crm_trace("%.8s not duplicate of %.8s: action %s vs. %s", op->id, other->id, op->action, other_action); continue; } if (pcmk__str_eq(op->client_name, other->client_name, pcmk__str_casei)) { crm_trace("%.8s not duplicate of %.8s: same client %s", op->id, other->id, op->client_name); continue; } if (pcmk__str_eq(other->target, other->originator, pcmk__str_casei)) { crm_trace("%.8s not duplicate of %.8s: self-fencing for %s", op->id, other->id, other->target); continue; } node = pcmk__get_node(0, other->originator, NULL, pcmk__node_search_cluster_member); if (!fencing_peer_active(node)) { crm_notice("Failing action '%s' targeting %s originating from " "client %s@%s: Originator is dead " CRM_XS " id=%.8s", other->action, other->target, other->client_name, other->originator, other->id); crm_trace("%.8s not duplicate of %.8s: originator dead", op->id, other->id); other->state = st_failed; continue; } if ((other->total_timeout > 0) && (now > (other->total_timeout + other->created))) { crm_trace("%.8s not duplicate of %.8s: old (%lld vs. %lld + %ds)", op->id, other->id, (long long)now, (long long)other->created, other->total_timeout); continue; } /* There is another in-flight request to fence the same host * Piggyback on that instead. If it fails, so do we. */ other->duplicates = g_list_append(other->duplicates, op); if (other->total_timeout == 0) { other->total_timeout = op->total_timeout = TIMEOUT_MULTIPLY_FACTOR * get_op_total_timeout(op, NULL); crm_trace("Best guess as to timeout used for %.8s: %ds", other->id, other->total_timeout); } crm_notice("Merging fencing action '%s' targeting %s originating from " "client %s with identical request from %s@%s " CRM_XS " original=%.8s duplicate=%.8s total_timeout=%ds", op->action, op->target, op->client_name, other->client_name, other->originator, op->id, other->id, other->total_timeout); report_timeout_period(op, other->total_timeout); op->state = st_duplicate; } } static uint32_t fencing_active_peers(void) { uint32_t count = 0; crm_node_t *entry; GHashTableIter gIter; g_hash_table_iter_init(&gIter, crm_peer_cache); while (g_hash_table_iter_next(&gIter, NULL, (void **)&entry)) { if(fencing_peer_active(entry)) { count++; } } return count; } /*! * \internal * \brief Process a manual confirmation of a pending fence action * * \param[in] client IPC client that sent confirmation * \param[in,out] msg Request XML with manual confirmation * * \return Standard Pacemaker return code */ int fenced_handle_manual_confirmation(const pcmk__client_t *client, xmlNode *msg) { remote_fencing_op_t *op = NULL; xmlNode *dev = get_xpath_object("//@" PCMK__XA_ST_TARGET, msg, LOG_ERR); CRM_CHECK(dev != NULL, return EPROTO); crm_notice("Received manual confirmation that %s has been fenced", pcmk__s(crm_element_value(dev, PCMK__XA_ST_TARGET), "unknown target")); op = initiate_remote_stonith_op(client, msg, TRUE); if (op == NULL) { return EPROTO; } op->state = st_done; set_fencing_completed(op); op->delegate = pcmk__str_copy("a human"); // For the fencer's purposes, the fencing operation is done pcmk__set_result(&op->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); finalize_op(op, msg, false); /* For the requester's purposes, the operation is still pending. The * actual result will be sent asynchronously via the operation's done_cb(). */ return EINPROGRESS; } /*! * \internal * \brief Create a new remote stonith operation * * \param[in] client ID of local stonith client that initiated the operation * \param[in] request The request from the client that started the operation * \param[in] peer TRUE if this operation is owned by another stonith peer * (an operation owned by one peer is stored on all peers, * but only the owner executes it; all nodes get the results * once the owner finishes execution) */ void * create_remote_stonith_op(const char *client, xmlNode *request, gboolean peer) { remote_fencing_op_t *op = NULL; xmlNode *dev = get_xpath_object("//@" PCMK__XA_ST_TARGET, request, LOG_NEVER); int rc = pcmk_rc_ok; const char *operation = NULL; init_stonith_remote_op_hash_table(&stonith_remote_op_list); /* If this operation is owned by another node, check to make * sure we haven't already created this operation. */ if (peer && dev) { const char *op_id = crm_element_value(dev, PCMK__XA_ST_REMOTE_OP); CRM_CHECK(op_id != NULL, return NULL); op = g_hash_table_lookup(stonith_remote_op_list, op_id); if (op) { crm_debug("Reusing existing remote fencing op %.8s for %s", op_id, ((client == NULL)? "unknown client" : client)); return op; } } op = pcmk__assert_alloc(1, sizeof(remote_fencing_op_t)); crm_element_value_int(request, PCMK__XA_ST_TIMEOUT, &(op->base_timeout)); // Value -1 means disable any static/random fencing delays crm_element_value_int(request, PCMK__XA_ST_DELAY, &(op->client_delay)); if (peer && dev) { op->id = crm_element_value_copy(dev, PCMK__XA_ST_REMOTE_OP); } else { op->id = crm_generate_uuid(); } g_hash_table_replace(stonith_remote_op_list, op->id, op); op->state = st_query; op->replies_expected = fencing_active_peers(); op->action = crm_element_value_copy(dev, PCMK__XA_ST_DEVICE_ACTION); /* The node initiating the stonith operation. If an operation is relayed, * this is the last node the operation lands on. When in standalone mode, * origin is the ID of the client that originated the operation. * * Or may be the name of the function that created the operation. */ op->originator = crm_element_value_copy(dev, PCMK__XA_ST_ORIGIN); if (op->originator == NULL) { /* Local or relayed request */ op->originator = pcmk__str_copy(stonith_our_uname); } // Delegate may not be set op->delegate = crm_element_value_copy(dev, PCMK__XA_ST_DELEGATE); op->created = time(NULL); CRM_LOG_ASSERT(client != NULL); op->client_id = pcmk__str_copy(client); /* For a RELAY operation, set fenced on the client. */ operation = crm_element_value(request, PCMK__XA_ST_OP); if (pcmk__str_eq(operation, STONITH_OP_RELAY, pcmk__str_none)) { op->client_name = crm_strdup_printf("%s.%lu", crm_system_name, (unsigned long) getpid()); } else { op->client_name = crm_element_value_copy(request, PCMK__XA_ST_CLIENTNAME); } op->target = crm_element_value_copy(dev, PCMK__XA_ST_TARGET); // @TODO Figure out how to avoid copying XML here op->request = pcmk__xml_copy(NULL, request); rc = pcmk__xe_get_flags(request, PCMK__XA_ST_CALLOPT, &(op->call_options), 0U); if (rc != pcmk_rc_ok) { crm_warn("Couldn't parse options from request %s: %s", op->id, pcmk_rc_str(rc)); } crm_element_value_int(request, PCMK__XA_ST_CALLID, &(op->client_callid)); crm_trace("%s new fencing op %s ('%s' targeting %s for client %s, " "base timeout %ds, %u %s expected)", (peer && dev)? "Recorded" : "Generated", op->id, op->action, op->target, op->client_name, op->base_timeout, op->replies_expected, pcmk__plural_alt(op->replies_expected, "reply", "replies")); if (op->call_options & st_opt_cs_nodeid) { int nodeid; crm_node_t *node; pcmk__scan_min_int(op->target, &nodeid, 0); node = pcmk__search_node_caches(nodeid, NULL, NULL, pcmk__node_search_any |pcmk__node_search_cluster_cib); /* Ensure the conversion only happens once */ stonith__clear_call_options(op->call_options, op->id, st_opt_cs_nodeid); if (node && node->uname) { pcmk__str_update(&(op->target), node->uname); } else { crm_warn("Could not expand nodeid '%s' into a host name", op->target); } } /* check to see if this is a duplicate operation of another in-flight operation */ merge_duplicates(op); if (op->state != st_duplicate) { /* kick history readers */ fenced_send_notification(PCMK__VALUE_ST_NOTIFY_HISTORY, NULL, NULL); } /* safe to trim as long as that doesn't touch pending ops */ stonith_fence_history_trim(); return op; } /*! * \internal * \brief Create a peer fencing operation from a request, and initiate it * * \param[in] client IPC client that made request (NULL to get from request) * \param[in] request Request XML * \param[in] manual_ack Whether this is a manual action confirmation * * \return Newly created operation on success, otherwise NULL */ remote_fencing_op_t * initiate_remote_stonith_op(const pcmk__client_t *client, xmlNode *request, gboolean manual_ack) { int query_timeout = 0; xmlNode *query = NULL; const char *client_id = NULL; remote_fencing_op_t *op = NULL; const char *relay_op_id = NULL; const char *operation = NULL; if (client) { client_id = client->id; } else { client_id = crm_element_value(request, PCMK__XA_ST_CLIENTID); } CRM_LOG_ASSERT(client_id != NULL); op = create_remote_stonith_op(client_id, request, FALSE); op->owner = TRUE; if (manual_ack) { return op; } CRM_CHECK(op->action, return NULL); if (advance_topology_level(op, true) != pcmk_rc_ok) { op->state = st_failed; } switch (op->state) { case st_failed: // advance_topology_level() exhausted levels pcmk__set_result(&op->result, CRM_EX_ERROR, PCMK_EXEC_ERROR, "All topology levels failed"); crm_warn("Could not request peer fencing (%s) targeting %s " CRM_XS " id=%.8s", op->action, op->target, op->id); finalize_op(op, NULL, false); return op; case st_duplicate: crm_info("Requesting peer fencing (%s) targeting %s (duplicate) " CRM_XS " id=%.8s", op->action, op->target, op->id); return op; default: crm_notice("Requesting peer fencing (%s) targeting %s " CRM_XS " id=%.8s state=%s base_timeout=%ds", op->action, op->target, op->id, stonith_op_state_str(op->state), op->base_timeout); } query = stonith_create_op(op->client_callid, op->id, STONITH_OP_QUERY, NULL, op->call_options); crm_xml_add(query, PCMK__XA_ST_REMOTE_OP, op->id); crm_xml_add(query, PCMK__XA_ST_TARGET, op->target); crm_xml_add(query, PCMK__XA_ST_DEVICE_ACTION, op_requested_action(op)); crm_xml_add(query, PCMK__XA_ST_ORIGIN, op->originator); crm_xml_add(query, PCMK__XA_ST_CLIENTID, op->client_id); crm_xml_add(query, PCMK__XA_ST_CLIENTNAME, op->client_name); crm_xml_add_int(query, PCMK__XA_ST_TIMEOUT, op->base_timeout); /* In case of RELAY operation, RELAY information is added to the query to delete the original operation of RELAY. */ operation = crm_element_value(request, PCMK__XA_ST_OP); if (pcmk__str_eq(operation, STONITH_OP_RELAY, pcmk__str_none)) { relay_op_id = crm_element_value(request, PCMK__XA_ST_REMOTE_OP); if (relay_op_id) { crm_xml_add(query, PCMK__XA_ST_REMOTE_OP_RELAY, relay_op_id); } } pcmk__cluster_send_message(NULL, crm_msg_stonith_ng, query); free_xml(query); query_timeout = op->base_timeout * TIMEOUT_MULTIPLY_FACTOR; op->query_timer = g_timeout_add((1000 * query_timeout), remote_op_query_timeout, op); return op; } enum find_best_peer_options { /*! Skip checking the target peer for capable fencing devices */ FIND_PEER_SKIP_TARGET = 0x0001, /*! Only check the target peer for capable fencing devices */ FIND_PEER_TARGET_ONLY = 0x0002, /*! Skip peers and devices that are not verified */ FIND_PEER_VERIFIED_ONLY = 0x0004, }; static bool is_watchdog_fencing(const remote_fencing_op_t *op, const char *device) { return (stonith_watchdog_timeout_ms > 0 // Only an explicit mismatch is considered not a watchdog fencing. && pcmk__str_eq(device, STONITH_WATCHDOG_ID, pcmk__str_null_matches) && pcmk__is_fencing_action(op->action) && node_does_watchdog_fencing(op->target)); } static peer_device_info_t * find_best_peer(const char *device, remote_fencing_op_t * op, enum find_best_peer_options options) { GList *iter = NULL; gboolean verified_devices_only = (options & FIND_PEER_VERIFIED_ONLY) ? TRUE : FALSE; if (!device && pcmk_is_set(op->call_options, st_opt_topology)) { return NULL; } for (iter = op->query_results; iter != NULL; iter = iter->next) { peer_device_info_t *peer = iter->data; crm_trace("Testing result from %s targeting %s with %d device%s: %d %x", peer->host, op->target, peer->ndevices, pcmk__plural_s(peer->ndevices), peer->tried, options); if ((options & FIND_PEER_SKIP_TARGET) && pcmk__str_eq(peer->host, op->target, pcmk__str_casei)) { continue; } if ((options & FIND_PEER_TARGET_ONLY) && !pcmk__str_eq(peer->host, op->target, pcmk__str_casei)) { continue; } if (pcmk_is_set(op->call_options, st_opt_topology)) { if (grab_peer_device(op, peer, device, verified_devices_only)) { return peer; } } else if (!peer->tried && count_peer_devices(op, peer, verified_devices_only, fenced_support_flag(op->action))) { /* No topology: Use the current best peer */ crm_trace("Simple fencing"); return peer; } } return NULL; } static peer_device_info_t * stonith_choose_peer(remote_fencing_op_t * op) { const char *device = NULL; peer_device_info_t *peer = NULL; uint32_t active = fencing_active_peers(); do { if (op->devices) { device = op->devices->data; crm_trace("Checking for someone to fence (%s) %s using %s", op->action, op->target, device); } else { crm_trace("Checking for someone to fence (%s) %s", op->action, op->target); } /* Best choice is a peer other than the target with verified access */ peer = find_best_peer(device, op, FIND_PEER_SKIP_TARGET|FIND_PEER_VERIFIED_ONLY); if (peer) { crm_trace("Found verified peer %s for %s", peer->host, device?device:""); return peer; } if(op->query_timer != 0 && op->replies < QB_MIN(op->replies_expected, active)) { crm_trace("Waiting before looking for unverified devices to fence %s", op->target); return NULL; } /* If no other peer has verified access, next best is unverified access */ peer = find_best_peer(device, op, FIND_PEER_SKIP_TARGET); if (peer) { crm_trace("Found best unverified peer %s", peer->host); return peer; } /* If no other peer can do it, last option is self-fencing * (which is never allowed for the "on" phase of a remapped reboot) */ if (op->phase != st_phase_on) { peer = find_best_peer(device, op, FIND_PEER_TARGET_ONLY); if (peer) { crm_trace("%s will fence itself", peer->host); return peer; } } /* Try the next fencing level if there is one (unless we're in the "on" * phase of a remapped "reboot", because we ignore errors in that case) */ } while ((op->phase != st_phase_on) && pcmk_is_set(op->call_options, st_opt_topology) && (advance_topology_level(op, false) == pcmk_rc_ok)); /* With a simple watchdog fencing configuration without a topology, * "device" is NULL here. Consider it should be done with watchdog fencing. */ if (is_watchdog_fencing(op, device)) { crm_info("Couldn't contact watchdog-fencing target-node (%s)", op->target); /* check_watchdog_fencing_and_wait will log additional info */ } else { crm_notice("Couldn't find anyone to fence (%s) %s using %s", op->action, op->target, (device? device : "any device")); } return NULL; } static int valid_fencing_timeout(int specified_timeout, bool action_specific, const remote_fencing_op_t *op, const char *device) { int timeout = specified_timeout; if (!is_watchdog_fencing(op, device)) { return timeout; } timeout = (int) QB_MIN(QB_MAX(specified_timeout, stonith_watchdog_timeout_ms / 1000), INT_MAX); if (timeout > specified_timeout) { if (action_specific) { crm_warn("pcmk_%s_timeout %ds for %s is too short (must be >= " PCMK_OPT_STONITH_WATCHDOG_TIMEOUT " %ds), using %ds " "instead", op->action, specified_timeout, device? device : "watchdog", timeout, timeout); } else { crm_warn("Fencing timeout %ds is too short (must be >= " PCMK_OPT_STONITH_WATCHDOG_TIMEOUT " %ds), using %ds " "instead", specified_timeout, timeout, timeout); } } return timeout; } static int get_device_timeout(const remote_fencing_op_t *op, const peer_device_info_t *peer, const char *device, bool with_delay) { - int timeout = op->base_timeout; + int timeout = valid_fencing_timeout(op->base_timeout, false, op, device); device_properties_t *props; - timeout = valid_fencing_timeout(op->base_timeout, false, op, device); - if (!peer || !device) { return timeout; } props = g_hash_table_lookup(peer->devices, device); if (!props) { return timeout; } if (props->custom_action_timeout[op->phase]) { timeout = valid_fencing_timeout(props->custom_action_timeout[op->phase], true, op, device); } // op->client_delay < 0 means disable any static/random fencing delays if (with_delay && (op->client_delay >= 0)) { // delay_base is eventually limited by delay_max timeout += (props->delay_max[op->phase] > 0 ? props->delay_max[op->phase] : props->delay_base[op->phase]); } return timeout; } struct timeout_data { const remote_fencing_op_t *op; const peer_device_info_t *peer; int total_timeout; }; /*! * \internal * \brief Add timeout to a total if device has not been executed yet * * \param[in] key GHashTable key (device ID) * \param[in] value GHashTable value (device properties) * \param[in,out] user_data Timeout data */ static void add_device_timeout(gpointer key, gpointer value, gpointer user_data) { const char *device_id = key; device_properties_t *props = value; struct timeout_data *timeout = user_data; if (!props->executed[timeout->op->phase] && !props->disallowed[timeout->op->phase]) { timeout->total_timeout += get_device_timeout(timeout->op, timeout->peer, device_id, true); } } static int get_peer_timeout(const remote_fencing_op_t *op, const peer_device_info_t *peer) { struct timeout_data timeout; timeout.op = op; timeout.peer = peer; timeout.total_timeout = 0; g_hash_table_foreach(peer->devices, add_device_timeout, &timeout); return (timeout.total_timeout? timeout.total_timeout : op->base_timeout); } static int get_op_total_timeout(const remote_fencing_op_t *op, const peer_device_info_t *chosen_peer) { long long total_timeout = 0; stonith_topology_t *tp = find_topology_for_host(op->target); if (pcmk_is_set(op->call_options, st_opt_topology) && tp) { int i; GList *device_list = NULL; GList *iter = NULL; GList *auto_list = NULL; if (pcmk__str_eq(op->action, PCMK_ACTION_ON, pcmk__str_none) && (op->automatic_list != NULL)) { auto_list = g_list_copy(op->automatic_list); } /* Yep, this looks scary, nested loops all over the place. * Here is what is going on. * Loop1: Iterate through fencing levels. * Loop2: If a fencing level has devices, loop through each device * Loop3: For each device in a fencing level, see what peer owns it * and what that peer has reported the timeout is for the device. */ for (i = 0; i < ST__LEVEL_COUNT; i++) { if (!tp->levels[i]) { continue; } for (device_list = tp->levels[i]; device_list; device_list = device_list->next) { bool found = false; for (iter = op->query_results; iter != NULL; iter = iter->next) { const peer_device_info_t *peer = iter->data; if (auto_list) { GList *match = g_list_find_custom(auto_list, device_list->data, sort_strings); if (match) { auto_list = g_list_remove(auto_list, match->data); } } if (find_peer_device(op, peer, device_list->data, fenced_support_flag(op->action))) { total_timeout += get_device_timeout(op, peer, device_list->data, true); found = true; break; } } /* End Loop3: match device with peer that owns device, find device's timeout period */ /* in case of watchdog-device we add the timeout to the budget if didn't get a reply */ if (!found && is_watchdog_fencing(op, device_list->data)) { total_timeout += stonith_watchdog_timeout_ms / 1000; } } /* End Loop2: iterate through devices at a specific level */ } /*End Loop1: iterate through fencing levels */ //Add only exists automatic_list device timeout if (auto_list) { for (iter = auto_list; iter != NULL; iter = iter->next) { GList *iter2 = NULL; - for (iter2 = op->query_results; iter2 != NULL; iter = iter2->next) { + for (iter2 = op->query_results; iter2 != NULL; iter2 = iter2->next) { peer_device_info_t *peer = iter2->data; if (find_peer_device(op, peer, iter->data, st_device_supports_on)) { total_timeout += get_device_timeout(op, peer, iter->data, true); break; } } } } g_list_free(auto_list); } else if (chosen_peer) { total_timeout = get_peer_timeout(op, chosen_peer); } else { total_timeout = valid_fencing_timeout(op->base_timeout, false, op, NULL); } if (total_timeout <= 0) { total_timeout = op->base_timeout; } /* Take any requested fencing delay into account to prevent it from eating * up the total timeout. */ if (op->client_delay > 0) { total_timeout += op->client_delay; } return (int) QB_MIN(total_timeout, INT_MAX); } static void report_timeout_period(remote_fencing_op_t * op, int op_timeout) { GList *iter = NULL; xmlNode *update = NULL; const char *client_node = NULL; const char *client_id = NULL; const char *call_id = NULL; if (op->call_options & st_opt_sync_call) { /* There is no reason to report the timeout for a synchronous call. It * is impossible to use the reported timeout to do anything when the client * is blocking for the response. This update is only important for * async calls that require a callback to report the results in. */ return; } else if (!op->request) { return; } crm_trace("Reporting timeout for %s (id=%.8s)", op->client_name, op->id); client_node = crm_element_value(op->request, PCMK__XA_ST_CLIENTNODE); call_id = crm_element_value(op->request, PCMK__XA_ST_CALLID); client_id = crm_element_value(op->request, PCMK__XA_ST_CLIENTID); if (!client_node || !call_id || !client_id) { return; } if (pcmk__str_eq(client_node, stonith_our_uname, pcmk__str_casei)) { // Client is connected to this node, so send update directly to them do_stonith_async_timeout_update(client_id, call_id, op_timeout); return; } /* The client is connected to another node, relay this update to them */ update = stonith_create_op(op->client_callid, op->id, STONITH_OP_TIMEOUT_UPDATE, NULL, 0); crm_xml_add(update, PCMK__XA_ST_REMOTE_OP, op->id); crm_xml_add(update, PCMK__XA_ST_CLIENTID, client_id); crm_xml_add(update, PCMK__XA_ST_CALLID, call_id); crm_xml_add_int(update, PCMK__XA_ST_TIMEOUT, op_timeout); pcmk__cluster_send_message(pcmk__get_node(0, client_node, NULL, pcmk__node_search_cluster_member), crm_msg_stonith_ng, update); free_xml(update); for (iter = op->duplicates; iter != NULL; iter = iter->next) { remote_fencing_op_t *dup = iter->data; crm_trace("Reporting timeout for duplicate %.8s to client %s", dup->id, dup->client_name); report_timeout_period(iter->data, op_timeout); } } /*! * \internal * \brief Advance an operation to the next device in its topology * * \param[in,out] op Fencer operation to advance * \param[in] device ID of device that just completed * \param[in,out] msg If not NULL, XML reply of last delegated operation */ static void advance_topology_device_in_level(remote_fencing_op_t *op, const char *device, xmlNode *msg) { /* Advance to the next device at this topology level, if any */ if (op->devices) { op->devices = op->devices->next; } /* Handle automatic unfencing if an "on" action was requested */ if ((op->phase == st_phase_requested) && pcmk__str_eq(op->action, PCMK_ACTION_ON, pcmk__str_none)) { /* If the device we just executed was required, it's not anymore */ remove_required_device(op, device); /* If there are no more devices at this topology level, run through any * remaining devices with automatic unfencing */ if (op->devices == NULL) { op->devices = op->automatic_list; } } if ((op->devices == NULL) && (op->phase == st_phase_off)) { /* We're done with this level and with required devices, but we had * remapped "reboot" to "off", so start over with "on". If any devices * need to be turned back on, op->devices will be non-NULL after this. */ op_phase_on(op); } // This function is only called if the previous device succeeded pcmk__set_result(&op->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); if (op->devices) { /* Necessary devices remain, so execute the next one */ crm_trace("Next targeting %s on behalf of %s@%s", op->target, op->client_name, op->originator); // The requested delay has been applied for the first device if (op->client_delay > 0) { op->client_delay = 0; } request_peer_fencing(op, NULL); } else { /* We're done with all devices and phases, so finalize operation */ crm_trace("Marking complex fencing op targeting %s as complete", op->target); op->state = st_done; finalize_op(op, msg, false); } } static gboolean check_watchdog_fencing_and_wait(remote_fencing_op_t * op) { if (node_does_watchdog_fencing(op->target)) { guint timeout_ms = QB_MIN(stonith_watchdog_timeout_ms, UINT_MAX); crm_notice("Waiting %s for %s to self-fence (%s) for " "client %s " CRM_XS " id=%.8s", pcmk__readable_interval(timeout_ms), op->target, op->action, op->client_name, op->id); if (op->op_timer_one) { g_source_remove(op->op_timer_one); } op->op_timer_one = g_timeout_add(timeout_ms, remote_op_watchdog_done, op); return TRUE; } else { crm_debug("Skipping fallback to watchdog-fencing as %s is " "not in host-list", op->target); } return FALSE; } /*! * \internal * \brief Ask a peer to execute a fencing operation * * \param[in,out] op Fencing operation to be executed * \param[in,out] peer If NULL or topology is in use, choose best peer to * execute the fencing, otherwise use this peer */ static void request_peer_fencing(remote_fencing_op_t *op, peer_device_info_t *peer) { const char *device = NULL; int timeout; CRM_CHECK(op != NULL, return); crm_trace("Action %.8s targeting %s for %s is %s", op->id, op->target, op->client_name, stonith_op_state_str(op->state)); if ((op->phase == st_phase_on) && (op->devices != NULL)) { /* We are in the "on" phase of a remapped topology reboot. If this * device has pcmk_reboot_action="off", or doesn't support the "on" * action, skip it. * * We can't check device properties at this point because we haven't * chosen a peer for this stage yet. Instead, we check the local node's * knowledge about the device. If different versions of the fence agent * are installed on different nodes, there's a chance this could be * mistaken, but the worst that could happen is we don't try turning the * node back on when we should. */ device = op->devices->data; if (pcmk__str_eq(fenced_device_reboot_action(device), PCMK_ACTION_OFF, pcmk__str_none)) { crm_info("Not turning %s back on using %s because the device is " "configured to stay off (pcmk_reboot_action='off')", op->target, device); advance_topology_device_in_level(op, device, NULL); return; } if (!fenced_device_supports_on(device)) { crm_info("Not turning %s back on using %s because the agent " "doesn't support 'on'", op->target, device); advance_topology_device_in_level(op, device, NULL); return; } } timeout = op->base_timeout; if ((peer == NULL) && !pcmk_is_set(op->call_options, st_opt_topology)) { peer = stonith_choose_peer(op); } if (!op->op_timer_total) { op->total_timeout = TIMEOUT_MULTIPLY_FACTOR * get_op_total_timeout(op, peer); op->op_timer_total = g_timeout_add(1000 * op->total_timeout, remote_op_timeout, op); report_timeout_period(op, op->total_timeout); crm_info("Total timeout set to %ds for peer's fencing targeting %s for %s " CRM_XS " id=%.8s", op->total_timeout, op->target, op->client_name, op->id); } if (pcmk_is_set(op->call_options, st_opt_topology) && op->devices) { /* Ignore the caller's peer preference if topology is in use, because * that peer might not have access to the required device. With * topology, stonith_choose_peer() removes the device from further * consideration, so the timeout must be calculated beforehand. * * @TODO Basing the total timeout on the caller's preferred peer (above) * is less than ideal. */ peer = stonith_choose_peer(op); device = op->devices->data; /* Fencing timeout sent to peer takes no delay into account. * The peer will add a dedicated timer for any delay upon * schedule_stonith_command(). */ timeout = get_device_timeout(op, peer, device, false); } if (peer) { int timeout_one = 0; xmlNode *remote_op = stonith_create_op(op->client_callid, op->id, STONITH_OP_FENCE, NULL, 0); const crm_node_t *peer_node = pcmk__get_node(0, peer->host, NULL, pcmk__node_search_cluster_member); if (op->client_delay > 0) { /* Take requested fencing delay into account to prevent it from * eating up the timeout. */ timeout_one = TIMEOUT_MULTIPLY_FACTOR * op->client_delay; } crm_xml_add(remote_op, PCMK__XA_ST_REMOTE_OP, op->id); crm_xml_add(remote_op, PCMK__XA_ST_TARGET, op->target); crm_xml_add(remote_op, PCMK__XA_ST_DEVICE_ACTION, op->action); crm_xml_add(remote_op, PCMK__XA_ST_ORIGIN, op->originator); crm_xml_add(remote_op, PCMK__XA_ST_CLIENTID, op->client_id); crm_xml_add(remote_op, PCMK__XA_ST_CLIENTNAME, op->client_name); crm_xml_add_int(remote_op, PCMK__XA_ST_TIMEOUT, timeout); crm_xml_add_int(remote_op, PCMK__XA_ST_CALLOPT, op->call_options); crm_xml_add_int(remote_op, PCMK__XA_ST_DELAY, op->client_delay); if (device) { timeout_one += TIMEOUT_MULTIPLY_FACTOR * get_device_timeout(op, peer, device, true); crm_notice("Requesting that %s perform '%s' action targeting %s " "using %s " CRM_XS " for client %s (%ds)", peer->host, op->action, op->target, device, op->client_name, timeout_one); crm_xml_add(remote_op, PCMK__XA_ST_DEVICE_ID, device); } else { timeout_one += TIMEOUT_MULTIPLY_FACTOR * get_peer_timeout(op, peer); crm_notice("Requesting that %s perform '%s' action targeting %s " CRM_XS " for client %s (%ds, %s)", peer->host, op->action, op->target, op->client_name, timeout_one, pcmk__readable_interval(stonith_watchdog_timeout_ms)); } op->state = st_exec; if (op->op_timer_one) { g_source_remove(op->op_timer_one); op->op_timer_one = 0; } if (!is_watchdog_fencing(op, device) || !check_watchdog_fencing_and_wait(op)) { /* Some thoughts about self-fencing cases reaching this point: - Actually check in check_watchdog_fencing_and_wait shouldn't fail if STONITH_WATCHDOG_ID is chosen as fencing-device and it being present implies watchdog-fencing is enabled anyway - If watchdog-fencing is disabled either in general or for a specific target - detected in check_watchdog_fencing_and_wait - for some other kind of self-fencing we can't expect a success answer but timeout is fine if the node doesn't come back in between - Delicate might be the case where we have watchdog-fencing enabled for a node but the watchdog-fencing-device isn't explicitly chosen for self-fencing. Local scheduler execution in sbd might detect the node as unclean and lead to timely self-fencing. Otherwise the selection of PCMK_OPT_STONITH_WATCHDOG_TIMEOUT at least is questionable. */ /* coming here we're not waiting for watchdog timeout - thus engage timer with timout evaluated before */ op->op_timer_one = g_timeout_add((1000 * timeout_one), remote_op_timeout_one, op); } pcmk__cluster_send_message(peer_node, crm_msg_stonith_ng, remote_op); peer->tried = TRUE; free_xml(remote_op); return; } else if (op->phase == st_phase_on) { /* A remapped "on" cannot be executed, but the node was already * turned off successfully, so ignore the error and continue. */ crm_warn("Ignoring %s 'on' failure (no capable peers) targeting %s " "after successful 'off'", device, op->target); advance_topology_device_in_level(op, device, NULL); return; } else if (op->owner == FALSE) { crm_err("Fencing (%s) targeting %s for client %s is not ours to control", op->action, op->target, op->client_name); } else if (op->query_timer == 0) { /* We've exhausted all available peers */ crm_info("No remaining peers capable of fencing (%s) %s for client %s " CRM_XS " state=%s", op->action, op->target, op->client_name, stonith_op_state_str(op->state)); CRM_CHECK(op->state < st_done, return); finalize_timed_out_op(op, "All nodes failed, or are unable, to " "fence target"); } else if(op->replies >= op->replies_expected || op->replies >= fencing_active_peers()) { /* if the operation never left the query state, * but we have all the expected replies, then no devices * are available to execute the fencing operation. */ if (is_watchdog_fencing(op, device) && check_watchdog_fencing_and_wait(op)) { /* Consider a watchdog fencing targeting an offline node executing * once it starts waiting for the target to self-fence. So that when * the query timer pops, remote_op_query_timeout() considers the * fencing already in progress. */ op->state = st_exec; return; } if (op->state == st_query) { crm_info("No peers (out of %d) have devices capable of fencing " "(%s) %s for client %s " CRM_XS " state=%s", op->replies, op->action, op->target, op->client_name, stonith_op_state_str(op->state)); pcmk__reset_result(&op->result); pcmk__set_result(&op->result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, NULL); } else { if (pcmk_is_set(op->call_options, st_opt_topology)) { pcmk__reset_result(&op->result); pcmk__set_result(&op->result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, NULL); } /* ... else use existing result from previous failed attempt * (topology is not in use, and no devices remain to be attempted). * Overwriting the result with PCMK_EXEC_NO_FENCE_DEVICE would * prevent finalize_op() from setting the correct delegate if * needed. */ crm_info("No peers (out of %d) are capable of fencing (%s) %s " "for client %s " CRM_XS " state=%s", op->replies, op->action, op->target, op->client_name, stonith_op_state_str(op->state)); } op->state = st_failed; finalize_op(op, NULL, false); } else { crm_info("Waiting for additional peers capable of fencing (%s) %s%s%s " "for client %s " CRM_XS " id=%.8s", op->action, op->target, (device? " using " : ""), (device? device : ""), op->client_name, op->id); } } /*! * \internal * \brief Comparison function for sorting query results * * \param[in] a GList item to compare * \param[in] b GList item to compare * * \return Per the glib documentation, "a negative integer if the first value * comes before the second, 0 if they are equal, or a positive integer * if the first value comes after the second." */ static gint sort_peers(gconstpointer a, gconstpointer b) { const peer_device_info_t *peer_a = a; const peer_device_info_t *peer_b = b; return (peer_b->ndevices - peer_a->ndevices); } /*! * \internal * \brief Determine if all the devices in the topology are found or not * * \param[in] op Fencing operation with topology to check */ static gboolean all_topology_devices_found(const remote_fencing_op_t *op) { GList *device = NULL; GList *iter = NULL; device_properties_t *match = NULL; stonith_topology_t *tp = NULL; gboolean skip_target = FALSE; int i; tp = find_topology_for_host(op->target); if (!tp) { return FALSE; } if (pcmk__is_fencing_action(op->action)) { /* Don't count the devices on the target node if we are killing * the target node. */ skip_target = TRUE; } for (i = 0; i < ST__LEVEL_COUNT; i++) { for (device = tp->levels[i]; device; device = device->next) { match = NULL; for (iter = op->query_results; iter && !match; iter = iter->next) { peer_device_info_t *peer = iter->data; if (skip_target && pcmk__str_eq(peer->host, op->target, pcmk__str_casei)) { continue; } match = find_peer_device(op, peer, device->data, st_device_supports_none); } if (!match) { return FALSE; } } } return TRUE; } /*! * \internal * \brief Parse action-specific device properties from XML * * \param[in] xml XML element containing the properties * \param[in] peer Name of peer that sent XML (for logs) * \param[in] device Device ID (for logs) * \param[in] action Action the properties relate to (for logs) * \param[in,out] op Fencing operation that properties are being parsed for * \param[in] phase Phase the properties relate to * \param[in,out] props Device properties to update */ static void parse_action_specific(const xmlNode *xml, const char *peer, const char *device, const char *action, remote_fencing_op_t *op, enum st_remap_phase phase, device_properties_t *props) { props->custom_action_timeout[phase] = 0; crm_element_value_int(xml, PCMK__XA_ST_ACTION_TIMEOUT, &props->custom_action_timeout[phase]); if (props->custom_action_timeout[phase]) { crm_trace("Peer %s with device %s returned %s action timeout %ds", peer, device, action, props->custom_action_timeout[phase]); } props->delay_max[phase] = 0; crm_element_value_int(xml, PCMK__XA_ST_DELAY_MAX, &props->delay_max[phase]); if (props->delay_max[phase]) { crm_trace("Peer %s with device %s returned maximum of random delay %ds for %s", peer, device, props->delay_max[phase], action); } props->delay_base[phase] = 0; crm_element_value_int(xml, PCMK__XA_ST_DELAY_BASE, &props->delay_base[phase]); if (props->delay_base[phase]) { crm_trace("Peer %s with device %s returned base delay %ds for %s", peer, device, props->delay_base[phase], action); } /* Handle devices with automatic unfencing */ if (pcmk__str_eq(action, PCMK_ACTION_ON, pcmk__str_none)) { int required = 0; crm_element_value_int(xml, PCMK__XA_ST_REQUIRED, &required); if (required) { crm_trace("Peer %s requires device %s to execute for action %s", peer, device, action); add_required_device(op, device); } } /* If a reboot is remapped to off+on, it's possible that a node is allowed * to perform one action but not another. */ if (pcmk__xe_attr_is_true(xml, PCMK__XA_ST_ACTION_DISALLOWED)) { props->disallowed[phase] = TRUE; crm_trace("Peer %s is disallowed from executing %s for device %s", peer, action, device); } } /*! * \internal * \brief Parse one device's properties from peer's XML query reply * * \param[in] xml XML node containing device properties * \param[in,out] op Operation that query and reply relate to * \param[in,out] peer Peer's device information * \param[in] device ID of device being parsed */ static void add_device_properties(const xmlNode *xml, remote_fencing_op_t *op, peer_device_info_t *peer, const char *device) { xmlNode *child; int verified = 0; device_properties_t *props = pcmk__assert_alloc(1, sizeof(device_properties_t)); int rc = pcmk_rc_ok; /* Add a new entry to this peer's devices list */ g_hash_table_insert(peer->devices, pcmk__str_copy(device), props); /* Peers with verified (monitored) access will be preferred */ crm_element_value_int(xml, PCMK__XA_ST_MONITOR_VERIFIED, &verified); if (verified) { crm_trace("Peer %s has confirmed a verified device %s", peer->host, device); props->verified = TRUE; } // Nodes <2.1.5 won't set this, so assume unfencing in that case rc = pcmk__xe_get_flags(xml, PCMK__XA_ST_DEVICE_SUPPORT_FLAGS, &(props->device_support_flags), st_device_supports_on); if (rc != pcmk_rc_ok) { crm_warn("Couldn't determine device support for %s " "(assuming unfencing): %s", device, pcmk_rc_str(rc)); } /* Parse action-specific device properties */ parse_action_specific(xml, peer->host, device, op_requested_action(op), op, st_phase_requested, props); for (child = pcmk__xe_first_child(xml, NULL, NULL, NULL); child != NULL; child = pcmk__xe_next(child)) { /* Replies for "reboot" operations will include the action-specific * values for "off" and "on" in child elements, just in case the reboot * winds up getting remapped. */ if (pcmk__str_eq(pcmk__xe_id(child), PCMK_ACTION_OFF, pcmk__str_none)) { parse_action_specific(child, peer->host, device, PCMK_ACTION_OFF, op, st_phase_off, props); } else if (pcmk__str_eq(pcmk__xe_id(child), PCMK_ACTION_ON, pcmk__str_none)) { parse_action_specific(child, peer->host, device, PCMK_ACTION_ON, op, st_phase_on, props); } } } /*! * \internal * \brief Parse a peer's XML query reply and add it to operation's results * * \param[in,out] op Operation that query and reply relate to * \param[in] host Name of peer that sent this reply * \param[in] ndevices Number of devices expected in reply * \param[in] xml XML node containing device list * * \return Newly allocated result structure with parsed reply */ static peer_device_info_t * add_result(remote_fencing_op_t *op, const char *host, int ndevices, const xmlNode *xml) { peer_device_info_t *peer = pcmk__assert_alloc(1, sizeof(peer_device_info_t)); xmlNode *child; peer->host = pcmk__str_copy(host); peer->devices = pcmk__strkey_table(free, free); /* Each child element describes one capable device available to the peer */ for (child = pcmk__xe_first_child(xml, NULL, NULL, NULL); child != NULL; child = pcmk__xe_next(child)) { const char *device = pcmk__xe_id(child); if (device) { add_device_properties(child, op, peer, device); } } peer->ndevices = g_hash_table_size(peer->devices); CRM_CHECK(ndevices == peer->ndevices, crm_err("Query claimed to have %d device%s but %d found", ndevices, pcmk__plural_s(ndevices), peer->ndevices)); op->query_results = g_list_insert_sorted(op->query_results, peer, sort_peers); return peer; } /*! * \internal * \brief Handle a peer's reply to our fencing query * * Parse a query result from XML and store it in the remote operation * table, and when enough replies have been received, issue a fencing request. * * \param[in] msg XML reply received * * \return pcmk_ok on success, -errno on error * * \note See initiate_remote_stonith_op() for how the XML query was initially * formed, and stonith_query() for how the peer formed its XML reply. */ int process_remote_stonith_query(xmlNode *msg) { int ndevices = 0; gboolean host_is_target = FALSE; gboolean have_all_replies = FALSE; const char *id = NULL; const char *host = NULL; remote_fencing_op_t *op = NULL; peer_device_info_t *peer = NULL; uint32_t replies_expected; xmlNode *dev = get_xpath_object("//@" PCMK__XA_ST_REMOTE_OP, msg, LOG_ERR); CRM_CHECK(dev != NULL, return -EPROTO); id = crm_element_value(dev, PCMK__XA_ST_REMOTE_OP); CRM_CHECK(id != NULL, return -EPROTO); dev = get_xpath_object("//@" PCMK__XA_ST_AVAILABLE_DEVICES, msg, LOG_ERR); CRM_CHECK(dev != NULL, return -EPROTO); crm_element_value_int(dev, PCMK__XA_ST_AVAILABLE_DEVICES, &ndevices); op = g_hash_table_lookup(stonith_remote_op_list, id); if (op == NULL) { crm_debug("Received query reply for unknown or expired operation %s", id); return -EOPNOTSUPP; } replies_expected = fencing_active_peers(); if (op->replies_expected < replies_expected) { replies_expected = op->replies_expected; } if ((++op->replies >= replies_expected) && (op->state == st_query)) { have_all_replies = TRUE; } host = crm_element_value(msg, PCMK__XA_SRC); host_is_target = pcmk__str_eq(host, op->target, pcmk__str_casei); crm_info("Query result %d of %d from %s for %s/%s (%d device%s) %s", op->replies, replies_expected, host, op->target, op->action, ndevices, pcmk__plural_s(ndevices), id); if (ndevices > 0) { peer = add_result(op, host, ndevices, dev); } pcmk__set_result(&op->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); if (pcmk_is_set(op->call_options, st_opt_topology)) { /* If we start the fencing before all the topology results are in, * it is possible fencing levels will be skipped because of the missing * query results. */ if (op->state == st_query && all_topology_devices_found(op)) { /* All the query results are in for the topology, start the fencing ops. */ crm_trace("All topology devices found"); request_peer_fencing(op, peer); } else if (have_all_replies) { crm_info("All topology query replies have arrived, continuing (%d expected/%d received) ", replies_expected, op->replies); request_peer_fencing(op, NULL); } } else if (op->state == st_query) { int nverified = count_peer_devices(op, peer, TRUE, fenced_support_flag(op->action)); /* We have a result for a non-topology fencing op that looks promising, * go ahead and start fencing before query timeout */ if ((peer != NULL) && !host_is_target && nverified) { /* we have a verified device living on a peer that is not the target */ crm_trace("Found %d verified device%s", nverified, pcmk__plural_s(nverified)); request_peer_fencing(op, peer); } else if (have_all_replies) { crm_info("All query replies have arrived, continuing (%d expected/%d received) ", replies_expected, op->replies); request_peer_fencing(op, NULL); } else { crm_trace("Waiting for more peer results before launching fencing operation"); } } else if ((peer != NULL) && (op->state == st_done)) { crm_info("Discarding query result from %s (%d device%s): " "Operation is %s", peer->host, peer->ndevices, pcmk__plural_s(peer->ndevices), stonith_op_state_str(op->state)); } return pcmk_ok; } /*! * \internal * \brief Handle a peer's reply to a fencing request * * Parse a fencing reply from XML, and either finalize the operation * or attempt another device as appropriate. * * \param[in] msg XML reply received */ void fenced_process_fencing_reply(xmlNode *msg) { const char *id = NULL; const char *device = NULL; remote_fencing_op_t *op = NULL; xmlNode *dev = get_xpath_object("//@" PCMK__XA_ST_REMOTE_OP, msg, LOG_ERR); pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; CRM_CHECK(dev != NULL, return); id = crm_element_value(dev, PCMK__XA_ST_REMOTE_OP); CRM_CHECK(id != NULL, return); dev = stonith__find_xe_with_result(msg); CRM_CHECK(dev != NULL, return); stonith__xe_get_result(dev, &result); device = crm_element_value(dev, PCMK__XA_ST_DEVICE_ID); if (stonith_remote_op_list) { op = g_hash_table_lookup(stonith_remote_op_list, id); } if ((op == NULL) && pcmk__result_ok(&result)) { /* Record successful fencing operations */ const char *client_id = crm_element_value(dev, PCMK__XA_ST_CLIENTID); op = create_remote_stonith_op(client_id, dev, TRUE); } if (op == NULL) { /* Could be for an event that began before we started */ /* TODO: Record the op for later querying */ crm_info("Received peer result of unknown or expired operation %s", id); pcmk__reset_result(&result); return; } pcmk__reset_result(&op->result); op->result = result; // The operation takes ownership of the result if (op->devices && device && !pcmk__str_eq(op->devices->data, device, pcmk__str_casei)) { crm_err("Received outdated reply for device %s (instead of %s) to " "fence (%s) %s. Operation already timed out at peer level.", device, (const char *) op->devices->data, op->action, op->target); return; } if (pcmk__str_eq(crm_element_value(msg, PCMK__XA_SUBT), PCMK__VALUE_BROADCAST, pcmk__str_none)) { if (pcmk__result_ok(&op->result)) { op->state = st_done; } else { op->state = st_failed; } finalize_op(op, msg, false); return; } else if (!pcmk__str_eq(op->originator, stonith_our_uname, pcmk__str_casei)) { /* If this isn't a remote level broadcast, and we are not the * originator of the operation, we should not be receiving this msg. */ crm_err("Received non-broadcast fencing result for operation %.8s " "we do not own (device %s targeting %s)", op->id, device, op->target); return; } if (pcmk_is_set(op->call_options, st_opt_topology)) { const char *device = NULL; const char *reason = op->result.exit_reason; /* We own the op, and it is complete. broadcast the result to all nodes * and notify our local clients. */ if (op->state == st_done) { finalize_op(op, msg, false); return; } device = crm_element_value(msg, PCMK__XA_ST_DEVICE_ID); if ((op->phase == 2) && !pcmk__result_ok(&op->result)) { /* A remapped "on" failed, but the node was already turned off * successfully, so ignore the error and continue. */ crm_warn("Ignoring %s 'on' failure (%s%s%s) targeting %s " "after successful 'off'", device, pcmk_exec_status_str(op->result.execution_status), (reason == NULL)? "" : ": ", (reason == NULL)? "" : reason, op->target); pcmk__set_result(&op->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); } else { crm_notice("Action '%s' targeting %s%s%s on behalf of %s@%s: " "%s%s%s%s", op->action, op->target, ((device == NULL)? "" : " using "), ((device == NULL)? "" : device), op->client_name, op->originator, pcmk_exec_status_str(op->result.execution_status), (reason == NULL)? "" : " (", (reason == NULL)? "" : reason, (reason == NULL)? "" : ")"); } if (pcmk__result_ok(&op->result)) { /* An operation completed successfully. Try another device if * necessary, otherwise mark the operation as done. */ advance_topology_device_in_level(op, device, msg); return; } else { /* This device failed, time to try another topology level. If no other * levels are available, mark this operation as failed and report results. */ if (advance_topology_level(op, false) != pcmk_rc_ok) { op->state = st_failed; finalize_op(op, msg, false); return; } } } else if (pcmk__result_ok(&op->result) && (op->devices == NULL)) { op->state = st_done; finalize_op(op, msg, false); return; } else if ((op->result.execution_status == PCMK_EXEC_TIMEOUT) && (op->devices == NULL)) { /* If the operation timed out don't bother retrying other peers. */ op->state = st_failed; finalize_op(op, msg, false); return; } else { /* fall-through and attempt other fencing action using another peer */ } /* Retry on failure */ crm_trace("Next for %s on behalf of %s@%s (result was: %s)", op->target, op->originator, op->client_name, pcmk_exec_status_str(op->result.execution_status)); request_peer_fencing(op, NULL); } gboolean stonith_check_fence_tolerance(int tolerance, const char *target, const char *action) { GHashTableIter iter; time_t now = time(NULL); remote_fencing_op_t *rop = NULL; if (tolerance <= 0 || !stonith_remote_op_list || target == NULL || action == NULL) { return FALSE; } g_hash_table_iter_init(&iter, stonith_remote_op_list); while (g_hash_table_iter_next(&iter, NULL, (void **)&rop)) { if (strcmp(rop->target, target) != 0) { continue; } else if (rop->state != st_done) { continue; /* We don't have to worry about remapped reboots here * because if state is done, any remapping has been undone */ } else if (strcmp(rop->action, action) != 0) { continue; } else if ((rop->completed + tolerance) < now) { continue; } crm_notice("Target %s was fenced (%s) less than %ds ago by %s on behalf of %s", target, action, tolerance, rop->delegate, rop->originator); return TRUE; } return FALSE; } diff --git a/devel/Makefile.am b/devel/Makefile.am index 15012f6f80..90750bec62 100644 --- a/devel/Makefile.am +++ b/devel/Makefile.am @@ -1,337 +1,337 @@ # -# Copyright 2020-2023 the Pacemaker project contributors +# Copyright 2020-2025 the Pacemaker project contributors # # The version control history for this file may have further details. # # This source code is licensed under the GNU General Public License version 2 # or later (GPLv2+) WITHOUT ANY WARRANTY. # include $(top_srcdir)/mk/common.mk include $(top_srcdir)/mk/release.mk # Coccinelle is a tool that takes special patch-like files (called semantic patches) and # applies them throughout a source tree. This is useful when refactoring, changing APIs, # catching dangerous or incorrect code, and other similar tasks. It's not especially # easy to write a semantic patch but most users should only be concerned about running # the target and inspecting the results. # # Documentation (including examples, which are the most useful): # https://coccinelle.gitlabpages.inria.fr/website/docs/ # # Run the "make cocci" target to just output what would be done, or "make cocci-inplace" # to apply the changes to the source tree. # # COCCI_FILES may be set on the command line, if you want to test just a single file # while it's under development. Otherwise, it is a list of all the files that are ready # to be run. # # ref-passed-variables-inited.cocci seems to be returning some false positives around # GHashTableIters, so it is disabled for the moment. COCCI_FILES ?= coccinelle/string-any-of.cocci \ coccinelle/string-empty.cocci \ coccinelle/string-null-matches.cocci \ coccinelle/use-func.cocci dist_noinst_SCRIPTS = coccinelle/test/testrunner.sh EXTRA_DIST = README \ gdbhelpers \ $(COCCI_FILES) \ coccinelle/ref-passed-variables-inited.cocci \ coccinelle/rename-fn.cocci \ coccinelle/test/ref-passed-variables-inited.input.c \ coccinelle/test/ref-passed-variables-inited.output # Any file in this list is allowed to use any of the pcmk__ internal functions. # Coccinelle can use any transformation that depends on "internal" to rewrite # code to use the internal functions. MAY_USE_INTERNAL_FILES = $(shell find .. -path "../lib/*.c" -o -path "../lib/*private.h" -o -path "../tools/*.c" -o -path "../daemons/*.c" -o -path '../include/pcmki/*h' -o -name '*internal.h') # And then any file in this list is public API, which may not use internal # functions. Thus, only those transformations that do not depend on "internal" # may be applied. OTHER_FILES = $(shell find ../include -name '*h' -a \! -name '*internal.h' -a \! -path '../include/pcmki/*') .PHONY: cocci cocci: -for cf in $(COCCI_FILES); do \ for f in $(MAY_USE_INTERNAL_FILES); do \ spatch $(_SPATCH_FLAGS) -D internal --very-quiet --local-includes --preprocess --sp-file $$cf $$f; \ done ; \ for f in $(OTHER_FILES); do \ spatch $(_SPATCH_FLAGS) --very-quiet --local-includes --preprocess --sp-file $$cf $$f; \ done ; \ done .PHONY: cocci-inplace cocci-inplace: $(MAKE) $(AM_MAKEFLAGS) _SPATCH_FLAGS=--in-place cocci .PHONY: cocci-test cocci-test: for f in coccinelle/test/*.c; do \ coccinelle/test/testrunner.sh $$f; \ done # # Static analysis # ## clang # See scan-build(1) for possible checkers (leave empty to use default set) CLANG_checkers ?= .PHONY: clang clang: OUT=$$(cd $(top_builddir) \ && scan-build $(CLANG_checkers:%=-enable-checker %) \ $(MAKE) $(AM_MAKEFLAGS) CFLAGS="-std=c99 $(CFLAGS)" \ clean all 2>&1); \ REPORT=$$(echo "$$OUT" \ | sed -n -e "s/.*'scan-view \(.*\)'.*/\1/p"); \ [ -z "$$REPORT" ] && echo "$$OUT" || scan-view "$$REPORT" ## coverity # Aggressiveness (low, medium, or high) COVLEVEL ?= low # Generated outputs COVERITY_DIR = $(abs_top_builddir)/coverity-$(TAG) COVTAR = $(abs_top_builddir)/$(PACKAGE)-coverity-$(TAG).tgz COVEMACS = $(abs_top_builddir)/$(TAG).coverity COVHTML = $(COVERITY_DIR)/output/errors # Coverity outputs are phony so they get rebuilt every invocation .PHONY: $(COVERITY_DIR) $(COVERITY_DIR): coverity-clean $(MAKE) $(AM_MAKEFLAGS) -C $(top_builddir) init core-clean $(AM_V_GEN)cd $(top_builddir) \ && cov-build --dir "$@" $(MAKE) $(AM_MAKEFLAGS) core # Public coverity instance .PHONY: $(COVTAR) $(COVTAR): $(COVERITY_DIR) $(AM_V_GEN)tar czf "$@" --transform="s@.*$(TAG)@cov-int@" "$<" .PHONY: coverity coverity: $(COVTAR) @echo "Now go to https://scan.coverity.com/users/sign_in and upload:" @echo " $(COVTAR)" @echo "then make clean at the top level" # Licensed coverity instance # # The prerequisites are a little hacky; rather than actually required, some # of them are designed so that things execute in the proper order (which is # not the same as GNU make's order-only prerequisites). .PHONY: coverity-analyze coverity-analyze: $(COVERITY_DIR) @echo "" @echo "Analyzing (waiting for coverity license if necessary) ..." cd $(top_builddir) && cov-analyze --dir "$<" --wait-for-license \ --security --aggressiveness-level "$(COVLEVEL)" .PHONY: $(COVEMACS) $(COVEMACS): coverity-analyze $(AM_V_GEN)cd $(top_builddir) \ && cov-format-errors --dir "$(COVERITY_DIR)" --emacs-style > "$@" .PHONY: $(COVHTML) $(COVHTML): $(COVEMACS) $(AM_V_GEN)cd $(top_builddir) \ && cov-format-errors --dir "$(COVERITY_DIR)" --html-output "$@" .PHONY: coverity-corp coverity-corp: $(COVHTML) $(MAKE) $(AM_MAKEFLAGS) -C $(top_builddir) core-clean @echo "Done. See:" @echo " file://$(COVHTML)/index.html" @echo "When no longer needed, make coverity-clean" # Remove all outputs regardless of tag .PHONY: coverity-clean coverity-clean: -rm -rf "$(abs_builddir)"/coverity-* \ "$(abs_builddir)"/$(PACKAGE)-coverity-*.tgz \ "$(abs_builddir)"/*.coverity ## cppcheck GLIB_CFLAGS ?= $(pkg-config --cflags glib-2.0) GLIB_INCL_DEF_CFLAGS = $(shell echo $(GLIB_CFLAGS) \ | tr ' ' '\n' | grep '^-[IDU]' | paste -d ' ') # Use CPPCHECK_ARGS to pass extra cppcheck options, e.g.: # --enable={warning,style,performance,portability,information,all} # --inconclusive --std=posix # -DBUILD_PUBLIC_LIBPACEMAKER -DDEFAULT_CONCURRENT_FENCING_TRUE CPPCHECK_ARGS ?= -CPPCHECK_DIRS = replace lib daemons tools +CPPCHECK_DIRS = lib daemons tools CPPCHECK_OUT = $(abs_top_builddir)/cppcheck.out .PHONY: cppcheck cppcheck: cppcheck $(CPPCHECK_ARGS) -I $(top_srcdir)/include \ --output-file=$(CPPCHECK_OUT) \ --max-configs=30 --inline-suppr -q \ --library=posix --library=gnu --library=gtk \ $(GLIB_INCL_DEF_CFLAGS) -D__GNUC__ \ $(foreach dir,$(CPPCHECK_DIRS),$(top_srcdir)/$(dir)) @echo "Done: See $(CPPCHECK_OUT)" @echo "When no longer needed, make cppcheck-clean" .PHONY: cppcheck-clean cppcheck-clean: -rm -f "$(CPPCHECK_OUT)" # # Coverage/profiling # COVERAGE_DIR = $(top_builddir)/coverage # Check coverage of unit tests .PHONY: coverage coverage: coverage-partial-clean cd $(top_builddir) \ && $(MAKE) $(AM_MAKEFLAGS) \ && lcov --no-external --exclude='*_test.c' -c -i -d . \ -o pacemaker_base.info \ && $(MAKE) $(AM_MAKEFLAGS) check \ && lcov --no-external --exclude='*_test.c' -c -d . \ -o pacemaker_test.info \ && lcov -a pacemaker_base.info -a pacemaker_test.info \ -o pacemaker_total.info \ && lcov --remove pacemaker_total.info -o pacemaker_filtered.info\ "$(abs_top_builddir)/tools/*" \ "$(abs_top_builddir)/daemons/*/*" \ "$(abs_top_builddir)/replace/*" \ "$(abs_top_builddir)/lib/gnu/*" genhtml $(top_builddir)/pacemaker_filtered.info -o $(COVERAGE_DIR) -s -t "Pacemaker code coverage" # Check coverage of CLI regression tests .PHONY: coverage-cts coverage-cts: coverage-partial-clean cd $(top_builddir) \ && $(MAKE) $(AM_MAKEFLAGS) \ && lcov --no-external -c -i -d tools -o pacemaker_base.info \ && cts/cts-cli \ && lcov --no-external -c -d tools -o pacemaker_test.info \ && lcov -a pacemaker_base.info -a pacemaker_test.info \ -o pacemaker_total.info genhtml $(top_builddir)/pacemaker_total.info -o $(COVERAGE_DIR) -s # Remove coverage-related files that aren't needed across runs .PHONY: coverage-partial-clean coverage-partial-clean: -rm -f $(top_builddir)/pacemaker_*.info -rm -rf $(COVERAGE_DIR) -find $(top_builddir) -name "*.gcda" -exec rm -f \{\} \; # This target removes all coverage-related files. It is only to be run when # done with coverage analysis and you are ready to go back to normal development, # starting with re-running ./configure. It is not to be run in between # "make coverage" runs. # # In particular, the *.gcno files are generated when the source is built. # Removing those files will break "make coverage" until the whole source tree # has been built and the *.gcno files generated again. .PHONY: coverage-clean coverage-clean: coverage-partial-clean -find $(top_builddir) -name "*.gcno" -exec rm -f \{\} \; # # indent cannot cope with all our exceptions and needs heavy manual editing # # indent target: Limit indent to these directories INDENT_DIRS ?= . # indent target: Extra options to pass to indent INDENT_OPTS ?= INDENT_IGNORE_PATHS = daemons/controld/controld_fsa.h \ lib/gnu/* INDENT_PACEMAKER_STYLE = --blank-lines-after-declarations \ --blank-lines-after-procedures \ --braces-after-func-def-line \ --braces-on-if-line \ --braces-on-struct-decl-line \ --break-before-boolean-operator \ --case-brace-indentation4 \ --case-indentation4 \ --comment-indentation0 \ --continuation-indentation4 \ --continue-at-parentheses \ --cuddle-do-while \ --cuddle-else \ --declaration-comment-column0 \ --declaration-indentation1 \ --else-endif-column0 \ --honour-newlines \ --indent-label0 \ --indent-level4 \ --line-comments-indentation0 \ --line-length80 \ --no-blank-lines-after-commas \ --no-comment-delimiters-on-blank-lines \ --no-space-after-function-call-names \ --no-space-after-parentheses \ --no-tabs \ --preprocessor-indentation2 \ --procnames-start-lines \ --space-after-cast \ --start-left-side-of-comments \ --swallow-optional-blank-lines \ --tab-size8 .PHONY: indent indent: VERSION_CONTROL=none \ find $(INDENT_DIRS) -type f -name "*.[ch]" \ $(INDENT_IGNORE_PATHS:%= ! -path '%') \ -exec indent $(INDENT_PACEMAKER_STYLE) $(INDENT_OPTS) \{\} \; # # Check whether copyrights have been updated appropriately # (Set COMMIT to desired commit or commit range to check, defaulting to HEAD, # or set it empty to check uncommitted changes) # YEAR = $(shell date +%Y) MODIFIED_FILES = $(shell case "$(COMMIT)" in \ [0-9a-f]*$(rparen) \ git diff-tree --no-commit-id \ --name-only "$(COMMIT)" -r ;; \ *$(rparen) \ cd "$(top_srcdir)"; \ git ls-files --modified ;; \ esac) .PHONY: copyright copyright: @cd "$(top_srcdir)" && for file in $(MODIFIED_FILES); do \ if ! grep 'opyright .*$(YEAR).* Pacemaker' "$$file" \ >/dev/null 2>&1; then \ echo "$$file"; \ fi; \ done # # Scratch file for ad-hoc testing # EXTRA_PROGRAMS = scratch nodist_scratch_SOURCES = scratch.c scratch_LDADD = $(top_builddir)/lib/common/libcrmcommon.la .PHONY: clean-local clean-local: coverage-clean coverity-clean cppcheck-clean -rm -f $(EXTRA_PROGRAMS) diff --git a/lib/cib/cib_remote.c b/lib/cib/cib_remote.c index b9b44298cd..baed17bc0d 100644 --- a/lib/cib/cib_remote.c +++ b/lib/cib/cib_remote.c @@ -1,694 +1,695 @@ /* - * Copyright 2008-2024 the Pacemaker project contributors + * Copyright 2008-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #ifdef HAVE_GNUTLS_GNUTLS_H # include // GnuTLS handshake timeout in seconds #define TLS_HANDSHAKE_TIMEOUT 5 static gnutls_anon_client_credentials_t anon_cred_c; static gboolean remote_gnutls_credentials_init = FALSE; #endif // HAVE_GNUTLS_GNUTLS_H #include typedef struct cib_remote_opaque_s { int port; char *server; char *user; char *passwd; gboolean encrypted; pcmk__remote_t command; pcmk__remote_t callback; pcmk__output_t *out; time_t start_time; int timeout_sec; } cib_remote_opaque_t; static int cib_remote_perform_op(cib_t *cib, const char *op, const char *host, const char *section, xmlNode *data, xmlNode **output_data, int call_options, const char *user_name) { int rc; int remaining_time = 0; time_t start_time; xmlNode *op_msg = NULL; xmlNode *op_reply = NULL; cib_remote_opaque_t *private = cib->variant_opaque; if (cib->state == cib_disconnected) { return -ENOTCONN; } if (output_data != NULL) { *output_data = NULL; } if (op == NULL) { crm_err("No operation specified"); return -EINVAL; } rc = cib__create_op(cib, op, host, section, data, call_options, user_name, NULL, &op_msg); if (rc != pcmk_ok) { return rc; } if (pcmk_is_set(call_options, cib_transaction)) { rc = cib__extend_transaction(cib, op_msg); free_xml(op_msg); return rc; } crm_trace("Sending %s message to the CIB manager", op); if (!(call_options & cib_sync_call)) { pcmk__remote_send_xml(&private->callback, op_msg); } else { pcmk__remote_send_xml(&private->command, op_msg); } free_xml(op_msg); if ((call_options & cib_discard_reply)) { crm_trace("Discarding reply"); return pcmk_ok; } else if (!(call_options & cib_sync_call)) { return cib->call_id; } crm_trace("Waiting for a synchronous reply"); start_time = time(NULL); remaining_time = cib->call_timeout ? cib->call_timeout : 60; rc = pcmk_rc_ok; while (remaining_time > 0 && (rc != ENOTCONN)) { int reply_id = -1; int msg_id = cib->call_id; rc = pcmk__read_remote_message(&private->command, remaining_time * 1000); op_reply = pcmk__remote_message_xml(&private->command); if (!op_reply) { break; } crm_element_value_int(op_reply, PCMK__XA_CIB_CALLID, &reply_id); if (reply_id == msg_id) { break; } else if (reply_id < msg_id) { crm_debug("Received old reply: %d (wanted %d)", reply_id, msg_id); crm_log_xml_trace(op_reply, "Old reply"); } else if ((reply_id - 10000) > msg_id) { /* wrap-around case */ crm_debug("Received old reply: %d (wanted %d)", reply_id, msg_id); crm_log_xml_trace(op_reply, "Old reply"); } else { crm_err("Received a __future__ reply:" " %d (wanted %d)", reply_id, msg_id); } free_xml(op_reply); op_reply = NULL; /* wasn't the right reply, try and read some more */ remaining_time = time(NULL) - start_time; } if (rc == ENOTCONN) { crm_err("Disconnected while waiting for reply."); return -ENOTCONN; } else if (op_reply == NULL) { crm_err("No reply message - empty"); return -ENOMSG; } crm_trace("Synchronous reply received"); /* Start processing the reply... */ if (crm_element_value_int(op_reply, PCMK__XA_CIB_RC, &rc) != 0) { rc = -EPROTO; } if (rc == -pcmk_err_diff_resync) { /* This is an internal value that clients do not and should not care about */ rc = pcmk_ok; } if (rc == pcmk_ok || rc == -EPERM) { crm_log_xml_debug(op_reply, "passed"); } else { crm_err("Call failed: %s", pcmk_strerror(rc)); crm_log_xml_warn(op_reply, "failed"); } if (output_data == NULL) { /* do nothing more */ } else if (!(call_options & cib_discard_reply)) { xmlNode *wrapper = pcmk__xe_first_child(op_reply, PCMK__XE_CIB_CALLDATA, NULL, NULL); xmlNode *tmp = pcmk__xe_first_child(wrapper, NULL, NULL, NULL); if (tmp == NULL) { crm_trace("No output in reply to \"%s\" command %d", op, cib->call_id - 1); } else { *output_data = pcmk__xml_copy(NULL, tmp); } } free_xml(op_reply); return rc; } static int cib_remote_callback_dispatch(gpointer user_data) { int rc; cib_t *cib = user_data; cib_remote_opaque_t *private = cib->variant_opaque; xmlNode *msg = NULL; const char *type = NULL; /* If start time is 0, we've previously handled a complete message and this * connection is being reused for a new message. Reset the start_time, * giving this new message timeout_sec from now to complete. */ if (private->start_time == 0) { private->start_time = time(NULL); } rc = pcmk__read_available_remote_data(&private->callback); switch (rc) { case pcmk_rc_ok: /* We have the whole message so process it */ break; case EAGAIN: /* Have we timed out? */ if (time(NULL) >= private->start_time + private->timeout_sec) { crm_info("Error reading from CIB manager connection: %s", pcmk_rc_str(ETIME)); return -1; } /* We haven't read the whole message yet */ return 0; default: /* Error */ crm_info("Error reading from CIB manager connection: %s", pcmk_rc_str(rc)); return -1; } + // coverity[tainted_data] This can't easily be changed right now msg = pcmk__remote_message_xml(&private->callback); if (msg == NULL) { private->start_time = 0; return 0; } type = crm_element_value(msg, PCMK__XA_T); crm_trace("Activating %s callbacks...", type); if (pcmk__str_eq(type, PCMK__VALUE_CIB, pcmk__str_none)) { cib_native_callback(cib, msg, 0, 0); } else if (pcmk__str_eq(type, PCMK__VALUE_CIB_NOTIFY, pcmk__str_none)) { g_list_foreach(cib->notify_list, cib_native_notify, msg); } else { crm_err("Unknown message type: %s", type); } free_xml(msg); private->start_time = 0; return 0; } static int cib_remote_command_dispatch(gpointer user_data) { int rc; cib_t *cib = user_data; cib_remote_opaque_t *private = cib->variant_opaque; /* See cib_remote_callback_dispatch */ if (private->start_time == 0) { private->start_time = time(NULL); } rc = pcmk__read_available_remote_data(&private->command); if (rc == EAGAIN) { /* Have we timed out? */ if (time(NULL) >= private->start_time + private->timeout_sec) { crm_info("Error reading from CIB manager connection: %s", pcmk_rc_str(ETIME)); return -1; } /* We haven't read the whole message yet */ return 0; } free(private->command.buffer); private->command.buffer = NULL; crm_err("received late reply for remote cib connection, discarding"); if (rc != pcmk_rc_ok) { crm_info("Error reading from CIB manager connection: %s", pcmk_rc_str(rc)); return -1; } private->start_time = 0; return 0; } static int cib_tls_close(cib_t *cib) { cib_remote_opaque_t *private = cib->variant_opaque; #ifdef HAVE_GNUTLS_GNUTLS_H if (private->encrypted) { if (private->command.tls_session) { gnutls_bye(*(private->command.tls_session), GNUTLS_SHUT_RDWR); gnutls_deinit(*(private->command.tls_session)); gnutls_free(private->command.tls_session); } if (private->callback.tls_session) { gnutls_bye(*(private->callback.tls_session), GNUTLS_SHUT_RDWR); gnutls_deinit(*(private->callback.tls_session)); gnutls_free(private->callback.tls_session); } private->command.tls_session = NULL; private->callback.tls_session = NULL; if (remote_gnutls_credentials_init) { gnutls_anon_free_client_credentials(anon_cred_c); gnutls_global_deinit(); remote_gnutls_credentials_init = FALSE; } } #endif if (private->command.tcp_socket) { shutdown(private->command.tcp_socket, SHUT_RDWR); /* no more receptions */ close(private->command.tcp_socket); } if (private->callback.tcp_socket) { shutdown(private->callback.tcp_socket, SHUT_RDWR); /* no more receptions */ close(private->callback.tcp_socket); } private->command.tcp_socket = 0; private->callback.tcp_socket = 0; free(private->command.buffer); free(private->callback.buffer); private->command.buffer = NULL; private->callback.buffer = NULL; return 0; } static void cib_remote_connection_destroy(gpointer user_data) { crm_err("Connection destroyed"); #ifdef HAVE_GNUTLS_GNUTLS_H cib_tls_close(user_data); #endif } static int cib_tls_signon(cib_t *cib, pcmk__remote_t *connection, gboolean event_channel) { cib_remote_opaque_t *private = cib->variant_opaque; int rc; xmlNode *answer = NULL; xmlNode *login = NULL; static struct mainloop_fd_callbacks cib_fd_callbacks = { 0, }; cib_fd_callbacks.dispatch = event_channel ? cib_remote_callback_dispatch : cib_remote_command_dispatch; cib_fd_callbacks.destroy = cib_remote_connection_destroy; connection->tcp_socket = -1; #ifdef HAVE_GNUTLS_GNUTLS_H connection->tls_session = NULL; #endif rc = pcmk__connect_remote(private->server, private->port, 0, NULL, &(connection->tcp_socket), NULL, NULL); if (rc != pcmk_rc_ok) { crm_info("Remote connection to %s:%d failed: %s " CRM_XS " rc=%d", private->server, private->port, pcmk_rc_str(rc), rc); return -ENOTCONN; } if (private->encrypted) { int tls_rc = GNUTLS_E_SUCCESS; /* initialize GnuTls lib */ #ifdef HAVE_GNUTLS_GNUTLS_H if (remote_gnutls_credentials_init == FALSE) { crm_gnutls_global_init(); gnutls_anon_allocate_client_credentials(&anon_cred_c); remote_gnutls_credentials_init = TRUE; } /* bind the socket to GnuTls lib */ connection->tls_session = pcmk__new_tls_session(connection->tcp_socket, GNUTLS_CLIENT, GNUTLS_CRD_ANON, anon_cred_c); if (connection->tls_session == NULL) { cib_tls_close(cib); return -1; } rc = pcmk__tls_client_handshake(connection, TLS_HANDSHAKE_TIMEOUT, &tls_rc); if (rc != pcmk_rc_ok) { crm_err("Remote CIB session creation for %s:%d failed: %s", private->server, private->port, (rc == EPROTO)? gnutls_strerror(tls_rc) : pcmk_rc_str(rc)); gnutls_deinit(*connection->tls_session); gnutls_free(connection->tls_session); connection->tls_session = NULL; cib_tls_close(cib); return -1; } #else return -EPROTONOSUPPORT; #endif } /* login to server */ login = pcmk__xe_create(NULL, PCMK__XE_CIB_COMMAND); crm_xml_add(login, PCMK_XA_OP, "authenticate"); crm_xml_add(login, PCMK_XA_USER, private->user); crm_xml_add(login, PCMK__XA_PASSWORD, private->passwd); crm_xml_add(login, PCMK__XA_HIDDEN, PCMK__VALUE_PASSWORD); pcmk__remote_send_xml(connection, login); free_xml(login); rc = pcmk_ok; if (pcmk__read_remote_message(connection, -1) == ENOTCONN) { rc = -ENOTCONN; } answer = pcmk__remote_message_xml(connection); crm_log_xml_trace(answer, "Reply"); if (answer == NULL) { rc = -EPROTO; } else { /* grab the token */ const char *msg_type = crm_element_value(answer, PCMK__XA_CIB_OP); const char *tmp_ticket = crm_element_value(answer, PCMK__XA_CIB_CLIENTID); if (!pcmk__str_eq(msg_type, CRM_OP_REGISTER, pcmk__str_casei)) { crm_err("Invalid registration message: %s", msg_type); rc = -EPROTO; } else if (tmp_ticket == NULL) { rc = -EPROTO; } else { connection->token = strdup(tmp_ticket); } } free_xml(answer); answer = NULL; if (rc != 0) { cib_tls_close(cib); return rc; } crm_trace("remote client connection established"); private->timeout_sec = 60; connection->source = mainloop_add_fd("cib-remote", G_PRIORITY_HIGH, connection->tcp_socket, cib, &cib_fd_callbacks); return rc; } static int cib_remote_signon(cib_t *cib, const char *name, enum cib_conn_type type) { int rc = pcmk_ok; cib_remote_opaque_t *private = cib->variant_opaque; xmlNode *hello = NULL; if (private->passwd == NULL) { if (private->out == NULL) { /* If no pcmk__output_t is set, just assume that a text prompt * is good enough. */ pcmk__text_prompt("Password", false, &(private->passwd)); } else { private->out->prompt("Password", false, &(private->passwd)); } } if (private->server == NULL || private->user == NULL) { rc = -EINVAL; } if (rc == pcmk_ok) { rc = cib_tls_signon(cib, &(private->command), FALSE); } if (rc == pcmk_ok) { rc = cib_tls_signon(cib, &(private->callback), TRUE); } if (rc == pcmk_ok) { rc = cib__create_op(cib, CRM_OP_REGISTER, NULL, NULL, NULL, cib_none, NULL, name, &hello); } if (rc == pcmk_ok) { rc = pcmk__remote_send_xml(&private->command, hello); rc = pcmk_rc2legacy(rc); free_xml(hello); } if (rc == pcmk_ok) { crm_info("Opened connection to %s:%d for %s", private->server, private->port, name); cib->state = cib_connected_command; cib->type = cib_command; } else { crm_info("Connection to %s:%d for %s failed: %s\n", private->server, private->port, name, pcmk_strerror(rc)); } return rc; } static int cib_remote_signoff(cib_t *cib) { int rc = pcmk_ok; crm_debug("Disconnecting from the CIB manager"); #ifdef HAVE_GNUTLS_GNUTLS_H cib_tls_close(cib); #endif cib->cmds->end_transaction(cib, false, cib_none); cib->state = cib_disconnected; cib->type = cib_no_connection; return rc; } static int cib_remote_free(cib_t *cib) { int rc = pcmk_ok; crm_warn("Freeing CIB"); if (cib->state != cib_disconnected) { rc = cib_remote_signoff(cib); if (rc == pcmk_ok) { cib_remote_opaque_t *private = cib->variant_opaque; free(private->server); free(private->user); free(private->passwd); free(cib->cmds); free(cib->user); free(private); free(cib); } } return rc; } static int cib_remote_inputfd(cib_t * cib) { cib_remote_opaque_t *private = cib->variant_opaque; return private->callback.tcp_socket; } static int cib_remote_register_notification(cib_t * cib, const char *callback, int enabled) { xmlNode *notify_msg = pcmk__xe_create(NULL, PCMK__XE_CIB_COMMAND); cib_remote_opaque_t *private = cib->variant_opaque; crm_xml_add(notify_msg, PCMK__XA_CIB_OP, PCMK__VALUE_CIB_NOTIFY); crm_xml_add(notify_msg, PCMK__XA_CIB_NOTIFY_TYPE, callback); crm_xml_add_int(notify_msg, PCMK__XA_CIB_NOTIFY_ACTIVATE, enabled); pcmk__remote_send_xml(&private->callback, notify_msg); free_xml(notify_msg); return pcmk_ok; } static int cib_remote_set_connection_dnotify(cib_t * cib, void (*dnotify) (gpointer user_data)) { return -EPROTONOSUPPORT; } /*! * \internal * \brief Get the given CIB connection's unique client identifiers * * These can be used to check whether this client requested the action that * triggered a CIB notification. * * \param[in] cib CIB connection * \param[out] async_id If not \p NULL, where to store asynchronous client ID * \param[out] sync_id If not \p NULL, where to store synchronous client ID * * \return Legacy Pacemaker return code (specifically, \p pcmk_ok) * * \note This is the \p cib_remote variant implementation of * \p cib_api_operations_t:client_id(). * \note The client IDs are assigned during CIB sign-on. */ static int cib_remote_client_id(const cib_t *cib, const char **async_id, const char **sync_id) { cib_remote_opaque_t *private = cib->variant_opaque; if (async_id != NULL) { // private->callback is the channel for async requests *async_id = private->callback.token; } if (sync_id != NULL) { // private->command is the channel for sync requests *sync_id = private->command.token; } return pcmk_ok; } cib_t * cib_remote_new(const char *server, const char *user, const char *passwd, int port, gboolean encrypted) { cib_remote_opaque_t *private = NULL; cib_t *cib = cib_new_variant(); if (cib == NULL) { return NULL; } private = calloc(1, sizeof(cib_remote_opaque_t)); if (private == NULL) { free(cib); return NULL; } cib->variant = cib_remote; cib->variant_opaque = private; private->server = pcmk__str_copy(server); private->user = pcmk__str_copy(user); private->passwd = pcmk__str_copy(passwd); private->port = port; private->encrypted = encrypted; /* assign variant specific ops */ cib->delegate_fn = cib_remote_perform_op; cib->cmds->signon = cib_remote_signon; cib->cmds->signoff = cib_remote_signoff; cib->cmds->free = cib_remote_free; cib->cmds->inputfd = cib_remote_inputfd; // Deprecated method cib->cmds->register_notification = cib_remote_register_notification; cib->cmds->set_connection_dnotify = cib_remote_set_connection_dnotify; cib->cmds->client_id = cib_remote_client_id; return cib; } void cib__set_output(cib_t *cib, pcmk__output_t *out) { cib_remote_opaque_t *private; if (cib->variant != cib_remote) { return; } private = cib->variant_opaque; private->out = out; } diff --git a/lib/cluster/Makefile.am b/lib/cluster/Makefile.am index 20b0028754..6d44e99df5 100644 --- a/lib/cluster/Makefile.am +++ b/lib/cluster/Makefile.am @@ -1,34 +1,34 @@ # # Copyright 2004-2024 the Pacemaker project contributors # # The version control history for this file may have further details. # # This source code is licensed under the GNU General Public License version 2 # or later (GPLv2+) WITHOUT ANY WARRANTY. # include $(top_srcdir)/mk/common.mk SUBDIRS = tests noinst_HEADERS = crmcluster_private.h ## libraries lib_LTLIBRARIES = libcrmcluster.la -libcrmcluster_la_LDFLAGS = -version-info 32:1:3 +libcrmcluster_la_LDFLAGS = -version-info 32:2:3 libcrmcluster_la_CFLAGS = $(CFLAGS_HARDENED_LIB) libcrmcluster_la_LDFLAGS += $(LDFLAGS_HARDENED_LIB) libcrmcluster_la_LIBADD = $(top_builddir)/lib/fencing/libstonithd.la libcrmcluster_la_LIBADD += $(top_builddir)/lib/common/libcrmcommon.la libcrmcluster_la_LIBADD += $(CLUSTERLIBS) ## Library sources (*must* use += format for bumplibs) libcrmcluster_la_SOURCES = cluster.c libcrmcluster_la_SOURCES += election.c libcrmcluster_la_SOURCES += membership.c if BUILD_CS_SUPPORT libcrmcluster_la_SOURCES += corosync.c libcrmcluster_la_SOURCES += cpg.c endif diff --git a/lib/cluster/election.c b/lib/cluster/election.c index c3e4cd126c..6ce672edc2 100644 --- a/lib/cluster/election.c +++ b/lib/cluster/election.c @@ -1,727 +1,727 @@ /* - * Copyright 2004-2024 the Pacemaker project contributors + * Copyright 2004-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #define STORM_INTERVAL 2 /* in seconds */ struct election_s { enum election_result state; guint count; // How many times local node has voted char *name; // Descriptive name for this election char *uname; // Local node's name GSourceFunc cb; // Function to call if election is won GHashTable *voted; // Key = node name, value = how node voted mainloop_timer_t *timeout; // When to abort if all votes not received int election_wins; // Track wins, for storm detection bool wrote_blackbox; // Write a storm blackbox at most once time_t expires; // When storm detection period ends time_t last_election_loss; // When dampening period ends }; static void election_complete(election_t *e) { e->state = election_won; if (e->cb != NULL) { e->cb(e); } election_reset(e); } static gboolean election_timer_cb(gpointer user_data) { election_t *e = user_data; crm_info("%s timed out, declaring local node as winner", e->name); election_complete(e); return FALSE; } /*! * \brief Get current state of an election * * \param[in] e Election object * * \return Current state of \e */ enum election_result election_state(const election_t *e) { return (e == NULL)? election_error : e->state; } /*! * \brief Create a new election object * * Every node that wishes to participate in an election must create an election * object. Typically, this should be done once, at start-up. A caller should * only create a single election object. * * \param[in] name Label for election (for logging) * \param[in] uname Local node's name * \param[in] period_ms How long to wait for all peers to vote * \param[in] cb Function to call if local node wins election * * \return Newly allocated election object on success, NULL on error * \note The caller is responsible for freeing the returned value using * election_fini(). */ election_t * election_init(const char *name, const char *uname, guint period_ms, GSourceFunc cb) { election_t *e = NULL; static guint count = 0; CRM_CHECK(uname != NULL, return NULL); e = calloc(1, sizeof(election_t)); if (e == NULL) { crm_perror(LOG_CRIT, "Cannot create election"); return NULL; } e->uname = strdup(uname); if (e->uname == NULL) { crm_perror(LOG_CRIT, "Cannot create election"); free(e); return NULL; } e->name = name? crm_strdup_printf("election-%s", name) : crm_strdup_printf("election-%u", count++); e->cb = cb; e->timeout = mainloop_timer_add(e->name, period_ms, FALSE, election_timer_cb, e); crm_trace("Created %s", e->name); return e; } /*! * \brief Disregard any previous vote by specified peer * * This discards any recorded vote from a specified peer. Election users should * call this whenever a voting peer becomes inactive. * * \param[in,out] e Election object * \param[in] uname Name of peer to disregard */ void election_remove(election_t *e, const char *uname) { if ((e != NULL) && (uname != NULL) && (e->voted != NULL)) { crm_trace("Discarding %s (no-)vote from lost peer %s", e->name, uname); g_hash_table_remove(e->voted, uname); } } /*! * \brief Stop election timer and disregard all votes * * \param[in,out] e Election object */ void election_reset(election_t *e) { if (e != NULL) { crm_trace("Resetting election %s", e->name); mainloop_timer_stop(e->timeout); if (e->voted) { crm_trace("Destroying voted cache with %d members", g_hash_table_size(e->voted)); g_hash_table_destroy(e->voted); e->voted = NULL; } } } /*! * \brief Free an election object * * Free all memory associated with an election object, stopping its * election timer (if running). * * \param[in,out] e Election object */ void election_fini(election_t *e) { if (e != NULL) { election_reset(e); crm_trace("Destroying %s", e->name); mainloop_timer_del(e->timeout); free(e->uname); free(e->name); free(e); } } static void election_timeout_start(election_t *e) { if (e != NULL) { mainloop_timer_start(e->timeout); } } /*! * \brief Stop an election's timer, if running * * \param[in,out] e Election object */ void election_timeout_stop(election_t *e) { if (e != NULL) { mainloop_timer_stop(e->timeout); } } /*! * \brief Change an election's timeout (restarting timer if running) * * \param[in,out] e Election object * \param[in] period New timeout */ void election_timeout_set_period(election_t *e, guint period) { if (e != NULL) { mainloop_timer_set_period(e->timeout, period); } else { crm_err("No election defined"); } } static int get_uptime(struct timeval *output) { static time_t expires = 0; static struct rusage info; time_t tm_now = time(NULL); if (expires < tm_now) { int rc = 0; info.ru_utime.tv_sec = 0; info.ru_utime.tv_usec = 0; rc = getrusage(RUSAGE_SELF, &info); output->tv_sec = 0; output->tv_usec = 0; if (rc < 0) { crm_perror(LOG_ERR, "Could not calculate the current uptime"); expires = 0; return -1; } crm_debug("Current CPU usage is: %lds, %ldus", (long)info.ru_utime.tv_sec, (long)info.ru_utime.tv_usec); } expires = tm_now + STORM_INTERVAL; /* N seconds after the last _access_ */ output->tv_sec = info.ru_utime.tv_sec; output->tv_usec = info.ru_utime.tv_usec; return 1; } static int compare_age(struct timeval your_age) { struct timeval our_age; get_uptime(&our_age); /* If an error occurred, our_age will be compared as {0,0} */ if (our_age.tv_sec > your_age.tv_sec) { crm_debug("Win: %ld vs %ld (seconds)", (long)our_age.tv_sec, (long)your_age.tv_sec); return 1; } else if (our_age.tv_sec < your_age.tv_sec) { crm_debug("Lose: %ld vs %ld (seconds)", (long)our_age.tv_sec, (long)your_age.tv_sec); return -1; } else if (our_age.tv_usec > your_age.tv_usec) { crm_debug("Win: %ld.%06ld vs %ld.%06ld (usec)", (long)our_age.tv_sec, (long)our_age.tv_usec, (long)your_age.tv_sec, (long)your_age.tv_usec); return 1; } else if (our_age.tv_usec < your_age.tv_usec) { crm_debug("Lose: %ld.%06ld vs %ld.%06ld (usec)", (long)our_age.tv_sec, (long)our_age.tv_usec, (long)your_age.tv_sec, (long)your_age.tv_usec); return -1; } return 0; } /*! * \brief Start a new election by offering local node's candidacy * * Broadcast a "vote" election message containing the local node's ID, * (incremented) election counter, and uptime, and start the election timer. * * \param[in,out] e Election object * * \note Any nodes agreeing to the candidacy will send a "no-vote" reply, and if * all active peers do so, or if the election times out, the local node * wins the election. (If we lose to any peer vote, we will stop the * timer, so a timeout means we did not lose -- either some peer did not * vote, or we did not call election_check() in time.) */ void election_vote(election_t *e) { struct timeval age; xmlNode *vote = NULL; crm_node_t *our_node; if (e == NULL) { crm_trace("Election vote requested, but no election available"); return; } our_node = pcmk__get_node(0, e->uname, NULL, pcmk__node_search_cluster_member); if (!pcmk__cluster_is_node_active(our_node)) { crm_trace("Cannot vote in %s yet: local node not connected to cluster", e->name); return; } election_reset(e); e->state = election_in_progress; vote = create_request(CRM_OP_VOTE, NULL, NULL, CRM_SYSTEM_CRMD, CRM_SYSTEM_CRMD, NULL); e->count++; crm_xml_add(vote, PCMK__XA_ELECTION_OWNER, pcmk__cluster_get_xml_id(our_node)); crm_xml_add_int(vote, PCMK__XA_ELECTION_ID, e->count); // Warning: PCMK__XA_ELECTION_AGE_NANO_SEC value is actually microseconds get_uptime(&age); crm_xml_add_timeval(vote, PCMK__XA_ELECTION_AGE_SEC, PCMK__XA_ELECTION_AGE_NANO_SEC, &age); pcmk__cluster_send_message(NULL, crm_msg_crmd, vote); free_xml(vote); crm_debug("Started %s round %d", e->name, e->count); election_timeout_start(e); return; } /*! * \brief Check whether local node has won an election * * If all known peers have sent no-vote messages, stop the election timer, set * the election state to won, and call any registered win callback. * * \param[in,out] e Election object * * \return TRUE if local node has won, FALSE otherwise * \note If all known peers have sent no-vote messages, but the election owner * does not call this function, the election will not be won (and the * callback will not be called) until the election times out. * \note This should be called when election_count_vote() returns * \c election_in_progress. */ bool election_check(election_t *e) { int voted_size = 0; int num_members = 0; if (e == NULL) { crm_trace("Election check requested, but no election available"); return FALSE; } if (e->voted == NULL) { crm_trace("%s check requested, but no votes received yet", e->name); return FALSE; } voted_size = g_hash_table_size(e->voted); num_members = pcmk__cluster_num_active_nodes(); /* in the case of #voted > #members, it is better to * wait for the timeout and give the cluster time to * stabilize */ if (voted_size >= num_members) { /* we won and everyone has voted */ election_timeout_stop(e); if (voted_size > num_members) { GHashTableIter gIter; const crm_node_t *node; char *key = NULL; crm_warn("Received too many votes in %s", e->name); g_hash_table_iter_init(&gIter, crm_peer_cache); while (g_hash_table_iter_next(&gIter, NULL, (gpointer *) & node)) { if (pcmk__cluster_is_node_active(node)) { crm_warn("* expected vote: %s", node->uname); } } g_hash_table_iter_init(&gIter, e->voted); while (g_hash_table_iter_next(&gIter, (gpointer *) & key, NULL)) { crm_warn("* actual vote: %s", key); } } crm_info("%s won by local node", e->name); election_complete(e); return TRUE; } else { crm_debug("%s still waiting on %d of %d votes", e->name, num_members - voted_size, num_members); } return FALSE; } #define LOSS_DAMPEN 2 /* in seconds */ struct vote { const char *op; const char *from; const char *version; const char *election_owner; int election_id; struct timeval age; }; /*! * \brief Unpack an election message * * \param[in] e Election object (for logging only) * \param[in] message Election message XML * \param[out] vote Parsed fields from message * * \return TRUE if election message and election are valid, FALSE otherwise * \note The parsed struct's pointer members are valid only for the lifetime of * the message argument. */ static bool parse_election_message(const election_t *e, const xmlNode *message, struct vote *vote) { CRM_CHECK(message && vote, return FALSE); vote->election_id = -1; vote->age.tv_sec = -1; vote->age.tv_usec = -1; vote->op = crm_element_value(message, PCMK__XA_CRM_TASK); vote->from = crm_element_value(message, PCMK__XA_SRC); vote->version = crm_element_value(message, PCMK_XA_VERSION); vote->election_owner = crm_element_value(message, PCMK__XA_ELECTION_OWNER); crm_element_value_int(message, PCMK__XA_ELECTION_ID, &(vote->election_id)); if ((vote->op == NULL) || (vote->from == NULL) || (vote->version == NULL) || (vote->election_owner == NULL) || (vote->election_id < 0)) { crm_warn("Invalid %s message from %s in %s ", (vote->op? vote->op : "election"), (vote->from? vote->from : "unspecified node"), (e? e->name : "election")); return FALSE; } // Op-specific validation if (pcmk__str_eq(vote->op, CRM_OP_VOTE, pcmk__str_none)) { /* Only vote ops have uptime. Warning: PCMK__XA_ELECTION_AGE_NANO_SEC value is in microseconds. */ crm_element_value_timeval(message, PCMK__XA_ELECTION_AGE_SEC, PCMK__XA_ELECTION_AGE_NANO_SEC, &(vote->age)); if ((vote->age.tv_sec < 0) || (vote->age.tv_usec < 0)) { crm_warn("Cannot count %s %s from %s because it is missing uptime", (e? e->name : "election"), vote->op, vote->from); return FALSE; } } else if (!pcmk__str_eq(vote->op, CRM_OP_NOVOTE, pcmk__str_none)) { crm_info("Cannot process %s message from %s because %s is not a known election op", (e? e->name : "election"), vote->from, vote->op); return FALSE; } // Election validation if (e == NULL) { crm_info("Cannot count %s from %s because no election available", vote->op, vote->from); return FALSE; } /* If the membership cache is NULL, we REALLY shouldn't be voting -- * the question is how we managed to get here. */ if (crm_peer_cache == NULL) { crm_info("Cannot count %s %s from %s because no peer information available", e->name, vote->op, vote->from); return FALSE; } return TRUE; } static void record_vote(election_t *e, struct vote *vote) { pcmk__assert(e && vote && vote->from && vote->op); if (e->voted == NULL) { e->voted = pcmk__strkey_table(free, free); } pcmk__insert_dup(e->voted, vote->from, vote->op); } static void send_no_vote(crm_node_t *peer, struct vote *vote) { // @TODO probably shouldn't hardcode CRM_SYSTEM_CRMD and crm_msg_crmd xmlNode *novote = create_request(CRM_OP_NOVOTE, NULL, vote->from, CRM_SYSTEM_CRMD, CRM_SYSTEM_CRMD, NULL); crm_xml_add(novote, PCMK__XA_ELECTION_OWNER, vote->election_owner); crm_xml_add_int(novote, PCMK__XA_ELECTION_ID, vote->election_id); pcmk__cluster_send_message(peer, crm_msg_crmd, novote); free_xml(novote); } /*! * \brief Process an election message (vote or no-vote) from a peer * * \param[in,out] e Election object * \param[in] message Election message XML from peer * \param[in] can_win Whether local node is eligible to win * * \return Election state after new vote is considered * \note If the peer message is a vote, and we prefer the peer to win, this will * send a no-vote reply to the peer. * \note The situations "we lost to this vote" from "this is a late no-vote * after we've already lost" both return election_lost. If a caller needs * to distinguish them, it should save the current state before calling * this function, and then compare the result. */ enum election_result election_count_vote(election_t *e, const xmlNode *message, bool can_win) { int log_level = LOG_INFO; gboolean done = FALSE; gboolean we_lose = FALSE; - const char *reason = "unknown"; + const char *reason = NULL; bool we_are_owner = FALSE; crm_node_t *our_node = NULL, *your_node = NULL; time_t tm_now = time(NULL); struct vote vote; CRM_CHECK(message != NULL, return election_error); if (parse_election_message(e, message, &vote) == FALSE) { return election_error; } your_node = pcmk__get_node(0, vote.from, NULL, pcmk__node_search_cluster_member); our_node = pcmk__get_node(0, e->uname, NULL, pcmk__node_search_cluster_member); we_are_owner = (our_node != NULL) && pcmk__str_eq(pcmk__cluster_get_xml_id(our_node), vote.election_owner, pcmk__str_none); if (!can_win) { reason = "Not eligible"; we_lose = TRUE; } else if (!pcmk__cluster_is_node_active(our_node)) { reason = "We are not part of the cluster"; log_level = LOG_ERR; we_lose = TRUE; } else if (we_are_owner && (vote.election_id != e->count)) { log_level = LOG_TRACE; reason = "Superseded"; done = TRUE; } else if (!pcmk__cluster_is_node_active(your_node)) { /* Possibly we cached the message in the FSA queue at a point that it wasn't */ reason = "Peer is not part of our cluster"; log_level = LOG_WARNING; done = TRUE; } else if (pcmk__str_eq(vote.op, CRM_OP_NOVOTE, pcmk__str_none) || pcmk__str_eq(vote.from, e->uname, pcmk__str_none)) { /* Receiving our own broadcast vote, or a no-vote from peer, is a vote * for us to win */ if (!we_are_owner) { crm_warn("Cannot count %s round %d %s from %s because we are not election owner (%s)", e->name, vote.election_id, vote.op, vote.from, vote.election_owner); return election_error; } if (e->state != election_in_progress) { // Should only happen if we already lost crm_debug("Not counting %s round %d %s from %s because no election in progress", e->name, vote.election_id, vote.op, vote.from); return e->state; } record_vote(e, &vote); reason = "Recorded"; done = TRUE; } else { // A peer vote requires a comparison to determine which node is better int age_result = compare_age(vote.age); int version_result = compare_version(vote.version, CRM_FEATURE_SET); if (version_result < 0) { reason = "Version"; we_lose = TRUE; } else if (version_result > 0) { reason = "Version"; } else if (age_result < 0) { reason = "Uptime"; we_lose = TRUE; } else if (age_result > 0) { reason = "Uptime"; } else if (strcasecmp(e->uname, vote.from) > 0) { reason = "Host name"; we_lose = TRUE; } else { reason = "Host name"; } } if (e->expires < tm_now) { e->election_wins = 0; e->expires = tm_now + STORM_INTERVAL; } else if (done == FALSE && we_lose == FALSE) { int peers = 1 + g_hash_table_size(crm_peer_cache); /* If every node has to vote down every other node, thats N*(N-1) total elections * Allow some leeway before _really_ complaining */ e->election_wins++; if (e->election_wins > (peers * peers)) { crm_warn("%s election storm detected: %d wins in %d seconds", e->name, e->election_wins, STORM_INTERVAL); e->election_wins = 0; e->expires = tm_now + STORM_INTERVAL; if (e->wrote_blackbox == FALSE) { /* It's questionable whether a black box (from every node in the * cluster) would be truly helpful in diagnosing an election * storm. It's also highly doubtful a production environment * would get multiple election storms from distinct causes, so * saving one blackbox per process lifetime should be * sufficient. Alternatives would be to save a timestamp of the * last blackbox write instead of a boolean, and write a new one * if some amount of time has passed; or to save a storm count, * write a blackbox on every Nth occurrence. */ crm_write_blackbox(0, NULL); e->wrote_blackbox = TRUE; } } } if (done) { do_crm_log(log_level + 1, "Processed %s round %d %s (current round %d) from %s (%s)", e->name, vote.election_id, vote.op, e->count, vote.from, reason); return e->state; } else if (we_lose == FALSE) { /* We track the time of the last election loss to implement an election * dampening period, reducing the likelihood of an election storm. If * this node has lost within the dampening period, don't start a new * election, even if we win against a peer's vote -- the peer we lost to * should win again. * * @TODO This has a problem case: if an election winner immediately * leaves the cluster, and a new election is immediately called, all * nodes could lose, with no new winner elected. The ideal solution * would be to tie the election structure with the peer caches, which * would allow us to clear the dampening when the previous winner * leaves (and would allow other improvements as well). */ if ((e->last_election_loss == 0) || ((tm_now - e->last_election_loss) > (time_t) LOSS_DAMPEN)) { do_crm_log(log_level, "%s round %d (owner node ID %s) pass: %s from %s (%s)", e->name, vote.election_id, vote.election_owner, vote.op, vote.from, reason); e->last_election_loss = 0; election_timeout_stop(e); /* Start a new election by voting down this, and other, peers */ e->state = election_start; return e->state; } else { char *loss_time = ctime(&e->last_election_loss); if (loss_time) { // Show only HH:MM:SS loss_time += 11; loss_time[8] = '\0'; } crm_info("Ignoring %s round %d (owner node ID %s) pass vs %s because we lost less than %ds ago at %s", e->name, vote.election_id, vote.election_owner, vote.from, LOSS_DAMPEN, (loss_time? loss_time : "unknown")); } } e->last_election_loss = tm_now; do_crm_log(log_level, "%s round %d (owner node ID %s) lost: %s from %s (%s)", e->name, vote.election_id, vote.election_owner, vote.op, vote.from, reason); election_reset(e); send_no_vote(your_node, &vote); e->state = election_lost; return e->state; } /*! * \brief Reset any election dampening currently in effect * * \param[in,out] e Election object to clear */ void election_clear_dampening(election_t *e) { e->last_election_loss = 0; } diff --git a/lib/common/Makefile.am b/lib/common/Makefile.am index b3be7eab21..030cec56c2 100644 --- a/lib/common/Makefile.am +++ b/lib/common/Makefile.am @@ -1,144 +1,144 @@ # # Copyright 2004-2024 the Pacemaker project contributors # # The version control history for this file may have further details. # # This source code is licensed under the GNU General Public License version 2 # or later (GPLv2+) WITHOUT ANY WARRANTY. # include $(top_srcdir)/mk/common.mk AM_CPPFLAGS += -I$(top_builddir)/lib/gnu \ -I$(top_srcdir)/lib/gnu ## libraries lib_LTLIBRARIES = libcrmcommon.la check_LTLIBRARIES = libcrmcommon_test.la # Disable -Wcast-qual if used, because we do some hacky casting, # and because libxml2 has some signatures that should be const but aren't # for backward compatibility reasons. # s390 needs -fPIC # s390-suse-linux/bin/ld: .libs/ipc.o: relocation R_390_PC32DBL against `__stack_chk_fail@@GLIBC_2.4' can not be used when making a shared object; recompile with -fPIC CFLAGS = $(CFLAGS_COPY:-Wcast-qual=) -fPIC # Without "." here, check-recursive will run through the subdirectories first # and then run "make check" here. This will fail, because there's things in # the subdirectories that need check_LTLIBRARIES built first. Adding "." here # changes the order so the subdirectories are processed afterwards. SUBDIRS = . tests noinst_HEADERS = crmcommon_private.h \ mock_private.h -libcrmcommon_la_LDFLAGS = -version-info 47:1:13 +libcrmcommon_la_LDFLAGS = -version-info 48:0:14 libcrmcommon_la_CFLAGS = $(CFLAGS_HARDENED_LIB) libcrmcommon_la_LDFLAGS += $(LDFLAGS_HARDENED_LIB) libcrmcommon_la_LIBADD = @LIBADD_DL@ \ $(top_builddir)/lib/gnu/libgnu.la # If configured with --with-profiling or --with-coverage, BUILD_PROFILING will # be set and -fno-builtin will be added to the CFLAGS. However, libcrmcommon # uses the fabs() function which is normally supplied by gcc as one of its # builtins. Therefore we need to explicitly link against libm here or the # tests won't link. if BUILD_PROFILING libcrmcommon_la_LIBADD += -lm endif ## Library sources (*must* use += format for bumplibs) libcrmcommon_la_SOURCES = libcrmcommon_la_SOURCES += acl.c libcrmcommon_la_SOURCES += actions.c libcrmcommon_la_SOURCES += agents.c libcrmcommon_la_SOURCES += alerts.c libcrmcommon_la_SOURCES += attrs.c libcrmcommon_la_SOURCES += cib.c if BUILD_CIBSECRETS libcrmcommon_la_SOURCES += cib_secrets.c endif libcrmcommon_la_SOURCES += cmdline.c libcrmcommon_la_SOURCES += digest.c libcrmcommon_la_SOURCES += health.c libcrmcommon_la_SOURCES += io.c libcrmcommon_la_SOURCES += ipc_attrd.c libcrmcommon_la_SOURCES += ipc_client.c libcrmcommon_la_SOURCES += ipc_common.c libcrmcommon_la_SOURCES += ipc_controld.c libcrmcommon_la_SOURCES += ipc_pacemakerd.c libcrmcommon_la_SOURCES += ipc_schedulerd.c libcrmcommon_la_SOURCES += ipc_server.c libcrmcommon_la_SOURCES += iso8601.c libcrmcommon_la_SOURCES += lists.c libcrmcommon_la_SOURCES += logging.c libcrmcommon_la_SOURCES += mainloop.c libcrmcommon_la_SOURCES += messages.c libcrmcommon_la_SOURCES += nodes.c libcrmcommon_la_SOURCES += nvpair.c libcrmcommon_la_SOURCES += options.c libcrmcommon_la_SOURCES += options_display.c libcrmcommon_la_SOURCES += output.c libcrmcommon_la_SOURCES += output_html.c libcrmcommon_la_SOURCES += output_log.c libcrmcommon_la_SOURCES += output_none.c libcrmcommon_la_SOURCES += output_text.c libcrmcommon_la_SOURCES += output_xml.c libcrmcommon_la_SOURCES += patchset.c libcrmcommon_la_SOURCES += patchset_display.c libcrmcommon_la_SOURCES += pid.c libcrmcommon_la_SOURCES += probes.c libcrmcommon_la_SOURCES += procfs.c libcrmcommon_la_SOURCES += remote.c libcrmcommon_la_SOURCES += resources.c libcrmcommon_la_SOURCES += results.c libcrmcommon_la_SOURCES += roles.c libcrmcommon_la_SOURCES += rules.c libcrmcommon_la_SOURCES += scheduler.c libcrmcommon_la_SOURCES += schemas.c libcrmcommon_la_SOURCES += scores.c libcrmcommon_la_SOURCES += strings.c libcrmcommon_la_SOURCES += utils.c libcrmcommon_la_SOURCES += watchdog.c libcrmcommon_la_SOURCES += xml.c libcrmcommon_la_SOURCES += xml_attr.c libcrmcommon_la_SOURCES += xml_comment.c libcrmcommon_la_SOURCES += xml_display.c libcrmcommon_la_SOURCES += xml_io.c libcrmcommon_la_SOURCES += xpath.c # # libcrmcommon_test is used only with unit tests, so we can mock system calls. # See mock.c for details. # include $(top_srcdir)/mk/tap.mk libcrmcommon_test_la_SOURCES = $(libcrmcommon_la_SOURCES) libcrmcommon_test_la_SOURCES += mock.c libcrmcommon_test_la_SOURCES += unittest.c libcrmcommon_test_la_LDFLAGS = $(libcrmcommon_la_LDFLAGS) \ -rpath $(libdir) \ $(LDFLAGS_WRAP) # If GCC emits a builtin function in place of something we've mocked up, that will # get used instead of the mocked version which leads to unexpected test results. So # disable all builtins. Older versions of GCC (at least, on RHEL7) will still emit # replacement code for strdup (and possibly other functions) unless -fno-inline is # also added. libcrmcommon_test_la_CFLAGS = $(libcrmcommon_la_CFLAGS) \ -DPCMK__UNIT_TESTING \ -fno-builtin \ -fno-inline # If -fno-builtin is used, -lm also needs to be added. See the comment at # BUILD_PROFILING above. libcrmcommon_test_la_LIBADD = $(libcrmcommon_la_LIBADD) if BUILD_COVERAGE libcrmcommon_test_la_LIBADD += -lgcov endif libcrmcommon_test_la_LIBADD += -lcmocka libcrmcommon_test_la_LIBADD += -lm nodist_libcrmcommon_test_la_SOURCES = $(nodist_libcrmcommon_la_SOURCES) diff --git a/lib/common/cmdline.c b/lib/common/cmdline.c index bc60f1ce93..288d83a1c4 100644 --- a/lib/common/cmdline.c +++ b/lib/common/cmdline.c @@ -1,379 +1,369 @@ /* - * Copyright 2019-2024 the Pacemaker project contributors + * Copyright 2019-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include static gboolean bump_verbosity(const gchar *option_name, const gchar *optarg, gpointer data, GError **error) { pcmk__common_args_t *common_args = (pcmk__common_args_t *) data; common_args->verbosity++; return TRUE; } pcmk__common_args_t * pcmk__new_common_args(const char *summary) { pcmk__common_args_t *args = NULL; - args = calloc(1, sizeof(pcmk__common_args_t)); - if (args == NULL) { - crm_exit(CRM_EX_OSERR); - } - - args->summary = strdup(summary); - if (args->summary == NULL) { - free(args); - args = NULL; - crm_exit(CRM_EX_OSERR); - } - + args = pcmk__assert_alloc(1, sizeof(pcmk__common_args_t)); + args->summary = pcmk__str_copy(summary); return args; } static void free_common_args(gpointer data) { pcmk__common_args_t *common_args = (pcmk__common_args_t *) data; free(common_args->summary); free(common_args->output_ty); free(common_args->output_dest); if (common_args->output_as_descr != NULL) { free(common_args->output_as_descr); } free(common_args); } GOptionContext * pcmk__build_arg_context(pcmk__common_args_t *common_args, const char *fmts, GOptionGroup **output_group, const char *param_string) { GOptionContext *context; GOptionGroup *main_group; GOptionEntry main_entries[3] = { { "version", '$', 0, G_OPTION_ARG_NONE, &(common_args->version), N_("Display software version and exit"), NULL }, { "verbose", 'V', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, bump_verbosity, N_("Increase debug output (may be specified multiple times)"), NULL }, { NULL } }; main_group = g_option_group_new(NULL, "Application Options:", NULL, common_args, free_common_args); g_option_group_add_entries(main_group, main_entries); context = g_option_context_new(param_string); g_option_context_set_summary(context, common_args->summary); g_option_context_set_description(context, "Report bugs to " PCMK__BUG_URL "\n"); g_option_context_set_main_group(context, main_group); if (fmts != NULL) { GOptionEntry output_entries[3] = { { "output-as", 0, 0, G_OPTION_ARG_STRING, &(common_args->output_ty), NULL, N_("FORMAT") }, { "output-to", 0, 0, G_OPTION_ARG_STRING, &(common_args->output_dest), N_( "Specify file name for output (or \"-\" for stdout)"), N_("DEST") }, { NULL } }; if (*output_group == NULL) { *output_group = g_option_group_new("output", N_("Output Options:"), N_("Show output help"), NULL, NULL); } common_args->output_as_descr = crm_strdup_printf("Specify output format as one of: %s", fmts); output_entries[0].description = common_args->output_as_descr; g_option_group_add_entries(*output_group, output_entries); g_option_context_add_group(context, *output_group); } // main_group is now owned by context, we don't free it here // cppcheck-suppress memleak return context; } void pcmk__free_arg_context(GOptionContext *context) { if (context == NULL) { return; } g_option_context_free(context); } void pcmk__add_main_args(GOptionContext *context, const GOptionEntry entries[]) { GOptionGroup *main_group = g_option_context_get_main_group(context); g_option_group_add_entries(main_group, entries); } void pcmk__add_arg_group(GOptionContext *context, const char *name, const char *header, const char *desc, const GOptionEntry entries[]) { GOptionGroup *group = NULL; group = g_option_group_new(name, header, desc, NULL, NULL); g_option_group_add_entries(group, entries); g_option_context_add_group(context, group); // group is now owned by context, we don't free it here // cppcheck-suppress memleak } static gchar * string_replace(gchar *str, const gchar *sub, const gchar *repl) { /* This function just replaces all occurrences of a substring * with some other string. It doesn't handle cases like overlapping, * so don't get clever with it. * * FIXME: When glib >= 2.68 is supported, we can get rid of this * function and use g_string_replace instead. */ gchar **split = g_strsplit(str, sub, 0); gchar *retval = g_strjoinv(repl, split); g_strfreev(split); return retval; } gchar * pcmk__quote_cmdline(gchar **argv) { GString *gs = NULL; if (argv == NULL || argv[0] == NULL) { return NULL; } gs = g_string_sized_new(100); for (int i = 0; argv[i] != NULL; i++) { if (i > 0) { g_string_append_c(gs, ' '); } if (strchr(argv[i], ' ') == NULL) { /* The arg does not contain a space. */ g_string_append(gs, argv[i]); } else if (strchr(argv[i], '\'') == NULL) { /* The arg contains a space, but not a single quote. */ pcmk__g_strcat(gs, "'", argv[i], "'", NULL); } else { /* The arg contains both a space and a single quote, which needs to * be replaced with an escaped version. We do this instead of counting * on libxml to handle the escaping for various reasons: * * (1) This keeps the string as valid shell. * (2) We don't want to use XML entities in formats besides XML and HTML. * (3) The string we are feeding to libxml is something like: "a b 'c d' e". * It won't escape the single quotes around 'c d' here because there is * no need to escape quotes inside a different form of quote. If we * change the string to "a b 'c'd' e", we haven't changed anything - it's * still single quotes inside double quotes. * * On the other hand, if we replace the single quote with "'", then * we have introduced an ampersand which libxml will escape. This leaves * us with "&apos;" which is not what we want. * * It's simplest to just escape with a backslash. */ gchar *repl = string_replace(argv[i], "'", "\\\'"); pcmk__g_strcat(gs, "'", repl, "'", NULL); g_free(repl); } } return g_string_free(gs, FALSE); } gchar ** pcmk__cmdline_preproc(char *const *argv, const char *special) { GPtrArray *arr = NULL; bool saw_dash_dash = false; bool copy_option = false; if (argv == NULL) { return NULL; } if (g_get_prgname() == NULL && argv && *argv) { gchar *basename = g_path_get_basename(*argv); g_set_prgname(basename); g_free(basename); } arr = g_ptr_array_new(); for (int i = 0; argv[i] != NULL; i++) { /* If this is the first time we saw "--" in the command line, set * a flag so we know to just copy everything after it over. We also * want to copy the "--" over so whatever actually parses the command * line when we're done knows where arguments end. */ if (saw_dash_dash == false && strcmp(argv[i], "--") == 0) { saw_dash_dash = true; } if (saw_dash_dash == true) { g_ptr_array_add(arr, g_strdup(argv[i])); continue; } if (copy_option == true) { g_ptr_array_add(arr, g_strdup(argv[i])); copy_option = false; continue; } /* This is just a dash by itself. That could indicate stdin/stdout, or * it could be user error. Copy it over and let glib figure it out. */ if (pcmk__str_eq(argv[i], "-", pcmk__str_casei)) { g_ptr_array_add(arr, g_strdup(argv[i])); continue; } /* "-INFINITY" is almost certainly meant as a string, not as an option * list */ if (strcmp(argv[i], "-INFINITY") == 0) { g_ptr_array_add(arr, g_strdup(argv[i])); continue; } /* This is a short argument, or perhaps several. Iterate over it * and explode them out into individual arguments. */ if (g_str_has_prefix(argv[i], "-") && !g_str_has_prefix(argv[i], "--")) { /* Skip over leading dash */ const char *ch = argv[i]+1; /* This looks like the start of a number, which means it is a negative * number. It's probably the argument to the preceeding option, but * we can't know that here. Copy it over and let whatever handles * arguments next figure it out. */ if (*ch != '\0' && *ch >= '1' && *ch <= '9') { bool is_numeric = true; while (*ch != '\0') { if (!isdigit(*ch)) { is_numeric = false; break; } ch++; } if (is_numeric) { g_ptr_array_add(arr, g_strdup_printf("%s", argv[i])); continue; } else { /* This argument wasn't entirely numeric. Reset ch to the * beginning so we can process it one character at a time. */ ch = argv[i]+1; } } while (*ch != '\0') { /* This is a special short argument that takes an option. getopt * allows values to be interspersed with a list of arguments, but * glib does not. Grab both the argument and its value and * separate them into a new argument. */ if (special != NULL && strchr(special, *ch) != NULL) { /* The argument does not occur at the end of this string of * arguments. Take everything through the end as its value. */ if (*(ch+1) != '\0') { fprintf(stderr, "Deprecated argument format '-%c%s' used.\n", *ch, ch+1); fprintf(stderr, "Please use '-%c %s' instead. " "Support will be removed in a future release.\n", *ch, ch+1); g_ptr_array_add(arr, g_strdup_printf("-%c", *ch)); g_ptr_array_add(arr, g_strdup(ch+1)); break; /* The argument occurs at the end of this string. Hopefully * whatever comes next in argv is its value. It may not be, * but that is not for us to decide. */ } else { g_ptr_array_add(arr, g_strdup_printf("-%c", *ch)); copy_option = true; ch++; } /* This is a regular short argument. Just copy it over. */ } else { g_ptr_array_add(arr, g_strdup_printf("-%c", *ch)); ch++; } } /* This is a long argument, or an option, or something else. * Copy it over - everything else is copied, so this keeps it easy for * the caller to know what to do with the memory when it's done. */ } else { g_ptr_array_add(arr, g_strdup(argv[i])); } } g_ptr_array_add(arr, NULL); return (char **) g_ptr_array_free(arr, FALSE); } G_GNUC_PRINTF(3, 4) gboolean pcmk__force_args(GOptionContext *context, GError **error, const char *format, ...) { int len = 0; char *buf = NULL; gchar **extra_args = NULL; va_list ap; gboolean retval = TRUE; va_start(ap, format); len = vasprintf(&buf, format, ap); pcmk__assert(len > 0); va_end(ap); if (!g_shell_parse_argv(buf, NULL, &extra_args, error)) { g_strfreev(extra_args); free(buf); return FALSE; } retval = g_option_context_parse_strv(context, &extra_args, error); g_strfreev(extra_args); free(buf); return retval; } diff --git a/lib/common/io.c b/lib/common/io.c index 6951b1fb40..9c1b767a45 100644 --- a/lib/common/io.c +++ b/lib/common/io.c @@ -1,662 +1,669 @@ /* - * Copyright 2004-2024 the Pacemaker project contributors + * Copyright 2004-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #ifndef _GNU_SOURCE # define _GNU_SOURCE #endif #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include /*! * \internal * \brief Create a directory, including any parent directories needed * * \param[in] path_c Pathname of the directory to create * \param[in] mode Permissions to be used (with current umask) when creating * * \return Standard Pacemaker return code */ int pcmk__build_path(const char *path_c, mode_t mode) { int offset = 1, len = 0; int rc = pcmk_rc_ok; char *path = strdup(path_c); // cppcheck seems not to understand the abort logic in CRM_CHECK // cppcheck-suppress memleak CRM_CHECK(path != NULL, return -ENOMEM); for (len = strlen(path); offset < len; offset++) { if (path[offset] == '/') { path[offset] = 0; if ((mkdir(path, mode) < 0) && (errno != EEXIST)) { rc = errno; goto done; } path[offset] = '/'; } } if ((mkdir(path, mode) < 0) && (errno != EEXIST)) { rc = errno; } done: free(path); return rc; } /*! * \internal * \brief Return canonicalized form of a path name * * \param[in] path Pathname to canonicalize * \param[out] resolved_path Where to store canonicalized pathname * * \return Standard Pacemaker return code * \note The caller is responsible for freeing \p resolved_path on success. * \note This function exists because not all C library versions of * realpath(path, resolved_path) support a NULL resolved_path. */ int pcmk__real_path(const char *path, char **resolved_path) { CRM_CHECK((path != NULL) && (resolved_path != NULL), return EINVAL); #if _POSIX_VERSION >= 200809L /* Recent C libraries can dynamically allocate memory as needed */ *resolved_path = realpath(path, NULL); return (*resolved_path == NULL)? errno : pcmk_rc_ok; #elif defined(PATH_MAX) /* Older implementations require pre-allocated memory */ /* (this is less desirable because PATH_MAX may be huge or not defined) */ *resolved_path = malloc(PATH_MAX); if ((*resolved_path == NULL) || (realpath(path, *resolved_path) == NULL)) { return errno; } return pcmk_rc_ok; #else *resolved_path = NULL; return ENOTSUP; #endif } /*! * \internal * \brief Create a file name using a sequence number * * \param[in] directory Directory that contains the file series * \param[in] series Start of file name * \param[in] sequence Sequence number * \param[in] bzip Whether to use ".bz2" instead of ".raw" as extension * * \return Newly allocated file path (asserts on error, so always non-NULL) * \note The caller is responsible for freeing the return value. */ char * pcmk__series_filename(const char *directory, const char *series, unsigned int sequence, bool bzip) { pcmk__assert((directory != NULL) && (series != NULL)); return crm_strdup_printf("%s/%s-%u.%s", directory, series, sequence, (bzip? "bz2" : "raw")); } /*! * \internal * \brief Read sequence number stored in a file series' .last file * * \param[in] directory Directory that contains the file series * \param[in] series Start of file name * \param[out] seq Where to store the sequence number * * \return Standard Pacemaker return code */ int pcmk__read_series_sequence(const char *directory, const char *series, unsigned int *seq) { int rc; FILE *fp = NULL; char *series_file = NULL; if ((directory == NULL) || (series == NULL) || (seq == NULL)) { return EINVAL; } series_file = crm_strdup_printf("%s/%s.last", directory, series); fp = fopen(series_file, "r"); if (fp == NULL) { rc = errno; crm_debug("Could not open series file %s: %s", series_file, strerror(rc)); free(series_file); return rc; } errno = 0; if (fscanf(fp, "%u", seq) != 1) { rc = (errno == 0)? ENODATA : errno; crm_debug("Could not read sequence number from series file %s: %s", series_file, pcmk_rc_str(rc)); fclose(fp); return rc; } fclose(fp); crm_trace("Found last sequence number %u in series file %s", *seq, series_file); free(series_file); return pcmk_rc_ok; } /*! * \internal * \brief Write sequence number to a file series' .last file * * \param[in] directory Directory that contains the file series * \param[in] series Start of file name * \param[in] sequence Sequence number to write * \param[in] max Maximum sequence value, after which it is reset to 0 * * \note This function logs some errors but does not return any to the caller */ void pcmk__write_series_sequence(const char *directory, const char *series, unsigned int sequence, int max) { int rc = 0; FILE *file_strm = NULL; char *series_file = NULL; CRM_CHECK(directory != NULL, return); CRM_CHECK(series != NULL, return); if (max == 0) { return; } if (max > 0 && sequence >= max) { sequence = 0; } series_file = crm_strdup_printf("%s/%s.last", directory, series); file_strm = fopen(series_file, "w"); if (file_strm != NULL) { rc = fprintf(file_strm, "%u", sequence); if (rc < 0) { crm_perror(LOG_ERR, "Cannot write to series file %s", series_file); } } else { crm_err("Cannot open series file %s for writing", series_file); } if (file_strm != NULL) { fflush(file_strm); fclose(file_strm); } crm_trace("Wrote %d to %s", sequence, series_file); free(series_file); } /*! * \internal * \brief Change the owner and group of a file series' .last file * * \param[in] directory Directory that contains series * \param[in] series Series to change * \param[in] uid User ID of desired file owner * \param[in] gid Group ID of desired file group * * \return Standard Pacemaker return code * \note The caller must have the appropriate privileges. */ int pcmk__chown_series_sequence(const char *directory, const char *series, uid_t uid, gid_t gid) { char *series_file = NULL; int rc = pcmk_rc_ok; if ((directory == NULL) || (series == NULL)) { return EINVAL; } series_file = crm_strdup_printf("%s/%s.last", directory, series); if (chown(series_file, uid, gid) < 0) { rc = errno; } free(series_file); return rc; } static bool pcmk__daemon_user_can_write(const char *target_name, struct stat *target_stat) { struct passwd *sys_user = NULL; errno = 0; sys_user = getpwnam(CRM_DAEMON_USER); if (sys_user == NULL) { crm_notice("Could not find user %s: %s", CRM_DAEMON_USER, pcmk_rc_str(errno)); return FALSE; } if (target_stat->st_uid != sys_user->pw_uid) { crm_notice("%s is not owned by user %s " CRM_XS " uid %d != %d", target_name, CRM_DAEMON_USER, sys_user->pw_uid, target_stat->st_uid); return FALSE; } if ((target_stat->st_mode & (S_IRUSR | S_IWUSR)) == 0) { crm_notice("%s is not readable and writable by user %s " CRM_XS " st_mode=0%lo", target_name, CRM_DAEMON_USER, (unsigned long) target_stat->st_mode); return FALSE; } return TRUE; } static bool pcmk__daemon_group_can_write(const char *target_name, struct stat *target_stat) { struct group *sys_grp = NULL; errno = 0; sys_grp = getgrnam(CRM_DAEMON_GROUP); if (sys_grp == NULL) { crm_notice("Could not find group %s: %s", CRM_DAEMON_GROUP, pcmk_rc_str(errno)); return FALSE; } if (target_stat->st_gid != sys_grp->gr_gid) { crm_notice("%s is not owned by group %s " CRM_XS " uid %d != %d", target_name, CRM_DAEMON_GROUP, sys_grp->gr_gid, target_stat->st_gid); return FALSE; } if ((target_stat->st_mode & (S_IRGRP | S_IWGRP)) == 0) { crm_notice("%s is not readable and writable by group %s " CRM_XS " st_mode=0%lo", target_name, CRM_DAEMON_GROUP, (unsigned long) target_stat->st_mode); return FALSE; } return TRUE; } /*! * \internal * \brief Check whether a directory or file is writable by the cluster daemon * * Return true if either the cluster daemon user or cluster daemon group has * write permission on a specified file or directory. * * \param[in] dir Directory to check (this argument must be specified, and * the directory must exist) * \param[in] file File to check (only the directory will be checked if this * argument is not specified or the file does not exist) * * \return true if target is writable by cluster daemon, false otherwise */ bool pcmk__daemon_can_write(const char *dir, const char *file) { int s_res = 0; struct stat buf; char *full_file = NULL; const char *target = NULL; // Caller must supply directory pcmk__assert(dir != NULL); // If file is given, check whether it exists as a regular file if (file != NULL) { full_file = crm_strdup_printf("%s/%s", dir, file); target = full_file; s_res = stat(full_file, &buf); if (s_res < 0) { crm_notice("%s not found: %s", target, pcmk_rc_str(errno)); free(full_file); full_file = NULL; target = NULL; } else if (S_ISREG(buf.st_mode) == FALSE) { crm_err("%s must be a regular file " CRM_XS " st_mode=0%lo", target, (unsigned long) buf.st_mode); free(full_file); return false; } } // If file is not given, ensure dir exists as directory if (target == NULL) { target = dir; s_res = stat(dir, &buf); if (s_res < 0) { crm_err("%s not found: %s", dir, pcmk_rc_str(errno)); return false; } else if (S_ISDIR(buf.st_mode) == FALSE) { crm_err("%s must be a directory " CRM_XS " st_mode=0%lo", dir, (unsigned long) buf.st_mode); return false; } } if (!pcmk__daemon_user_can_write(target, &buf) && !pcmk__daemon_group_can_write(target, &buf)) { crm_err("%s must be owned and writable by either user %s or group %s " CRM_XS " st_mode=0%lo", target, CRM_DAEMON_USER, CRM_DAEMON_GROUP, (unsigned long) buf.st_mode); free(full_file); return false; } free(full_file); return true; } /*! * \internal * \brief Flush and sync a directory to disk * * \param[in] name Directory to flush and sync * \note This function logs errors but does not return them to the caller */ void pcmk__sync_directory(const char *name) { int fd; DIR *directory; directory = opendir(name); if (directory == NULL) { crm_perror(LOG_ERR, "Could not open %s for syncing", name); return; } fd = dirfd(directory); if (fd < 0) { crm_perror(LOG_ERR, "Could not obtain file descriptor for %s", name); return; } if (fsync(fd) < 0) { crm_perror(LOG_ERR, "Could not sync %s", name); } if (closedir(directory) < 0) { crm_perror(LOG_ERR, "Could not close %s after fsync", name); } } /*! * \internal * \brief Read the contents of a file * * \param[in] filename Name of file to read * \param[out] contents Where to store file contents * * \return Standard Pacemaker return code * \note On success, the caller is responsible for freeing contents. */ int pcmk__file_contents(const char *filename, char **contents) { FILE *fp; int length, read_len; int rc = pcmk_rc_ok; if ((filename == NULL) || (contents == NULL)) { return EINVAL; } fp = fopen(filename, "r"); if ((fp == NULL) || (fseek(fp, 0L, SEEK_END) < 0)) { rc = errno; goto bail; } length = ftell(fp); if (length < 0) { rc = errno; goto bail; } if (length == 0) { *contents = NULL; } else { *contents = calloc(length + 1, sizeof(char)); if (*contents == NULL) { rc = errno; goto bail; } + + errno = 0; + rewind(fp); + if (errno != 0) { + rc = errno; + goto bail; + } read_len = fread(*contents, 1, length, fp); if (read_len != length) { free(*contents); *contents = NULL; rc = EIO; } else { /* Coverity thinks *contents isn't null-terminated. It doesn't * understand calloc(). */ (*contents)[length] = '\0'; } } bail: if (fp != NULL) { fclose(fp); } return rc; } /*! * \internal * \brief Write text to a file, flush and sync it to disk, then close the file * * \param[in] fd File descriptor opened for writing * \param[in] contents String to write to file * * \return Standard Pacemaker return code */ int pcmk__write_sync(int fd, const char *contents) { int rc = 0; FILE *fp = fdopen(fd, "w"); if (fp == NULL) { return errno; } if ((contents != NULL) && (fprintf(fp, "%s", contents) < 0)) { rc = EIO; } if (fflush(fp) != 0) { rc = errno; } if (fsync(fileno(fp)) < 0) { rc = errno; } fclose(fp); return rc; } /*! * \internal * \brief Set a file descriptor to non-blocking * * \param[in] fd File descriptor to use * * \return Standard Pacemaker return code */ int pcmk__set_nonblocking(int fd) { int flag = fcntl(fd, F_GETFL); if (flag < 0) { return errno; } if (fcntl(fd, F_SETFL, flag | O_NONBLOCK) < 0) { return errno; } return pcmk_rc_ok; } /*! * \internal * \brief Get directory name for temporary files * * Return the value of the TMPDIR environment variable if it is set to a * full path, otherwise return "/tmp". * * \return Name of directory to be used for temporary files */ const char * pcmk__get_tmpdir(void) { const char *dir = getenv("TMPDIR"); return (dir && (*dir == '/'))? dir : "/tmp"; } /*! * \internal * \brief Close open file descriptors * * Close all file descriptors (except optionally stdin, stdout, and stderr), * which is a best practice for a new child process forked for the purpose of * executing an external program. * * \param[in] bool If true, close stdin, stdout, and stderr as well */ void pcmk__close_fds_in_child(bool all) { DIR *dir; struct rlimit rlim; rlim_t max_fd; int min_fd = (all? 0 : (STDERR_FILENO + 1)); /* Find the current process's (soft) limit for open files. getrlimit() * should always work, but have a fallback just in case. */ if (getrlimit(RLIMIT_NOFILE, &rlim) == 0) { max_fd = rlim.rlim_cur - 1; } else { long conf_max = sysconf(_SC_OPEN_MAX); max_fd = (conf_max > 0)? conf_max : 1024; } /* /proc/self/fd (on Linux) or /dev/fd (on most OSes) contains symlinks to * all open files for the current process, named as the file descriptor. * Use this if available, because it's more efficient than a shotgun * approach to closing descriptors. */ #if HAVE_LINUX_PROCFS dir = opendir("/proc/self/fd"); if (dir == NULL) { dir = opendir("/dev/fd"); } #else dir = opendir("/dev/fd"); #endif // HAVE_LINUX_PROCFS if (dir != NULL) { struct dirent *entry; int dir_fd = dirfd(dir); while ((entry = readdir(dir)) != NULL) { int lpc = atoi(entry->d_name); /* How could one of these entries be higher than max_fd, you ask? * It isn't possible in normal operation, but when run under * valgrind, valgrind can open high-numbered file descriptors for * its own use that are higher than the process's soft limit. * These will show up in the fd directory but aren't closable. */ if ((lpc >= min_fd) && (lpc <= max_fd) && (lpc != dir_fd)) { close(lpc); } } closedir(dir); return; } /* If no fd directory is available, iterate over all possible descriptors. * This is less efficient due to the overhead of many system calls. */ for (int lpc = max_fd; lpc >= min_fd; lpc--) { close(lpc); } } /*! * \brief Duplicate a file path, inserting a prefix if not absolute * * \param[in] filename File path to duplicate * \param[in] dirname If filename is not absolute, prefix to add * * \return Newly allocated memory with full path (guaranteed non-NULL) */ char * pcmk__full_path(const char *filename, const char *dirname) { pcmk__assert(filename != NULL); if (filename[0] == '/') { return pcmk__str_copy(filename); } pcmk__assert(dirname != NULL); return crm_strdup_printf("%s/%s", dirname, filename); } // Deprecated functions kept only for backward API compatibility // LCOV_EXCL_START #include void crm_build_path(const char *path_c, mode_t mode) { int rc = pcmk__build_path(path_c, mode); if (rc != pcmk_rc_ok) { crm_err("Could not create directory '%s': %s", path_c, pcmk_rc_str(rc)); } } // LCOV_EXCL_STOP // End deprecated API diff --git a/lib/common/iso8601.c b/lib/common/iso8601.c index dd2e75de6c..477289b508 100644 --- a/lib/common/iso8601.c +++ b/lib/common/iso8601.c @@ -1,2251 +1,2255 @@ /* - * Copyright 2005-2024 the Pacemaker project contributors + * Copyright 2005-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ /* * References: * https://en.wikipedia.org/wiki/ISO_8601 * http://www.staff.science.uu.nl/~gent0113/calendar/isocalendar.htm */ #include #include #include #include #include #include // INT_MIN, INT_MAX #include #include #include #include #include "crmcommon_private.h" /* * Andrew's code was originally written for OSes whose "struct tm" contains: * long tm_gmtoff; :: Seconds east of UTC * const char *tm_zone; :: Timezone abbreviation * Some OSes lack these, instead having: * time_t (or long) timezone; :: "difference between UTC and local standard time" * char *tzname[2] = { "...", "..." }; * I (David Lee) confess to not understanding the details. So my attempted * generalisations for where their use is necessary may be flawed. * * 1. Does "difference between ..." subtract the same or opposite way? * 2. Should it use "altzone" instead of "timezone"? * 3. Should it use tzname[0] or tzname[1]? Interaction with timezone/altzone? */ #if defined(HAVE_STRUCT_TM_TM_GMTOFF) # define GMTOFF(tm) ((tm)->tm_gmtoff) #else /* Note: extern variable; macro argument not actually used. */ # define GMTOFF(tm) (-timezone+daylight) #endif #define HOUR_SECONDS (60 * 60) #define DAY_SECONDS (HOUR_SECONDS * 24) /*! * \internal * \brief Validate a seconds/microseconds tuple * * The microseconds value must be in the correct range, and if both are nonzero * they must have the same sign. * * \param[in] sec Seconds * \param[in] usec Microseconds * * \return true if the seconds/microseconds tuple is valid, or false otherwise */ #define valid_sec_usec(sec, usec) \ ((QB_ABS(usec) < QB_TIME_US_IN_SEC) \ && (((sec) == 0) || ((usec) == 0) || (((sec) < 0) == ((usec) < 0)))) // A date/time or duration struct crm_time_s { int years; // Calendar year (date/time) or number of years (duration) int months; // Number of months (duration only) int days; // Ordinal day of year (date/time) or number of days (duration) int seconds; // Seconds of day (date/time) or number of seconds (duration) int offset; // Seconds offset from UTC (date/time only) bool duration; // True if duration }; static crm_time_t *parse_date(const char *date_str); static crm_time_t * crm_get_utc_time(const crm_time_t *dt) { crm_time_t *utc = NULL; if (dt == NULL) { errno = EINVAL; return NULL; } utc = crm_time_new_undefined(); utc->years = dt->years; utc->days = dt->days; utc->seconds = dt->seconds; utc->offset = 0; if (dt->offset) { crm_time_add_seconds(utc, -dt->offset); } else { /* Durations (which are the only things that can include months, never have a timezone */ utc->months = dt->months; } crm_time_log(LOG_TRACE, "utc-source", dt, crm_time_log_date | crm_time_log_timeofday | crm_time_log_with_timezone); crm_time_log(LOG_TRACE, "utc-target", utc, crm_time_log_date | crm_time_log_timeofday | crm_time_log_with_timezone); return utc; } crm_time_t * crm_time_new(const char *date_time) { tzset(); if (date_time == NULL) { return pcmk__copy_timet(time(NULL)); } return parse_date(date_time); } /*! * \brief Allocate memory for an uninitialized time object * * \return Newly allocated time object * \note The caller is responsible for freeing the return value using * crm_time_free(). */ crm_time_t * crm_time_new_undefined(void) { return (crm_time_t *) pcmk__assert_alloc(1, sizeof(crm_time_t)); } /*! * \brief Check whether a time object has been initialized yet * * \param[in] t Time object to check * * \return TRUE if time object has been initialized, FALSE otherwise */ bool crm_time_is_defined(const crm_time_t *t) { // Any nonzero member indicates something has been done to t return (t != NULL) && (t->years || t->months || t->days || t->seconds || t->offset || t->duration); } void crm_time_free(crm_time_t * dt) { if (dt == NULL) { return; } free(dt); } static int year_days(int year) { int d = 365; if (crm_time_leapyear(year)) { d++; } return d; } /* From http://myweb.ecu.edu/mccartyr/ISOwdALG.txt : * * 5. Find the Jan1Weekday for Y (Monday=1, Sunday=7) * YY = (Y-1) % 100 * C = (Y-1) - YY * G = YY + YY/4 * Jan1Weekday = 1 + (((((C / 100) % 4) x 5) + G) % 7) */ int crm_time_january1_weekday(int year) { int YY = (year - 1) % 100; int C = (year - 1) - YY; int G = YY + YY / 4; int jan1 = 1 + (((((C / 100) % 4) * 5) + G) % 7); crm_trace("YY=%d, C=%d, G=%d", YY, C, G); crm_trace("January 1 %.4d: %d", year, jan1); return jan1; } int crm_time_weeks_in_year(int year) { int weeks = 52; int jan1 = crm_time_january1_weekday(year); /* if jan1 == thursday */ if (jan1 == 4) { weeks++; } else { jan1 = crm_time_january1_weekday(year + 1); /* if dec31 == thursday aka. jan1 of next year is a friday */ if (jan1 == 5) { weeks++; } } return weeks; } // Jan-Dec plus Feb of leap years static int month_days[13] = { 31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31, 29 }; /*! * \brief Return number of days in given month of given year * * \param[in] month Ordinal month (1-12) * \param[in] year Gregorian year * * \return Number of days in given month (0 if given month or year is invalid) */ int crm_time_days_in_month(int month, int year) { if ((month < 1) || (month > 12) || (year < 1)) { return 0; } if ((month == 2) && crm_time_leapyear(year)) { month = 13; } return month_days[month - 1]; } bool crm_time_leapyear(int year) { gboolean is_leap = FALSE; if (year % 4 == 0) { is_leap = TRUE; } if (year % 100 == 0 && year % 400 != 0) { is_leap = FALSE; } return is_leap; } /*! * \internal * \brief Get ordinal day number of year corresponding to given date * * \param[in] y Year * \param[in] m Month (1-12) * \param[in] d Day of month (1-31) * * \return Day number of year \p y corresponding to month \p m and day \p d, * or 0 for invalid arguments */ static int get_ordinal_days(uint32_t y, uint32_t m, uint32_t d) { int result = 0; CRM_CHECK((y > 0) && (y <= INT_MAX) && (m >= 1) && (m <= 12) && (d >= 1) && (d <= 31), return 0); result = d; for (int lpc = 1; lpc < m; lpc++) { result += crm_time_days_in_month(lpc, y); } return result; } void crm_time_log_alias(int log_level, const char *file, const char *function, int line, const char *prefix, const crm_time_t *date_time, int flags) { char *date_s = crm_time_as_string(date_time, flags); if (log_level == LOG_STDOUT) { printf("%s%s%s\n", (prefix? prefix : ""), (prefix? ": " : ""), date_s); } else { do_crm_log_alias(log_level, file, function, line, "%s%s%s", (prefix? prefix : ""), (prefix? ": " : ""), date_s); } free(date_s); } static void crm_time_get_sec(int sec, uint32_t *h, uint32_t *m, uint32_t *s) { uint32_t hours, minutes, seconds; seconds = QB_ABS(sec); hours = seconds / HOUR_SECONDS; seconds -= HOUR_SECONDS * hours; minutes = seconds / 60; seconds -= 60 * minutes; crm_trace("%d == %.2" PRIu32 ":%.2" PRIu32 ":%.2" PRIu32, sec, hours, minutes, seconds); *h = hours; *m = minutes; *s = seconds; } int crm_time_get_timeofday(const crm_time_t *dt, uint32_t *h, uint32_t *m, uint32_t *s) { crm_time_get_sec(dt->seconds, h, m, s); return TRUE; } int crm_time_get_timezone(const crm_time_t *dt, uint32_t *h, uint32_t *m) { uint32_t s; crm_time_get_sec(dt->seconds, h, m, &s); return TRUE; } long long crm_time_get_seconds(const crm_time_t *dt) { int lpc; crm_time_t *utc = NULL; long long in_seconds = 0; if (dt == NULL) { return 0; } utc = crm_get_utc_time(dt); if (utc == NULL) { return 0; } for (lpc = 1; lpc < utc->years; lpc++) { long long dmax = year_days(lpc); in_seconds += DAY_SECONDS * dmax; } /* utc->months is an offset that can only be set for a duration. * By definition, the value is variable depending on the date to * which it is applied. * * Force 30-day months so that something vaguely sane happens * for anyone that tries to use a month in this way. */ if (utc->months > 0) { in_seconds += DAY_SECONDS * 30 * (long long) (utc->months); } if (utc->days > 0) { in_seconds += DAY_SECONDS * (long long) (utc->days - 1); } in_seconds += utc->seconds; crm_time_free(utc); return in_seconds; } #define EPOCH_SECONDS 62135596800ULL /* Calculated using crm_time_get_seconds() */ long long crm_time_get_seconds_since_epoch(const crm_time_t *dt) { return (dt == NULL)? 0 : (crm_time_get_seconds(dt) - EPOCH_SECONDS); } int crm_time_get_gregorian(const crm_time_t *dt, uint32_t *y, uint32_t *m, uint32_t *d) { int months = 0; int days = dt->days; if(dt->years != 0) { for (months = 1; months <= 12 && days > 0; months++) { int mdays = crm_time_days_in_month(months, dt->years); if (mdays >= days) { break; } else { days -= mdays; } } } else if (dt->months) { /* This is a duration including months, don't convert the days field */ months = dt->months; } else { /* This is a duration not including months, still don't convert the days field */ } *y = dt->years; *m = months; *d = days; crm_trace("%.4d-%.3d -> %.4d-%.2d-%.2d", dt->years, dt->days, dt->years, months, days); return TRUE; } int crm_time_get_ordinal(const crm_time_t *dt, uint32_t *y, uint32_t *d) { *y = dt->years; *d = dt->days; return TRUE; } int crm_time_get_isoweek(const crm_time_t *dt, uint32_t *y, uint32_t *w, uint32_t *d) { /* * Monday 29 December 2008 is written "2009-W01-1" * Sunday 3 January 2010 is written "2009-W53-7" */ int year_num = 0; int jan1 = crm_time_january1_weekday(dt->years); int h = -1; CRM_CHECK(dt->days > 0, return FALSE); /* 6. Find the Weekday for Y M D */ h = dt->days + jan1 - 1; *d = 1 + ((h - 1) % 7); /* 7. Find if Y M D falls in YearNumber Y-1, WeekNumber 52 or 53 */ if (dt->days <= (8 - jan1) && jan1 > 4) { crm_trace("year--, jan1=%d", jan1); year_num = dt->years - 1; *w = crm_time_weeks_in_year(year_num); } else { year_num = dt->years; } /* 8. Find if Y M D falls in YearNumber Y+1, WeekNumber 1 */ if (year_num == dt->years) { int dmax = year_days(year_num); int correction = 4 - *d; if ((dmax - dt->days) < correction) { crm_trace("year++, jan1=%d, i=%d vs. %d", jan1, dmax - dt->days, correction); year_num = dt->years + 1; *w = 1; } } /* 9. Find if Y M D falls in YearNumber Y, WeekNumber 1 through 53 */ if (year_num == dt->years) { int j = dt->days + (7 - *d) + (jan1 - 1); *w = j / 7; if (jan1 > 4) { *w -= 1; } } *y = year_num; crm_trace("Converted %.4d-%.3d to %.4" PRIu32 "-W%.2" PRIu32 "-%" PRIu32, dt->years, dt->days, *y, *w, *d); return TRUE; } #define DATE_MAX 128 /*! * \internal * \brief Print "." to a buffer * * \param[in] sec Seconds * \param[in] usec Microseconds (must be of same sign as \p sec and of * absolute value less than \p QB_TIME_US_IN_SEC) * \param[in,out] buf Result buffer * \param[in,out] offset Current offset within \p buf */ static inline void sec_usec_as_string(long long sec, int usec, char *buf, size_t *offset) { *offset += snprintf(buf + *offset, DATE_MAX - *offset, "%s%lld.%06d", ((sec == 0) && (usec < 0))? "-" : "", sec, QB_ABS(usec)); } /*! * \internal * \brief Get a string representation of a duration * * \param[in] dt Time object to interpret as a duration * \param[in] usec Microseconds to add to \p dt * \param[in] show_usec Whether to include microseconds in \p result * \param[out] result Where to store the result string */ static void crm_duration_as_string(const crm_time_t *dt, int usec, bool show_usec, char *result) { size_t offset = 0; pcmk__assert(valid_sec_usec(dt->seconds, usec)); if (dt->years) { offset += snprintf(result + offset, DATE_MAX - offset, "%4d year%s ", dt->years, pcmk__plural_s(dt->years)); } if (dt->months) { offset += snprintf(result + offset, DATE_MAX - offset, "%2d month%s ", dt->months, pcmk__plural_s(dt->months)); } if (dt->days) { offset += snprintf(result + offset, DATE_MAX - offset, "%2d day%s ", dt->days, pcmk__plural_s(dt->days)); } // At least print seconds (and optionally usecs) if ((offset == 0) || (dt->seconds != 0) || (show_usec && (usec != 0))) { if (show_usec) { sec_usec_as_string(dt->seconds, usec, result, &offset); } else { offset += snprintf(result + offset, DATE_MAX - offset, "%d", dt->seconds); } offset += snprintf(result + offset, DATE_MAX - offset, " second%s", pcmk__plural_s(dt->seconds)); } // More than one minute, so provide a more readable breakdown into units if (QB_ABS(dt->seconds) >= 60) { uint32_t h = 0; uint32_t m = 0; uint32_t s = 0; uint32_t u = QB_ABS(usec); bool print_sec_component = false; crm_time_get_sec(dt->seconds, &h, &m, &s); print_sec_component = ((s != 0) || (show_usec && (u != 0))); offset += snprintf(result + offset, DATE_MAX - offset, " ("); if (h) { offset += snprintf(result + offset, DATE_MAX - offset, "%" PRIu32 " hour%s%s", h, pcmk__plural_s(h), ((m != 0) || print_sec_component)? " " : ""); } if (m) { offset += snprintf(result + offset, DATE_MAX - offset, "%" PRIu32 " minute%s%s", m, pcmk__plural_s(m), print_sec_component? " " : ""); } if (print_sec_component) { if (show_usec) { sec_usec_as_string(s, u, result, &offset); } else { offset += snprintf(result + offset, DATE_MAX - offset, "%" PRIu32, s); } offset += snprintf(result + offset, DATE_MAX - offset, " second%s", pcmk__plural_s(dt->seconds)); } offset += snprintf(result + offset, DATE_MAX - offset, ")"); } } /*! * \internal * \brief Get a string representation of a time object * * \param[in] dt Time to convert to string * \param[in] usec Microseconds to add to \p dt * \param[in] flags Group of \p crm_time_* string format options * \param[out] result Where to store the result string * * \note \p result must be of size \p DATE_MAX or larger. */ static void time_as_string_common(const crm_time_t *dt, int usec, uint32_t flags, char *result) { crm_time_t *utc = NULL; size_t offset = 0; if (!crm_time_is_defined(dt)) { strcpy(result, ""); return; } pcmk__assert(valid_sec_usec(dt->seconds, usec)); /* Simple cases: as duration, seconds, or seconds since epoch. * These never depend on time zone. */ if (pcmk_is_set(flags, crm_time_log_duration)) { crm_duration_as_string(dt, usec, pcmk_is_set(flags, crm_time_usecs), result); return; } if (pcmk_any_flags_set(flags, crm_time_seconds|crm_time_epoch)) { long long seconds = 0; if (pcmk_is_set(flags, crm_time_seconds)) { seconds = crm_time_get_seconds(dt); } else { seconds = crm_time_get_seconds_since_epoch(dt); } if (pcmk_is_set(flags, crm_time_usecs)) { sec_usec_as_string(seconds, usec, result, &offset); } else { snprintf(result, DATE_MAX, "%lld", seconds); } return; } // Convert to UTC if local timezone was not requested if ((dt->offset != 0) && !pcmk_is_set(flags, crm_time_log_with_timezone)) { crm_trace("UTC conversion"); utc = crm_get_utc_time(dt); dt = utc; } // As readable string if (pcmk_is_set(flags, crm_time_log_date)) { if (pcmk_is_set(flags, crm_time_weeks)) { // YYYY-WW-D uint32_t y = 0; uint32_t w = 0; uint32_t d = 0; if (crm_time_get_isoweek(dt, &y, &w, &d)) { offset += snprintf(result + offset, DATE_MAX - offset, "%" PRIu32 "-W%.2" PRIu32 "-%" PRIu32, y, w, d); } } else if (pcmk_is_set(flags, crm_time_ordinal)) { // YYYY-DDD uint32_t y = 0; uint32_t d = 0; if (crm_time_get_ordinal(dt, &y, &d)) { offset += snprintf(result + offset, DATE_MAX - offset, "%" PRIu32 "-%.3" PRIu32, y, d); } } else { // YYYY-MM-DD uint32_t y = 0; uint32_t m = 0; uint32_t d = 0; if (crm_time_get_gregorian(dt, &y, &m, &d)) { offset += snprintf(result + offset, DATE_MAX - offset, "%.4" PRIu32 "-%.2" PRIu32 "-%.2" PRIu32, y, m, d); } } } if (pcmk_is_set(flags, crm_time_log_timeofday)) { uint32_t h = 0, m = 0, s = 0; if (offset > 0) { offset += snprintf(result + offset, DATE_MAX - offset, " "); } if (crm_time_get_timeofday(dt, &h, &m, &s)) { offset += snprintf(result + offset, DATE_MAX - offset, "%.2" PRIu32 ":%.2" PRIu32 ":%.2" PRIu32, h, m, s); if (pcmk_is_set(flags, crm_time_usecs)) { offset += snprintf(result + offset, DATE_MAX - offset, ".%06" PRIu32, QB_ABS(usec)); } } if (pcmk_is_set(flags, crm_time_log_with_timezone) && (dt->offset != 0)) { crm_time_get_sec(dt->offset, &h, &m, &s); offset += snprintf(result + offset, DATE_MAX - offset, " %c%.2" PRIu32 ":%.2" PRIu32, ((dt->offset < 0)? '-' : '+'), h, m); } else { offset += snprintf(result + offset, DATE_MAX - offset, "Z"); } } crm_time_free(utc); } /*! * \brief Get a string representation of a \p crm_time_t object * * \param[in] dt Time to convert to string * \param[in] flags Group of \p crm_time_* string format options * * \note The caller is responsible for freeing the return value using \p free(). */ char * crm_time_as_string(const crm_time_t *dt, int flags) { char result[DATE_MAX] = { '\0', }; time_as_string_common(dt, 0, flags, result); return pcmk__str_copy(result); } /*! * \internal * \brief Determine number of seconds from an hour:minute:second string * * \param[in] time_str Time specification string * \param[out] result Number of seconds equivalent to time_str * * \return TRUE if specification was valid, FALSE (and set errno) otherwise * \note This may return the number of seconds in a day (which is out of bounds * for a time object) if given 24:00:00. */ static bool crm_time_parse_sec(const char *time_str, int *result) { int rc; uint32_t hour = 0; uint32_t minute = 0; uint32_t second = 0; *result = 0; // Must have at least hour, but minutes and seconds are optional rc = sscanf(time_str, "%" SCNu32 ":%" SCNu32 ":%" SCNu32, &hour, &minute, &second); if (rc == 1) { rc = sscanf(time_str, "%2" SCNu32 "%2" SCNu32 "%2" SCNu32, &hour, &minute, &second); } if (rc == 0) { crm_err("%s is not a valid ISO 8601 time specification", time_str); errno = EINVAL; return FALSE; } crm_trace("Got valid time: %.2" PRIu32 ":%.2" PRIu32 ":%.2" PRIu32, hour, minute, second); if ((hour == 24) && (minute == 0) && (second == 0)) { // Equivalent to 00:00:00 of next day, return number of seconds in day } else if (hour >= 24) { crm_err("%s is not a valid ISO 8601 time specification " "because %" PRIu32 " is not a valid hour", time_str, hour); errno = EINVAL; return FALSE; } if (minute >= 60) { crm_err("%s is not a valid ISO 8601 time specification " "because %" PRIu32 " is not a valid minute", time_str, minute); errno = EINVAL; return FALSE; } if (second >= 60) { crm_err("%s is not a valid ISO 8601 time specification " "because %" PRIu32 " is not a valid second", time_str, second); errno = EINVAL; return FALSE; } *result = (hour * HOUR_SECONDS) + (minute * 60) + second; return TRUE; } static bool crm_time_parse_offset(const char *offset_str, int *offset) { tzset(); if (offset_str == NULL) { // Use local offset #if defined(HAVE_STRUCT_TM_TM_GMTOFF) time_t now = time(NULL); struct tm *now_tm = localtime(&now); #endif int h_offset = GMTOFF(now_tm) / HOUR_SECONDS; int m_offset = (GMTOFF(now_tm) - (HOUR_SECONDS * h_offset)) / 60; if (h_offset < 0 && m_offset < 0) { m_offset = 0 - m_offset; } *offset = (HOUR_SECONDS * h_offset) + (60 * m_offset); return TRUE; } if (offset_str[0] == 'Z') { // @TODO invalid if anything after? *offset = 0; return TRUE; } *offset = 0; if ((offset_str[0] == '+') || (offset_str[0] == '-') || isdigit((int)offset_str[0])) { gboolean negate = FALSE; if (offset_str[0] == '+') { offset_str++; } else if (offset_str[0] == '-') { negate = TRUE; offset_str++; } if (crm_time_parse_sec(offset_str, offset) == FALSE) { return FALSE; } if (negate) { *offset = 0 - *offset; } } // @TODO else invalid? return TRUE; } /*! * \internal * \brief Parse the time portion of an ISO 8601 date/time string * * \param[in] time_str Time portion of specification (after any 'T') * \param[in,out] a_time Time object to parse into * * \return TRUE if valid time was parsed, FALSE (and set errno) otherwise * \note This may add a day to a_time (if the time is 24:00:00). */ static bool crm_time_parse(const char *time_str, crm_time_t *a_time) { uint32_t h, m, s; char *offset_s = NULL; tzset(); if (time_str) { if (crm_time_parse_sec(time_str, &(a_time->seconds)) == FALSE) { return FALSE; } offset_s = strstr(time_str, "Z"); if (offset_s == NULL) { offset_s = strstr(time_str, " "); if (offset_s) { while (isspace(offset_s[0])) { offset_s++; } } } } if (crm_time_parse_offset(offset_s, &(a_time->offset)) == FALSE) { return FALSE; } crm_time_get_sec(a_time->offset, &h, &m, &s); crm_trace("Got tz: %c%2." PRIu32 ":%.2" PRIu32, (a_time->offset < 0)? '-' : '+', h, m); if (a_time->seconds == DAY_SECONDS) { // 24:00:00 == 00:00:00 of next day a_time->seconds = 0; crm_time_add_days(a_time, 1); } return TRUE; } /* * \internal * \brief Parse a time object from an ISO 8601 date/time specification * * \param[in] date_str ISO 8601 date/time specification (or * \c PCMK__VALUE_EPOCH) * * \return New time object on success, NULL (and set errno) otherwise */ static crm_time_t * parse_date(const char *date_str) { const char *time_s = NULL; crm_time_t *dt = NULL; uint32_t year = 0U; uint32_t month = 0U; uint32_t day = 0U; uint32_t week = 0U; int rc = 0; if (pcmk__str_empty(date_str)) { crm_err("No ISO 8601 date/time specification given"); goto invalid; } if ((date_str[0] == 'T') || ((strlen(date_str) > 2) && (date_str[2] == ':'))) { /* Just a time supplied - Infer current date */ dt = crm_time_new(NULL); if (date_str[0] == 'T') { time_s = date_str + 1; } else { time_s = date_str; } goto parse_time; } dt = crm_time_new_undefined(); if ((strncasecmp(PCMK__VALUE_EPOCH, date_str, 5) == 0) && ((date_str[5] == '\0') || (date_str[5] == '/') || isspace(date_str[5]))) { dt->days = 1; dt->years = 1970; crm_time_log(LOG_TRACE, "Unpacked", dt, crm_time_log_date | crm_time_log_timeofday); return dt; } /* YYYY-MM-DD */ rc = sscanf(date_str, "%" SCNu32 "-%" SCNu32 "-%" SCNu32 "", &year, &month, &day); if (rc == 1) { /* YYYYMMDD */ rc = sscanf(date_str, "%4" SCNu32 "%2" SCNu32 "%2" SCNu32 "", &year, &month, &day); } if (rc == 3) { if ((month < 1U) || (month > 12U)) { crm_err("'%s' is not a valid ISO 8601 date/time specification " "because '%" PRIu32 "' is not a valid month", date_str, month); goto invalid; } else if ((year < 1U) || (year > INT_MAX)) { crm_err("'%s' is not a valid ISO 8601 date/time specification " "because '%" PRIu32 "' is not a valid year", date_str, year); goto invalid; } else if ((day < 1) || (day > INT_MAX) || (day > crm_time_days_in_month(month, year))) { crm_err("'%s' is not a valid ISO 8601 date/time specification " "because '%" PRIu32 "' is not a valid day of the month", date_str, day); goto invalid; } else { dt->years = year; dt->days = get_ordinal_days(year, month, day); crm_trace("Parsed Gregorian date '%.4" PRIu32 "-%.3d' " "from date string '%s'", year, dt->days, date_str); } goto parse_time; } /* YYYY-DDD */ rc = sscanf(date_str, "%" SCNu32 "-%" SCNu32, &year, &day); if (rc == 2) { if ((year < 1U) || (year > INT_MAX)) { crm_err("'%s' is not a valid ISO 8601 date/time specification " "because '%" PRIu32 "' is not a valid year", date_str, year); goto invalid; } else if ((day < 1U) || (day > INT_MAX) || (day > year_days(year))) { crm_err("'%s' is not a valid ISO 8601 date/time specification " "because '%" PRIu32 "' is not a valid day of year %" PRIu32 " (1-%d)", date_str, day, year, year_days(year)); goto invalid; } crm_trace("Parsed ordinal year %d and days %d from date string '%s'", year, day, date_str); dt->days = day; dt->years = year; goto parse_time; } /* YYYY-Www-D */ rc = sscanf(date_str, "%" SCNu32 "-W%" SCNu32 "-%" SCNu32, &year, &week, &day); if (rc == 3) { if ((week < 1U) || (week > crm_time_weeks_in_year(year))) { crm_err("'%s' is not a valid ISO 8601 date/time specification " "because '%" PRIu32 "' is not a valid week of year %" PRIu32 " (1-%d)", date_str, week, year, crm_time_weeks_in_year(year)); goto invalid; } else if ((day < 1U) || (day > 7U)) { crm_err("'%s' is not a valid ISO 8601 date/time specification " "because '%" PRIu32 "' is not a valid day of the week", date_str, day); goto invalid; } else { /* * See https://en.wikipedia.org/wiki/ISO_week_date * * Monday 29 December 2008 is written "2009-W01-1" * Sunday 3 January 2010 is written "2009-W53-7" * Saturday 27 September 2008 is written "2008-W37-6" * * If 1 January is on a Monday, Tuesday, Wednesday or Thursday, it * is in week 1. If 1 January is on a Friday, Saturday or Sunday, * it is in week 52 or 53 of the previous year. */ int jan1 = crm_time_january1_weekday(year); crm_trace("Parsed year %" PRIu32 " (Jan 1 = %d), week %" PRIu32 ", and day %" PRIu32 " from date string '%s'", year, jan1, week, day, date_str); dt->years = year; crm_time_add_days(dt, (week - 1) * 7); if (jan1 <= 4) { crm_time_add_days(dt, 1 - jan1); } else { crm_time_add_days(dt, 8 - jan1); } crm_time_add_days(dt, day); } goto parse_time; } crm_err("'%s' is not a valid ISO 8601 date/time specification", date_str); goto invalid; parse_time: if (time_s == NULL) { time_s = date_str + strspn(date_str, "0123456789-W"); if ((time_s[0] == ' ') || (time_s[0] == 'T')) { ++time_s; } else { time_s = NULL; } } if ((time_s != NULL) && (crm_time_parse(time_s, dt) == FALSE)) { goto invalid; } crm_time_log(LOG_TRACE, "Unpacked", dt, crm_time_log_date | crm_time_log_timeofday); if (crm_time_check(dt) == FALSE) { crm_err("'%s' is not a valid ISO 8601 date/time specification", date_str); goto invalid; } return dt; invalid: crm_time_free(dt); errno = EINVAL; return NULL; } // Parse an ISO 8601 numeric value and return number of characters consumed static int parse_int(const char *str, int *result) { unsigned int lpc; int offset = (str[0] == 'T')? 1 : 0; bool negate = false; *result = 0; // @TODO This cannot handle combinations of these characters switch (str[offset]) { case '.': case ',': return 0; // Fractions are not supported case '-': negate = true; offset++; break; case '+': case ':': offset++; break; default: break; } for (lpc = 0; (lpc < 10) && isdigit(str[offset]); lpc++) { const int digit = str[offset++] - '0'; if ((*result * 10LL + digit) > INT_MAX) { return 0; // Overflow } *result = *result * 10 + digit; } if (negate) { *result = 0 - *result; } return (lpc > 0)? offset : 0; } /*! * \brief Parse a time duration from an ISO 8601 duration specification * * \param[in] period_s ISO 8601 duration specification (optionally followed by * whitespace, after which the rest of the string will be * ignored) * * \return New time object on success, NULL (and set errno) otherwise * \note It is the caller's responsibility to return the result using * crm_time_free(). */ crm_time_t * crm_time_parse_duration(const char *period_s) { gboolean is_time = FALSE; crm_time_t *diff = NULL; if (pcmk__str_empty(period_s)) { crm_err("No ISO 8601 time duration given"); goto invalid; } if (period_s[0] != 'P') { crm_err("'%s' is not a valid ISO 8601 time duration " "because it does not start with a 'P'", period_s); goto invalid; } if ((period_s[1] == '\0') || isspace(period_s[1])) { crm_err("'%s' is not a valid ISO 8601 time duration " "because nothing follows 'P'", period_s); goto invalid; } diff = crm_time_new_undefined(); for (const char *current = period_s + 1; current[0] && (current[0] != '/') && !isspace(current[0]); ++current) { int an_int = 0, rc; long long result = 0LL; if (current[0] == 'T') { /* A 'T' separates year/month/day from hour/minute/seconds. We don't * require it strictly, but just use it to differentiate month from * minutes. */ is_time = TRUE; continue; } // An integer must be next rc = parse_int(current, &an_int); if (rc == 0) { crm_err("'%s' is not a valid ISO 8601 time duration " "because no valid integer at '%s'", period_s, current); goto invalid; } current += rc; // A time unit must be next (we're not strict about the order) switch (current[0]) { case 'Y': diff->years = an_int; break; case 'M': if (!is_time) { // Months diff->months = an_int; } else { // Minutes result = diff->seconds + an_int * 60LL; if ((result < INT_MIN) || (result > INT_MAX)) { crm_err("'%s' is not a valid ISO 8601 time duration " "because integer at '%s' is too %s", period_s, current - rc, ((result > 0)? "large" : "small")); goto invalid; } else { diff->seconds = (int) result; } } break; case 'W': result = diff->days + an_int * 7LL; if ((result < INT_MIN) || (result > INT_MAX)) { crm_err("'%s' is not a valid ISO 8601 time duration " "because integer at '%s' is too %s", period_s, current - rc, ((result > 0)? "large" : "small")); goto invalid; } else { diff->days = (int) result; } break; case 'D': result = diff->days + (long long) an_int; if ((result < INT_MIN) || (result > INT_MAX)) { crm_err("'%s' is not a valid ISO 8601 time duration " "because integer at '%s' is too %s", period_s, current - rc, ((result > 0)? "large" : "small")); goto invalid; } else { diff->days = (int) result; } break; case 'H': result = diff->seconds + (long long) an_int * HOUR_SECONDS; if ((result < INT_MIN) || (result > INT_MAX)) { crm_err("'%s' is not a valid ISO 8601 time duration " "because integer at '%s' is too %s", period_s, current - rc, ((result > 0)? "large" : "small")); goto invalid; } else { diff->seconds = (int) result; } break; case 'S': result = diff->seconds + (long long) an_int; if ((result < INT_MIN) || (result > INT_MAX)) { crm_err("'%s' is not a valid ISO 8601 time duration " "because integer at '%s' is too %s", period_s, current - rc, ((result > 0)? "large" : "small")); goto invalid; } else { diff->seconds = (int) result; } break; case '\0': crm_err("'%s' is not a valid ISO 8601 time duration " "because no units after %d", period_s, an_int); goto invalid; default: crm_err("'%s' is not a valid ISO 8601 time duration " "because '%c' is not a valid time unit", period_s, current[0]); goto invalid; } } if (!crm_time_is_defined(diff)) { crm_err("'%s' is not a valid ISO 8601 time duration " "because no amounts and units given", period_s); goto invalid; } diff->duration = TRUE; return diff; invalid: crm_time_free(diff); errno = EINVAL; return NULL; } /*! * \brief Parse a time period from an ISO 8601 interval specification * * \param[in] period_str ISO 8601 interval specification (start/end, * start/duration, or duration/end) * * \return New time period object on success, NULL (and set errno) otherwise * \note The caller is responsible for freeing the result using * crm_time_free_period(). */ crm_time_period_t * crm_time_parse_period(const char *period_str) { const char *original = period_str; crm_time_period_t *period = NULL; if (pcmk__str_empty(period_str)) { crm_err("No ISO 8601 time period given"); goto invalid; } tzset(); period = pcmk__assert_alloc(1, sizeof(crm_time_period_t)); if (period_str[0] == 'P') { period->diff = crm_time_parse_duration(period_str); if (period->diff == NULL) { goto error; } } else { period->start = parse_date(period_str); if (period->start == NULL) { goto error; } } period_str = strstr(original, "/"); if (period_str) { ++period_str; if (period_str[0] == 'P') { if (period->diff != NULL) { crm_err("'%s' is not a valid ISO 8601 time period " "because it has two durations", original); goto invalid; } period->diff = crm_time_parse_duration(period_str); if (period->diff == NULL) { goto error; } } else { period->end = parse_date(period_str); if (period->end == NULL) { goto error; } } } else if (period->diff != NULL) { // Only duration given, assume start is now period->start = crm_time_new(NULL); } else { // Only start given crm_err("'%s' is not a valid ISO 8601 time period " "because it has no duration or ending time", original); goto invalid; } if (period->start == NULL) { period->start = crm_time_subtract(period->end, period->diff); } else if (period->end == NULL) { period->end = crm_time_add(period->start, period->diff); } if (crm_time_check(period->start) == FALSE) { crm_err("'%s' is not a valid ISO 8601 time period " "because the start is invalid", period_str); goto invalid; } if (crm_time_check(period->end) == FALSE) { crm_err("'%s' is not a valid ISO 8601 time period " "because the end is invalid", period_str); goto invalid; } return period; invalid: errno = EINVAL; error: crm_time_free_period(period); return NULL; } /*! * \brief Free a dynamically allocated time period object * * \param[in,out] period Time period to free */ void crm_time_free_period(crm_time_period_t *period) { if (period) { crm_time_free(period->start); crm_time_free(period->end); crm_time_free(period->diff); free(period); } } void crm_time_set(crm_time_t *target, const crm_time_t *source) { crm_trace("target=%p, source=%p", target, source); CRM_CHECK(target != NULL && source != NULL, return); target->years = source->years; target->days = source->days; target->months = source->months; /* Only for durations */ target->seconds = source->seconds; target->offset = source->offset; crm_time_log(LOG_TRACE, "source", source, crm_time_log_date | crm_time_log_timeofday | crm_time_log_with_timezone); crm_time_log(LOG_TRACE, "target", target, crm_time_log_date | crm_time_log_timeofday | crm_time_log_with_timezone); } static void ha_set_tm_time(crm_time_t *target, const struct tm *source) { int h_offset = 0; int m_offset = 0; /* Ensure target is fully initialized */ target->years = 0; target->months = 0; target->days = 0; target->seconds = 0; target->offset = 0; target->duration = FALSE; if (source->tm_year > 0) { /* years since 1900 */ target->years = 1900; crm_time_add_years(target, source->tm_year); } if (source->tm_yday >= 0) { /* days since January 1 [0-365] */ target->days = 1 + source->tm_yday; } if (source->tm_hour >= 0) { target->seconds += HOUR_SECONDS * source->tm_hour; } if (source->tm_min >= 0) { target->seconds += 60 * source->tm_min; } if (source->tm_sec >= 0) { target->seconds += source->tm_sec; } /* tm_gmtoff == offset from UTC in seconds */ h_offset = GMTOFF(source) / HOUR_SECONDS; m_offset = (GMTOFF(source) - (HOUR_SECONDS * h_offset)) / 60; crm_trace("Time offset is %lds (%.2d:%.2d)", GMTOFF(source), h_offset, m_offset); target->offset += HOUR_SECONDS * h_offset; target->offset += 60 * m_offset; } void crm_time_set_timet(crm_time_t *target, const time_t *source) { ha_set_tm_time(target, localtime(source)); } /*! * \internal * \brief Set one time object to another if the other is earlier * * \param[in,out] target Time object to set * \param[in] source Time object to use if earlier */ void pcmk__set_time_if_earlier(crm_time_t *target, const crm_time_t *source) { if ((target != NULL) && (source != NULL) && (!crm_time_is_defined(target) || (crm_time_compare(source, target) < 0))) { crm_time_set(target, source); } } crm_time_t * pcmk_copy_time(const crm_time_t *source) { crm_time_t *target = crm_time_new_undefined(); crm_time_set(target, source); return target; } /*! * \internal * \brief Convert a \p time_t time to a \p crm_time_t time * * \param[in] source Time to convert * * \return A \p crm_time_t object representing \p source */ crm_time_t * pcmk__copy_timet(time_t source) { crm_time_t *target = crm_time_new_undefined(); crm_time_set_timet(target, &source); return target; } crm_time_t * crm_time_add(const crm_time_t *dt, const crm_time_t *value) { crm_time_t *utc = NULL; crm_time_t *answer = NULL; if ((dt == NULL) || (value == NULL)) { errno = EINVAL; return NULL; } answer = pcmk_copy_time(dt); utc = crm_get_utc_time(value); if (utc == NULL) { crm_time_free(answer); return NULL; } crm_time_add_years(answer, utc->years); crm_time_add_months(answer, utc->months); crm_time_add_days(answer, utc->days); crm_time_add_seconds(answer, utc->seconds); crm_time_free(utc); return answer; } /*! * \internal * \brief Return the XML attribute name corresponding to a time component * * \param[in] component Component to check * * \return XML attribute name corresponding to \p component, or NULL if * \p component is invalid */ const char * pcmk__time_component_attr(enum pcmk__time_component component) { switch (component) { case pcmk__time_years: return PCMK_XA_YEARS; case pcmk__time_months: return PCMK_XA_MONTHS; case pcmk__time_weeks: return PCMK_XA_WEEKS; case pcmk__time_days: return PCMK_XA_DAYS; case pcmk__time_hours: return PCMK_XA_HOURS; case pcmk__time_minutes: return PCMK_XA_MINUTES; case pcmk__time_seconds: return PCMK_XA_SECONDS; default: return NULL; } } typedef void (*component_fn_t)(crm_time_t *, int); /*! * \internal * \brief Get the addition function corresponding to a time component * \param[in] component Component to check * * \return Addition function corresponding to \p component, or NULL if * \p component is invalid */ static component_fn_t component_fn(enum pcmk__time_component component) { switch (component) { case pcmk__time_years: return crm_time_add_years; case pcmk__time_months: return crm_time_add_months; case pcmk__time_weeks: return crm_time_add_weeks; case pcmk__time_days: return crm_time_add_days; case pcmk__time_hours: return crm_time_add_hours; case pcmk__time_minutes: return crm_time_add_minutes; case pcmk__time_seconds: return crm_time_add_seconds; default: return NULL; } } /*! * \internal * \brief Add the value of an XML attribute to a time object * * \param[in,out] t Time object to add to * \param[in] component Component of \p t to add to * \param[in] xml XML with value to add * * \return Standard Pacemaker return code */ int pcmk__add_time_from_xml(crm_time_t *t, enum pcmk__time_component component, const xmlNode *xml) { long long value; const char *attr = pcmk__time_component_attr(component); component_fn_t add = component_fn(component); if ((t == NULL) || (attr == NULL) || (add == NULL)) { return EINVAL; } if (xml == NULL) { return pcmk_rc_ok; } if (pcmk__scan_ll(crm_element_value(xml, attr), &value, 0LL) != pcmk_rc_ok) { return pcmk_rc_unpack_error; } if ((value < INT_MIN) || (value > INT_MAX)) { return ERANGE; } if (value != 0LL) { add(t, (int) value); } return pcmk_rc_ok; } crm_time_t * crm_time_calculate_duration(const crm_time_t *dt, const crm_time_t *value) { crm_time_t *utc = NULL; crm_time_t *answer = NULL; if ((dt == NULL) || (value == NULL)) { errno = EINVAL; return NULL; } utc = crm_get_utc_time(value); if (utc == NULL) { return NULL; } answer = crm_get_utc_time(dt); if (answer == NULL) { crm_time_free(utc); return NULL; } answer->duration = TRUE; crm_time_add_years(answer, -utc->years); if(utc->months != 0) { crm_time_add_months(answer, -utc->months); } crm_time_add_days(answer, -utc->days); crm_time_add_seconds(answer, -utc->seconds); crm_time_free(utc); return answer; } crm_time_t * crm_time_subtract(const crm_time_t *dt, const crm_time_t *value) { crm_time_t *utc = NULL; crm_time_t *answer = NULL; if ((dt == NULL) || (value == NULL)) { errno = EINVAL; return NULL; } utc = crm_get_utc_time(value); if (utc == NULL) { return NULL; } answer = pcmk_copy_time(dt); crm_time_add_years(answer, -utc->years); if(utc->months != 0) { crm_time_add_months(answer, -utc->months); } crm_time_add_days(answer, -utc->days); crm_time_add_seconds(answer, -utc->seconds); crm_time_free(utc); return answer; } /*! * \brief Check whether a time object represents a sensible date/time * * \param[in] dt Date/time object to check * * \return \c true if years, days, and seconds are sensible, \c false otherwise */ bool crm_time_check(const crm_time_t *dt) { return (dt != NULL) && (dt->days > 0) && (dt->days <= year_days(dt->years)) && (dt->seconds >= 0) && (dt->seconds < DAY_SECONDS); } #define do_cmp_field(l, r, field) \ if(rc == 0) { \ if(l->field > r->field) { \ crm_trace("%s: %d > %d", \ #field, l->field, r->field); \ rc = 1; \ } else if(l->field < r->field) { \ crm_trace("%s: %d < %d", \ #field, l->field, r->field); \ rc = -1; \ } \ } int crm_time_compare(const crm_time_t *a, const crm_time_t *b) { int rc = 0; crm_time_t *t1 = crm_get_utc_time(a); crm_time_t *t2 = crm_get_utc_time(b); if ((t1 == NULL) && (t2 == NULL)) { rc = 0; } else if (t1 == NULL) { rc = -1; } else if (t2 == NULL) { rc = 1; } else { do_cmp_field(t1, t2, years); do_cmp_field(t1, t2, days); do_cmp_field(t1, t2, seconds); } crm_time_free(t1); crm_time_free(t2); return rc; } /*! * \brief Add a given number of seconds to a date/time or duration * * \param[in,out] a_time Date/time or duration to add seconds to * \param[in] extra Number of seconds to add */ void crm_time_add_seconds(crm_time_t *a_time, int extra) { int days = extra / DAY_SECONDS; pcmk__assert(a_time != NULL); crm_trace("Adding %d seconds (including %d whole day%s) to %d", extra, days, pcmk__plural_s(days), a_time->seconds); a_time->seconds += extra % DAY_SECONDS; // Check whether the addition crossed a day boundary if (a_time->seconds > DAY_SECONDS) { ++days; a_time->seconds -= DAY_SECONDS; } else if (a_time->seconds < 0) { --days; a_time->seconds += DAY_SECONDS; } crm_time_add_days(a_time, days); } #define ydays(t) (crm_time_leapyear((t)->years)? 366 : 365) /*! * \brief Add days to a date/time * * \param[in,out] a_time Time to modify * \param[in] extra Number of days to add (may be negative to subtract) */ void crm_time_add_days(crm_time_t *a_time, int extra) { pcmk__assert(a_time != NULL); crm_trace("Adding %d days to %.4d-%.3d", extra, a_time->years, a_time->days); if (extra > 0) { while ((a_time->days + (long long) extra) > ydays(a_time)) { if ((a_time->years + 1LL) > INT_MAX) { a_time->days = ydays(a_time); // Clip to latest we can handle return; } extra -= ydays(a_time); a_time->years++; } } else if (extra < 0) { const int min_days = a_time->duration? 0 : 1; while ((a_time->days + (long long) extra) < min_days) { if ((a_time->years - 1) < 1) { a_time->days = 1; // Clip to earliest we can handle (no BCE) return; } a_time->years--; extra += ydays(a_time); } } a_time->days += extra; } void crm_time_add_months(crm_time_t * a_time, int extra) { int lpc; uint32_t y, m, d, dmax; crm_time_get_gregorian(a_time, &y, &m, &d); crm_trace("Adding %d months to %.4" PRIu32 "-%.2" PRIu32 "-%.2" PRIu32, extra, y, m, d); if (extra > 0) { for (lpc = extra; lpc > 0; lpc--) { m++; if (m == 13) { m = 1; y++; } } } else { for (lpc = -extra; lpc > 0; lpc--) { m--; if (m == 0) { m = 12; y--; } } } dmax = crm_time_days_in_month(m, y); if (dmax < d) { /* Preserve day-of-month unless the month doesn't have enough days */ d = dmax; } crm_trace("Calculated %.4" PRIu32 "-%.2" PRIu32 "-%.2" PRIu32, y, m, d); a_time->years = y; a_time->days = get_ordinal_days(y, m, d); crm_time_get_gregorian(a_time, &y, &m, &d); crm_trace("Got %.4" PRIu32 "-%.2" PRIu32 "-%.2" PRIu32, y, m, d); } void crm_time_add_minutes(crm_time_t * a_time, int extra) { crm_time_add_seconds(a_time, extra * 60); } void crm_time_add_hours(crm_time_t * a_time, int extra) { crm_time_add_seconds(a_time, extra * HOUR_SECONDS); } void crm_time_add_weeks(crm_time_t * a_time, int extra) { crm_time_add_days(a_time, extra * 7); } void crm_time_add_years(crm_time_t * a_time, int extra) { pcmk__assert(a_time != NULL); if ((extra > 0) && ((a_time->years + (long long) extra) > INT_MAX)) { a_time->years = INT_MAX; } else if ((extra < 0) && ((a_time->years + (long long) extra) < 1)) { a_time->years = 1; // Clip to earliest we can handle (no BCE) } else { a_time->years += extra; } } static void ha_get_tm_time(struct tm *target, const crm_time_t *source) { *target = (struct tm) { .tm_year = source->years - 1900, .tm_mday = source->days, .tm_sec = source->seconds % 60, .tm_min = ( source->seconds / 60 ) % 60, .tm_hour = source->seconds / HOUR_SECONDS, .tm_isdst = -1, /* don't adjust */ #if defined(HAVE_STRUCT_TM_TM_GMTOFF) .tm_gmtoff = source->offset #endif }; mktime(target); } /* The high-resolution variant of time object was added to meet an immediate * need, and is kept internal API. * * @TODO The long-term goal is to come up with a clean, unified design for a * time type (or types) that meets all the various needs, to replace * crm_time_t, pcmk__time_hr_t, and struct timespec (in lrmd_cmd_t). */ pcmk__time_hr_t * pcmk__time_hr_convert(pcmk__time_hr_t *target, const crm_time_t *dt) { pcmk__time_hr_t *hr_dt = NULL; if (dt) { hr_dt = target; if (hr_dt == NULL) { hr_dt = pcmk__assert_alloc(1, sizeof(pcmk__time_hr_t)); } *hr_dt = (pcmk__time_hr_t) { .years = dt->years, .months = dt->months, .days = dt->days, .seconds = dt->seconds, .offset = dt->offset, .duration = dt->duration }; } return hr_dt; } void pcmk__time_set_hr_dt(crm_time_t *target, const pcmk__time_hr_t *hr_dt) { pcmk__assert((target != NULL) && (hr_dt != NULL)); *target = (crm_time_t) { .years = hr_dt->years, .months = hr_dt->months, .days = hr_dt->days, .seconds = hr_dt->seconds, .offset = hr_dt->offset, .duration = hr_dt->duration }; } /*! * \internal * \brief Return the current time as a high-resolution time * * \param[out] epoch If not NULL, this will be set to seconds since epoch * * \return Newly allocated high-resolution time set to the current time */ pcmk__time_hr_t * pcmk__time_hr_now(time_t *epoch) { struct timespec tv; crm_time_t dt; pcmk__time_hr_t *hr; qb_util_timespec_from_epoch_get(&tv); if (epoch != NULL) { *epoch = tv.tv_sec; } crm_time_set_timet(&dt, &(tv.tv_sec)); hr = pcmk__time_hr_convert(NULL, &dt); if (hr != NULL) { hr->useconds = tv.tv_nsec / QB_TIME_NS_IN_USEC; } return hr; } pcmk__time_hr_t * pcmk__time_hr_new(const char *date_time) { pcmk__time_hr_t *hr_dt = NULL; if (date_time == NULL) { hr_dt = pcmk__time_hr_now(NULL); } else { crm_time_t *dt; dt = parse_date(date_time); hr_dt = pcmk__time_hr_convert(NULL, dt); crm_time_free(dt); } return hr_dt; } void pcmk__time_hr_free(pcmk__time_hr_t * hr_dt) { free(hr_dt); } /*! * \internal * \brief Expand a date/time format string, including %N for nanoseconds * * \param[in] format Date/time format string as per strftime(3) with the * addition of %N for nanoseconds * \param[in] hr_dt Time value to format * * \return Newly allocated string with formatted string */ char * pcmk__time_format_hr(const char *format, const pcmk__time_hr_t *hr_dt) { int scanned_pos = 0; // How many characters of format have been parsed int printed_pos = 0; // How many characters of format have been processed size_t date_len = 0; char nano_s[10] = { '\0', }; char date_s[128] = { '\0', }; struct tm tm = { 0, }; crm_time_t dt = { 0, }; if (format == NULL) { return NULL; } pcmk__time_set_hr_dt(&dt, hr_dt); ha_get_tm_time(&tm, &dt); sprintf(nano_s, "%06d000", hr_dt->useconds); while (format[scanned_pos] != '\0') { int fmt_pos; // Index after last character to pass as-is int nano_digits = 0; // Length of %N field width (if any) char *tmp_fmt_s = NULL; size_t nbytes = 0; // Look for next format specifier const char *mark_s = strchr(&format[scanned_pos], '%'); if (mark_s == NULL) { // No more specifiers, so pass remaining string to strftime() as-is scanned_pos = strlen(format); fmt_pos = scanned_pos; } else { fmt_pos = mark_s - format; // Index of % // Skip % and any field width scanned_pos = fmt_pos + 1; while (isdigit(format[scanned_pos])) { scanned_pos++; } switch (format[scanned_pos]) { case '\0': // Literal % and possibly digits at end of string fmt_pos = scanned_pos; // Pass remaining string as-is break; case 'N': // %[width]N scanned_pos++; // Parse field width nano_digits = atoi(&format[fmt_pos + 1]); nano_digits = QB_MAX(nano_digits, 0); nano_digits = QB_MIN(nano_digits, 6); break; default: // Some other specifier if (format[++scanned_pos] != '\0') { // More to parse continue; } fmt_pos = scanned_pos; // Pass remaining string as-is break; } } if (date_len >= sizeof(date_s)) { return NULL; // No room for remaining string } tmp_fmt_s = strndup(&format[printed_pos], fmt_pos - printed_pos); + if (tmp_fmt_s == NULL) { + return NULL; + } + #ifdef HAVE_FORMAT_NONLITERAL #pragma GCC diagnostic push #pragma GCC diagnostic ignored "-Wformat-nonliteral" #endif nbytes = strftime(&date_s[date_len], sizeof(date_s) - date_len, tmp_fmt_s, &tm); #ifdef HAVE_FORMAT_NONLITERAL #pragma GCC diagnostic pop #endif free(tmp_fmt_s); if (nbytes == 0) { // Would overflow buffer return NULL; } date_len += nbytes; printed_pos = scanned_pos; if (nano_digits != 0) { int nc = 0; if (date_len >= sizeof(date_s)) { return NULL; // No room to add nanoseconds } nc = snprintf(&date_s[date_len], sizeof(date_s) - date_len, "%.*s", nano_digits, nano_s); if ((nc < 0) || (nc == (sizeof(date_s) - date_len))) { return NULL; // Error or would overflow buffer } date_len += nc; } } return (date_len == 0)? NULL : pcmk__str_copy(date_s); } /*! * \internal * \brief Return a human-friendly string corresponding to an epoch time value * * \param[in] source Pointer to epoch time value (or \p NULL for current time) * \param[in] flags Group of \p crm_time_* flags controlling display format * (0 to use \p ctime() with newline removed) * * \return String representation of \p source on success (may be empty depending * on \p flags; guaranteed not to be \p NULL) * * \note The caller is responsible for freeing the return value using \p free(). */ char * pcmk__epoch2str(const time_t *source, uint32_t flags) { time_t epoch_time = (source == NULL)? time(NULL) : *source; if (flags == 0) { return pcmk__str_copy(pcmk__trim(ctime(&epoch_time))); } else { crm_time_t dt; crm_time_set_timet(&dt, &epoch_time); return crm_time_as_string(&dt, flags); } } /*! * \internal * \brief Return a human-friendly string corresponding to seconds-and- * nanoseconds value * * Time is shown with microsecond resolution if \p crm_time_usecs is in \p * flags. * * \param[in] ts Time in seconds and nanoseconds (or \p NULL for current * time) * \param[in] flags Group of \p crm_time_* flags controlling display format * * \return String representation of \p ts on success (may be empty depending on * \p flags; guaranteed not to be \p NULL) * * \note The caller is responsible for freeing the return value using \p free(). */ char * pcmk__timespec2str(const struct timespec *ts, uint32_t flags) { struct timespec tmp_ts; crm_time_t dt; char result[DATE_MAX] = { 0 }; if (ts == NULL) { qb_util_timespec_from_epoch_get(&tmp_ts); ts = &tmp_ts; } crm_time_set_timet(&dt, &ts->tv_sec); time_as_string_common(&dt, ts->tv_nsec / QB_TIME_NS_IN_USEC, flags, result); return pcmk__str_copy(result); } /*! * \internal * \brief Given a millisecond interval, return a log-friendly string * * \param[in] interval_ms Interval in milliseconds * * \return Readable version of \p interval_ms * * \note The return value is a pointer to static memory that will be * overwritten by later calls to this function. */ const char * pcmk__readable_interval(guint interval_ms) { #define MS_IN_S (1000) #define MS_IN_M (MS_IN_S * 60) #define MS_IN_H (MS_IN_M * 60) #define MS_IN_D (MS_IN_H * 24) #define MAXSTR sizeof("..d..h..m..s...ms") static char str[MAXSTR]; int offset = 0; str[0] = '\0'; if (interval_ms >= MS_IN_D) { offset += snprintf(str + offset, MAXSTR - offset, "%ud", interval_ms / MS_IN_D); interval_ms -= (interval_ms / MS_IN_D) * MS_IN_D; } if (interval_ms >= MS_IN_H) { offset += snprintf(str + offset, MAXSTR - offset, "%uh", interval_ms / MS_IN_H); interval_ms -= (interval_ms / MS_IN_H) * MS_IN_H; } if (interval_ms >= MS_IN_M) { offset += snprintf(str + offset, MAXSTR - offset, "%um", interval_ms / MS_IN_M); interval_ms -= (interval_ms / MS_IN_M) * MS_IN_M; } // Ns, N.NNNs, or NNNms if (interval_ms >= MS_IN_S) { offset += snprintf(str + offset, MAXSTR - offset, "%u", interval_ms / MS_IN_S); interval_ms -= (interval_ms / MS_IN_S) * MS_IN_S; if (interval_ms > 0) { offset += snprintf(str + offset, MAXSTR - offset, ".%03u", interval_ms); } (void) snprintf(str + offset, MAXSTR - offset, "s"); } else if (interval_ms > 0) { (void) snprintf(str + offset, MAXSTR - offset, "%ums", interval_ms); } else if (str[0] == '\0') { strcpy(str, "0s"); } return str; } diff --git a/lib/common/schemas.c b/lib/common/schemas.c index 8f9fdfa558..8ae2128f2f 100644 --- a/lib/common/schemas.c +++ b/lib/common/schemas.c @@ -1,1755 +1,1758 @@ /* - * Copyright 2004-2024 the Pacemaker project contributors + * Copyright 2004-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include /* PCMK__XML_LOG_BASE */ #include "crmcommon_private.h" #define SCHEMA_ZERO { .v = { 0, 0 } } #define schema_strdup_printf(prefix, version, suffix) \ crm_strdup_printf(prefix "%u.%u" suffix, (version).v[0], (version).v[1]) typedef struct { xmlRelaxNGPtr rng; xmlRelaxNGValidCtxtPtr valid; xmlRelaxNGParserCtxtPtr parser; } relaxng_ctx_cache_t; static GList *known_schemas = NULL; static bool initialized = false; static bool silent_logging = FALSE; static void G_GNUC_PRINTF(2, 3) xml_log(int priority, const char *fmt, ...) { va_list ap; va_start(ap, fmt); if (silent_logging == FALSE) { /* XXX should not this enable dechunking as well? */ PCMK__XML_LOG_BASE(priority, FALSE, 0, NULL, fmt, ap); } va_end(ap); } static int xml_latest_schema_index(void) { /* This function assumes that crm_schema_init() has been called beforehand, * so we have at least three schemas (one real schema, the "pacemaker-next" * schema, and the "none" schema). * * @COMPAT: pacemaker-next is deprecated since 2.1.5 and none since 2.1.8. * Update this when we drop those. */ return g_list_length(known_schemas) - 3; } /*! * \internal * \brief Return the schema entry of the highest-versioned schema * * \return Schema entry of highest-versioned schema (or NULL on error) */ static GList * get_highest_schema(void) { /* The highest numerically versioned schema is the one before pacemaker-next * * @COMPAT pacemaker-next is deprecated since 2.1.5 */ GList *entry = pcmk__get_schema("pacemaker-next"); pcmk__assert((entry != NULL) && (entry->prev != NULL)); return entry->prev; } /*! * \internal * \brief Return the name of the highest-versioned schema * * \return Name of highest-versioned schema (or NULL on error) */ const char * pcmk__highest_schema_name(void) { GList *entry = get_highest_schema(); return ((pcmk__schema_t *)(entry->data))->name; } /*! * \internal * \brief Find first entry of highest major schema version series * * \return Schema entry of first schema with highest major version */ GList * pcmk__find_x_0_schema(void) { #if defined(PCMK__UNIT_TESTING) /* If we're unit testing, this can't be static because it'll stick * around from one test run to the next. It needs to be cleared out * every time. */ GList *x_0_entry = NULL; #else static GList *x_0_entry = NULL; #endif pcmk__schema_t *highest_schema = NULL; if (x_0_entry != NULL) { return x_0_entry; } x_0_entry = get_highest_schema(); highest_schema = x_0_entry->data; for (GList *iter = x_0_entry->prev; iter != NULL; iter = iter->prev) { pcmk__schema_t *schema = iter->data; /* We've found a schema in an older major version series. Return * the index of the first one in the same major version series as * the highest schema. */ if (schema->version.v[0] < highest_schema->version.v[0]) { x_0_entry = iter->next; break; } /* We're out of list to examine. This probably means there was only * one major version series, so return the first schema entry. */ if (iter->prev == NULL) { x_0_entry = known_schemas->data; break; } } return x_0_entry; } static inline bool version_from_filename(const char *filename, pcmk__schema_version_t *version) { if (pcmk__ends_with(filename, ".rng")) { return sscanf(filename, "pacemaker-%hhu.%hhu.rng", &(version->v[0]), &(version->v[1])) == 2; } else { return sscanf(filename, "pacemaker-%hhu.%hhu", &(version->v[0]), &(version->v[1])) == 2; } } static int schema_filter(const struct dirent *a) { int rc = 0; pcmk__schema_version_t version = SCHEMA_ZERO; if (strstr(a->d_name, "pacemaker-") != a->d_name) { /* crm_trace("%s - wrong prefix", a->d_name); */ } else if (!pcmk__ends_with_ext(a->d_name, ".rng")) { /* crm_trace("%s - wrong suffix", a->d_name); */ } else if (!version_from_filename(a->d_name, &version)) { /* crm_trace("%s - wrong format", a->d_name); */ } else { /* crm_debug("%s - candidate", a->d_name); */ rc = 1; } return rc; } static int schema_cmp(pcmk__schema_version_t a_version, pcmk__schema_version_t b_version) { for (int i = 0; i < 2; ++i) { if (a_version.v[i] < b_version.v[i]) { return -1; } else if (a_version.v[i] > b_version.v[i]) { return 1; } } return 0; } static int schema_cmp_directory(const struct dirent **a, const struct dirent **b) { pcmk__schema_version_t a_version = SCHEMA_ZERO; pcmk__schema_version_t b_version = SCHEMA_ZERO; if (!version_from_filename(a[0]->d_name, &a_version) || !version_from_filename(b[0]->d_name, &b_version)) { // Shouldn't be possible, but makes static analysis happy return 0; } return schema_cmp(a_version, b_version); } /*! * \internal * \brief Add given schema + auxiliary data to internal bookkeeping. * * \note When providing \p version, should not be called directly but * through \c add_schema_by_version. */ static void add_schema(enum pcmk__schema_validator validator, const pcmk__schema_version_t *version, const char *name, const char *transform, const char *transform_enter, bool transform_onleave) { pcmk__schema_t *schema = NULL; schema = pcmk__assert_alloc(1, sizeof(pcmk__schema_t)); schema->validator = validator; schema->version.v[0] = version->v[0]; schema->version.v[1] = version->v[1]; schema->transform_onleave = transform_onleave; // schema->schema_index is set after all schemas are loaded and sorted if (version->v[0] || version->v[1]) { schema->name = schema_strdup_printf("pacemaker-", *version, ""); } else { schema->name = pcmk__str_copy(name); } if (transform) { schema->transform = pcmk__str_copy(transform); } if (transform_enter) { schema->transform_enter = pcmk__str_copy(transform_enter); } known_schemas = g_list_prepend(known_schemas, schema); } /*! * \internal * \brief Add version-specified schema + auxiliary data to internal bookkeeping. * \return Standard Pacemaker return value (the only possible values are * \c ENOENT when no upgrade schema is associated, or \c pcmk_rc_ok otherwise. * * \note There's no reliance on the particular order of schemas entering here. * * \par A bit of theory * We track 3 XSLT stylesheets that differ per usage: * - "upgrade": * . sparsely spread over the sequence of all available schemas, * as they are only relevant when major version of the schema * is getting bumped -- in that case, it MUST be set * . name convention: upgrade-X.Y.xsl * - "upgrade-enter": * . may only accompany "upgrade" occurrence, but doesn't need to * be present anytime such one is, i.e., it MAY not be set when * "upgrade" is * . name convention: upgrade-X.Y-enter.xsl, * when not present: upgrade-enter.xsl * - "upgrade-leave": * . like "upgrade-enter", but SHOULD be present whenever * "upgrade-enter" is (and vice versa, but that's only * to prevent confusion based on observing the files, * it would get ignored regardless) * . name convention: (see "upgrade-enter") */ static int add_schema_by_version(const pcmk__schema_version_t *version, bool transform_expected) { bool transform_onleave = FALSE; int rc = pcmk_rc_ok; struct stat s; char *xslt = NULL, *transform_upgrade = NULL, *transform_enter = NULL; /* prologue for further transform_expected handling */ if (transform_expected) { /* check if there's suitable "upgrade" stylesheet */ transform_upgrade = schema_strdup_printf("upgrade-", *version, ); xslt = pcmk__xml_artefact_path(pcmk__xml_artefact_ns_legacy_xslt, transform_upgrade); } if (!transform_expected) { /* jump directly to the end */ } else if (stat(xslt, &s) == 0) { /* perhaps there's also a targeted "upgrade-enter" stylesheet */ transform_enter = schema_strdup_printf("upgrade-", *version, "-enter"); free(xslt); xslt = pcmk__xml_artefact_path(pcmk__xml_artefact_ns_legacy_xslt, transform_enter); if (stat(xslt, &s) != 0) { /* or initially, at least a generic one */ crm_debug("Upgrade-enter transform %s.xsl not found", xslt); free(xslt); free(transform_enter); transform_enter = strdup("upgrade-enter"); xslt = pcmk__xml_artefact_path(pcmk__xml_artefact_ns_legacy_xslt, transform_enter); if (stat(xslt, &s) != 0) { crm_debug("Upgrade-enter transform %s.xsl not found, either", xslt); free(xslt); xslt = NULL; } } /* xslt contains full path to "upgrade-enter" stylesheet */ if (xslt != NULL) { /* then there should be "upgrade-leave" counterpart (enter->leave) */ + // coverity[returned_null] Not worth correcting this code memcpy(strrchr(xslt, '-') + 1, "leave", sizeof("leave") - 1); transform_onleave = (stat(xslt, &s) == 0); free(xslt); } else { free(transform_enter); transform_enter = NULL; } } else { crm_err("Upgrade transform %s not found", xslt); free(xslt); free(transform_upgrade); transform_upgrade = NULL; rc = ENOENT; } add_schema(pcmk__schema_validator_rng, version, NULL, transform_upgrade, transform_enter, transform_onleave); free(transform_upgrade); free(transform_enter); return rc; } static void wrap_libxslt(bool finalize) { static xsltSecurityPrefsPtr secprefs; int ret = 0; /* security framework preferences */ if (!finalize) { pcmk__assert(secprefs == NULL); secprefs = xsltNewSecurityPrefs(); ret = xsltSetSecurityPrefs(secprefs, XSLT_SECPREF_WRITE_FILE, xsltSecurityForbid) | xsltSetSecurityPrefs(secprefs, XSLT_SECPREF_CREATE_DIRECTORY, xsltSecurityForbid) | xsltSetSecurityPrefs(secprefs, XSLT_SECPREF_READ_NETWORK, xsltSecurityForbid) | xsltSetSecurityPrefs(secprefs, XSLT_SECPREF_WRITE_NETWORK, xsltSecurityForbid); if (ret != 0) { return; } } else { xsltFreeSecurityPrefs(secprefs); secprefs = NULL; } /* cleanup only */ if (finalize) { xsltCleanupGlobals(); } } void pcmk__load_schemas_from_dir(const char *dir) { int lpc, max; struct dirent **namelist = NULL; max = scandir(dir, &namelist, schema_filter, schema_cmp_directory); if (max < 0) { crm_warn("Could not load schemas from %s: %s", dir, strerror(errno)); return; } for (lpc = 0; lpc < max; lpc++) { bool transform_expected = false; pcmk__schema_version_t version = SCHEMA_ZERO; if (!version_from_filename(namelist[lpc]->d_name, &version)) { // Shouldn't be possible, but makes static analysis happy crm_warn("Skipping schema '%s': could not parse version", namelist[lpc]->d_name); continue; } if ((lpc + 1) < max) { pcmk__schema_version_t next_version = SCHEMA_ZERO; if (version_from_filename(namelist[lpc+1]->d_name, &next_version) && (version.v[0] < next_version.v[0])) { transform_expected = true; } } if (add_schema_by_version(&version, transform_expected) != pcmk_rc_ok) { break; } } for (lpc = 0; lpc < max; lpc++) { free(namelist[lpc]); } free(namelist); } static gint schema_sort_GCompareFunc(gconstpointer a, gconstpointer b) { const pcmk__schema_t *schema_a = a; const pcmk__schema_t *schema_b = b; // @COMPAT pacemaker-next is deprecated since 2.1.5 and none since 2.1.8 if (pcmk__str_eq(schema_a->name, "pacemaker-next", pcmk__str_none)) { if (pcmk__str_eq(schema_b->name, PCMK_VALUE_NONE, pcmk__str_none)) { return -1; } else { return 1; } } else if (pcmk__str_eq(schema_a->name, PCMK_VALUE_NONE, pcmk__str_none)) { return 1; } else if (pcmk__str_eq(schema_b->name, "pacemaker-next", pcmk__str_none)) { return -1; } else { return schema_cmp(schema_a->version, schema_b->version); } } /*! * \internal * \brief Sort the list of known schemas such that all pacemaker-X.Y are in * version order, then pacemaker-next, then none * * This function should be called whenever additional schemas are loaded using * pcmk__load_schemas_from_dir(), after the initial sets in crm_schema_init(). */ void pcmk__sort_schemas(void) { known_schemas = g_list_sort(known_schemas, schema_sort_GCompareFunc); } /*! * \internal * \brief Load pacemaker schemas into cache * * \note This currently also serves as an entry point for the * generic initialization of the libxslt library. */ void crm_schema_init(void) { if (!initialized) { const char *remote_schema_dir = pcmk__remote_schema_dir(); char *base = pcmk__xml_artefact_root(pcmk__xml_artefact_ns_legacy_rng); const pcmk__schema_version_t zero = SCHEMA_ZERO; int schema_index = 0; initialized = true; wrap_libxslt(false); pcmk__load_schemas_from_dir(base); pcmk__load_schemas_from_dir(remote_schema_dir); free(base); // @COMPAT: Deprecated since 2.1.5 add_schema(pcmk__schema_validator_rng, &zero, "pacemaker-next", NULL, NULL, FALSE); // @COMPAT Deprecated since 2.1.8 add_schema(pcmk__schema_validator_none, &zero, PCMK_VALUE_NONE, NULL, NULL, FALSE); /* add_schema() prepends items to the list, so in the simple case, this * just reverses the list. However if there were any remote schemas, * sorting is necessary. */ pcmk__sort_schemas(); // Now set the schema indexes and log the final result for (GList *iter = known_schemas; iter != NULL; iter = iter->next) { pcmk__schema_t *schema = iter->data; if (schema->transform == NULL) { crm_debug("Loaded schema %d: %s", schema_index, schema->name); } else { crm_debug("Loaded schema %d: %s (upgrades with %s.xsl)", schema_index, schema->name, schema->transform); } schema->schema_index = schema_index++; } } } static bool validate_with_relaxng(xmlDocPtr doc, xmlRelaxNGValidityErrorFunc error_handler, void *error_handler_context, const char *relaxng_file, relaxng_ctx_cache_t **cached_ctx) { int rc = 0; bool valid = true; relaxng_ctx_cache_t *ctx = NULL; CRM_CHECK(doc != NULL, return false); CRM_CHECK(relaxng_file != NULL, return false); if (cached_ctx && *cached_ctx) { ctx = *cached_ctx; } else { crm_debug("Creating RNG parser context"); ctx = pcmk__assert_alloc(1, sizeof(relaxng_ctx_cache_t)); ctx->parser = xmlRelaxNGNewParserCtxt(relaxng_file); CRM_CHECK(ctx->parser != NULL, goto cleanup); if (error_handler) { xmlRelaxNGSetParserErrors(ctx->parser, (xmlRelaxNGValidityErrorFunc) error_handler, (xmlRelaxNGValidityWarningFunc) error_handler, error_handler_context); } else { xmlRelaxNGSetParserErrors(ctx->parser, (xmlRelaxNGValidityErrorFunc) fprintf, (xmlRelaxNGValidityWarningFunc) fprintf, stderr); } ctx->rng = xmlRelaxNGParse(ctx->parser); CRM_CHECK(ctx->rng != NULL, crm_err("Could not find/parse %s", relaxng_file); goto cleanup); ctx->valid = xmlRelaxNGNewValidCtxt(ctx->rng); CRM_CHECK(ctx->valid != NULL, goto cleanup); if (error_handler) { xmlRelaxNGSetValidErrors(ctx->valid, (xmlRelaxNGValidityErrorFunc) error_handler, (xmlRelaxNGValidityWarningFunc) error_handler, error_handler_context); } else { xmlRelaxNGSetValidErrors(ctx->valid, (xmlRelaxNGValidityErrorFunc) fprintf, (xmlRelaxNGValidityWarningFunc) fprintf, stderr); } } rc = xmlRelaxNGValidateDoc(ctx->valid, doc); if (rc > 0) { valid = false; } else if (rc < 0) { crm_err("Internal libxml error during validation"); } cleanup: if (cached_ctx) { *cached_ctx = ctx; } else { if (ctx->parser != NULL) { xmlRelaxNGFreeParserCtxt(ctx->parser); } if (ctx->valid != NULL) { xmlRelaxNGFreeValidCtxt(ctx->valid); } if (ctx->rng != NULL) { xmlRelaxNGFree(ctx->rng); } free(ctx); } return valid; } static void free_schema(gpointer data) { pcmk__schema_t *schema = data; relaxng_ctx_cache_t *ctx = NULL; switch (schema->validator) { case pcmk__schema_validator_none: // not cached break; case pcmk__schema_validator_rng: // cached ctx = (relaxng_ctx_cache_t *) schema->cache; if (ctx == NULL) { break; } if (ctx->parser != NULL) { xmlRelaxNGFreeParserCtxt(ctx->parser); } if (ctx->valid != NULL) { xmlRelaxNGFreeValidCtxt(ctx->valid); } if (ctx->rng != NULL) { xmlRelaxNGFree(ctx->rng); } free(ctx); schema->cache = NULL; break; } free(schema->name); free(schema->transform); free(schema->transform_enter); free(schema); } /*! * \internal * \brief Clean up global memory associated with XML schemas */ void crm_schema_cleanup(void) { if (known_schemas != NULL) { g_list_free_full(known_schemas, free_schema); known_schemas = NULL; } initialized = false; wrap_libxslt(true); } /*! * \internal * \brief Get schema list entry corresponding to a schema name * * \param[in] name Name of schema to get * * \return Schema list entry corresponding to \p name, or NULL if unknown */ GList * pcmk__get_schema(const char *name) { // @COMPAT Not specifying a schema name is deprecated since 2.1.8 if (name == NULL) { name = PCMK_VALUE_NONE; } for (GList *iter = known_schemas; iter != NULL; iter = iter->next) { pcmk__schema_t *schema = iter->data; if (pcmk__str_eq(name, schema->name, pcmk__str_casei)) { return iter; } } return NULL; } /*! * \internal * \brief Compare two schema version numbers given the schema names * * \param[in] schema1 Name of first schema to compare * \param[in] schema2 Name of second schema to compare * * \return Standard comparison result (negative integer if \p schema1 has the * lower version number, positive integer if \p schema1 has the higher * version number, of 0 if the version numbers are equal) */ int pcmk__cmp_schemas_by_name(const char *schema1_name, const char *schema2_name) { GList *entry1 = pcmk__get_schema(schema1_name); GList *entry2 = pcmk__get_schema(schema2_name); if (entry1 == NULL) { return (entry2 == NULL)? 0 : -1; } else if (entry2 == NULL) { return 1; } else { pcmk__schema_t *schema1 = entry1->data; pcmk__schema_t *schema2 = entry2->data; return schema1->schema_index - schema2->schema_index; } } static bool validate_with(xmlNode *xml, pcmk__schema_t *schema, xmlRelaxNGValidityErrorFunc error_handler, void *error_handler_context) { bool valid = false; char *file = NULL; relaxng_ctx_cache_t **cache = NULL; if (schema == NULL) { return false; } if (schema->validator == pcmk__schema_validator_none) { return true; } file = pcmk__xml_artefact_path(pcmk__xml_artefact_ns_legacy_rng, schema->name); crm_trace("Validating with %s (type=%d)", pcmk__s(file, "missing schema"), schema->validator); switch (schema->validator) { case pcmk__schema_validator_rng: cache = (relaxng_ctx_cache_t **) &(schema->cache); valid = validate_with_relaxng(xml->doc, error_handler, error_handler_context, file, cache); break; default: crm_err("Unknown validator type: %d", schema->validator); break; } free(file); return valid; } static bool validate_with_silent(xmlNode *xml, pcmk__schema_t *schema) { bool rc, sl_backup = silent_logging; silent_logging = TRUE; rc = validate_with(xml, schema, (xmlRelaxNGValidityErrorFunc) xml_log, GUINT_TO_POINTER(LOG_ERR)); silent_logging = sl_backup; return rc; } bool pcmk__validate_xml(xmlNode *xml_blob, const char *validation, xmlRelaxNGValidityErrorFunc error_handler, void *error_handler_context) { GList *entry = NULL; pcmk__schema_t *schema = NULL; CRM_CHECK((xml_blob != NULL) && (xml_blob->doc != NULL), return false); if (validation == NULL) { validation = crm_element_value(xml_blob, PCMK_XA_VALIDATE_WITH); } pcmk__warn_if_schema_deprecated(validation); // @COMPAT Not specifying a schema name is deprecated since 2.1.8 if (validation == NULL) { bool valid = false; for (entry = known_schemas; entry != NULL; entry = entry->next) { schema = entry->data; if (validate_with(xml_blob, schema, NULL, NULL)) { valid = true; crm_xml_add(xml_blob, PCMK_XA_VALIDATE_WITH, schema->name); crm_info("XML validated against %s", schema->name); } } return valid; } entry = pcmk__get_schema(validation); if (entry == NULL) { pcmk__config_err("Cannot validate CIB with " PCMK_XA_VALIDATE_WITH " set to an unknown schema such as '%s' (manually" " edit to use a known schema)", validation); return false; } schema = entry->data; return validate_with(xml_blob, schema, error_handler, error_handler_context); } /*! * \internal * \brief Validate XML using its configured schema (and send errors to logs) * * \param[in] xml XML to validate * * \return true if XML validates, otherwise false */ bool pcmk__configured_schema_validates(xmlNode *xml) { return pcmk__validate_xml(xml, NULL, (xmlRelaxNGValidityErrorFunc) xml_log, GUINT_TO_POINTER(LOG_ERR)); } /* With this arrangement, an attempt to identify the message severity as explicitly signalled directly from XSLT is performed in rather a smart way (no reliance on formatting string + arguments being always specified as ["%s", purposeful_string], as it can also be ["%s: %s", some_prefix, purposeful_string] etc. so every argument pertaining %s specifier is investigated), and if such a mark found, the respective level is determined and, when the messages are to go to the native logs, the mark itself gets dropped (by the means of string shift). NOTE: whether the native logging is the right sink is decided per the ctx parameter -- NULL denotes this case, otherwise it carries a pointer to the numeric expression of the desired target logging level (messages with higher level will be suppressed) NOTE: on some architectures, this string shift may not have any effect, but that's an acceptable tradeoff The logging level for not explicitly designated messages (suspicious, likely internal errors or some runaways) is LOG_WARNING. */ static void G_GNUC_PRINTF(2, 3) cib_upgrade_err(void *ctx, const char *fmt, ...) { va_list ap, aq; char *arg_cur; bool found = FALSE; const char *fmt_iter = fmt; uint8_t msg_log_level = LOG_WARNING; /* default for runaway messages */ const unsigned * log_level = (const unsigned *) ctx; enum { escan_seennothing, escan_seenpercent, } scan_state = escan_seennothing; va_start(ap, fmt); va_copy(aq, ap); while (!found && *fmt_iter != '\0') { /* while casing schema borrowed from libqb:qb_vsnprintf_serialize */ switch (*fmt_iter++) { case '%': if (scan_state == escan_seennothing) { scan_state = escan_seenpercent; } else if (scan_state == escan_seenpercent) { scan_state = escan_seennothing; } break; case 's': if (scan_state == escan_seenpercent) { scan_state = escan_seennothing; arg_cur = va_arg(aq, char *); if (arg_cur != NULL) { switch (arg_cur[0]) { case 'W': if (!strncmp(arg_cur, "WARNING: ", sizeof("WARNING: ") - 1)) { msg_log_level = LOG_WARNING; } if (ctx == NULL) { memmove(arg_cur, arg_cur + sizeof("WARNING: ") - 1, strlen(arg_cur + sizeof("WARNING: ") - 1) + 1); } found = TRUE; break; case 'I': if (!strncmp(arg_cur, "INFO: ", sizeof("INFO: ") - 1)) { msg_log_level = LOG_INFO; } if (ctx == NULL) { memmove(arg_cur, arg_cur + sizeof("INFO: ") - 1, strlen(arg_cur + sizeof("INFO: ") - 1) + 1); } found = TRUE; break; case 'D': if (!strncmp(arg_cur, "DEBUG: ", sizeof("DEBUG: ") - 1)) { msg_log_level = LOG_DEBUG; } if (ctx == NULL) { memmove(arg_cur, arg_cur + sizeof("DEBUG: ") - 1, strlen(arg_cur + sizeof("DEBUG: ") - 1) + 1); } found = TRUE; break; } } } break; case '#': case '-': case ' ': case '+': case '\'': case 'I': case '.': case '0': case '1': case '2': case '3': case '4': case '5': case '6': case '7': case '8': case '9': case '*': break; case 'l': case 'z': case 't': case 'j': case 'd': case 'i': case 'o': case 'u': case 'x': case 'X': case 'e': case 'E': case 'f': case 'F': case 'g': case 'G': case 'a': case 'A': case 'c': case 'p': if (scan_state == escan_seenpercent) { (void) va_arg(aq, void *); /* skip forward */ scan_state = escan_seennothing; } break; default: scan_state = escan_seennothing; break; } } if (log_level != NULL) { /* intention of the following offset is: cibadmin -V -> start showing INFO labelled messages */ if (*log_level + 4 >= msg_log_level) { vfprintf(stderr, fmt, ap); } } else { PCMK__XML_LOG_BASE(msg_log_level, TRUE, 0, "CIB upgrade: ", fmt, ap); } va_end(aq); va_end(ap); } /*! * \internal * \brief Apply a single XSL transformation to given XML * * \param[in] xml XML to transform * \param[in] transform XSL name * \param[in] to_logs If false, certain validation errors will be sent to * stderr rather than logged * * \return Transformed XML on success, otherwise NULL */ static xmlNode * apply_transformation(const xmlNode *xml, const char *transform, gboolean to_logs) { char *xform = NULL; xmlNode *out = NULL; xmlDocPtr res = NULL; xsltStylesheet *xslt = NULL; xform = pcmk__xml_artefact_path(pcmk__xml_artefact_ns_legacy_xslt, transform); /* for capturing, e.g., what's emitted via */ if (to_logs) { xsltSetGenericErrorFunc(NULL, cib_upgrade_err); } else { xsltSetGenericErrorFunc(&crm_log_level, cib_upgrade_err); } xslt = xsltParseStylesheetFile((pcmkXmlStr) xform); CRM_CHECK(xslt != NULL, goto cleanup); /* Caller allocates private data for final result document. Intermediate * result documents are temporary and don't need private data. */ res = xsltApplyStylesheet(xslt, xml->doc, NULL); CRM_CHECK(res != NULL, goto cleanup); xsltSetGenericErrorFunc(NULL, NULL); /* restore default one */ out = xmlDocGetRootElement(res); cleanup: if (xslt) { xsltFreeStylesheet(xslt); } free(xform); return out; } /*! * \internal * \brief Perform all transformations needed to upgrade XML to next schema * * A schema upgrade can require up to three XSL transformations: an "enter" * transform, the main upgrade transform, and a "leave" transform. Perform * all needed transforms to upgrade given XML to the next schema. * * \param[in] original_xml XML to transform * \param[in] schema_index Index of schema that successfully validates * \p original_xml * \param[in] to_logs If false, certain validation errors will be sent to * stderr rather than logged * * \return XML result of schema transforms if successful, otherwise NULL */ static xmlNode * apply_upgrade(const xmlNode *original_xml, int schema_index, gboolean to_logs) { pcmk__schema_t *schema = g_list_nth_data(known_schemas, schema_index); pcmk__schema_t *upgraded_schema = g_list_nth_data(known_schemas, schema_index + 1); bool transform_onleave = false; char *transform_leave; const xmlNode *xml = original_xml; xmlNode *upgrade = NULL; xmlNode *final = NULL; xmlRelaxNGValidityErrorFunc error_handler = NULL; pcmk__assert((schema != NULL) && (upgraded_schema != NULL)); if (to_logs) { error_handler = (xmlRelaxNGValidityErrorFunc) xml_log; } transform_onleave = schema->transform_onleave; if (schema->transform_enter != NULL) { crm_debug("Upgrading schema from %s to %s: " "applying pre-upgrade XSL transform %s", schema->name, upgraded_schema->name, schema->transform_enter); upgrade = apply_transformation(xml, schema->transform_enter, to_logs); if (upgrade == NULL) { crm_warn("Pre-upgrade XSL transform %s failed, " "will skip post-upgrade transform", schema->transform_enter); transform_onleave = FALSE; } else { xml = upgrade; } } crm_debug("Upgrading schema from %s to %s: " "applying upgrade XSL transform %s", schema->name, upgraded_schema->name, schema->transform); final = apply_transformation(xml, schema->transform, to_logs); if (upgrade != xml) { free_xml(upgrade); /* upgrade = NULL; */ // Static analysis dislikes this, so be careful } if ((final != NULL) && transform_onleave) { upgrade = final; /* following condition ensured in add_schema_by_version */ pcmk__assert(schema->transform_enter != NULL); - transform_leave = strdup(schema->transform_enter); + transform_leave = pcmk__str_copy(schema->transform_enter); /* enter -> leave */ + // coverity[returned_null] Not worth correcting this code memcpy(strrchr(transform_leave, '-') + 1, "leave", sizeof("leave") - 1); crm_debug("Upgrading schema from %s to %s: " "applying post-upgrade XSL transform %s", schema->name, upgraded_schema->name, transform_leave); final = apply_transformation(upgrade, transform_leave, to_logs); if (final == NULL) { crm_warn("Ignoring failure of post-upgrade XSL transform %s", transform_leave); final = upgrade; } else { free_xml(upgrade); } free(transform_leave); } if (final == NULL) { return NULL; } // Final result document from upgrade pipeline needs private data pcmk__xml_new_private_data((xmlNode *) final->doc); // Ensure result validates with its new schema if (!validate_with(final, upgraded_schema, error_handler, GUINT_TO_POINTER(LOG_ERR))) { crm_err("Schema upgrade from %s to %s failed: " "XSL transform %s produced an invalid configuration", schema->name, upgraded_schema->name, schema->transform); crm_log_xml_debug(final, "bad-transform-result"); free_xml(final); return NULL; } crm_info("Schema upgrade from %s to %s succeeded", schema->name, upgraded_schema->name); return final; } /*! * \internal * \brief Get the schema list entry corresponding to XML configuration * * \param[in] xml CIB XML to check * * \return List entry of schema configured in \p xml */ static GList * get_configured_schema(const xmlNode *xml) { const char *schema_name = crm_element_value(xml, PCMK_XA_VALIDATE_WITH); pcmk__warn_if_schema_deprecated(schema_name); if (schema_name == NULL) { return NULL; } return pcmk__get_schema(schema_name); } /*! * \brief Update CIB XML to latest schema that validates it * * \param[in,out] xml XML to update (may be freed and replaced * after being transformed) * \param[in] max_schema_name If not NULL, do not update \p xml to any * schema later than this one * \param[in] transform If false, do not update \p xml to any schema * that requires an XSL transform * \param[in] to_logs If false, certain validation errors will be * sent to stderr rather than logged * * \return Standard Pacemaker return code */ int pcmk__update_schema(xmlNode **xml, const char *max_schema_name, bool transform, bool to_logs) { int max_stable_schemas = xml_latest_schema_index(); int max_schema_index = 0; int rc = pcmk_rc_ok; GList *entry = NULL; pcmk__schema_t *best_schema = NULL; pcmk__schema_t *original_schema = NULL; xmlRelaxNGValidityErrorFunc error_handler = to_logs ? (xmlRelaxNGValidityErrorFunc) xml_log : NULL; CRM_CHECK((xml != NULL) && (*xml != NULL) && ((*xml)->doc != NULL), return EINVAL); if (max_schema_name != NULL) { GList *max_entry = pcmk__get_schema(max_schema_name); if (max_entry != NULL) { pcmk__schema_t *max_schema = max_entry->data; max_schema_index = max_schema->schema_index; } } if ((max_schema_index < 1) || (max_schema_index > max_stable_schemas)) { max_schema_index = max_stable_schemas; } entry = get_configured_schema(*xml); if (entry == NULL) { // @COMPAT Not specifying a schema name is deprecated since 2.1.8 entry = known_schemas; } else { original_schema = entry->data; if (original_schema->schema_index >= max_schema_index) { return pcmk_rc_ok; } } for (; entry != NULL; entry = entry->next) { pcmk__schema_t *current_schema = entry->data; xmlNode *upgrade = NULL; if (current_schema->schema_index > max_schema_index) { break; } if (!validate_with(*xml, current_schema, error_handler, GUINT_TO_POINTER(LOG_ERR))) { crm_debug("Schema %s does not validate", current_schema->name); if (best_schema != NULL) { /* we've satisfied the validation, no need to check further */ break; } rc = pcmk_rc_schema_validation; continue; // Try again with the next higher schema } crm_debug("Schema %s validates", current_schema->name); rc = pcmk_rc_ok; best_schema = current_schema; if (current_schema->schema_index == max_schema_index) { break; // No further transformations possible } + // coverity[null_field] The index check ensures entry->next is not NULL if (!transform || (current_schema->transform == NULL) || validate_with_silent(*xml, entry->next->data)) { /* The next schema either doesn't require a transform or validates * successfully even without the transform. Skip the transform and * try the next schema with the same XML. */ continue; } upgrade = apply_upgrade(*xml, current_schema->schema_index, to_logs); if (upgrade == NULL) { /* The transform failed, so this schema can't be used. Later * schemas are unlikely to validate, but try anyway until we * run out of options. */ rc = pcmk_rc_transform_failed; } else { best_schema = current_schema; free_xml(*xml); *xml = upgrade; } } if (best_schema != NULL) { if ((original_schema == NULL) || (best_schema->schema_index > original_schema->schema_index)) { crm_info("%s the configuration schema to %s", (transform? "Transformed" : "Upgraded"), best_schema->name); crm_xml_add(*xml, PCMK_XA_VALIDATE_WITH, best_schema->name); } } return rc; } int pcmk_update_configured_schema(xmlNode **xml) { return pcmk__update_configured_schema(xml, true); } /*! * \brief Update XML from its configured schema to the latest major series * * \param[in,out] xml XML to update * \param[in] to_logs If false, certain validation errors will be * sent to stderr rather than logged * * \return Standard Pacemaker return code */ int pcmk__update_configured_schema(xmlNode **xml, bool to_logs) { int rc = pcmk_rc_ok; char *original_schema_name = NULL; // @COMPAT Not specifying a schema name is deprecated since 2.1.8 const char *effective_original_name = "the first"; int orig_version = -1; pcmk__schema_t *x_0_schema = pcmk__find_x_0_schema()->data; GList *entry = NULL; CRM_CHECK(xml != NULL, return EINVAL); original_schema_name = crm_element_value_copy(*xml, PCMK_XA_VALIDATE_WITH); pcmk__warn_if_schema_deprecated(original_schema_name); entry = pcmk__get_schema(original_schema_name); if (entry != NULL) { pcmk__schema_t *original_schema = entry->data; effective_original_name = original_schema->name; orig_version = original_schema->schema_index; } if (orig_version < x_0_schema->schema_index) { // Current configuration schema is not acceptable, try to update xmlNode *converted = NULL; const char *new_schema_name = NULL; pcmk__schema_t *schema = NULL; entry = NULL; converted = pcmk__xml_copy(NULL, *xml); if (pcmk__update_schema(&converted, NULL, true, to_logs) == pcmk_rc_ok) { new_schema_name = crm_element_value(converted, PCMK_XA_VALIDATE_WITH); entry = pcmk__get_schema(new_schema_name); } schema = (entry == NULL)? NULL : entry->data; if ((schema == NULL) || (schema->schema_index < x_0_schema->schema_index)) { // Updated configuration schema is still not acceptable if ((orig_version == -1) || (schema == NULL) || (schema->schema_index < orig_version)) { // We couldn't validate any schema at all if (to_logs) { pcmk__config_err("Cannot upgrade configuration (claiming " "%s schema) to at least %s because it " "does not validate with any schema from " "%s to the latest", pcmk__s(original_schema_name, "no"), x_0_schema->name, effective_original_name); } else { fprintf(stderr, "Cannot upgrade configuration (claiming " "%s schema) to at least %s because it " "does not validate with any schema from " "%s to the latest\n", pcmk__s(original_schema_name, "no"), x_0_schema->name, effective_original_name); } } else { // We updated configuration successfully, but still too low if (to_logs) { pcmk__config_err("Cannot upgrade configuration (claiming " "%s schema) to at least %s because it " "would not upgrade past %s", pcmk__s(original_schema_name, "no"), x_0_schema->name, pcmk__s(new_schema_name, "unspecified version")); } else { fprintf(stderr, "Cannot upgrade configuration (claiming " "%s schema) to at least %s because it " "would not upgrade past %s\n", pcmk__s(original_schema_name, "no"), x_0_schema->name, pcmk__s(new_schema_name, "unspecified version")); } } free_xml(converted); converted = NULL; rc = pcmk_rc_transform_failed; } else { // Updated configuration schema is acceptable free_xml(*xml); *xml = converted; if (schema->schema_index < xml_latest_schema_index()) { if (to_logs) { pcmk__config_warn("Configuration with %s schema was " "internally upgraded to acceptable (but " "not most recent) %s", pcmk__s(original_schema_name, "no"), schema->name); } } else if (to_logs) { crm_info("Configuration with %s schema was internally " "upgraded to latest version %s", pcmk__s(original_schema_name, "no"), schema->name); } } } else { // @COMPAT the none schema is deprecated since 2.1.8 pcmk__schema_t *none_schema = NULL; entry = pcmk__get_schema(PCMK_VALUE_NONE); pcmk__assert((entry != NULL) && (entry->data != NULL)); none_schema = entry->data; if (!to_logs && (orig_version >= none_schema->schema_index)) { fprintf(stderr, "Schema validation of configuration is " "disabled (support for " PCMK_XA_VALIDATE_WITH " set to \"" PCMK_VALUE_NONE "\" is deprecated" " and will be removed in a future release)\n"); } } free(original_schema_name); return rc; } /*! * \internal * \brief Return a list of all schema files and any associated XSLT files * later than the given one * \brief Return a list of all schema versions later than the given one * * \param[in] schema The schema to compare against (for example, * "pacemaker-3.1.rng" or "pacemaker-3.1") * * \note The caller is responsible for freeing both the returned list and * the elements of the list */ GList * pcmk__schema_files_later_than(const char *name) { GList *lst = NULL; pcmk__schema_version_t ver; if (!version_from_filename(name, &ver)) { return lst; } for (GList *iter = g_list_nth(known_schemas, xml_latest_schema_index()); iter != NULL; iter = iter->prev) { pcmk__schema_t *schema = iter->data; char *s = NULL; if (schema_cmp(ver, schema->version) != -1) { continue; } s = crm_strdup_printf("%s.rng", schema->name); lst = g_list_prepend(lst, s); if (schema->transform != NULL) { char *xform = crm_strdup_printf("%s.xsl", schema->transform); lst = g_list_prepend(lst, xform); } if (schema->transform_enter != NULL) { char *enter = crm_strdup_printf("%s.xsl", schema->transform_enter); lst = g_list_prepend(lst, enter); if (schema->transform_onleave) { int last_dash = strrchr(enter, '-') - enter; char *leave = crm_strdup_printf("%.*s-leave.xsl", last_dash, enter); lst = g_list_prepend(lst, leave); } } } return lst; } static void append_href(xmlNode *xml, void *user_data) { GList **list = user_data; char *href = crm_element_value_copy(xml, "href"); if (href == NULL) { return; } *list = g_list_prepend(*list, href); } static void external_refs_in_schema(GList **list, const char *contents) { /* local-name()= is needed to ignore the xmlns= setting at the top of * the XML file. Otherwise, the xpath query will always return nothing. */ const char *search = "//*[local-name()='externalRef'] | //*[local-name()='include']"; xmlNode *xml = pcmk__xml_parse(contents); crm_foreach_xpath_result(xml, search, append_href, list); free_xml(xml); } static int read_file_contents(const char *file, char **contents) { int rc = pcmk_rc_ok; char *path = NULL; if (pcmk__ends_with(file, ".rng")) { path = pcmk__xml_artefact_path(pcmk__xml_artefact_ns_legacy_rng, file); } else { path = pcmk__xml_artefact_path(pcmk__xml_artefact_ns_legacy_xslt, file); } rc = pcmk__file_contents(path, contents); free(path); return rc; } static void add_schema_file_to_xml(xmlNode *parent, const char *file, GList **already_included) { char *contents = NULL; char *path = NULL; xmlNode *file_node = NULL; GList *includes = NULL; int rc = pcmk_rc_ok; /* If we already included this file, don't do so again. */ if (g_list_find_custom(*already_included, file, (GCompareFunc) strcmp) != NULL) { return; } /* Ensure whatever file we were given has a suffix we know about. If not, * just assume it's an RNG file. */ if (!pcmk__ends_with(file, ".rng") && !pcmk__ends_with(file, ".xsl")) { path = crm_strdup_printf("%s.rng", file); } else { path = pcmk__str_copy(file); } rc = read_file_contents(path, &contents); if (rc != pcmk_rc_ok || contents == NULL) { crm_warn("Could not read schema file %s: %s", file, pcmk_rc_str(rc)); free(path); return; } /* Create a new node with the contents of the file * as a CDATA block underneath it. */ file_node = pcmk__xe_create(parent, PCMK_XA_FILE); crm_xml_add(file_node, PCMK_XA_PATH, path); *already_included = g_list_prepend(*already_included, path); xmlAddChild(file_node, xmlNewCDataBlock(parent->doc, (pcmkXmlStr) contents, strlen(contents))); /* Scan the file for any or nodes and build up * a list of the files they reference. */ external_refs_in_schema(&includes, contents); /* For each referenced file, recurse to add it (and potentially anything it * references, ...) to the XML. */ for (GList *iter = includes; iter != NULL; iter = iter->next) { add_schema_file_to_xml(parent, iter->data, already_included); } free(contents); g_list_free_full(includes, free); } /*! * \internal * \brief Add an XML schema file and all the files it references as children * of a given XML node * * \param[in,out] parent The parent XML node * \param[in] name The schema version to compare against * (for example, "pacemaker-3.1" or "pacemaker-3.1.rng") * \param[in,out] already_included A list of names that have already been added * to the parent node. * * \note The caller is responsible for freeing both the returned list and * the elements of the list */ void pcmk__build_schema_xml_node(xmlNode *parent, const char *name, GList **already_included) { xmlNode *schema_node = pcmk__xe_create(parent, PCMK__XA_SCHEMA); crm_xml_add(schema_node, PCMK_XA_VERSION, name); add_schema_file_to_xml(schema_node, name, already_included); if (schema_node->children == NULL) { // Not needed if empty. May happen if name was invalid, for example. free_xml(schema_node); } } /*! * \internal * \brief Return the directory containing any extra schema files that a * Pacemaker Remote node fetched from the cluster */ const char * pcmk__remote_schema_dir(void) { const char *dir = pcmk__env_option(PCMK__ENV_REMOTE_SCHEMA_DIRECTORY); if (pcmk__str_empty(dir)) { return PCMK__REMOTE_SCHEMA_DIR; } return dir; } /*! * \internal * \brief Warn if a given validation schema is deprecated * * \param[in] Schema name to check */ void pcmk__warn_if_schema_deprecated(const char *schema) { if ((schema == NULL) || pcmk__strcase_any_of(schema, "pacemaker-next", PCMK_VALUE_NONE, NULL)) { pcmk__config_warn("Support for " PCMK_XA_VALIDATE_WITH "='%s' is " "deprecated and will be removed in a future release " "without the possibility of upgrades (manually edit " "to use a supported schema)", pcmk__s(schema, "")); } } // Deprecated functions kept only for backward API compatibility // LCOV_EXCL_START #include const char * xml_latest_schema(void) { return pcmk__highest_schema_name(); } const char * get_schema_name(int version) { pcmk__schema_t *schema = g_list_nth_data(known_schemas, version); return (schema != NULL)? schema->name : "unknown"; } int get_schema_version(const char *name) { int lpc = 0; if (name == NULL) { name = PCMK_VALUE_NONE; } for (GList *iter = known_schemas; iter != NULL; iter = iter->next) { pcmk__schema_t *schema = iter->data; if (pcmk__str_eq(name, schema->name, pcmk__str_casei)) { return lpc; } lpc++; } return -1; } int update_validation(xmlNode **xml, int *best, int max, gboolean transform, gboolean to_logs) { int rc = pcmk__update_schema(xml, get_schema_name(max), transform, to_logs); if ((best != NULL) && (xml != NULL) && (rc == pcmk_rc_ok)) { const char *schema_name = crm_element_value(*xml, PCMK_XA_VALIDATE_WITH); GList *schema_entry = pcmk__get_schema(schema_name); if (schema_entry != NULL) { *best = ((pcmk__schema_t *)(schema_entry->data))->schema_index; } } return pcmk_rc2legacy(rc); } gboolean validate_xml(xmlNode *xml_blob, const char *validation, gboolean to_logs) { bool rc = pcmk__validate_xml(xml_blob, validation, to_logs? (xmlRelaxNGValidityErrorFunc) xml_log : NULL, GUINT_TO_POINTER(LOG_ERR)); return rc? TRUE : FALSE; } static void dump_file(const char *filename) { FILE *fp = NULL; int ch, line = 0; CRM_CHECK(filename != NULL, return); fp = fopen(filename, "r"); if (fp == NULL) { crm_perror(LOG_ERR, "Could not open %s for reading", filename); return; } fprintf(stderr, "%4d ", ++line); do { ch = getc(fp); if (ch == EOF) { putc('\n', stderr); break; } else if (ch == '\n') { fprintf(stderr, "\n%4d ", ++line); } else { putc(ch, stderr); } } while (1); fclose(fp); } gboolean validate_xml_verbose(const xmlNode *xml_blob) { int fd = 0; xmlDoc *doc = NULL; xmlNode *xml = NULL; gboolean rc = FALSE; char *filename = NULL; filename = crm_strdup_printf("%s/cib-invalid.XXXXXX", pcmk__get_tmpdir()); umask(S_IWGRP | S_IWOTH | S_IROTH); fd = mkstemp(filename); pcmk__xml_write_fd(xml_blob, filename, fd, false, NULL); dump_file(filename); doc = xmlReadFile(filename, NULL, 0); xml = xmlDocGetRootElement(doc); rc = pcmk__validate_xml(xml, NULL, NULL, NULL); free_xml(xml); unlink(filename); free(filename); return rc? TRUE : FALSE; } gboolean cli_config_update(xmlNode **xml, int *best_version, gboolean to_logs) { int rc = pcmk__update_configured_schema(xml, to_logs); if (best_version != NULL) { const char *name = crm_element_value(*xml, PCMK_XA_VALIDATE_WITH); if (name == NULL) { *best_version = -1; } else { GList *entry = pcmk__get_schema(name); pcmk__schema_t *schema = (entry == NULL)? NULL : entry->data; *best_version = (schema == NULL)? -1 : schema->schema_index; } } return (rc == pcmk_rc_ok)? TRUE: FALSE; } // LCOV_EXCL_STOP // End deprecated API diff --git a/lib/common/xml_io.c b/lib/common/xml_io.c index 41d3434ecb..0ce8837ab0 100644 --- a/lib/common/xml_io.c +++ b/lib/common/xml_io.c @@ -1,842 +1,846 @@ /* - * Copyright 2004-2024 the Pacemaker project contributors + * Copyright 2004-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include // xmlOutputBuffer* #include #include #include #include "crmcommon_private.h" /* @COMPAT XML_PARSE_RECOVER allows some XML errors to be silently worked around * by libxml2, which is potentially ambiguous and dangerous. We should drop it * when we can break backward compatibility with configurations that might be * relying on it (i.e. pacemaker 3.0.0). */ #define PCMK__XML_PARSE_OPTS_WITHOUT_RECOVER (XML_PARSE_NOBLANKS) #define PCMK__XML_PARSE_OPTS_WITH_RECOVER (XML_PARSE_NOBLANKS \ |XML_PARSE_RECOVER) /*! * \internal * \brief Read from \c stdin until EOF or error * * \return Newly allocated string containing the bytes read from \c stdin, or * \c NULL on error * * \note The caller is responsible for freeing the return value using \c free(). */ static char * read_stdin(void) { char *buf = NULL; size_t length = 0; do { buf = pcmk__realloc(buf, length + PCMK__BUFFER_SIZE + 1); length += fread(buf + length, 1, PCMK__BUFFER_SIZE, stdin); } while ((feof(stdin) == 0) && (ferror(stdin) == 0)); if (ferror(stdin) != 0) { crm_err("Error reading input from stdin"); free(buf); buf = NULL; } else { buf[length] = '\0'; } clearerr(stdin); return buf; } /*! * \internal * \brief Decompress a bzip2-compressed file into a string buffer * * \param[in] filename Name of file to decompress * * \return Newly allocated string with the decompressed contents of \p filename, * or \c NULL on error. * * \note The caller is responsible for freeing the return value using \c free(). */ static char * decompress_file(const char *filename) { char *buffer = NULL; int rc = pcmk_rc_ok; size_t length = 0; BZFILE *bz_file = NULL; FILE *input = fopen(filename, "r"); if (input == NULL) { crm_perror(LOG_ERR, "Could not open %s for reading", filename); return NULL; } bz_file = BZ2_bzReadOpen(&rc, input, 0, 0, NULL, 0); rc = pcmk__bzlib2rc(rc); if (rc != pcmk_rc_ok) { crm_err("Could not prepare to read compressed %s: %s " CRM_XS " rc=%d", filename, pcmk_rc_str(rc), rc); goto done; } // cppcheck seems not to understand the abort-logic in pcmk__realloc // cppcheck-suppress memleak do { int read_len = 0; buffer = pcmk__realloc(buffer, length + PCMK__BUFFER_SIZE + 1); read_len = BZ2_bzRead(&rc, bz_file, buffer + length, PCMK__BUFFER_SIZE); if ((rc == BZ_OK) || (rc == BZ_STREAM_END)) { crm_trace("Read %ld bytes from file: %d", (long) read_len, rc); length += read_len; } } while (rc == BZ_OK); rc = pcmk__bzlib2rc(rc); if (rc != pcmk_rc_ok) { rc = pcmk__bzlib2rc(rc); crm_err("Could not read compressed %s: %s " CRM_XS " rc=%d", filename, pcmk_rc_str(rc), rc); free(buffer); buffer = NULL; } else { buffer[length] = '\0'; } done: BZ2_bzReadClose(&rc, bz_file); fclose(input); return buffer; } // @COMPAT Remove macro at 3.0.0 when we drop XML_PARSE_RECOVER /*! * \internal * \brief Try to parse XML first without and then with recovery enabled * * \param[out] result Where to store the resulting XML doc (xmlDoc **) * \param[in] fn XML parser function * \param[in] ... All arguments for \p fn except the final one (an * \c xmlParserOption group) */ #define parse_xml_recover(result, fn, ...) do { \ *result = fn(__VA_ARGS__, PCMK__XML_PARSE_OPTS_WITHOUT_RECOVER); \ if (*result == NULL) { \ *result = fn(__VA_ARGS__, PCMK__XML_PARSE_OPTS_WITH_RECOVER); \ \ if (*result != NULL) { \ crm_warn("Successfully recovered from XML errors " \ "(note: a future release will treat this as a " \ "fatal failure)"); \ } \ } \ } while (0); /*! * \internal * \brief Parse XML from a file * * \param[in] filename Name of file containing XML (\c NULL or \c "-" for * \c stdin); if \p filename ends in \c ".bz2", the file * will be decompressed using \c bzip2 * * \return XML tree parsed from the given file; may be \c NULL or only partial * on error */ xmlNode * pcmk__xml_read(const char *filename) { bool use_stdin = pcmk__str_eq(filename, "-", pcmk__str_null_matches); xmlNode *xml = NULL; xmlDoc *output = NULL; xmlParserCtxt *ctxt = NULL; const xmlError *last_error = NULL; // Create a parser context ctxt = xmlNewParserCtxt(); CRM_CHECK(ctxt != NULL, return NULL); xmlCtxtResetLastError(ctxt); xmlSetGenericErrorFunc(ctxt, pcmk__log_xmllib_err); if (use_stdin) { /* @COMPAT After dropping XML_PARSE_RECOVER, we can avoid capturing * stdin into a buffer and instead call * xmlCtxtReadFd(ctxt, STDIN_FILENO, NULL, NULL, XML_PARSE_NOBLANKS); * * For now we have to save the input so that we can use it twice. */ char *input = read_stdin(); if (input != NULL) { + // coverity[unsafe_xml_parse_config] parse_xml_recover(&output, xmlCtxtReadDoc, ctxt, (pcmkXmlStr) input, NULL, NULL); free(input); } } else if (pcmk__ends_with_ext(filename, ".bz2")) { char *input = decompress_file(filename); if (input != NULL) { + // coverity[unsafe_xml_parse_config] parse_xml_recover(&output, xmlCtxtReadDoc, ctxt, (pcmkXmlStr) input, NULL, NULL); free(input); } } else { + // coverity[unsafe_xml_parse_config] parse_xml_recover(&output, xmlCtxtReadFile, ctxt, filename, NULL); } if (output != NULL) { pcmk__xml_new_private_data((xmlNode *) output); xml = xmlDocGetRootElement(output); if (xml != NULL) { /* @TODO Should we really be stripping out text? This seems like an * overly broad way to get rid of whitespace, if that's the goal. * Text nodes may be invalid in most or all Pacemaker inputs, but * stripping them in a generic "parse XML from file" function may * not be the best way to ignore them. */ pcmk__strip_xml_text(xml); } } // @COMPAT At 3.0.0, free xml and return NULL if xml != NULL on error last_error = xmlCtxtGetLastError(ctxt); if (last_error != NULL) { if (xml != NULL) { crm_log_xml_info(xml, "Partial"); } } xmlFreeParserCtxt(ctxt); return xml; } /*! * \internal * \brief Parse XML from a string * * \param[in] input String to parse * * \return XML tree parsed from the given string; may be \c NULL or only partial * on error */ xmlNode * pcmk__xml_parse(const char *input) { xmlNode *xml = NULL; xmlDoc *output = NULL; xmlParserCtxt *ctxt = NULL; const xmlError *last_error = NULL; if (input == NULL) { return NULL; } ctxt = xmlNewParserCtxt(); if (ctxt == NULL) { return NULL; } xmlCtxtResetLastError(ctxt); xmlSetGenericErrorFunc(ctxt, pcmk__log_xmllib_err); + // coverity[unsafe_xml_parse_config] parse_xml_recover(&output, xmlCtxtReadDoc, ctxt, (pcmkXmlStr) input, NULL, NULL); if (output != NULL) { pcmk__xml_new_private_data((xmlNode *) output); xml = xmlDocGetRootElement(output); } // @COMPAT At 3.0.0, free xml and return NULL if xml != NULL; update doxygen last_error = xmlCtxtGetLastError(ctxt); if (last_error != NULL) { if (xml != NULL) { crm_log_xml_info(xml, "Partial"); } } xmlFreeParserCtxt(ctxt); return xml; } /*! * \internal * \brief Append a string representation of an XML element to a buffer * * \param[in] data XML whose representation to append * \param[in] options Group of \p pcmk__xml_fmt_options flags * \param[in,out] buffer Where to append the content (must not be \p NULL) * \param[in] depth Current indentation level */ static void dump_xml_element(const xmlNode *data, uint32_t options, GString *buffer, int depth) { bool pretty = pcmk_is_set(options, pcmk__xml_fmt_pretty); bool filtered = pcmk_is_set(options, pcmk__xml_fmt_filtered); int spaces = pretty? (2 * depth) : 0; for (int lpc = 0; lpc < spaces; lpc++) { g_string_append_c(buffer, ' '); } pcmk__g_strcat(buffer, "<", data->name, NULL); for (const xmlAttr *attr = pcmk__xe_first_attr(data); attr != NULL; attr = attr->next) { if (!filtered || !pcmk__xa_filterable((const char *) (attr->name))) { pcmk__dump_xml_attr(attr, buffer); } } if (data->children == NULL) { g_string_append(buffer, "/>"); } else { g_string_append_c(buffer, '>'); } if (pretty) { g_string_append_c(buffer, '\n'); } if (data->children) { for (const xmlNode *child = data->children; child != NULL; child = child->next) { pcmk__xml_string(child, options, buffer, depth + 1); } for (int lpc = 0; lpc < spaces; lpc++) { g_string_append_c(buffer, ' '); } pcmk__g_strcat(buffer, "name, ">", NULL); if (pretty) { g_string_append_c(buffer, '\n'); } } } /*! * \internal * \brief Append XML text content to a buffer * * \param[in] data XML whose content to append * \param[in] options Group of \p xml_log_options flags * \param[in,out] buffer Where to append the content (must not be \p NULL) * \param[in] depth Current indentation level */ static void dump_xml_text(const xmlNode *data, uint32_t options, GString *buffer, int depth) { bool pretty = pcmk_is_set(options, pcmk__xml_fmt_pretty); int spaces = pretty? (2 * depth) : 0; const char *content = (const char *) data->content; gchar *content_esc = NULL; if (pcmk__xml_needs_escape(content, pcmk__xml_escape_text)) { content_esc = pcmk__xml_escape(content, pcmk__xml_escape_text); content = content_esc; } for (int lpc = 0; lpc < spaces; lpc++) { g_string_append_c(buffer, ' '); } g_string_append(buffer, content); if (pretty) { g_string_append_c(buffer, '\n'); } g_free(content_esc); } /*! * \internal * \brief Append XML CDATA content to a buffer * * \param[in] data XML whose content to append * \param[in] options Group of \p pcmk__xml_fmt_options flags * \param[in,out] buffer Where to append the content (must not be \p NULL) * \param[in] depth Current indentation level */ static void dump_xml_cdata(const xmlNode *data, uint32_t options, GString *buffer, int depth) { bool pretty = pcmk_is_set(options, pcmk__xml_fmt_pretty); int spaces = pretty? (2 * depth) : 0; for (int lpc = 0; lpc < spaces; lpc++) { g_string_append_c(buffer, ' '); } pcmk__g_strcat(buffer, "content, "]]>", NULL); if (pretty) { g_string_append_c(buffer, '\n'); } } /*! * \internal * \brief Append an XML comment to a buffer * * \param[in] data XML whose content to append * \param[in] options Group of \p pcmk__xml_fmt_options flags * \param[in,out] buffer Where to append the content (must not be \p NULL) * \param[in] depth Current indentation level */ static void dump_xml_comment(const xmlNode *data, uint32_t options, GString *buffer, int depth) { bool pretty = pcmk_is_set(options, pcmk__xml_fmt_pretty); int spaces = pretty? (2 * depth) : 0; for (int lpc = 0; lpc < spaces; lpc++) { g_string_append_c(buffer, ' '); } pcmk__g_strcat(buffer, "", NULL); if (pretty) { g_string_append_c(buffer, '\n'); } } /*! * \internal * \brief Get a string representation of an XML element type * * \param[in] type XML element type * * \return String representation of \p type */ static const char * xml_element_type_text(xmlElementType type) { static const char *const element_type_names[] = { [XML_ELEMENT_NODE] = "element", [XML_ATTRIBUTE_NODE] = "attribute", [XML_TEXT_NODE] = "text", [XML_CDATA_SECTION_NODE] = "CDATA section", [XML_ENTITY_REF_NODE] = "entity reference", [XML_ENTITY_NODE] = "entity", [XML_PI_NODE] = "PI", [XML_COMMENT_NODE] = "comment", [XML_DOCUMENT_NODE] = "document", [XML_DOCUMENT_TYPE_NODE] = "document type", [XML_DOCUMENT_FRAG_NODE] = "document fragment", [XML_NOTATION_NODE] = "notation", [XML_HTML_DOCUMENT_NODE] = "HTML document", [XML_DTD_NODE] = "DTD", [XML_ELEMENT_DECL] = "element declaration", [XML_ATTRIBUTE_DECL] = "attribute declaration", [XML_ENTITY_DECL] = "entity declaration", [XML_NAMESPACE_DECL] = "namespace declaration", [XML_XINCLUDE_START] = "XInclude start", [XML_XINCLUDE_END] = "XInclude end", }; if ((type < 0) || (type >= PCMK__NELEM(element_type_names))) { return "unrecognized type"; } return element_type_names[type]; } /*! * \internal * \brief Create a string representation of an XML object * * libxml2's \c xmlNodeDumpOutput() doesn't allow filtering, doesn't escape * special characters thoroughly, and doesn't allow a const argument. * * \param[in] data XML to convert * \param[in] options Group of \p pcmk__xml_fmt_options flags * \param[in,out] buffer Where to store the text (must not be \p NULL) * \param[in] depth Current indentation level * * \todo Create a wrapper that doesn't require \p depth. Only used with * recursive calls currently. */ void pcmk__xml_string(const xmlNode *data, uint32_t options, GString *buffer, int depth) { if (data == NULL) { crm_trace("Nothing to dump"); return; } pcmk__assert(buffer != NULL); CRM_CHECK(depth >= 0, depth = 0); switch(data->type) { case XML_ELEMENT_NODE: /* Handle below */ dump_xml_element(data, options, buffer, depth); break; case XML_TEXT_NODE: if (pcmk_is_set(options, pcmk__xml_fmt_text)) { dump_xml_text(data, options, buffer, depth); } break; case XML_COMMENT_NODE: dump_xml_comment(data, options, buffer, depth); break; case XML_CDATA_SECTION_NODE: dump_xml_cdata(data, options, buffer, depth); break; default: crm_warn("Cannot convert XML %s node to text " CRM_XS " type=%d", xml_element_type_text(data->type), data->type); break; } } /*! * \internal * \brief Write a string to a file stream, compressed using \c bzip2 * * \param[in] text String to write * \param[in] filename Name of file being written (for logging only) * \param[in,out] stream Open file stream to write to * \param[out] bytes_out Number of bytes written (valid only on success) * * \return Standard Pacemaker return code */ static int write_compressed_stream(char *text, const char *filename, FILE *stream, unsigned int *bytes_out) { unsigned int bytes_in = 0; int rc = pcmk_rc_ok; // (5, 0, 0): (intermediate block size, silent, default workFactor) BZFILE *bz_file = BZ2_bzWriteOpen(&rc, stream, 5, 0, 0); rc = pcmk__bzlib2rc(rc); if (rc != pcmk_rc_ok) { crm_warn("Not compressing %s: could not prepare file stream: %s " CRM_XS " rc=%d", filename, pcmk_rc_str(rc), rc); goto done; } BZ2_bzWrite(&rc, bz_file, text, strlen(text)); rc = pcmk__bzlib2rc(rc); if (rc != pcmk_rc_ok) { crm_warn("Not compressing %s: could not compress data: %s " CRM_XS " rc=%d errno=%d", filename, pcmk_rc_str(rc), rc, errno); goto done; } BZ2_bzWriteClose(&rc, bz_file, 0, &bytes_in, bytes_out); bz_file = NULL; rc = pcmk__bzlib2rc(rc); if (rc != pcmk_rc_ok) { crm_warn("Not compressing %s: could not write compressed data: %s " CRM_XS " rc=%d errno=%d", filename, pcmk_rc_str(rc), rc, errno); goto done; } crm_trace("Compressed XML for %s from %u bytes to %u", filename, bytes_in, *bytes_out); done: if (bz_file != NULL) { BZ2_bzWriteClose(&rc, bz_file, 0, NULL, NULL); } return rc; } /*! * \internal * \brief Write XML to a file stream * * \param[in] xml XML to write * \param[in] filename Name of file being written (for logging only) * \param[in,out] stream Open file stream corresponding to filename (closed * when this function returns) * \param[in] compress Whether to compress XML before writing * \param[out] nbytes Number of bytes written * * \return Standard Pacemaker return code */ static int write_xml_stream(const xmlNode *xml, const char *filename, FILE *stream, bool compress, unsigned int *nbytes) { // @COMPAT Drop nbytes as arg when we drop write_xml_fd()/write_xml_file() GString *buffer = g_string_sized_new(1024); unsigned int bytes_out = 0; int rc = pcmk_rc_ok; pcmk__xml_string(xml, pcmk__xml_fmt_pretty, buffer, 0); CRM_CHECK(!pcmk__str_empty(buffer->str), crm_log_xml_info(xml, "dump-failed"); rc = pcmk_rc_error; goto done); crm_log_xml_trace(xml, "writing"); if (compress && (write_compressed_stream(buffer->str, filename, stream, &bytes_out) == pcmk_rc_ok)) { goto done; } rc = fprintf(stream, "%s", buffer->str); if (rc < 0) { rc = EIO; crm_perror(LOG_ERR, "writing %s", filename); goto done; } bytes_out = (unsigned int) rc; rc = pcmk_rc_ok; done: if (fflush(stream) != 0) { rc = errno; crm_perror(LOG_ERR, "flushing %s", filename); } // Don't report error if the file does not support synchronization if ((fsync(fileno(stream)) < 0) && (errno != EROFS) && (errno != EINVAL)) { rc = errno; crm_perror(LOG_ERR, "synchronizing %s", filename); } fclose(stream); crm_trace("Saved %u bytes to %s as XML", bytes_out, filename); if (nbytes != NULL) { *nbytes = bytes_out; } g_string_free(buffer, TRUE); return rc; } /*! * \internal * \brief Write XML to a file descriptor * * \param[in] xml XML to write * \param[in] filename Name of file being written (for logging only) * \param[in] fd Open file descriptor corresponding to \p filename * \param[in] compress If \c true, compress XML before writing * \param[out] nbytes Number of bytes written (can be \c NULL) * * \return Standard Pacemaker return code */ int pcmk__xml_write_fd(const xmlNode *xml, const char *filename, int fd, bool compress, unsigned int *nbytes) { // @COMPAT Drop compress and nbytes arguments when we drop write_xml_fd() FILE *stream = NULL; CRM_CHECK((xml != NULL) && (fd > 0), return EINVAL); stream = fdopen(fd, "w"); if (stream == NULL) { return errno; } return write_xml_stream(xml, pcmk__s(filename, "unnamed file"), stream, compress, nbytes); } /*! * \internal * \brief Write XML to a file * * \param[in] xml XML to write * \param[in] filename Name of file to write * \param[in] compress If \c true, compress XML before writing * \param[out] nbytes Number of bytes written (can be \c NULL) * * \return Standard Pacemaker return code */ int pcmk__xml_write_file(const xmlNode *xml, const char *filename, bool compress, unsigned int *nbytes) { // @COMPAT Drop nbytes argument when we drop write_xml_fd() FILE *stream = NULL; CRM_CHECK((xml != NULL) && (filename != NULL), return EINVAL); stream = fopen(filename, "w"); if (stream == NULL) { return errno; } return write_xml_stream(xml, filename, stream, compress, nbytes); } /*! * \internal * \brief Serialize XML (using libxml) into provided descriptor * * \param[in] fd File descriptor to (piece-wise) write to * \param[in] cur XML subtree to proceed * * \return a standard Pacemaker return code */ int pcmk__xml2fd(int fd, xmlNode *cur) { bool success; xmlOutputBuffer *fd_out = xmlOutputBufferCreateFd(fd, NULL); pcmk__mem_assert(fd_out); xmlNodeDumpOutput(fd_out, cur->doc, cur, 0, pcmk__xml_fmt_pretty, NULL); success = xmlOutputBufferWrite(fd_out, sizeof("\n") - 1, "\n") != -1; success = xmlOutputBufferClose(fd_out) != -1 && success; if (!success) { return EIO; } fsync(fd); return pcmk_rc_ok; } void save_xml_to_file(const xmlNode *xml, const char *desc, const char *filename) { char *f = NULL; if (filename == NULL) { char *uuid = crm_generate_uuid(); f = crm_strdup_printf("%s/%s", pcmk__get_tmpdir(), uuid); filename = f; free(uuid); } crm_info("Saving %s to %s", desc, filename); pcmk__xml_write_file(xml, filename, false, NULL); free(f); } // Deprecated functions kept only for backward API compatibility // LCOV_EXCL_START #include xmlNode * filename2xml(const char *filename) { return pcmk__xml_read(filename); } xmlNode * stdin2xml(void) { return pcmk__xml_read(NULL); } xmlNode * string2xml(const char *input) { return pcmk__xml_parse(input); } char * dump_xml_formatted(const xmlNode *xml) { char *str = NULL; GString *buffer = g_string_sized_new(1024); pcmk__xml_string(xml, pcmk__xml_fmt_pretty, buffer, 0); str = pcmk__str_copy(buffer->str); g_string_free(buffer, TRUE); return str; } char * dump_xml_formatted_with_text(const xmlNode *xml) { char *str = NULL; GString *buffer = g_string_sized_new(1024); pcmk__xml_string(xml, pcmk__xml_fmt_pretty|pcmk__xml_fmt_text, buffer, 0); str = pcmk__str_copy(buffer->str); g_string_free(buffer, TRUE); return str; } char * dump_xml_unformatted(const xmlNode *xml) { char *str = NULL; GString *buffer = g_string_sized_new(1024); pcmk__xml_string(xml, 0, buffer, 0); str = pcmk__str_copy(buffer->str); g_string_free(buffer, TRUE); return str; } int write_xml_fd(const xmlNode *xml, const char *filename, int fd, gboolean compress) { unsigned int nbytes = 0; int rc = pcmk__xml_write_fd(xml, filename, fd, compress, &nbytes); if (rc != pcmk_rc_ok) { return pcmk_rc2legacy(rc); } return (int) nbytes; } int write_xml_file(const xmlNode *xml, const char *filename, gboolean compress) { unsigned int nbytes = 0; int rc = pcmk__xml_write_file(xml, filename, compress, &nbytes); if (rc != pcmk_rc_ok) { return pcmk_rc2legacy(rc); } return (int) nbytes; } // LCOV_EXCL_STOP // End deprecated API diff --git a/lib/fencing/Makefile.am b/lib/fencing/Makefile.am index 512e908444..742d486259 100644 --- a/lib/fencing/Makefile.am +++ b/lib/fencing/Makefile.am @@ -1,32 +1,32 @@ # # Original Author: Sun Jiang Dong # Copyright 2004 International Business Machines # # with later changes copyright 2004-2022 the Pacemaker project contributors. # The version control history for this file may have further details. # # This source code is licensed under the GNU General Public License version 2 # or later (GPLv2+) WITHOUT ANY WARRANTY. # include $(top_srcdir)/mk/common.mk noinst_HEADERS = fencing_private.h lib_LTLIBRARIES = libstonithd.la -libstonithd_la_LDFLAGS = -version-info 35:0:9 +libstonithd_la_LDFLAGS = -version-info 35:1:9 libstonithd_la_CFLAGS = $(CFLAGS_HARDENED_LIB) libstonithd_la_LDFLAGS += $(LDFLAGS_HARDENED_LIB) libstonithd_la_LIBADD = $(top_builddir)/lib/services/libcrmservice.la libstonithd_la_LIBADD += $(top_builddir)/lib/common/libcrmcommon.la ## Library sources (*must* use += format for bumplibs) libstonithd_la_SOURCES = st_actions.c libstonithd_la_SOURCES += st_client.c if BUILD_LHA_SUPPORT libstonithd_la_SOURCES += st_lha.c endif libstonithd_la_SOURCES += st_output.c libstonithd_la_SOURCES += st_rhcs.c diff --git a/lib/fencing/st_client.c b/lib/fencing/st_client.c index 23710fdd0d..c3a9698484 100644 --- a/lib/fencing/st_client.c +++ b/lib/fencing/st_client.c @@ -1,2720 +1,2728 @@ /* - * Copyright 2004-2024 the Pacemaker project contributors + * Copyright 2004-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include "fencing_private.h" CRM_TRACE_INIT_DATA(stonith); // Used as stonith_t:st_private typedef struct stonith_private_s { char *token; crm_ipc_t *ipc; mainloop_io_t *source; GHashTable *stonith_op_callback_table; GList *notify_list; int notify_refcnt; bool notify_deletes; void (*op_callback) (stonith_t * st, stonith_callback_data_t * data); } stonith_private_t; // Used as stonith_event_t:opaque struct event_private { pcmk__action_result_t result; }; typedef struct stonith_notify_client_s { const char *event; const char *obj_id; /* implement one day */ const char *obj_type; /* implement one day */ void (*notify) (stonith_t * st, stonith_event_t * e); bool delete; } stonith_notify_client_t; typedef struct stonith_callback_client_s { void (*callback) (stonith_t * st, stonith_callback_data_t * data); const char *id; void *user_data; gboolean only_success; gboolean allow_timeout_updates; struct timer_rec_s *timer; } stonith_callback_client_t; struct notify_blob_s { stonith_t *stonith; xmlNode *xml; }; struct timer_rec_s { int call_id; int timeout; guint ref; stonith_t *stonith; }; typedef int (*stonith_op_t) (const char *, int, const char *, xmlNode *, xmlNode *, xmlNode *, xmlNode **, xmlNode **); bool stonith_dispatch(stonith_t * st); xmlNode *stonith_create_op(int call_id, const char *token, const char *op, xmlNode * data, int call_options); static int stonith_send_command(stonith_t *stonith, const char *op, xmlNode *data, xmlNode **output_data, int call_options, int timeout); static void stonith_connection_destroy(gpointer user_data); static void stonith_send_notification(gpointer data, gpointer user_data); static int stonith_api_del_notification(stonith_t *stonith, const char *event); /*! * \brief Get agent namespace by name * * \param[in] namespace_s Name of namespace as string * * \return Namespace as enum value */ enum stonith_namespace stonith_text2namespace(const char *namespace_s) { if (pcmk__str_eq(namespace_s, "any", pcmk__str_null_matches)) { return st_namespace_any; } else if (!strcmp(namespace_s, "redhat") || !strcmp(namespace_s, "stonith-ng")) { return st_namespace_rhcs; } else if (!strcmp(namespace_s, "internal")) { return st_namespace_internal; } else if (!strcmp(namespace_s, "heartbeat")) { return st_namespace_lha; } return st_namespace_invalid; } /*! * \brief Get agent namespace name * * \param[in] namespace Namespace as enum value * * \return Namespace name as string */ const char * stonith_namespace2text(enum stonith_namespace st_namespace) { switch (st_namespace) { case st_namespace_any: return "any"; case st_namespace_rhcs: return "stonith-ng"; case st_namespace_internal: return "internal"; case st_namespace_lha: return "heartbeat"; default: break; } return "unsupported"; } /*! * \brief Determine namespace of a fence agent * * \param[in] agent Fence agent type * \param[in] namespace_s Name of agent namespace as string, if known * * \return Namespace of specified agent, as enum value */ enum stonith_namespace stonith_get_namespace(const char *agent, const char *namespace_s) { if (pcmk__str_eq(namespace_s, "internal", pcmk__str_none)) { return st_namespace_internal; } if (stonith__agent_is_rhcs(agent)) { return st_namespace_rhcs; } #if HAVE_STONITH_STONITH_H if (stonith__agent_is_lha(agent)) { return st_namespace_lha; } #endif crm_err("Unknown fence agent: %s", agent); return st_namespace_invalid; } gboolean stonith__watchdog_fencing_enabled_for_node_api(stonith_t *st, const char *node) { gboolean rv = FALSE; stonith_t *stonith_api = st?st:stonith_api_new(); char *list = NULL; if(stonith_api) { if (stonith_api->state == stonith_disconnected) { int rc = stonith_api->cmds->connect(stonith_api, "stonith-api", NULL); if (rc != pcmk_ok) { crm_err("Failed connecting to Stonith-API for watchdog-fencing-query."); } } if (stonith_api->state != stonith_disconnected) { /* caveat!!! * this might fail when when stonithd is just updating the device-list * probably something we should fix as well for other api-calls */ int rc = stonith_api->cmds->list(stonith_api, st_opt_sync_call, STONITH_WATCHDOG_ID, &list, 0); if ((rc != pcmk_ok) || (list == NULL)) { /* due to the race described above it can happen that * we drop in here - so as not to make remote nodes * panic on that answer */ if (rc == -ENODEV) { crm_notice("Cluster does not have watchdog fencing device"); } else { crm_warn("Could not check for watchdog fencing device: %s", pcmk_strerror(rc)); } } else if (list[0] == '\0') { rv = TRUE; } else { GList *targets = stonith__parse_targets(list); rv = pcmk__str_in_list(node, targets, pcmk__str_casei); g_list_free_full(targets, free); } free(list); if (!st) { /* if we're provided the api we still might have done the * connection - but let's assume the caller won't bother */ stonith_api->cmds->disconnect(stonith_api); } } if (!st) { stonith_api_delete(stonith_api); } } else { crm_err("Stonith-API for watchdog-fencing-query couldn't be created."); } crm_trace("Pacemaker assumes node %s %sto do watchdog-fencing.", node, rv?"":"not "); return rv; } gboolean stonith__watchdog_fencing_enabled_for_node(const char *node) { return stonith__watchdog_fencing_enabled_for_node_api(NULL, node); } /* when cycling through the list we don't want to delete items so just mark them and when we know nobody is using the list loop over it to remove the marked items */ static void foreach_notify_entry (stonith_private_t *private, GFunc func, gpointer user_data) { private->notify_refcnt++; g_list_foreach(private->notify_list, func, user_data); private->notify_refcnt--; if ((private->notify_refcnt == 0) && private->notify_deletes) { GList *list_item = private->notify_list; private->notify_deletes = FALSE; while (list_item != NULL) { stonith_notify_client_t *list_client = list_item->data; GList *next = g_list_next(list_item); if (list_client->delete) { free(list_client); private->notify_list = g_list_delete_link(private->notify_list, list_item); } list_item = next; } } } static void stonith_connection_destroy(gpointer user_data) { stonith_t *stonith = user_data; stonith_private_t *native = NULL; struct notify_blob_s blob; crm_trace("Sending destroyed notification"); blob.stonith = stonith; blob.xml = pcmk__xe_create(NULL, PCMK__XE_NOTIFY); native = stonith->st_private; native->ipc = NULL; native->source = NULL; free(native->token); native->token = NULL; stonith->state = stonith_disconnected; crm_xml_add(blob.xml, PCMK__XA_T, PCMK__VALUE_ST_NOTIFY); crm_xml_add(blob.xml, PCMK__XA_SUBT, PCMK__VALUE_ST_NOTIFY_DISCONNECT); foreach_notify_entry(native, stonith_send_notification, &blob); free_xml(blob.xml); } xmlNode * create_device_registration_xml(const char *id, enum stonith_namespace namespace, const char *agent, const stonith_key_value_t *params, const char *rsc_provides) { xmlNode *data = pcmk__xe_create(NULL, PCMK__XE_ST_DEVICE_ID); xmlNode *args = pcmk__xe_create(data, PCMK__XE_ATTRIBUTES); #if HAVE_STONITH_STONITH_H if (namespace == st_namespace_any) { namespace = stonith_get_namespace(agent, NULL); } if (namespace == st_namespace_lha) { hash2field((gpointer) "plugin", (gpointer) agent, args); agent = "fence_legacy"; } #endif crm_xml_add(data, PCMK_XA_ID, id); crm_xml_add(data, PCMK__XA_ST_ORIGIN, __func__); crm_xml_add(data, PCMK_XA_AGENT, agent); if ((namespace != st_namespace_any) && (namespace != st_namespace_invalid)) { crm_xml_add(data, PCMK__XA_NAMESPACE, stonith_namespace2text(namespace)); } if (rsc_provides) { crm_xml_add(data, PCMK__XA_RSC_PROVIDES, rsc_provides); } for (; params; params = params->next) { hash2field((gpointer) params->key, (gpointer) params->value, args); } return data; } static int stonith_api_register_device(stonith_t *st, int call_options, const char *id, const char *namespace_s, const char *agent, const stonith_key_value_t *params) { int rc = 0; xmlNode *data = NULL; data = create_device_registration_xml(id, stonith_text2namespace(namespace_s), agent, params, NULL); rc = stonith_send_command(st, STONITH_OP_DEVICE_ADD, data, NULL, call_options, 0); free_xml(data); return rc; } static int stonith_api_remove_device(stonith_t * st, int call_options, const char *name) { int rc = 0; xmlNode *data = NULL; data = pcmk__xe_create(NULL, PCMK__XE_ST_DEVICE_ID); crm_xml_add(data, PCMK__XA_ST_ORIGIN, __func__); crm_xml_add(data, PCMK_XA_ID, name); rc = stonith_send_command(st, STONITH_OP_DEVICE_DEL, data, NULL, call_options, 0); free_xml(data); return rc; } static int stonith_api_remove_level_full(stonith_t *st, int options, const char *node, const char *pattern, const char *attr, const char *value, int level) { int rc = 0; xmlNode *data = NULL; CRM_CHECK(node || pattern || (attr && value), return -EINVAL); data = pcmk__xe_create(NULL, PCMK_XE_FENCING_LEVEL); crm_xml_add(data, PCMK__XA_ST_ORIGIN, __func__); if (node) { crm_xml_add(data, PCMK_XA_TARGET, node); } else if (pattern) { crm_xml_add(data, PCMK_XA_TARGET_PATTERN, pattern); } else { crm_xml_add(data, PCMK_XA_TARGET_ATTRIBUTE, attr); crm_xml_add(data, PCMK_XA_TARGET_VALUE, value); } crm_xml_add_int(data, PCMK_XA_INDEX, level); rc = stonith_send_command(st, STONITH_OP_LEVEL_DEL, data, NULL, options, 0); free_xml(data); return rc; } static int stonith_api_remove_level(stonith_t * st, int options, const char *node, int level) { return stonith_api_remove_level_full(st, options, node, NULL, NULL, NULL, level); } /*! * \internal * \brief Create XML for fence topology level registration request * * \param[in] node If not NULL, target level by this node name * \param[in] pattern If not NULL, target by node name using this regex * \param[in] attr If not NULL, target by this node attribute * \param[in] value If not NULL, target by this node attribute value * \param[in] level Index number of level to register * \param[in] device_list List of devices in level * * \return Newly allocated XML tree on success, NULL otherwise * * \note The caller should set only one of node, pattern or attr/value. */ xmlNode * create_level_registration_xml(const char *node, const char *pattern, const char *attr, const char *value, int level, const stonith_key_value_t *device_list) { GString *list = NULL; xmlNode *data; CRM_CHECK(node || pattern || (attr && value), return NULL); data = pcmk__xe_create(NULL, PCMK_XE_FENCING_LEVEL); crm_xml_add(data, PCMK__XA_ST_ORIGIN, __func__); crm_xml_add_int(data, PCMK_XA_ID, level); crm_xml_add_int(data, PCMK_XA_INDEX, level); if (node) { crm_xml_add(data, PCMK_XA_TARGET, node); } else if (pattern) { crm_xml_add(data, PCMK_XA_TARGET_PATTERN, pattern); } else { crm_xml_add(data, PCMK_XA_TARGET_ATTRIBUTE, attr); crm_xml_add(data, PCMK_XA_TARGET_VALUE, value); } for (; device_list; device_list = device_list->next) { pcmk__add_separated_word(&list, 1024, device_list->value, ","); } if (list != NULL) { crm_xml_add(data, PCMK_XA_DEVICES, (const char *) list->str); g_string_free(list, TRUE); } return data; } static int stonith_api_register_level_full(stonith_t *st, int options, const char *node, const char *pattern, const char *attr, const char *value, int level, const stonith_key_value_t *device_list) { int rc = 0; xmlNode *data = create_level_registration_xml(node, pattern, attr, value, level, device_list); CRM_CHECK(data != NULL, return -EINVAL); rc = stonith_send_command(st, STONITH_OP_LEVEL_ADD, data, NULL, options, 0); free_xml(data); return rc; } static int stonith_api_register_level(stonith_t * st, int options, const char *node, int level, const stonith_key_value_t * device_list) { return stonith_api_register_level_full(st, options, node, NULL, NULL, NULL, level, device_list); } static int stonith_api_device_list(stonith_t *stonith, int call_options, const char *namespace_s, stonith_key_value_t **devices, int timeout) { int count = 0; enum stonith_namespace ns = stonith_text2namespace(namespace_s); if (devices == NULL) { crm_err("Parameter error: stonith_api_device_list"); return -EFAULT; } #if HAVE_STONITH_STONITH_H // Include Linux-HA agents if requested if ((ns == st_namespace_any) || (ns == st_namespace_lha)) { count += stonith__list_lha_agents(devices); } #endif // Include Red Hat agents if requested if ((ns == st_namespace_any) || (ns == st_namespace_rhcs)) { count += stonith__list_rhcs_agents(devices); } return count; } // See stonith_api_operations_t:metadata() documentation static int stonith_api_device_metadata(stonith_t *stonith, int call_options, const char *agent, const char *namespace_s, char **output, int timeout_sec) { /* By executing meta-data directly, we can get it from stonith_admin when * the cluster is not running, which is important for higher-level tools. */ enum stonith_namespace ns = stonith_get_namespace(agent, namespace_s); if (timeout_sec <= 0) { timeout_sec = PCMK_DEFAULT_METADATA_TIMEOUT_MS; } crm_trace("Looking up metadata for %s agent %s", stonith_namespace2text(ns), agent); switch (ns) { case st_namespace_rhcs: return stonith__rhcs_metadata(agent, timeout_sec, output); #if HAVE_STONITH_STONITH_H case st_namespace_lha: return stonith__lha_metadata(agent, timeout_sec, output); #endif default: crm_err("Can't get fence agent '%s' meta-data: No such agent", agent); break; } return -ENODEV; } static int stonith_api_query(stonith_t * stonith, int call_options, const char *target, stonith_key_value_t ** devices, int timeout) { int rc = 0, lpc = 0, max = 0; xmlNode *data = NULL; xmlNode *output = NULL; xmlXPathObjectPtr xpathObj = NULL; CRM_CHECK(devices != NULL, return -EINVAL); data = pcmk__xe_create(NULL, PCMK__XE_ST_DEVICE_ID); crm_xml_add(data, PCMK__XA_ST_ORIGIN, __func__); crm_xml_add(data, PCMK__XA_ST_TARGET, target); crm_xml_add(data, PCMK__XA_ST_DEVICE_ACTION, PCMK_ACTION_OFF); rc = stonith_send_command(stonith, STONITH_OP_QUERY, data, &output, call_options, timeout); if (rc < 0) { return rc; } xpathObj = xpath_search(output, "//@agent"); if (xpathObj) { max = numXpathResults(xpathObj); for (lpc = 0; lpc < max; lpc++) { xmlNode *match = getXpathResult(xpathObj, lpc); CRM_LOG_ASSERT(match != NULL); if(match != NULL) { xmlChar *match_path = xmlGetNodePath(match); crm_info("%s[%d] = %s", "//@agent", lpc, match_path); free(match_path); *devices = stonith_key_value_add(*devices, NULL, crm_element_value(match, PCMK_XA_ID)); } } freeXpathObject(xpathObj); } free_xml(output); free_xml(data); return max; } /*! * \internal * \brief Make a STONITH_OP_EXEC request * * \param[in,out] stonith Fencer connection * \param[in] call_options Bitmask of \c stonith_call_options * \param[in] id Fence device ID that request is for * \param[in] action Agent action to request (list, status, monitor) * \param[in] target Name of target node for requested action * \param[in] timeout_sec Error if not completed within this many seconds * \param[out] output Where to set agent output */ static int stonith_api_call(stonith_t *stonith, int call_options, const char *id, const char *action, const char *target, int timeout_sec, xmlNode **output) { int rc = 0; xmlNode *data = NULL; data = pcmk__xe_create(NULL, PCMK__XE_ST_DEVICE_ID); crm_xml_add(data, PCMK__XA_ST_ORIGIN, __func__); crm_xml_add(data, PCMK__XA_ST_DEVICE_ID, id); crm_xml_add(data, PCMK__XA_ST_DEVICE_ACTION, action); crm_xml_add(data, PCMK__XA_ST_TARGET, target); rc = stonith_send_command(stonith, STONITH_OP_EXEC, data, output, call_options, timeout_sec); free_xml(data); return rc; } static int stonith_api_list(stonith_t * stonith, int call_options, const char *id, char **list_info, int timeout) { int rc; xmlNode *output = NULL; rc = stonith_api_call(stonith, call_options, id, PCMK_ACTION_LIST, NULL, timeout, &output); if (output && list_info) { const char *list_str; list_str = crm_element_value(output, PCMK__XA_ST_OUTPUT); if (list_str) { *list_info = strdup(list_str); } } if (output) { free_xml(output); } return rc; } static int stonith_api_monitor(stonith_t * stonith, int call_options, const char *id, int timeout) { return stonith_api_call(stonith, call_options, id, PCMK_ACTION_MONITOR, NULL, timeout, NULL); } static int stonith_api_status(stonith_t * stonith, int call_options, const char *id, const char *port, int timeout) { return stonith_api_call(stonith, call_options, id, PCMK_ACTION_STATUS, port, timeout, NULL); } static int stonith_api_fence_with_delay(stonith_t * stonith, int call_options, const char *node, const char *action, int timeout, int tolerance, int delay) { int rc = 0; xmlNode *data = NULL; data = pcmk__xe_create(NULL, __func__); crm_xml_add(data, PCMK__XA_ST_TARGET, node); crm_xml_add(data, PCMK__XA_ST_DEVICE_ACTION, action); crm_xml_add_int(data, PCMK__XA_ST_TIMEOUT, timeout); crm_xml_add_int(data, PCMK__XA_ST_TOLERANCE, tolerance); crm_xml_add_int(data, PCMK__XA_ST_DELAY, delay); rc = stonith_send_command(stonith, STONITH_OP_FENCE, data, NULL, call_options, timeout); free_xml(data); return rc; } static int stonith_api_fence(stonith_t * stonith, int call_options, const char *node, const char *action, int timeout, int tolerance) { return stonith_api_fence_with_delay(stonith, call_options, node, action, timeout, tolerance, 0); } static int stonith_api_confirm(stonith_t * stonith, int call_options, const char *target) { stonith__set_call_options(call_options, target, st_opt_manual_ack); return stonith_api_fence(stonith, call_options, target, PCMK_ACTION_OFF, 0, 0); } static int stonith_api_history(stonith_t * stonith, int call_options, const char *node, stonith_history_t ** history, int timeout) { int rc = 0; xmlNode *data = NULL; xmlNode *output = NULL; stonith_history_t *last = NULL; *history = NULL; if (node) { data = pcmk__xe_create(NULL, __func__); crm_xml_add(data, PCMK__XA_ST_TARGET, node); } stonith__set_call_options(call_options, node, st_opt_sync_call); rc = stonith_send_command(stonith, STONITH_OP_FENCE_HISTORY, data, &output, call_options, timeout); free_xml(data); if (rc == 0) { xmlNode *op = NULL; xmlNode *reply = get_xpath_object("//" PCMK__XE_ST_HISTORY, output, LOG_NEVER); for (op = pcmk__xe_first_child(reply, NULL, NULL, NULL); op != NULL; op = pcmk__xe_next(op)) { stonith_history_t *kvp; long long completed; long long completed_nsec = 0L; kvp = pcmk__assert_alloc(1, sizeof(stonith_history_t)); kvp->target = crm_element_value_copy(op, PCMK__XA_ST_TARGET); kvp->action = crm_element_value_copy(op, PCMK__XA_ST_DEVICE_ACTION); kvp->origin = crm_element_value_copy(op, PCMK__XA_ST_ORIGIN); kvp->delegate = crm_element_value_copy(op, PCMK__XA_ST_DELEGATE); kvp->client = crm_element_value_copy(op, PCMK__XA_ST_CLIENTNAME); crm_element_value_ll(op, PCMK__XA_ST_DATE, &completed); kvp->completed = (time_t) completed; crm_element_value_ll(op, PCMK__XA_ST_DATE_NSEC, &completed_nsec); kvp->completed_nsec = completed_nsec; crm_element_value_int(op, PCMK__XA_ST_STATE, &kvp->state); kvp->exit_reason = crm_element_value_copy(op, PCMK_XA_EXIT_REASON); if (last) { last->next = kvp; } else { *history = kvp; } last = kvp; } } free_xml(output); return rc; } void stonith_history_free(stonith_history_t *history) { stonith_history_t *hp, *hp_old; for (hp = history; hp; hp_old = hp, hp = hp->next, free(hp_old)) { free(hp->target); free(hp->action); free(hp->origin); free(hp->delegate); free(hp->client); free(hp->exit_reason); } } static gint stonithlib_GCompareFunc(gconstpointer a, gconstpointer b) { int rc = 0; const stonith_notify_client_t *a_client = a; const stonith_notify_client_t *b_client = b; if (a_client->delete || b_client->delete) { /* make entries marked for deletion not findable */ return -1; } CRM_CHECK(a_client->event != NULL && b_client->event != NULL, return 0); rc = strcmp(a_client->event, b_client->event); if (rc == 0) { if (a_client->notify == NULL || b_client->notify == NULL) { return 0; } else if (a_client->notify == b_client->notify) { return 0; } else if (((long)a_client->notify) < ((long)b_client->notify)) { crm_err("callbacks for %s are not equal: %p vs. %p", a_client->event, a_client->notify, b_client->notify); return -1; } crm_err("callbacks for %s are not equal: %p vs. %p", a_client->event, a_client->notify, b_client->notify); return 1; } return rc; } xmlNode * stonith_create_op(int call_id, const char *token, const char *op, xmlNode * data, int call_options) { xmlNode *op_msg = NULL; CRM_CHECK(token != NULL, return NULL); op_msg = pcmk__xe_create(NULL, PCMK__XE_STONITH_COMMAND); crm_xml_add(op_msg, PCMK__XA_T, PCMK__VALUE_STONITH_NG); crm_xml_add(op_msg, PCMK__XA_ST_OP, op); crm_xml_add_int(op_msg, PCMK__XA_ST_CALLID, call_id); crm_trace("Sending call options: %.8lx, %d", (long)call_options, call_options); crm_xml_add_int(op_msg, PCMK__XA_ST_CALLOPT, call_options); if (data != NULL) { xmlNode *wrapper = pcmk__xe_create(op_msg, PCMK__XE_ST_CALLDATA); pcmk__xml_copy(wrapper, data); } return op_msg; } static void stonith_destroy_op_callback(gpointer data) { stonith_callback_client_t *blob = data; if (blob->timer && blob->timer->ref > 0) { g_source_remove(blob->timer->ref); } free(blob->timer); free(blob); } static int stonith_api_signoff(stonith_t * stonith) { stonith_private_t *native = stonith->st_private; crm_debug("Disconnecting from the fencer"); if (native->source != NULL) { /* Attached to mainloop */ mainloop_del_ipc_client(native->source); native->source = NULL; native->ipc = NULL; } else if (native->ipc) { /* Not attached to mainloop */ crm_ipc_t *ipc = native->ipc; native->ipc = NULL; crm_ipc_close(ipc); crm_ipc_destroy(ipc); } free(native->token); native->token = NULL; stonith->state = stonith_disconnected; return pcmk_ok; } static int stonith_api_del_callback(stonith_t * stonith, int call_id, bool all_callbacks) { stonith_private_t *private = stonith->st_private; if (all_callbacks) { private->op_callback = NULL; g_hash_table_destroy(private->stonith_op_callback_table); private->stonith_op_callback_table = pcmk__intkey_table(stonith_destroy_op_callback); } else if (call_id == 0) { private->op_callback = NULL; } else { pcmk__intkey_table_remove(private->stonith_op_callback_table, call_id); } return pcmk_ok; } /*! * \internal * \brief Invoke a (single) specified fence action callback * * \param[in,out] st Fencer API connection * \param[in] call_id If positive, call ID of completed fence action, * otherwise legacy return code for early failure * \param[in,out] result Full result for action * \param[in,out] userdata User data to pass to callback * \param[in] callback Fence action callback to invoke */ static void invoke_fence_action_callback(stonith_t *st, int call_id, pcmk__action_result_t *result, void *userdata, void (*callback) (stonith_t *st, stonith_callback_data_t *data)) { stonith_callback_data_t data = { 0, }; data.call_id = call_id; data.rc = pcmk_rc2legacy(stonith__result2rc(result)); data.userdata = userdata; data.opaque = (void *) result; callback(st, &data); } /*! * \internal * \brief Invoke any callbacks registered for a specified fence action result * * Given a fence action result from the fencer, invoke any callback registered * for that action, as well as any global callback registered. * * \param[in,out] stonith Fencer API connection * \param[in] msg If non-NULL, fencer reply * \param[in] call_id If \p msg is NULL, call ID of action that timed out */ static void invoke_registered_callbacks(stonith_t *stonith, const xmlNode *msg, int call_id) { stonith_private_t *private = NULL; stonith_callback_client_t *cb_info = NULL; pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; CRM_CHECK(stonith != NULL, return); CRM_CHECK(stonith->st_private != NULL, return); private = stonith->st_private; if (msg == NULL) { // Fencer didn't reply in time pcmk__set_result(&result, CRM_EX_ERROR, PCMK_EXEC_TIMEOUT, "Fencer accepted request but did not reply in time"); CRM_LOG_ASSERT(call_id > 0); } else { // We have the fencer reply if ((crm_element_value_int(msg, PCMK__XA_ST_CALLID, &call_id) != 0) || (call_id <= 0)) { crm_log_xml_warn(msg, "Bad fencer reply"); } stonith__xe_get_result(msg, &result); } if (call_id > 0) { cb_info = pcmk__intkey_table_lookup(private->stonith_op_callback_table, call_id); } if ((cb_info != NULL) && (cb_info->callback != NULL) && (pcmk__result_ok(&result) || !(cb_info->only_success))) { crm_trace("Invoking callback %s for call %d", pcmk__s(cb_info->id, "without ID"), call_id); invoke_fence_action_callback(stonith, call_id, &result, cb_info->user_data, cb_info->callback); } else if ((private->op_callback == NULL) && !pcmk__result_ok(&result)) { crm_warn("Fencing action without registered callback failed: %d (%s%s%s)", result.exit_status, pcmk_exec_status_str(result.execution_status), ((result.exit_reason == NULL)? "" : ": "), ((result.exit_reason == NULL)? "" : result.exit_reason)); crm_log_xml_debug(msg, "Failed fence update"); } if (private->op_callback != NULL) { crm_trace("Invoking global callback for call %d", call_id); invoke_fence_action_callback(stonith, call_id, &result, NULL, private->op_callback); } if (cb_info != NULL) { stonith_api_del_callback(stonith, call_id, FALSE); } pcmk__reset_result(&result); } static gboolean stonith_async_timeout_handler(gpointer data) { struct timer_rec_s *timer = data; crm_err("Async call %d timed out after %dms", timer->call_id, timer->timeout); invoke_registered_callbacks(timer->stonith, NULL, timer->call_id); /* Always return TRUE, never remove the handler * We do that in stonith_del_callback() */ return TRUE; } static void set_callback_timeout(stonith_callback_client_t * callback, stonith_t * stonith, int call_id, int timeout) { struct timer_rec_s *async_timer = callback->timer; if (timeout <= 0) { return; } if (!async_timer) { async_timer = pcmk__assert_alloc(1, sizeof(struct timer_rec_s)); callback->timer = async_timer; } async_timer->stonith = stonith; async_timer->call_id = call_id; /* Allow a fair bit of grace to allow the server to tell us of a timeout * This is only a fallback */ async_timer->timeout = (timeout + 60) * 1000; if (async_timer->ref) { g_source_remove(async_timer->ref); } async_timer->ref = g_timeout_add(async_timer->timeout, stonith_async_timeout_handler, async_timer); } static void update_callback_timeout(int call_id, int timeout, stonith_t * st) { stonith_callback_client_t *callback = NULL; stonith_private_t *private = st->st_private; callback = pcmk__intkey_table_lookup(private->stonith_op_callback_table, call_id); if (!callback || !callback->allow_timeout_updates) { return; } set_callback_timeout(callback, st, call_id, timeout); } static int stonith_dispatch_internal(const char *buffer, ssize_t length, gpointer userdata) { const char *type = NULL; struct notify_blob_s blob; stonith_t *st = userdata; stonith_private_t *private = NULL; pcmk__assert(st != NULL); private = st->st_private; blob.stonith = st; blob.xml = pcmk__xml_parse(buffer); if (blob.xml == NULL) { crm_warn("Received malformed message from fencer: %s", buffer); return 0; } /* do callbacks */ type = crm_element_value(blob.xml, PCMK__XA_T); crm_trace("Activating %s callbacks...", type); if (pcmk__str_eq(type, PCMK__VALUE_STONITH_NG, pcmk__str_none)) { invoke_registered_callbacks(st, blob.xml, 0); } else if (pcmk__str_eq(type, PCMK__VALUE_ST_NOTIFY, pcmk__str_none)) { foreach_notify_entry(private, stonith_send_notification, &blob); } else if (pcmk__str_eq(type, PCMK__VALUE_ST_ASYNC_TIMEOUT_VALUE, pcmk__str_none)) { int call_id = 0; int timeout = 0; crm_element_value_int(blob.xml, PCMK__XA_ST_TIMEOUT, &timeout); crm_element_value_int(blob.xml, PCMK__XA_ST_CALLID, &call_id); update_callback_timeout(call_id, timeout, st); } else { crm_err("Unknown message type: %s", type); crm_log_xml_warn(blob.xml, "BadReply"); } free_xml(blob.xml); return 1; } static int stonith_api_signon(stonith_t * stonith, const char *name, int *stonith_fd) { int rc = pcmk_ok; stonith_private_t *native = NULL; const char *display_name = name? name : "client"; struct ipc_client_callbacks st_callbacks = { .dispatch = stonith_dispatch_internal, .destroy = stonith_connection_destroy }; CRM_CHECK(stonith != NULL, return -EINVAL); native = stonith->st_private; pcmk__assert(native != NULL); crm_debug("Attempting fencer connection by %s with%s mainloop", display_name, (stonith_fd? "out" : "")); stonith->state = stonith_connected_command; if (stonith_fd) { /* No mainloop */ native->ipc = crm_ipc_new("stonith-ng", 0); if (native->ipc != NULL) { rc = pcmk__connect_generic_ipc(native->ipc); if (rc == pcmk_rc_ok) { rc = pcmk__ipc_fd(native->ipc, stonith_fd); if (rc != pcmk_rc_ok) { crm_debug("Couldn't get file descriptor for IPC: %s", pcmk_rc_str(rc)); } } if (rc != pcmk_rc_ok) { crm_ipc_close(native->ipc); crm_ipc_destroy(native->ipc); native->ipc = NULL; } } } else { /* With mainloop */ native->source = mainloop_add_ipc_client("stonith-ng", G_PRIORITY_MEDIUM, 0, stonith, &st_callbacks); native->ipc = mainloop_get_ipc_client(native->source); } if (native->ipc == NULL) { rc = -ENOTCONN; } else { xmlNode *reply = NULL; xmlNode *hello = pcmk__xe_create(NULL, PCMK__XE_STONITH_COMMAND); crm_xml_add(hello, PCMK__XA_T, PCMK__VALUE_STONITH_NG); crm_xml_add(hello, PCMK__XA_ST_OP, CRM_OP_REGISTER); crm_xml_add(hello, PCMK__XA_ST_CLIENTNAME, name); rc = crm_ipc_send(native->ipc, hello, crm_ipc_client_response, -1, &reply); if (rc < 0) { crm_debug("Couldn't register with the fencer: %s " CRM_XS " rc=%d", pcmk_strerror(rc), rc); rc = -ECOMM; } else if (reply == NULL) { crm_debug("Couldn't register with the fencer: no reply"); rc = -EPROTO; } else { const char *msg_type = crm_element_value(reply, PCMK__XA_ST_OP); native->token = crm_element_value_copy(reply, PCMK__XA_ST_CLIENTID); if (!pcmk__str_eq(msg_type, CRM_OP_REGISTER, pcmk__str_none)) { crm_debug("Couldn't register with the fencer: invalid reply type '%s'", (msg_type? msg_type : "(missing)")); crm_log_xml_debug(reply, "Invalid fencer reply"); rc = -EPROTO; } else if (native->token == NULL) { crm_debug("Couldn't register with the fencer: no token in reply"); crm_log_xml_debug(reply, "Invalid fencer reply"); rc = -EPROTO; } else { crm_debug("Connection to fencer by %s succeeded (registration token: %s)", display_name, native->token); rc = pcmk_ok; } } free_xml(reply); free_xml(hello); } if (rc != pcmk_ok) { crm_debug("Connection attempt to fencer by %s failed: %s " CRM_XS " rc=%d", display_name, pcmk_strerror(rc), rc); stonith->cmds->disconnect(stonith); } return rc; } static int stonith_set_notification(stonith_t * stonith, const char *callback, int enabled) { int rc = pcmk_ok; xmlNode *notify_msg = pcmk__xe_create(NULL, __func__); stonith_private_t *native = stonith->st_private; if (stonith->state != stonith_disconnected) { crm_xml_add(notify_msg, PCMK__XA_ST_OP, STONITH_OP_NOTIFY); if (enabled) { crm_xml_add(notify_msg, PCMK__XA_ST_NOTIFY_ACTIVATE, callback); } else { crm_xml_add(notify_msg, PCMK__XA_ST_NOTIFY_DEACTIVATE, callback); } rc = crm_ipc_send(native->ipc, notify_msg, crm_ipc_client_response, -1, NULL); if (rc < 0) { crm_perror(LOG_DEBUG, "Couldn't register for fencing notifications: %d", rc); rc = -ECOMM; } else { rc = pcmk_ok; } } free_xml(notify_msg); return rc; } static int stonith_api_add_notification(stonith_t * stonith, const char *event, void (*callback) (stonith_t * stonith, stonith_event_t * e)) { GList *list_item = NULL; stonith_notify_client_t *new_client = NULL; stonith_private_t *private = NULL; private = stonith->st_private; crm_trace("Adding callback for %s events (%d)", event, g_list_length(private->notify_list)); new_client = pcmk__assert_alloc(1, sizeof(stonith_notify_client_t)); new_client->event = event; new_client->notify = callback; list_item = g_list_find_custom(private->notify_list, new_client, stonithlib_GCompareFunc); if (list_item != NULL) { crm_warn("Callback already present"); free(new_client); return -ENOTUNIQ; } else { private->notify_list = g_list_append(private->notify_list, new_client); stonith_set_notification(stonith, event, 1); crm_trace("Callback added (%d)", g_list_length(private->notify_list)); } return pcmk_ok; } static void del_notify_entry(gpointer data, gpointer user_data) { stonith_notify_client_t *entry = data; stonith_t * stonith = user_data; if (!entry->delete) { crm_debug("Removing callback for %s events", entry->event); stonith_api_del_notification(stonith, entry->event); } } static int stonith_api_del_notification(stonith_t * stonith, const char *event) { GList *list_item = NULL; stonith_notify_client_t *new_client = NULL; stonith_private_t *private = stonith->st_private; if (event == NULL) { foreach_notify_entry(private, del_notify_entry, stonith); crm_trace("Removed callback"); return pcmk_ok; } crm_debug("Removing callback for %s events", event); new_client = pcmk__assert_alloc(1, sizeof(stonith_notify_client_t)); new_client->event = event; new_client->notify = NULL; list_item = g_list_find_custom(private->notify_list, new_client, stonithlib_GCompareFunc); stonith_set_notification(stonith, event, 0); if (list_item != NULL) { stonith_notify_client_t *list_client = list_item->data; if (private->notify_refcnt) { list_client->delete = TRUE; private->notify_deletes = TRUE; } else { private->notify_list = g_list_remove(private->notify_list, list_client); free(list_client); } crm_trace("Removed callback"); } else { crm_trace("Callback not present"); } free(new_client); return pcmk_ok; } static int stonith_api_add_callback(stonith_t * stonith, int call_id, int timeout, int options, void *user_data, const char *callback_name, void (*callback) (stonith_t * st, stonith_callback_data_t * data)) { stonith_callback_client_t *blob = NULL; stonith_private_t *private = NULL; CRM_CHECK(stonith != NULL, return -EINVAL); CRM_CHECK(stonith->st_private != NULL, return -EINVAL); private = stonith->st_private; if (call_id == 0) { // Add global callback private->op_callback = callback; } else if (call_id < 0) { // Call failed immediately, so call callback now if (!(options & st_opt_report_only_success)) { pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; crm_trace("Call failed, calling %s: %s", callback_name, pcmk_strerror(call_id)); pcmk__set_result(&result, CRM_EX_ERROR, stonith__legacy2status(call_id), NULL); invoke_fence_action_callback(stonith, call_id, &result, user_data, callback); } else { crm_warn("Fencer call failed: %s", pcmk_strerror(call_id)); } return FALSE; } blob = pcmk__assert_alloc(1, sizeof(stonith_callback_client_t)); blob->id = callback_name; blob->only_success = (options & st_opt_report_only_success) ? TRUE : FALSE; blob->user_data = user_data; blob->callback = callback; blob->allow_timeout_updates = (options & st_opt_timeout_updates) ? TRUE : FALSE; if (timeout > 0) { set_callback_timeout(blob, stonith, call_id, timeout); } pcmk__intkey_table_insert(private->stonith_op_callback_table, call_id, blob); crm_trace("Added callback to %s for call %d", callback_name, call_id); return TRUE; } static void stonith_dump_pending_op(gpointer key, gpointer value, gpointer user_data) { int call = GPOINTER_TO_INT(key); stonith_callback_client_t *blob = value; crm_debug("Call %d (%s): pending", call, pcmk__s(blob->id, "no ID")); } void stonith_dump_pending_callbacks(stonith_t * stonith) { stonith_private_t *private = stonith->st_private; if (private->stonith_op_callback_table == NULL) { return; } return g_hash_table_foreach(private->stonith_op_callback_table, stonith_dump_pending_op, NULL); } /*! * \internal * \brief Get the data section of a fencer notification * * \param[in] msg Notification XML * \param[in] ntype Notification type */ static xmlNode * get_event_data_xml(xmlNode *msg, const char *ntype) { char *data_addr = crm_strdup_printf("//%s", ntype); xmlNode *data = get_xpath_object(data_addr, msg, LOG_DEBUG); free(data_addr); return data; } /* */ static stonith_event_t * xml_to_event(xmlNode *msg) { stonith_event_t *event = pcmk__assert_alloc(1, sizeof(stonith_event_t)); struct event_private *event_private = NULL; event->opaque = pcmk__assert_alloc(1, sizeof(struct event_private)); event_private = (struct event_private *) event->opaque; crm_log_xml_trace(msg, "stonith_notify"); // All notification types have the operation result and notification subtype stonith__xe_get_result(msg, &event_private->result); event->operation = crm_element_value_copy(msg, PCMK__XA_ST_OP); // @COMPAT The API originally provided the result as a legacy return code event->result = pcmk_rc2legacy(stonith__result2rc(&event_private->result)); // Some notification subtypes have additional information if (pcmk__str_eq(event->operation, PCMK__VALUE_ST_NOTIFY_FENCE, pcmk__str_none)) { xmlNode *data = get_event_data_xml(msg, event->operation); if (data == NULL) { crm_err("No data for %s event", event->operation); crm_log_xml_notice(msg, "BadEvent"); } else { event->origin = crm_element_value_copy(data, PCMK__XA_ST_ORIGIN); event->action = crm_element_value_copy(data, PCMK__XA_ST_DEVICE_ACTION); event->target = crm_element_value_copy(data, PCMK__XA_ST_TARGET); event->executioner = crm_element_value_copy(data, PCMK__XA_ST_DELEGATE); event->id = crm_element_value_copy(data, PCMK__XA_ST_REMOTE_OP); event->client_origin = crm_element_value_copy(data, PCMK__XA_ST_CLIENTNAME); event->device = crm_element_value_copy(data, PCMK__XA_ST_DEVICE_ID); } } else if (pcmk__str_any_of(event->operation, STONITH_OP_DEVICE_ADD, STONITH_OP_DEVICE_DEL, STONITH_OP_LEVEL_ADD, STONITH_OP_LEVEL_DEL, NULL)) { xmlNode *data = get_event_data_xml(msg, event->operation); if (data == NULL) { crm_err("No data for %s event", event->operation); crm_log_xml_notice(msg, "BadEvent"); } else { event->device = crm_element_value_copy(data, PCMK__XA_ST_DEVICE_ID); } } return event; } static void event_free(stonith_event_t * event) { struct event_private *event_private = event->opaque; free(event->id); free(event->type); free(event->message); free(event->operation); free(event->origin); free(event->action); free(event->target); free(event->executioner); free(event->device); free(event->client_origin); pcmk__reset_result(&event_private->result); free(event->opaque); free(event); } static void stonith_send_notification(gpointer data, gpointer user_data) { struct notify_blob_s *blob = user_data; stonith_notify_client_t *entry = data; stonith_event_t *st_event = NULL; const char *event = NULL; if (blob->xml == NULL) { crm_warn("Skipping callback - NULL message"); return; } event = crm_element_value(blob->xml, PCMK__XA_SUBT); if (entry == NULL) { crm_warn("Skipping callback - NULL callback client"); return; } else if (entry->delete) { crm_trace("Skipping callback - marked for deletion"); return; } else if (entry->notify == NULL) { crm_warn("Skipping callback - NULL callback"); return; } else if (!pcmk__str_eq(entry->event, event, pcmk__str_none)) { crm_trace("Skipping callback - event mismatch %p/%s vs. %s", entry, entry->event, event); return; } st_event = xml_to_event(blob->xml); crm_trace("Invoking callback for %p/%s event...", entry, event); + // coverity[null_field] entry->notify(blob->stonith, st_event); crm_trace("Callback invoked..."); event_free(st_event); } /*! * \internal * \brief Create and send an API request * * \param[in,out] stonith Stonith connection * \param[in] op API operation to request * \param[in] data Data to attach to request * \param[out] output_data If not NULL, will be set to reply if synchronous * \param[in] call_options Bitmask of stonith_call_options to use * \param[in] timeout Error if not completed within this many seconds * * \return pcmk_ok (for synchronous requests) or positive call ID * (for asynchronous requests) on success, -errno otherwise */ static int stonith_send_command(stonith_t * stonith, const char *op, xmlNode * data, xmlNode ** output_data, int call_options, int timeout) { int rc = 0; int reply_id = -1; xmlNode *op_msg = NULL; xmlNode *op_reply = NULL; stonith_private_t *native = NULL; pcmk__assert((stonith != NULL) && (stonith->st_private != NULL) && (op != NULL)); native = stonith->st_private; if (output_data != NULL) { *output_data = NULL; } if ((stonith->state == stonith_disconnected) || (native->token == NULL)) { return -ENOTCONN; } /* Increment the call ID, which must be positive to avoid conflicting with * error codes. This shouldn't be a problem unless the client mucked with * it or the counter wrapped around. */ stonith->call_id++; if (stonith->call_id < 1) { stonith->call_id = 1; } op_msg = stonith_create_op(stonith->call_id, native->token, op, data, call_options); if (op_msg == NULL) { return -EINVAL; } crm_xml_add_int(op_msg, PCMK__XA_ST_TIMEOUT, timeout); crm_trace("Sending %s message to fencer with timeout %ds", op, timeout); if (data) { const char *delay_s = crm_element_value(data, PCMK__XA_ST_DELAY); if (delay_s) { crm_xml_add(op_msg, PCMK__XA_ST_DELAY, delay_s); } } { enum crm_ipc_flags ipc_flags = crm_ipc_flags_none; if (call_options & st_opt_sync_call) { pcmk__set_ipc_flags(ipc_flags, "stonith command", crm_ipc_client_response); } rc = crm_ipc_send(native->ipc, op_msg, ipc_flags, 1000 * (timeout + 60), &op_reply); } free_xml(op_msg); if (rc < 0) { crm_perror(LOG_ERR, "Couldn't perform %s operation (timeout=%ds): %d", op, timeout, rc); rc = -ECOMM; goto done; } crm_log_xml_trace(op_reply, "Reply"); if (!(call_options & st_opt_sync_call)) { crm_trace("Async call %d, returning", stonith->call_id); free_xml(op_reply); return stonith->call_id; } crm_element_value_int(op_reply, PCMK__XA_ST_CALLID, &reply_id); if (reply_id == stonith->call_id) { pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; crm_trace("Synchronous reply %d received", reply_id); stonith__xe_get_result(op_reply, &result); rc = pcmk_rc2legacy(stonith__result2rc(&result)); pcmk__reset_result(&result); if ((call_options & st_opt_discard_reply) || output_data == NULL) { crm_trace("Discarding reply"); } else { *output_data = op_reply; op_reply = NULL; /* Prevent subsequent free */ } } else if (reply_id <= 0) { crm_err("Received bad reply: No id set"); crm_log_xml_err(op_reply, "Bad reply"); free_xml(op_reply); op_reply = NULL; rc = -ENOMSG; } else { crm_err("Received bad reply: %d (wanted %d)", reply_id, stonith->call_id); crm_log_xml_err(op_reply, "Old reply"); free_xml(op_reply); op_reply = NULL; rc = -ENOMSG; } done: if (!crm_ipc_connected(native->ipc)) { crm_err("Fencer disconnected"); free(native->token); native->token = NULL; stonith->state = stonith_disconnected; } free_xml(op_reply); return rc; } /* Not used with mainloop */ bool stonith_dispatch(stonith_t * st) { gboolean stay_connected = TRUE; stonith_private_t *private = NULL; pcmk__assert(st != NULL); private = st->st_private; while (crm_ipc_ready(private->ipc)) { if (crm_ipc_read(private->ipc) > 0) { const char *msg = crm_ipc_buffer(private->ipc); stonith_dispatch_internal(msg, strlen(msg), st); } if (!crm_ipc_connected(private->ipc)) { crm_err("Connection closed"); stay_connected = FALSE; } } return stay_connected; } static int stonith_api_free(stonith_t * stonith) { int rc = pcmk_ok; crm_trace("Destroying %p", stonith); if (stonith->state != stonith_disconnected) { crm_trace("Unregistering notifications and disconnecting %p first", stonith); stonith->cmds->remove_notification(stonith, NULL); rc = stonith->cmds->disconnect(stonith); } if (stonith->state == stonith_disconnected) { stonith_private_t *private = stonith->st_private; crm_trace("Removing %d callbacks", g_hash_table_size(private->stonith_op_callback_table)); g_hash_table_destroy(private->stonith_op_callback_table); crm_trace("Destroying %d notification clients", g_list_length(private->notify_list)); g_list_free_full(private->notify_list, free); free(stonith->st_private); free(stonith->cmds); free(stonith); } else { crm_err("Not free'ing active connection: %s (%d)", pcmk_strerror(rc), rc); } return rc; } void stonith_api_delete(stonith_t * stonith) { crm_trace("Destroying %p", stonith); if(stonith) { stonith->cmds->free(stonith); } } static int stonith_api_validate(stonith_t *st, int call_options, const char *rsc_id, const char *namespace_s, const char *agent, const stonith_key_value_t *params, int timeout_sec, char **output, char **error_output) { /* Validation should be done directly via the agent, so we can get it from * stonith_admin when the cluster is not running, which is important for * higher-level tools. */ int rc = pcmk_ok; /* Use a dummy node name in case the agent requires a target. We assume the * actual target doesn't matter for validation purposes (if in practice, * that is incorrect, we will need to allow the caller to pass the target). */ const char *target = "node1"; const char *host_arg = NULL; GHashTable *params_table = pcmk__strkey_table(free, free); // Convert parameter list to a hash table for (; params; params = params->next) { if (pcmk__str_eq(params->key, PCMK_STONITH_HOST_ARGUMENT, pcmk__str_none)) { host_arg = params->value; } if (!pcmk_stonith_param(params->key)) { pcmk__insert_dup(params_table, params->key, params->value); } } #if SUPPORT_CIBSECRETS rc = pcmk__substitute_secrets(rsc_id, params_table); if (rc != pcmk_rc_ok) { crm_warn("Could not replace secret parameters for validation of %s: %s", agent, pcmk_rc_str(rc)); // rc is standard return value, don't return it in this function } #endif if (output) { *output = NULL; } if (error_output) { *error_output = NULL; } if (timeout_sec <= 0) { timeout_sec = PCMK_DEFAULT_METADATA_TIMEOUT_MS; // Questionable } switch (stonith_get_namespace(agent, namespace_s)) { case st_namespace_rhcs: rc = stonith__rhcs_validate(st, call_options, target, agent, params_table, host_arg, timeout_sec, output, error_output); break; #if HAVE_STONITH_STONITH_H case st_namespace_lha: rc = stonith__lha_validate(st, call_options, target, agent, params_table, timeout_sec, output, error_output); break; #endif case st_namespace_invalid: errno = ENOENT; rc = -errno; if (error_output) { *error_output = crm_strdup_printf("Agent %s not found", agent); } else { crm_err("Agent %s not found", agent); } break; default: errno = EOPNOTSUPP; rc = -errno; if (error_output) { *error_output = crm_strdup_printf("Agent %s does not support validation", agent); } else { crm_err("Agent %s does not support validation", agent); } break; } g_hash_table_destroy(params_table); return rc; } stonith_t * stonith_api_new(void) { stonith_t *new_stonith = NULL; stonith_private_t *private = NULL; new_stonith = calloc(1, sizeof(stonith_t)); if (new_stonith == NULL) { return NULL; } private = calloc(1, sizeof(stonith_private_t)); if (private == NULL) { free(new_stonith); return NULL; } new_stonith->st_private = private; private->stonith_op_callback_table = pcmk__intkey_table(stonith_destroy_op_callback); private->notify_list = NULL; private->notify_refcnt = 0; private->notify_deletes = FALSE; new_stonith->call_id = 1; new_stonith->state = stonith_disconnected; new_stonith->cmds = calloc(1, sizeof(stonith_api_operations_t)); if (new_stonith->cmds == NULL) { free(new_stonith->st_private); free(new_stonith); return NULL; } /* *INDENT-OFF* */ new_stonith->cmds->free = stonith_api_free; new_stonith->cmds->connect = stonith_api_signon; new_stonith->cmds->disconnect = stonith_api_signoff; new_stonith->cmds->list = stonith_api_list; new_stonith->cmds->monitor = stonith_api_monitor; new_stonith->cmds->status = stonith_api_status; new_stonith->cmds->fence = stonith_api_fence; new_stonith->cmds->fence_with_delay = stonith_api_fence_with_delay; new_stonith->cmds->confirm = stonith_api_confirm; new_stonith->cmds->history = stonith_api_history; new_stonith->cmds->list_agents = stonith_api_device_list; new_stonith->cmds->metadata = stonith_api_device_metadata; new_stonith->cmds->query = stonith_api_query; new_stonith->cmds->remove_device = stonith_api_remove_device; new_stonith->cmds->register_device = stonith_api_register_device; new_stonith->cmds->remove_level = stonith_api_remove_level; new_stonith->cmds->remove_level_full = stonith_api_remove_level_full; new_stonith->cmds->register_level = stonith_api_register_level; new_stonith->cmds->register_level_full = stonith_api_register_level_full; new_stonith->cmds->remove_callback = stonith_api_del_callback; new_stonith->cmds->register_callback = stonith_api_add_callback; new_stonith->cmds->remove_notification = stonith_api_del_notification; new_stonith->cmds->register_notification = stonith_api_add_notification; new_stonith->cmds->validate = stonith_api_validate; /* *INDENT-ON* */ return new_stonith; } /*! * \brief Make a blocking connection attempt to the fencer * * \param[in,out] st Fencer API object * \param[in] name Client name to use with fencer * \param[in] max_attempts Return error if this many attempts fail * * \return pcmk_ok on success, result of last attempt otherwise */ int stonith_api_connect_retry(stonith_t *st, const char *name, int max_attempts) { int rc = -EINVAL; // if max_attempts is not positive for (int attempt = 1; attempt <= max_attempts; attempt++) { rc = st->cmds->connect(st, name, NULL); if (rc == pcmk_ok) { return pcmk_ok; } else if (attempt < max_attempts) { crm_notice("Fencer connection attempt %d of %d failed (retrying in 2s): %s " CRM_XS " rc=%d", attempt, max_attempts, pcmk_strerror(rc), rc); sleep(2); } } crm_notice("Could not connect to fencer: %s " CRM_XS " rc=%d", pcmk_strerror(rc), rc); return rc; } stonith_key_value_t * stonith_key_value_add(stonith_key_value_t * head, const char *key, const char *value) { stonith_key_value_t *p, *end; p = pcmk__assert_alloc(1, sizeof(stonith_key_value_t)); p->key = pcmk__str_copy(key); p->value = pcmk__str_copy(value); end = head; while (end && end->next) { end = end->next; } if (end) { end->next = p; } else { head = p; } return head; } void stonith_key_value_freeall(stonith_key_value_t * head, int keys, int values) { stonith_key_value_t *p; while (head) { p = head->next; if (keys) { free(head->key); } if (values) { free(head->value); } free(head); head = p; } } #define api_log_open() openlog("stonith-api", LOG_CONS | LOG_NDELAY | LOG_PID, LOG_DAEMON) #define api_log(level, fmt, args...) syslog(level, "%s: "fmt, __func__, args) int stonith_api_kick(uint32_t nodeid, const char *uname, int timeout, bool off) { int rc = pcmk_ok; stonith_t *st = stonith_api_new(); const char *action = off? PCMK_ACTION_OFF : PCMK_ACTION_REBOOT; api_log_open(); if (st == NULL) { api_log(LOG_ERR, "API initialization failed, could not kick (%s) node %u/%s", action, nodeid, uname); return -EPROTO; } rc = st->cmds->connect(st, "stonith-api", NULL); if (rc != pcmk_ok) { api_log(LOG_ERR, "Connection failed, could not kick (%s) node %u/%s : %s (%d)", action, nodeid, uname, pcmk_strerror(rc), rc); } else { char *name = (uname == NULL)? pcmk__itoa(nodeid) : strdup(uname); int opts = 0; stonith__set_call_options(opts, name, st_opt_sync_call|st_opt_allow_self_fencing); if ((uname == NULL) && (nodeid > 0)) { stonith__set_call_options(opts, name, st_opt_cs_nodeid); } rc = st->cmds->fence(st, opts, name, action, timeout, 0); free(name); if (rc != pcmk_ok) { api_log(LOG_ERR, "Could not kick (%s) node %u/%s : %s (%d)", action, nodeid, uname, pcmk_strerror(rc), rc); } else { api_log(LOG_NOTICE, "Node %u/%s kicked: %s", nodeid, uname, action); } } stonith_api_delete(st); return rc; } time_t stonith_api_time(uint32_t nodeid, const char *uname, bool in_progress) { int rc = pcmk_ok; time_t when = 0; stonith_t *st = stonith_api_new(); stonith_history_t *history = NULL, *hp = NULL; if (st == NULL) { api_log(LOG_ERR, "Could not retrieve fence history for %u/%s: " "API initialization failed", nodeid, uname); return when; } rc = st->cmds->connect(st, "stonith-api", NULL); if (rc != pcmk_ok) { api_log(LOG_NOTICE, "Connection failed: %s (%d)", pcmk_strerror(rc), rc); } else { int entries = 0; int progress = 0; int completed = 0; int opts = 0; char *name = (uname == NULL)? pcmk__itoa(nodeid) : strdup(uname); stonith__set_call_options(opts, name, st_opt_sync_call); if ((uname == NULL) && (nodeid > 0)) { stonith__set_call_options(opts, name, st_opt_cs_nodeid); } rc = st->cmds->history(st, opts, name, &history, 120); free(name); for (hp = history; hp; hp = hp->next) { entries++; if (in_progress) { progress++; if (hp->state != st_done && hp->state != st_failed) { when = time(NULL); } } else if (hp->state == st_done) { completed++; if (hp->completed > when) { when = hp->completed; } } } stonith_history_free(history); if(rc == pcmk_ok) { api_log(LOG_INFO, "Found %d entries for %u/%s: %d in progress, %d completed", entries, nodeid, uname, progress, completed); } else { api_log(LOG_ERR, "Could not retrieve fence history for %u/%s: %s (%d)", nodeid, uname, pcmk_strerror(rc), rc); } } stonith_api_delete(st); if(when) { api_log(LOG_INFO, "Node %u/%s last kicked at: %ld", nodeid, uname, (long int)when); } return when; } bool stonith_agent_exists(const char *agent, int timeout) { stonith_t *st = NULL; stonith_key_value_t *devices = NULL; stonith_key_value_t *dIter = NULL; bool rc = FALSE; if (agent == NULL) { return rc; } st = stonith_api_new(); if (st == NULL) { crm_err("Could not list fence agents: API memory allocation failed"); return FALSE; } st->cmds->list_agents(st, st_opt_sync_call, NULL, &devices, timeout == 0 ? 120 : timeout); for (dIter = devices; dIter != NULL; dIter = dIter->next) { if (pcmk__str_eq(dIter->value, agent, pcmk__str_none)) { rc = TRUE; break; } } stonith_key_value_freeall(devices, 1, 1); stonith_api_delete(st); return rc; } const char * stonith_action_str(const char *action) { if (action == NULL) { return "fencing"; } else if (strcmp(action, PCMK_ACTION_ON) == 0) { return "unfencing"; } else if (strcmp(action, PCMK_ACTION_OFF) == 0) { return "turning off"; } else { return action; } } /*! * \internal * \brief Parse a target name from one line of a target list string * * \param[in] line One line of a target list string * \param[in] len String length of line * \param[in,out] output List to add newly allocated target name to */ static void parse_list_line(const char *line, int len, GList **output) { size_t i = 0; size_t entry_start = 0; + if (line == NULL) { + return; + } + /* Skip complaints about additional parameters device doesn't understand * * @TODO Document or eliminate the implied restriction of target names */ if (strstr(line, "invalid") || strstr(line, "variable")) { crm_debug("Skipping list output line: %s", line); return; } // Process line content, character by character for (i = 0; i <= len; i++) { if (isspace(line[i]) || (line[i] == ',') || (line[i] == ';') || (line[i] == '\0')) { // We've found a separator (i.e. the end of an entry) int rc = 0; char *entry = NULL; if (i == entry_start) { // Skip leading and sequential separators entry_start = i + 1; continue; } entry = pcmk__assert_alloc(i - entry_start + 1, sizeof(char)); /* Read entry, stopping at first separator * * @TODO Document or eliminate these character restrictions */ rc = sscanf(line + entry_start, "%[a-zA-Z0-9_-.]", entry); if (rc != 1) { crm_warn("Could not parse list output entry: %s " CRM_XS " entry_start=%d position=%d", line + entry_start, entry_start, i); free(entry); } else if (pcmk__strcase_any_of(entry, PCMK_ACTION_ON, PCMK_ACTION_OFF, NULL)) { /* Some agents print the target status in the list output, * though none are known now (the separate list-status command * is used for this, but it can also print "UNKNOWN"). To handle * this possibility, skip such entries. * * @TODO Document or eliminate the implied restriction of target * names. */ free(entry); } else { // We have a valid entry *output = g_list_append(*output, entry); } entry_start = i + 1; } } } /*! * \internal * \brief Parse a list of targets from a string * * \param[in] list_output Target list as a string * * \return List of target names * \note The target list string format is flexible, to allow for user-specified * lists such pcmk_host_list and the output of an agent's list action * (whether direct or via the API, which escapes newlines). There may be * multiple lines, separated by either a newline or an escaped newline * (backslash n). Each line may have one or more target names, separated * by any combination of whitespace, commas, and semi-colons. Lines * containing "invalid" or "variable" will be ignored entirely. Target * names "on" or "off" (case-insensitive) will be ignored. Target names * may contain only alphanumeric characters, underbars (_), dashes (-), * and dots (.) (if any other character occurs in the name, it and all * subsequent characters in the name will be ignored). * \note The caller is responsible for freeing the result with * g_list_free_full(result, free). */ GList * stonith__parse_targets(const char *target_spec) { GList *targets = NULL; if (target_spec != NULL) { size_t out_len = strlen(target_spec); size_t line_start = 0; // Starting index of line being processed for (size_t i = 0; i <= out_len; ++i) { if ((target_spec[i] == '\n') || (target_spec[i] == '\0') || ((target_spec[i] == '\\') && (target_spec[i + 1] == 'n'))) { // We've reached the end of one line of output int len = i - line_start; if (len > 0) { char *line = strndup(target_spec + line_start, len); + pcmk__assert(line != NULL); + + // cppcheck-suppress nullPointerOutOfMemory line[len] = '\0'; // Because it might be a newline parse_list_line(line, len, &targets); free(line); } if (target_spec[i] == '\\') { ++i; // backslash-n takes up two positions } line_start = i + 1; } } } return targets; } /*! * \internal * \brief Check whether a fencing failure was followed by an equivalent success * * \param[in] event Fencing failure * \param[in] top_history Complete fencing history (must be sorted by * stonith__sort_history() beforehand) * * \return The name of the node that executed the fencing if a later successful * event exists, or NULL if no such event exists */ const char * stonith__later_succeeded(const stonith_history_t *event, const stonith_history_t *top_history) { const char *other = NULL; for (const stonith_history_t *prev_hp = top_history; prev_hp != NULL; prev_hp = prev_hp->next) { if (prev_hp == event) { break; } if ((prev_hp->state == st_done) && pcmk__str_eq(event->target, prev_hp->target, pcmk__str_casei) && pcmk__str_eq(event->action, prev_hp->action, pcmk__str_none) && ((event->completed < prev_hp->completed) || ((event->completed == prev_hp->completed) && (event->completed_nsec < prev_hp->completed_nsec)))) { if ((event->delegate == NULL) || pcmk__str_eq(event->delegate, prev_hp->delegate, pcmk__str_casei)) { // Prefer equivalent fencing by same executioner return prev_hp->delegate; } else if (other == NULL) { // Otherwise remember first successful executioner other = (prev_hp->delegate == NULL)? "some node" : prev_hp->delegate; } } } return other; } /*! * \internal * \brief Sort fencing history, pending first then by most recently completed * * \param[in,out] history List of stonith actions * * \return New head of sorted \p history */ stonith_history_t * stonith__sort_history(stonith_history_t *history) { stonith_history_t *new = NULL, *pending = NULL, *hp, *np, *tmp; for (hp = history; hp; ) { tmp = hp->next; if ((hp->state == st_done) || (hp->state == st_failed)) { /* sort into new */ if ((!new) || (hp->completed > new->completed) || ((hp->completed == new->completed) && (hp->completed_nsec > new->completed_nsec))) { hp->next = new; new = hp; } else { np = new; do { if ((!np->next) || (hp->completed > np->next->completed) || ((hp->completed == np->next->completed) && (hp->completed_nsec > np->next->completed_nsec))) { hp->next = np->next; np->next = hp; break; } np = np->next; } while (1); } } else { /* put into pending */ hp->next = pending; pending = hp; } hp = tmp; } /* pending actions don't have a completed-stamp so make them go front */ if (pending) { stonith_history_t *last_pending = pending; while (last_pending->next) { last_pending = last_pending->next; } last_pending->next = new; new = pending; } return new; } /*! * \brief Return string equivalent of an operation state value * * \param[in] state Fencing operation state value * * \return Human-friendly string equivalent of state */ const char * stonith_op_state_str(enum op_state state) { switch (state) { case st_query: return "querying"; case st_exec: return "executing"; case st_done: return "completed"; case st_duplicate: return "duplicate"; case st_failed: return "failed"; } return "unknown"; } stonith_history_t * stonith__first_matching_event(stonith_history_t *history, bool (*matching_fn)(stonith_history_t *, void *), void *user_data) { for (stonith_history_t *hp = history; hp; hp = hp->next) { if (matching_fn(hp, user_data)) { return hp; } } return NULL; } bool stonith__event_state_pending(stonith_history_t *history, void *user_data) { return history->state != st_failed && history->state != st_done; } bool stonith__event_state_eq(stonith_history_t *history, void *user_data) { return history->state == GPOINTER_TO_INT(user_data); } bool stonith__event_state_neq(stonith_history_t *history, void *user_data) { return history->state != GPOINTER_TO_INT(user_data); } void stonith__device_parameter_flags(uint32_t *device_flags, const char *device_name, xmlNode *metadata) { xmlXPathObjectPtr xpath = NULL; int max = 0; int lpc = 0; CRM_CHECK((device_flags != NULL) && (metadata != NULL), return); xpath = xpath_search(metadata, "//" PCMK_XE_PARAMETER); max = numXpathResults(xpath); if (max <= 0) { freeXpathObject(xpath); return; } for (lpc = 0; lpc < max; lpc++) { const char *parameter = NULL; xmlNode *match = getXpathResult(xpath, lpc); CRM_LOG_ASSERT(match != NULL); if (match == NULL) { continue; } parameter = crm_element_value(match, PCMK_XA_NAME); if (pcmk__str_eq(parameter, "plug", pcmk__str_casei)) { stonith__set_device_flags(*device_flags, device_name, st_device_supports_parameter_plug); } else if (pcmk__str_eq(parameter, "port", pcmk__str_casei)) { stonith__set_device_flags(*device_flags, device_name, st_device_supports_parameter_port); } } freeXpathObject(xpath); } /*! * \internal * \brief Retrieve fence agent meta-data asynchronously * * \param[in] agent Agent to execute * \param[in] timeout_sec Error if not complete within this time * \param[in] callback Function to call with result (this will always be * called, whether by this function directly or * later via the main loop, and on success the * metadata will be in its result argument's * action_stdout) * \param[in,out] user_data User data to pass to callback * * \return Standard Pacemaker return code * \note The caller must use a main loop. This function is not a * stonith_api_operations_t method because it does not need a stonith_t * object and does not go through the fencer, but executes the agent * directly. */ int stonith__metadata_async(const char *agent, int timeout_sec, void (*callback)(int pid, const pcmk__action_result_t *result, void *user_data), void *user_data) { switch (stonith_get_namespace(agent, NULL)) { case st_namespace_rhcs: { stonith_action_t *action = NULL; int rc = pcmk_ok; action = stonith__action_create(agent, PCMK_ACTION_METADATA, NULL, 0, timeout_sec, NULL, NULL, NULL); rc = stonith__execute_async(action, user_data, callback, NULL); if (rc != pcmk_ok) { callback(0, stonith__action_result(action), user_data); stonith__destroy_action(action); } return pcmk_legacy2rc(rc); } #if HAVE_STONITH_STONITH_H case st_namespace_lha: // LHA metadata is simply synthesized, so simulate async { pcmk__action_result_t result = { .exit_status = CRM_EX_OK, .execution_status = PCMK_EXEC_DONE, .exit_reason = NULL, .action_stdout = NULL, .action_stderr = NULL, }; stonith__lha_metadata(agent, timeout_sec, &result.action_stdout); callback(0, &result, user_data); pcmk__reset_result(&result); return pcmk_rc_ok; } #endif default: { pcmk__action_result_t result = { .exit_status = CRM_EX_NOSUCH, .execution_status = PCMK_EXEC_ERROR_HARD, .exit_reason = crm_strdup_printf("No such agent '%s'", agent), .action_stdout = NULL, .action_stderr = NULL, }; callback(0, &result, user_data); pcmk__reset_result(&result); return ENOENT; } } } /*! * \internal * \brief Return the exit status from an async action callback * * \param[in] data Callback data * * \return Exit status from callback data */ int stonith__exit_status(const stonith_callback_data_t *data) { if ((data == NULL) || (data->opaque == NULL)) { return CRM_EX_ERROR; } return ((pcmk__action_result_t *) data->opaque)->exit_status; } /*! * \internal * \brief Return the execution status from an async action callback * * \param[in] data Callback data * * \return Execution status from callback data */ int stonith__execution_status(const stonith_callback_data_t *data) { if ((data == NULL) || (data->opaque == NULL)) { return PCMK_EXEC_UNKNOWN; } return ((pcmk__action_result_t *) data->opaque)->execution_status; } /*! * \internal * \brief Return the exit reason from an async action callback * * \param[in] data Callback data * * \return Exit reason from callback data */ const char * stonith__exit_reason(const stonith_callback_data_t *data) { if ((data == NULL) || (data->opaque == NULL)) { return NULL; } return ((pcmk__action_result_t *) data->opaque)->exit_reason; } /*! * \internal * \brief Return the exit status from an event notification * * \param[in] event Event * * \return Exit status from event */ int stonith__event_exit_status(const stonith_event_t *event) { if ((event == NULL) || (event->opaque == NULL)) { return CRM_EX_ERROR; } else { struct event_private *event_private = event->opaque; return event_private->result.exit_status; } } /*! * \internal * \brief Return the execution status from an event notification * * \param[in] event Event * * \return Execution status from event */ int stonith__event_execution_status(const stonith_event_t *event) { if ((event == NULL) || (event->opaque == NULL)) { return PCMK_EXEC_UNKNOWN; } else { struct event_private *event_private = event->opaque; return event_private->result.execution_status; } } /*! * \internal * \brief Return the exit reason from an event notification * * \param[in] event Event * * \return Exit reason from event */ const char * stonith__event_exit_reason(const stonith_event_t *event) { if ((event == NULL) || (event->opaque == NULL)) { return NULL; } else { struct event_private *event_private = event->opaque; return event_private->result.exit_reason; } } /*! * \internal * \brief Return a human-friendly description of a fencing event * * \param[in] event Event to describe * * \return Newly allocated string with description of \p event * \note The caller is responsible for freeing the return value. * This function asserts on memory errors and never returns NULL. */ char * stonith__event_description(const stonith_event_t *event) { // Use somewhat readable defaults const char *origin = pcmk__s(event->client_origin, "a client"); const char *origin_node = pcmk__s(event->origin, "a node"); const char *executioner = pcmk__s(event->executioner, "the cluster"); const char *device = pcmk__s(event->device, "unknown"); const char *action = pcmk__s(event->action, event->operation); const char *target = pcmk__s(event->target, "no node"); const char *reason = stonith__event_exit_reason(event); const char *status; if (action == NULL) { action = "(unknown)"; } if (stonith__event_execution_status(event) != PCMK_EXEC_DONE) { status = pcmk_exec_status_str(stonith__event_execution_status(event)); } else if (stonith__event_exit_status(event) != CRM_EX_OK) { status = pcmk_exec_status_str(PCMK_EXEC_ERROR); } else { status = crm_exit_str(CRM_EX_OK); } if (pcmk__str_eq(event->operation, PCMK__VALUE_ST_NOTIFY_HISTORY, pcmk__str_none)) { return crm_strdup_printf("Fencing history may have changed"); } else if (pcmk__str_eq(event->operation, STONITH_OP_DEVICE_ADD, pcmk__str_none)) { return crm_strdup_printf("A fencing device (%s) was added", device); } else if (pcmk__str_eq(event->operation, STONITH_OP_DEVICE_DEL, pcmk__str_none)) { return crm_strdup_printf("A fencing device (%s) was removed", device); } else if (pcmk__str_eq(event->operation, STONITH_OP_LEVEL_ADD, pcmk__str_none)) { return crm_strdup_printf("A fencing topology level (%s) was added", device); } else if (pcmk__str_eq(event->operation, STONITH_OP_LEVEL_DEL, pcmk__str_none)) { return crm_strdup_printf("A fencing topology level (%s) was removed", device); } // event->operation should be PCMK__VALUE_ST_NOTIFY_FENCE at this point return crm_strdup_printf("Operation %s of %s by %s for %s@%s: %s%s%s%s (ref=%s)", action, target, executioner, origin, origin_node, status, ((reason == NULL)? "" : " ("), pcmk__s(reason, ""), ((reason == NULL)? "" : ")"), pcmk__s(event->id, "(none)")); } // Deprecated functions kept only for backward API compatibility // LCOV_EXCL_START const char *get_stonith_provider(const char *agent, const char *provider); const char * get_stonith_provider(const char *agent, const char *provider) { return stonith_namespace2text(stonith_get_namespace(agent, provider)); } // LCOV_EXCL_STOP // End deprecated API diff --git a/lib/fencing/st_rhcs.c b/lib/fencing/st_rhcs.c index ebf3399bf6..bf5b0f7f31 100644 --- a/lib/fencing/st_rhcs.c +++ b/lib/fencing/st_rhcs.c @@ -1,330 +1,330 @@ /* - * Copyright 2004-2024 the Pacemaker project contributors + * Copyright 2004-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include #include "fencing_private.h" #define RH_STONITH_PREFIX "fence_" /*! * \internal * \brief Add available RHCS-compatible agents to a list * * \param[in,out] List to add to * * \return Number of agents added */ int stonith__list_rhcs_agents(stonith_key_value_t **devices) { // Essentially: ls -1 @sbin_dir@/fence_* int count = 0, i; - struct dirent **namelist; + struct dirent **namelist = NULL; const int file_num = scandir(PCMK__FENCE_BINDIR, &namelist, 0, alphasort); #if _POSIX_C_SOURCE < 200809L && !(defined(O_SEARCH) || defined(O_PATH)) char buffer[FILENAME_MAX + 1]; #elif defined(O_SEARCH) const int dirfd = open(PCMK__FENCE_BINDIR, O_SEARCH); #else const int dirfd = open(PCMK__FENCE_BINDIR, O_PATH); #endif for (i = 0; i < file_num; i++) { struct stat prop; if (pcmk__starts_with(namelist[i]->d_name, RH_STONITH_PREFIX)) { #if _POSIX_C_SOURCE < 200809L && !(defined(O_SEARCH) || defined(O_PATH)) snprintf(buffer, sizeof(buffer), "%s/%s", PCMK__FENCE_BINDIR, namelist[i]->d_name); if (stat(buffer, &prop) == 0 && S_ISREG(prop.st_mode)) { #else if (dirfd == -1) { if (i == 0) { crm_notice("Problem with listing %s directory " CRM_XS " errno=%d", RH_STONITH_PREFIX, errno); } free(namelist[i]); continue; } /* note: we can possibly prevent following symlinks here, which may be a good idea, but fall on the nose when these agents are moved elsewhere & linked back */ if (fstatat(dirfd, namelist[i]->d_name, &prop, 0) == 0 && S_ISREG(prop.st_mode)) { #endif *devices = stonith_key_value_add(*devices, NULL, namelist[i]->d_name); count++; } } free(namelist[i]); } if (file_num > 0) { free(namelist); } #if _POSIX_C_SOURCE >= 200809L || defined(O_SEARCH) || defined(O_PATH) if (dirfd >= 0) { close(dirfd); } #endif return count; } static void stonith_rhcs_parameter_not_required(xmlNode *metadata, const char *parameter) { char *xpath = NULL; xmlXPathObject *xpathObj = NULL; CRM_CHECK(metadata != NULL, return); CRM_CHECK(parameter != NULL, return); xpath = crm_strdup_printf("//" PCMK_XE_PARAMETER "[@" PCMK_XA_NAME "='%s']", parameter); /* Fudge metadata so that the parameter isn't required in config * Pacemaker handles and adds it */ xpathObj = xpath_search(metadata, xpath); if (numXpathResults(xpathObj) > 0) { xmlNode *tmp = getXpathResult(xpathObj, 0); crm_xml_add(tmp, "required", "0"); } freeXpathObject(xpathObj); free(xpath); } /*! * \brief Execute RHCS-compatible agent's metadata action * * \param[in] agent Agent to execute * \param[in] timeout_sec Action timeout * \param[out] metadata Where to store output xmlNode (or NULL to ignore) */ static int stonith__rhcs_get_metadata(const char *agent, int timeout_sec, xmlNode **metadata) { xmlNode *xml = NULL; xmlNode *actions = NULL; xmlXPathObject *xpathObj = NULL; stonith_action_t *action = stonith__action_create(agent, PCMK_ACTION_METADATA, NULL, 0, timeout_sec, NULL, NULL, NULL); int rc = stonith__execute(action); pcmk__action_result_t *result = stonith__action_result(action); if (result == NULL) { if (rc < 0) { crm_warn("Could not execute metadata action for %s: %s " CRM_XS " rc=%d", agent, pcmk_strerror(rc), rc); } stonith__destroy_action(action); return rc; } if (result->execution_status != PCMK_EXEC_DONE) { crm_warn("Could not execute metadata action for %s: %s", agent, pcmk_exec_status_str(result->execution_status)); rc = pcmk_rc2legacy(stonith__result2rc(result)); stonith__destroy_action(action); return rc; } if (!pcmk__result_ok(result)) { crm_warn("Metadata action for %s returned error code %d", agent, result->exit_status); rc = pcmk_rc2legacy(stonith__result2rc(result)); stonith__destroy_action(action); return rc; } if (result->action_stdout == NULL) { crm_warn("Metadata action for %s returned no data", agent); stonith__destroy_action(action); return -ENODATA; } xml = pcmk__xml_parse(result->action_stdout); stonith__destroy_action(action); if (xml == NULL) { crm_warn("Metadata for %s is invalid", agent); return -pcmk_err_schema_validation; } xpathObj = xpath_search(xml, "//" PCMK_XE_ACTIONS); if (numXpathResults(xpathObj) > 0) { actions = getXpathResult(xpathObj, 0); } freeXpathObject(xpathObj); // Add start and stop (implemented by pacemaker, not agent) to meta-data xpathObj = xpath_search(xml, "//" PCMK_XE_ACTION "[@" PCMK_XA_NAME "='" PCMK_ACTION_STOP "']"); if (numXpathResults(xpathObj) <= 0) { xmlNode *tmp = NULL; const char *timeout_str = NULL; timeout_str = pcmk__readable_interval(PCMK_DEFAULT_ACTION_TIMEOUT_MS); tmp = pcmk__xe_create(actions, PCMK_XE_ACTION); crm_xml_add(tmp, PCMK_XA_NAME, PCMK_ACTION_STOP); crm_xml_add(tmp, PCMK_META_TIMEOUT, timeout_str); tmp = pcmk__xe_create(actions, PCMK_XE_ACTION); crm_xml_add(tmp, PCMK_XA_NAME, PCMK_ACTION_START); crm_xml_add(tmp, PCMK_META_TIMEOUT, timeout_str); } freeXpathObject(xpathObj); // Fudge metadata so parameters are not required in config (pacemaker adds them) stonith_rhcs_parameter_not_required(xml, "action"); stonith_rhcs_parameter_not_required(xml, "plug"); stonith_rhcs_parameter_not_required(xml, "port"); if (metadata) { *metadata = xml; } else { free_xml(xml); } return pcmk_ok; } /*! * \brief Retrieve metadata for RHCS-compatible fence agent * * \param[in] agent Agent to execute * \param[in] timeout_sec Action timeout * \param[out] output Where to store action output (or NULL to ignore) */ int stonith__rhcs_metadata(const char *agent, int timeout_sec, char **output) { GString *buffer = NULL; xmlNode *xml = NULL; int rc = stonith__rhcs_get_metadata(agent, timeout_sec, &xml); if (rc != pcmk_ok) { goto done; } buffer = g_string_sized_new(1024); pcmk__xml_string(xml, pcmk__xml_fmt_pretty|pcmk__xml_fmt_text, buffer, 0); if (pcmk__str_empty(buffer->str)) { rc = -pcmk_err_schema_validation; goto done; } if (output != NULL) { pcmk__str_update(output, buffer->str); } done: if (buffer != NULL) { g_string_free(buffer, TRUE); } free_xml(xml); return rc; } bool stonith__agent_is_rhcs(const char *agent) { struct stat prop; char *buffer = crm_strdup_printf(PCMK__FENCE_BINDIR "/%s", agent); int rc = stat(buffer, &prop); free(buffer); return (rc >= 0) && S_ISREG(prop.st_mode); } int stonith__rhcs_validate(stonith_t *st, int call_options, const char *target, const char *agent, GHashTable *params, const char * host_arg, int timeout, char **output, char **error_output) { int rc = pcmk_ok; int remaining_timeout = timeout; xmlNode *metadata = NULL; stonith_action_t *action = NULL; pcmk__action_result_t *result = NULL; if (host_arg == NULL) { time_t start_time = time(NULL); rc = stonith__rhcs_get_metadata(agent, remaining_timeout, &metadata); if (rc == pcmk_ok) { uint32_t device_flags = 0; stonith__device_parameter_flags(&device_flags, agent, metadata); if (pcmk_is_set(device_flags, st_device_supports_parameter_port)) { host_arg = "port"; } else if (pcmk_is_set(device_flags, st_device_supports_parameter_plug)) { host_arg = "plug"; } } free_xml(metadata); remaining_timeout -= time(NULL) - start_time; if (rc == -ETIME || remaining_timeout <= 0 ) { return -ETIME; } } else if (pcmk__str_eq(host_arg, PCMK_VALUE_NONE, pcmk__str_casei)) { host_arg = NULL; } action = stonith__action_create(agent, PCMK_ACTION_VALIDATE_ALL, target, 0, remaining_timeout, params, NULL, host_arg); rc = stonith__execute(action); result = stonith__action_result(action); if (result != NULL) { rc = pcmk_rc2legacy(stonith__result2rc(result)); // Take ownership of output so stonith__destroy_action() doesn't free it if (output != NULL) { *output = result->action_stdout; result->action_stdout = NULL; } if (error_output != NULL) { *error_output = result->action_stderr; result->action_stderr = NULL; } } stonith__destroy_action(action); return rc; } diff --git a/lib/lrmd/Makefile.am b/lib/lrmd/Makefile.am index a8b716c03d..202fcb5c12 100644 --- a/lib/lrmd/Makefile.am +++ b/lib/lrmd/Makefile.am @@ -1,26 +1,26 @@ # # Copyright 2012-2023 the Pacemaker project contributors # # The version control history for this file may have further details. # # This source code is licensed under the GNU Lesser General Public License # version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. # include $(top_srcdir)/mk/common.mk lib_LTLIBRARIES = liblrmd.la -liblrmd_la_LDFLAGS = -version-info 31:1:3 +liblrmd_la_LDFLAGS = -version-info 31:2:3 liblrmd_la_CFLAGS = $(CFLAGS_HARDENED_LIB) liblrmd_la_LDFLAGS += $(LDFLAGS_HARDENED_LIB) liblrmd_la_LIBADD = $(top_builddir)/lib/fencing/libstonithd.la liblrmd_la_LIBADD += $(top_builddir)/lib/services/libcrmservice.la liblrmd_la_LIBADD += $(top_builddir)/lib/common/libcrmcommon.la ## Library sources (*must* use += format for bumplibs) liblrmd_la_SOURCES = lrmd_alerts.c liblrmd_la_SOURCES += lrmd_client.c liblrmd_la_SOURCES += lrmd_output.c liblrmd_la_SOURCES += proxy_common.c diff --git a/lib/lrmd/lrmd_client.c b/lib/lrmd/lrmd_client.c index 5fcdf2377e..c0f4e4db1c 100644 --- a/lib/lrmd/lrmd_client.c +++ b/lib/lrmd/lrmd_client.c @@ -1,2731 +1,2737 @@ /* * Copyright 2012-2024 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include // uint32_t, uint64_t #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include // stonith__* #ifdef HAVE_GNUTLS_GNUTLS_H # include #endif #include #include #include #include #include #define MAX_TLS_RECV_WAIT 10000 CRM_TRACE_INIT_DATA(lrmd); static int lrmd_api_disconnect(lrmd_t * lrmd); static int lrmd_api_is_connected(lrmd_t * lrmd); /* IPC proxy functions */ int lrmd_internal_proxy_send(lrmd_t * lrmd, xmlNode *msg); static void lrmd_internal_proxy_dispatch(lrmd_t *lrmd, xmlNode *msg); void lrmd_internal_set_proxy_callback(lrmd_t * lrmd, void *userdata, void (*callback)(lrmd_t *lrmd, void *userdata, xmlNode *msg)); #ifdef HAVE_GNUTLS_GNUTLS_H // GnuTLS client handshake timeout in seconds #define TLS_HANDSHAKE_TIMEOUT 5 gnutls_psk_client_credentials_t psk_cred_s; static void lrmd_tls_disconnect(lrmd_t * lrmd); static int global_remote_msg_id = 0; static void lrmd_tls_connection_destroy(gpointer userdata); static int add_tls_to_mainloop(lrmd_t *lrmd, bool do_handshake); #endif static void report_async_connection_result(lrmd_t * lrmd, int rc); typedef struct lrmd_private_s { uint64_t type; char *token; mainloop_io_t *source; /* IPC parameters */ crm_ipc_t *ipc; pcmk__remote_t *remote; /* Extra TLS parameters */ char *remote_nodename; #ifdef HAVE_GNUTLS_GNUTLS_H char *server; int port; gnutls_psk_client_credentials_t psk_cred_c; /* while the async connection is occurring, this is the id * of the connection timeout timer. */ int async_timer; int sock; /* since tls requires a round trip across the network for a * request/reply, there are times where we just want to be able * to send a request from the client and not wait around (or even care * about) what the reply is. */ int expected_late_replies; GList *pending_notify; crm_trigger_t *process_notify; crm_trigger_t *handshake_trigger; #endif lrmd_event_callback callback; /* Internal IPC proxy msg passing for remote guests */ void (*proxy_callback)(lrmd_t *lrmd, void *userdata, xmlNode *msg); void *proxy_callback_userdata; char *peer_version; } lrmd_private_t; static lrmd_list_t * lrmd_list_add(lrmd_list_t * head, const char *value) { lrmd_list_t *p, *end; p = pcmk__assert_alloc(1, sizeof(lrmd_list_t)); p->val = strdup(value); end = head; while (end && end->next) { end = end->next; } if (end) { end->next = p; } else { head = p; } return head; } void lrmd_list_freeall(lrmd_list_t * head) { lrmd_list_t *p; while (head) { char *val = (char *)head->val; p = head->next; free(val); free(head); head = p; } } lrmd_key_value_t * lrmd_key_value_add(lrmd_key_value_t * head, const char *key, const char *value) { lrmd_key_value_t *p, *end; p = pcmk__assert_alloc(1, sizeof(lrmd_key_value_t)); p->key = strdup(key); p->value = strdup(value); end = head; while (end && end->next) { end = end->next; } if (end) { end->next = p; } else { head = p; } return head; } void lrmd_key_value_freeall(lrmd_key_value_t * head) { lrmd_key_value_t *p; while (head) { p = head->next; free(head->key); free(head->value); free(head); head = p; } } /*! * \brief Create a new lrmd_event_data_t object * * \param[in] rsc_id ID of resource involved in event * \param[in] task Action name * \param[in] interval_ms Action interval * * \return Newly allocated and initialized lrmd_event_data_t * \note This functions asserts on memory errors, so the return value is * guaranteed to be non-NULL. The caller is responsible for freeing the * result with lrmd_free_event(). */ lrmd_event_data_t * lrmd_new_event(const char *rsc_id, const char *task, guint interval_ms) { lrmd_event_data_t *event = pcmk__assert_alloc(1, sizeof(lrmd_event_data_t)); // lrmd_event_data_t has (const char *) members that lrmd_free_event() frees event->rsc_id = pcmk__str_copy(rsc_id); event->op_type = pcmk__str_copy(task); event->interval_ms = interval_ms; return event; } lrmd_event_data_t * lrmd_copy_event(lrmd_event_data_t * event) { lrmd_event_data_t *copy = NULL; copy = pcmk__assert_alloc(1, sizeof(lrmd_event_data_t)); copy->type = event->type; // lrmd_event_data_t has (const char *) members that lrmd_free_event() frees copy->rsc_id = pcmk__str_copy(event->rsc_id); copy->op_type = pcmk__str_copy(event->op_type); copy->user_data = pcmk__str_copy(event->user_data); copy->output = pcmk__str_copy(event->output); copy->remote_nodename = pcmk__str_copy(event->remote_nodename); copy->exit_reason = pcmk__str_copy(event->exit_reason); copy->call_id = event->call_id; copy->timeout = event->timeout; copy->interval_ms = event->interval_ms; copy->start_delay = event->start_delay; copy->rsc_deleted = event->rsc_deleted; copy->rc = event->rc; copy->op_status = event->op_status; copy->t_run = event->t_run; copy->t_rcchange = event->t_rcchange; copy->exec_time = event->exec_time; copy->queue_time = event->queue_time; copy->connection_rc = event->connection_rc; copy->params = pcmk__str_table_dup(event->params); return copy; } /*! * \brief Free an executor event * * \param[in,out] Executor event object to free */ void lrmd_free_event(lrmd_event_data_t *event) { if (event == NULL) { return; } // @TODO Why are these const char *? free((void *) event->rsc_id); free((void *) event->op_type); free((void *) event->user_data); free((void *) event->remote_nodename); lrmd__reset_result(event); if (event->params != NULL) { g_hash_table_destroy(event->params); } free(event); } static void lrmd_dispatch_internal(gpointer data, gpointer user_data) { xmlNode *msg = data; lrmd_t *lrmd = user_data; const char *type; const char *proxy_session = crm_element_value(msg, PCMK__XA_LRMD_IPC_SESSION); lrmd_private_t *native = lrmd->lrmd_private; lrmd_event_data_t event = { 0, }; if (proxy_session != NULL) { /* this is proxy business */ lrmd_internal_proxy_dispatch(lrmd, msg); return; } else if (!native->callback) { /* no callback set */ crm_trace("notify event received but client has not set callback"); return; } event.remote_nodename = native->remote_nodename; type = crm_element_value(msg, PCMK__XA_LRMD_OP); crm_element_value_int(msg, PCMK__XA_LRMD_CALLID, &event.call_id); event.rsc_id = crm_element_value(msg, PCMK__XA_LRMD_RSC_ID); if (pcmk__str_eq(type, LRMD_OP_RSC_REG, pcmk__str_none)) { event.type = lrmd_event_register; } else if (pcmk__str_eq(type, LRMD_OP_RSC_UNREG, pcmk__str_none)) { event.type = lrmd_event_unregister; } else if (pcmk__str_eq(type, LRMD_OP_RSC_EXEC, pcmk__str_none)) { int rc = 0; int exec_time = 0; int queue_time = 0; time_t epoch = 0; crm_element_value_int(msg, PCMK__XA_LRMD_TIMEOUT, &event.timeout); crm_element_value_ms(msg, PCMK__XA_LRMD_RSC_INTERVAL, &event.interval_ms); crm_element_value_int(msg, PCMK__XA_LRMD_RSC_START_DELAY, &event.start_delay); crm_element_value_int(msg, PCMK__XA_LRMD_EXEC_RC, &rc); event.rc = (enum ocf_exitcode) rc; crm_element_value_int(msg, PCMK__XA_LRMD_EXEC_OP_STATUS, &event.op_status); crm_element_value_int(msg, PCMK__XA_LRMD_RSC_DELETED, &event.rsc_deleted); crm_element_value_epoch(msg, PCMK__XA_LRMD_RUN_TIME, &epoch); // coverity[store_truncates_time_t] event.t_run = (unsigned int) epoch; crm_element_value_epoch(msg, PCMK__XA_LRMD_RCCHANGE_TIME, &epoch); // coverity[store_truncates_time_t] event.t_rcchange = (unsigned int) epoch; crm_element_value_int(msg, PCMK__XA_LRMD_EXEC_TIME, &exec_time); CRM_LOG_ASSERT(exec_time >= 0); event.exec_time = QB_MAX(0, exec_time); crm_element_value_int(msg, PCMK__XA_LRMD_QUEUE_TIME, &queue_time); CRM_LOG_ASSERT(queue_time >= 0); event.queue_time = QB_MAX(0, queue_time); event.op_type = crm_element_value(msg, PCMK__XA_LRMD_RSC_ACTION); event.user_data = crm_element_value(msg, PCMK__XA_LRMD_RSC_USERDATA_STR); event.type = lrmd_event_exec_complete; /* output and exit_reason may be freed by a callback */ event.output = crm_element_value_copy(msg, PCMK__XA_LRMD_RSC_OUTPUT); lrmd__set_result(&event, event.rc, event.op_status, crm_element_value(msg, PCMK__XA_LRMD_RSC_EXIT_REASON)); event.params = xml2list(msg); } else if (pcmk__str_eq(type, LRMD_OP_NEW_CLIENT, pcmk__str_none)) { event.type = lrmd_event_new_client; } else if (pcmk__str_eq(type, LRMD_OP_POKE, pcmk__str_none)) { event.type = lrmd_event_poke; } else { return; } crm_trace("op %s notify event received", type); native->callback(&event); if (event.params) { g_hash_table_destroy(event.params); } lrmd__reset_result(&event); } // \return Always 0, to indicate that IPC mainloop source should be kept static int lrmd_ipc_dispatch(const char *buffer, ssize_t length, gpointer userdata) { lrmd_t *lrmd = userdata; lrmd_private_t *native = lrmd->lrmd_private; if (native->callback != NULL) { xmlNode *msg = pcmk__xml_parse(buffer); lrmd_dispatch_internal(msg, lrmd); free_xml(msg); } return 0; } #ifdef HAVE_GNUTLS_GNUTLS_H static void lrmd_free_xml(gpointer userdata) { free_xml((xmlNode *) userdata); } static bool remote_executor_connected(lrmd_t * lrmd) { lrmd_private_t *native = lrmd->lrmd_private; return (native->remote->tls_session != NULL); } static void handle_remote_msg(xmlNode *xml, lrmd_t *lrmd) { lrmd_private_t *native = lrmd->lrmd_private; const char *msg_type = NULL; msg_type = crm_element_value(xml, PCMK__XA_LRMD_REMOTE_MSG_TYPE); if (pcmk__str_eq(msg_type, "notify", pcmk__str_casei)) { lrmd_dispatch_internal(xml, lrmd); } else if (pcmk__str_eq(msg_type, "reply", pcmk__str_casei)) { if (native->expected_late_replies > 0) { native->expected_late_replies--; } else { int reply_id = 0; crm_element_value_int(xml, PCMK__XA_LRMD_CALLID, &reply_id); /* if this happens, we want to know about it */ crm_err("Got outdated Pacemaker Remote reply %d", reply_id); } } } static void process_pending_notifies(lrmd_t *lrmd) { lrmd_private_t *native = lrmd->lrmd_private; if (native->pending_notify == NULL) { return; } crm_trace("Processing pending notifies"); g_list_foreach(native->pending_notify, lrmd_dispatch_internal, lrmd); g_list_free_full(native->pending_notify, lrmd_free_xml); native->pending_notify = NULL; } /*! * \internal * \brief TLS dispatch function (for both trigger and file descriptor sources) * * \param[in,out] userdata API connection * * \return Always return a nonnegative value, which as a file descriptor * dispatch function means keep the mainloop source, and as a * trigger dispatch function, 0 means remove the trigger from the * mainloop while 1 means keep it (and job completed) */ static int lrmd_tls_dispatch(gpointer userdata) { lrmd_t *lrmd = userdata; lrmd_private_t *native = lrmd->lrmd_private; xmlNode *xml = NULL; int rc = pcmk_rc_ok; if (!remote_executor_connected(lrmd)) { crm_trace("TLS dispatch triggered after disconnect"); return 0; } crm_trace("TLS dispatch triggered"); /* First check if there are any pending notifies to process that came * while we were waiting for replies earlier. */ process_pending_notifies(lrmd); /* Next read the current buffer and see if there are any messages to handle. */ rc = pcmk__remote_ready(native->remote, 0); if (rc == pcmk_rc_ok) { rc = pcmk__read_remote_message(native->remote, -1); } if (rc != pcmk_rc_ok && rc != ETIME) { crm_info("Lost %s executor connection while reading data", (native->remote_nodename? native->remote_nodename : "local")); lrmd_tls_disconnect(lrmd); return 0; } /* If rc is ETIME, there was nothing to read but we may already have a * full message in the buffer */ xml = pcmk__remote_message_xml(native->remote); if (xml == NULL) { return 1; } handle_remote_msg(xml, lrmd); free_xml(xml); return 1; } #endif /* Not used with mainloop */ int lrmd_poll(lrmd_t * lrmd, int timeout) { lrmd_private_t *native = lrmd->lrmd_private; switch (native->type) { case pcmk__client_ipc: return crm_ipc_ready(native->ipc); #ifdef HAVE_GNUTLS_GNUTLS_H case pcmk__client_tls: if (native->pending_notify) { return 1; } else { int rc = pcmk__remote_ready(native->remote, 0); switch (rc) { case pcmk_rc_ok: return 1; case ETIME: return 0; default: return pcmk_rc2legacy(rc); } } #endif default: crm_err("Unsupported executor connection type (bug?): %d", native->type); return -EPROTONOSUPPORT; } } /* Not used with mainloop */ bool lrmd_dispatch(lrmd_t * lrmd) { lrmd_private_t *private = NULL; pcmk__assert(lrmd != NULL); private = lrmd->lrmd_private; switch (private->type) { case pcmk__client_ipc: while (crm_ipc_ready(private->ipc)) { if (crm_ipc_read(private->ipc) > 0) { const char *msg = crm_ipc_buffer(private->ipc); lrmd_ipc_dispatch(msg, strlen(msg), lrmd); } } break; #ifdef HAVE_GNUTLS_GNUTLS_H case pcmk__client_tls: lrmd_tls_dispatch(lrmd); break; #endif default: crm_err("Unsupported executor connection type (bug?): %d", private->type); } if (lrmd_api_is_connected(lrmd) == FALSE) { crm_err("Connection closed"); return FALSE; } return TRUE; } static xmlNode * lrmd_create_op(const char *token, const char *op, xmlNode *data, int timeout, enum lrmd_call_options options) { xmlNode *op_msg = NULL; CRM_CHECK(token != NULL, return NULL); op_msg = pcmk__xe_create(NULL, PCMK__XE_LRMD_COMMAND); crm_xml_add(op_msg, PCMK__XA_T, PCMK__VALUE_LRMD); crm_xml_add(op_msg, PCMK__XA_LRMD_OP, op); crm_xml_add_int(op_msg, PCMK__XA_LRMD_TIMEOUT, timeout); crm_xml_add_int(op_msg, PCMK__XA_LRMD_CALLOPT, options); if (data != NULL) { xmlNode *wrapper = pcmk__xe_create(op_msg, PCMK__XE_LRMD_CALLDATA); pcmk__xml_copy(wrapper, data); } crm_trace("Created executor %s command with call options %.8lx (%d)", op, (long)options, options); return op_msg; } static void lrmd_ipc_connection_destroy(gpointer userdata) { lrmd_t *lrmd = userdata; lrmd_private_t *native = lrmd->lrmd_private; switch (native->type) { case pcmk__client_ipc: crm_info("Disconnected from local executor"); break; #ifdef HAVE_GNUTLS_GNUTLS_H case pcmk__client_tls: crm_info("Disconnected from remote executor on %s", native->remote_nodename); break; #endif default: crm_err("Unsupported executor connection type %d (bug?)", native->type); } /* Prevent these from being cleaned up in lrmd_api_disconnect() */ native->ipc = NULL; native->source = NULL; if (native->callback) { lrmd_event_data_t event = { 0, }; event.type = lrmd_event_disconnect; event.remote_nodename = native->remote_nodename; native->callback(&event); } } #ifdef HAVE_GNUTLS_GNUTLS_H static void lrmd_tls_connection_destroy(gpointer userdata) { lrmd_t *lrmd = userdata; lrmd_private_t *native = lrmd->lrmd_private; crm_info("TLS connection destroyed"); if (native->remote->tls_session) { gnutls_bye(*native->remote->tls_session, GNUTLS_SHUT_RDWR); gnutls_deinit(*native->remote->tls_session); gnutls_free(native->remote->tls_session); native->remote->tls_session = NULL; } if (native->psk_cred_c) { gnutls_psk_free_client_credentials(native->psk_cred_c); } if (native->sock) { close(native->sock); } if (native->process_notify) { mainloop_destroy_trigger(native->process_notify); native->process_notify = NULL; } if (native->pending_notify) { g_list_free_full(native->pending_notify, lrmd_free_xml); native->pending_notify = NULL; } if (native->handshake_trigger != NULL) { mainloop_destroy_trigger(native->handshake_trigger); native->handshake_trigger = NULL; } free(native->remote->buffer); free(native->remote->start_state); native->remote->buffer = NULL; native->remote->start_state = NULL; native->source = 0; native->sock = 0; native->psk_cred_c = NULL; native->sock = 0; if (native->callback) { lrmd_event_data_t event = { 0, }; event.remote_nodename = native->remote_nodename; event.type = lrmd_event_disconnect; native->callback(&event); } return; } // \return Standard Pacemaker return code int lrmd__remote_send_xml(pcmk__remote_t *session, xmlNode *msg, uint32_t id, const char *msg_type) { crm_xml_add_int(msg, PCMK__XA_LRMD_REMOTE_MSG_ID, id); crm_xml_add(msg, PCMK__XA_LRMD_REMOTE_MSG_TYPE, msg_type); return pcmk__remote_send_xml(session, msg); } // \return Standard Pacemaker return code static int read_remote_reply(lrmd_t *lrmd, int total_timeout, int expected_reply_id, xmlNode **reply) { lrmd_private_t *native = lrmd->lrmd_private; time_t start = time(NULL); const char *msg_type = NULL; int reply_id = 0; int remaining_timeout = 0; int rc = pcmk_rc_ok; /* A timeout of 0 here makes no sense. We have to wait a period of time * for the response to come back. If -1 or 0, default to 10 seconds. */ if (total_timeout <= 0 || total_timeout > MAX_TLS_RECV_WAIT) { total_timeout = MAX_TLS_RECV_WAIT; } for (*reply = NULL; *reply == NULL; ) { *reply = pcmk__remote_message_xml(native->remote); if (*reply == NULL) { /* read some more off the tls buffer if we still have time left. */ if (remaining_timeout) { remaining_timeout = total_timeout - ((time(NULL) - start) * 1000); } else { remaining_timeout = total_timeout; } if (remaining_timeout <= 0) { return ETIME; } rc = pcmk__read_remote_message(native->remote, remaining_timeout); if (rc != pcmk_rc_ok) { return rc; } *reply = pcmk__remote_message_xml(native->remote); if (*reply == NULL) { return ENOMSG; } } crm_element_value_int(*reply, PCMK__XA_LRMD_REMOTE_MSG_ID, &reply_id); msg_type = crm_element_value(*reply, PCMK__XA_LRMD_REMOTE_MSG_TYPE); if (!msg_type) { crm_err("Empty msg type received while waiting for reply"); free_xml(*reply); *reply = NULL; } else if (pcmk__str_eq(msg_type, "notify", pcmk__str_casei)) { /* got a notify while waiting for reply, trigger the notify to be processed later */ crm_info("queueing notify"); native->pending_notify = g_list_append(native->pending_notify, *reply); if (native->process_notify) { crm_info("notify trigger set."); mainloop_set_trigger(native->process_notify); } *reply = NULL; } else if (!pcmk__str_eq(msg_type, "reply", pcmk__str_casei)) { /* msg isn't a reply, make some noise */ crm_err("Expected a reply, got %s", msg_type); free_xml(*reply); *reply = NULL; } else if (reply_id != expected_reply_id) { if (native->expected_late_replies > 0) { native->expected_late_replies--; } else { crm_err("Got outdated reply, expected id %d got id %d", expected_reply_id, reply_id); } free_xml(*reply); *reply = NULL; } } if (native->remote->buffer && native->process_notify) { mainloop_set_trigger(native->process_notify); } return rc; } // \return Standard Pacemaker return code static int send_remote_message(lrmd_t *lrmd, xmlNode *msg) { int rc = pcmk_rc_ok; lrmd_private_t *native = lrmd->lrmd_private; global_remote_msg_id++; if (global_remote_msg_id <= 0) { global_remote_msg_id = 1; } rc = lrmd__remote_send_xml(native->remote, msg, global_remote_msg_id, "request"); if (rc != pcmk_rc_ok) { crm_err("Disconnecting because TLS message could not be sent to " "Pacemaker Remote: %s", pcmk_rc_str(rc)); lrmd_tls_disconnect(lrmd); } return rc; } static int lrmd_tls_send_recv(lrmd_t * lrmd, xmlNode * msg, int timeout, xmlNode ** reply) { int rc = 0; xmlNode *xml = NULL; if (!remote_executor_connected(lrmd)) { return -ENOTCONN; } rc = send_remote_message(lrmd, msg); if (rc != pcmk_rc_ok) { return pcmk_rc2legacy(rc); } rc = read_remote_reply(lrmd, timeout, global_remote_msg_id, &xml); if (rc != pcmk_rc_ok) { crm_err("Disconnecting remote after request %d reply not received: %s " CRM_XS " rc=%d timeout=%dms", global_remote_msg_id, pcmk_rc_str(rc), rc, timeout); lrmd_tls_disconnect(lrmd); } if (reply) { *reply = xml; } else { free_xml(xml); } return pcmk_rc2legacy(rc); } #endif static int lrmd_send_xml(lrmd_t * lrmd, xmlNode * msg, int timeout, xmlNode ** reply) { int rc = pcmk_ok; lrmd_private_t *native = lrmd->lrmd_private; switch (native->type) { case pcmk__client_ipc: rc = crm_ipc_send(native->ipc, msg, crm_ipc_client_response, timeout, reply); break; #ifdef HAVE_GNUTLS_GNUTLS_H case pcmk__client_tls: rc = lrmd_tls_send_recv(lrmd, msg, timeout, reply); break; #endif default: crm_err("Unsupported executor connection type (bug?): %d", native->type); rc = -EPROTONOSUPPORT; } return rc; } static int lrmd_send_xml_no_reply(lrmd_t * lrmd, xmlNode * msg) { int rc = pcmk_ok; lrmd_private_t *native = lrmd->lrmd_private; switch (native->type) { case pcmk__client_ipc: rc = crm_ipc_send(native->ipc, msg, crm_ipc_flags_none, 0, NULL); break; #ifdef HAVE_GNUTLS_GNUTLS_H case pcmk__client_tls: rc = send_remote_message(lrmd, msg); if (rc == pcmk_rc_ok) { /* we don't want to wait around for the reply, but * since the request/reply protocol needs to behave the same * as libqb, a reply will eventually come later anyway. */ native->expected_late_replies++; } rc = pcmk_rc2legacy(rc); break; #endif default: crm_err("Unsupported executor connection type (bug?): %d", native->type); rc = -EPROTONOSUPPORT; } return rc; } static int lrmd_api_is_connected(lrmd_t * lrmd) { lrmd_private_t *native = lrmd->lrmd_private; switch (native->type) { case pcmk__client_ipc: return crm_ipc_connected(native->ipc); #ifdef HAVE_GNUTLS_GNUTLS_H case pcmk__client_tls: return remote_executor_connected(lrmd); #endif default: crm_err("Unsupported executor connection type (bug?): %d", native->type); return 0; } } /*! * \internal * \brief Send a prepared API command to the executor * * \param[in,out] lrmd Existing connection to the executor * \param[in] op Name of API command to send * \param[in] data Command data XML to add to the sent command * \param[out] output_data If expecting a reply, it will be stored here * \param[in] timeout Timeout in milliseconds (if 0, defaults to * a sensible value per the type of connection, * standard vs. pacemaker remote); * also propagated to the command XML * \param[in] call_options Call options to pass to server when sending * \param[in] expect_reply If TRUE, wait for a reply from the server; * must be TRUE for IPC (as opposed to TLS) clients * * \return pcmk_ok on success, -errno on error */ static int lrmd_send_command(lrmd_t *lrmd, const char *op, xmlNode *data, xmlNode **output_data, int timeout, enum lrmd_call_options options, gboolean expect_reply) { int rc = pcmk_ok; lrmd_private_t *native = lrmd->lrmd_private; xmlNode *op_msg = NULL; xmlNode *op_reply = NULL; if (!lrmd_api_is_connected(lrmd)) { return -ENOTCONN; } if (op == NULL) { crm_err("No operation specified"); return -EINVAL; } CRM_CHECK(native->token != NULL,; ); crm_trace("Sending %s op to executor", op); op_msg = lrmd_create_op(native->token, op, data, timeout, options); if (op_msg == NULL) { return -EINVAL; } if (expect_reply) { rc = lrmd_send_xml(lrmd, op_msg, timeout, &op_reply); } else { rc = lrmd_send_xml_no_reply(lrmd, op_msg); goto done; } if (rc < 0) { crm_perror(LOG_ERR, "Couldn't perform %s operation (timeout=%d): %d", op, timeout, rc); goto done; } else if(op_reply == NULL) { rc = -ENOMSG; goto done; } rc = pcmk_ok; crm_trace("%s op reply received", op); if (crm_element_value_int(op_reply, PCMK__XA_LRMD_RC, &rc) != 0) { rc = -ENOMSG; goto done; } crm_log_xml_trace(op_reply, "Reply"); if (output_data) { *output_data = op_reply; op_reply = NULL; /* Prevent subsequent free */ } done: if (lrmd_api_is_connected(lrmd) == FALSE) { crm_err("Executor disconnected"); } free_xml(op_msg); free_xml(op_reply); return rc; } static int lrmd_api_poke_connection(lrmd_t * lrmd) { int rc; lrmd_private_t *native = lrmd->lrmd_private; xmlNode *data = pcmk__xe_create(NULL, PCMK__XE_LRMD_RSC); crm_xml_add(data, PCMK__XA_LRMD_ORIGIN, __func__); rc = lrmd_send_command(lrmd, LRMD_OP_POKE, data, NULL, 0, 0, (native->type == pcmk__client_ipc)); free_xml(data); return rc < 0 ? rc : pcmk_ok; } // \return Standard Pacemaker return code int lrmd__validate_remote_settings(lrmd_t *lrmd, GHashTable *hash) { int rc = pcmk_rc_ok; const char *value; lrmd_private_t *native = lrmd->lrmd_private; xmlNode *data = pcmk__xe_create(NULL, PCMK__XA_LRMD_OP); crm_xml_add(data, PCMK__XA_LRMD_ORIGIN, __func__); value = g_hash_table_lookup(hash, PCMK_OPT_STONITH_WATCHDOG_TIMEOUT); if ((value) && (stonith__watchdog_fencing_enabled_for_node(native->remote_nodename))) { crm_xml_add(data, PCMK__XA_LRMD_WATCHDOG, value); } rc = lrmd_send_command(lrmd, LRMD_OP_CHECK, data, NULL, 0, 0, (native->type == pcmk__client_ipc)); free_xml(data); return (rc < 0)? pcmk_legacy2rc(rc) : pcmk_rc_ok; } static int lrmd_handshake(lrmd_t * lrmd, const char *name) { int rc = pcmk_ok; lrmd_private_t *native = lrmd->lrmd_private; xmlNode *reply = NULL; xmlNode *hello = pcmk__xe_create(NULL, PCMK__XE_LRMD_COMMAND); crm_xml_add(hello, PCMK__XA_T, PCMK__VALUE_LRMD); crm_xml_add(hello, PCMK__XA_LRMD_OP, CRM_OP_REGISTER); crm_xml_add(hello, PCMK__XA_LRMD_CLIENTNAME, name); crm_xml_add(hello, PCMK__XA_LRMD_PROTOCOL_VERSION, LRMD_PROTOCOL_VERSION); /* advertise that we are a proxy provider */ if (native->proxy_callback) { pcmk__xe_set_bool_attr(hello, PCMK__XA_LRMD_IS_IPC_PROVIDER, true); } rc = lrmd_send_xml(lrmd, hello, -1, &reply); if (rc < 0) { crm_perror(LOG_DEBUG, "Couldn't complete registration with the executor API: %d", rc); rc = -ECOMM; } else if (reply == NULL) { crm_err("Did not receive registration reply"); rc = -EPROTO; } else { const char *version = crm_element_value(reply, PCMK__XA_LRMD_PROTOCOL_VERSION); const char *msg_type = crm_element_value(reply, PCMK__XA_LRMD_OP); const char *tmp_ticket = crm_element_value(reply, PCMK__XA_LRMD_CLIENTID); const char *start_state = crm_element_value(reply, PCMK__XA_NODE_START_STATE); long long uptime = -1; crm_element_value_int(reply, PCMK__XA_LRMD_RC, &rc); /* The remote executor may add its uptime to the XML reply, which is * useful in handling transient attributes when the connection to the * remote node unexpectedly drops. If no parameter is given, just * default to -1. */ crm_element_value_ll(reply, PCMK__XA_UPTIME, &uptime); native->remote->uptime = uptime; if (start_state) { native->remote->start_state = strdup(start_state); } if (rc == -EPROTO) { crm_err("Executor protocol version mismatch between client (%s) and server (%s)", LRMD_PROTOCOL_VERSION, version); crm_log_xml_err(reply, "Protocol Error"); } else if (!pcmk__str_eq(msg_type, CRM_OP_REGISTER, pcmk__str_casei)) { crm_err("Invalid registration message: %s", msg_type); crm_log_xml_err(reply, "Bad reply"); rc = -EPROTO; } else if (tmp_ticket == NULL) { crm_err("No registration token provided"); crm_log_xml_err(reply, "Bad reply"); rc = -EPROTO; } else { crm_trace("Obtained registration token: %s", tmp_ticket); native->token = strdup(tmp_ticket); native->peer_version = strdup(version?version:"1.0"); /* Included since 1.1 */ rc = pcmk_ok; } } free_xml(reply); free_xml(hello); if (rc != pcmk_ok) { lrmd_api_disconnect(lrmd); } return rc; } static int lrmd_ipc_connect(lrmd_t * lrmd, int *fd) { int rc = pcmk_ok; lrmd_private_t *native = lrmd->lrmd_private; struct ipc_client_callbacks lrmd_callbacks = { .dispatch = lrmd_ipc_dispatch, .destroy = lrmd_ipc_connection_destroy }; crm_info("Connecting to executor"); if (fd) { /* No mainloop */ native->ipc = crm_ipc_new(CRM_SYSTEM_LRMD, 0); if (native->ipc != NULL) { rc = pcmk__connect_generic_ipc(native->ipc); if (rc == pcmk_rc_ok) { rc = pcmk__ipc_fd(native->ipc, fd); } if (rc != pcmk_rc_ok) { crm_err("Connection to executor failed: %s", pcmk_rc_str(rc)); rc = -ENOTCONN; } } } else { native->source = mainloop_add_ipc_client(CRM_SYSTEM_LRMD, G_PRIORITY_HIGH, 0, lrmd, &lrmd_callbacks); native->ipc = mainloop_get_ipc_client(native->source); } if (native->ipc == NULL) { crm_debug("Could not connect to the executor API"); rc = -ENOTCONN; } return rc; } #ifdef HAVE_GNUTLS_GNUTLS_H static void copy_gnutls_datum(gnutls_datum_t *dest, gnutls_datum_t *source) { pcmk__assert((dest != NULL) && (source != NULL) && (source->data != NULL)); dest->data = gnutls_malloc(source->size); pcmk__mem_assert(dest->data); memcpy(dest->data, source->data, source->size); dest->size = source->size; } static void clear_gnutls_datum(gnutls_datum_t *datum) { gnutls_free(datum->data); datum->data = NULL; datum->size = 0; } #define KEY_READ_LEN 256 // Chunk size for reading key from file // \return Standard Pacemaker return code static int read_gnutls_key(const char *location, gnutls_datum_t *key) { FILE *stream = NULL; size_t buf_len = KEY_READ_LEN; if ((location == NULL) || (key == NULL)) { return EINVAL; } stream = fopen(location, "r"); if (stream == NULL) { return errno; } key->data = gnutls_malloc(buf_len); key->size = 0; while (!feof(stream)) { int next = fgetc(stream); if (next == EOF) { if (!feof(stream)) { crm_warn("Pacemaker Remote key read was partially successful " "(copy in memory may be corrupted)"); } break; } if (key->size == buf_len) { buf_len = key->size + KEY_READ_LEN; key->data = gnutls_realloc(key->data, buf_len); pcmk__assert(key->data); } key->data[key->size++] = (unsigned char) next; } fclose(stream); if (key->size == 0) { clear_gnutls_datum(key); return ENOKEY; } return pcmk_rc_ok; } // Cache the most recently used Pacemaker Remote authentication key struct key_cache_s { time_t updated; // When cached key was read (valid for 1 minute) const char *location; // Where cached key was read from gnutls_datum_t key; // Cached key }; static bool key_is_cached(struct key_cache_s *key_cache) { return key_cache->updated != 0; } static bool key_cache_expired(struct key_cache_s *key_cache) { return (time(NULL) - key_cache->updated) >= 60; } static void clear_key_cache(struct key_cache_s *key_cache) { clear_gnutls_datum(&(key_cache->key)); if ((key_cache->updated != 0) || (key_cache->location != NULL)) { key_cache->updated = 0; key_cache->location = NULL; crm_debug("Cleared Pacemaker Remote key cache"); } } static void get_cached_key(struct key_cache_s *key_cache, gnutls_datum_t *key) { copy_gnutls_datum(key, &(key_cache->key)); crm_debug("Using cached Pacemaker Remote key from %s", pcmk__s(key_cache->location, "unknown location")); } static void cache_key(struct key_cache_s *key_cache, gnutls_datum_t *key, const char *location) { key_cache->updated = time(NULL); key_cache->location = location; copy_gnutls_datum(&(key_cache->key), key); crm_debug("Using (and cacheing) Pacemaker Remote key from %s", pcmk__s(location, "unknown location")); } /*! * \internal * \brief Get Pacemaker Remote authentication key from file or cache * * \param[in] location Path to key file to try (this memory must * persist across all calls of this function) * \param[out] key Key from location or cache * * \return Standard Pacemaker return code */ static int get_remote_key(const char *location, gnutls_datum_t *key) { static struct key_cache_s key_cache = { 0, }; int rc = pcmk_rc_ok; if ((location == NULL) || (key == NULL)) { return EINVAL; } if (key_is_cached(&key_cache)) { if (key_cache_expired(&key_cache)) { clear_key_cache(&key_cache); } else { get_cached_key(&key_cache, key); return pcmk_rc_ok; } } rc = read_gnutls_key(location, key); if (rc != pcmk_rc_ok) { return rc; } cache_key(&key_cache, key, location); return pcmk_rc_ok; } /*! * \internal * \brief Initialize the Pacemaker Remote authentication key * * Try loading the Pacemaker Remote authentication key from cache if available, * otherwise from these locations, in order of preference: the value of the * PCMK_authkey_location environment variable, if set; the Pacemaker default key * file location; or (for historical reasons) /etc/corosync/authkey. * * \param[out] key Where to store key * * \return Standard Pacemaker return code */ int lrmd__init_remote_key(gnutls_datum_t *key) { static const char *env_location = NULL; static bool need_env = true; int env_rc = pcmk_rc_ok; int default_rc = pcmk_rc_ok; int alt_rc = pcmk_rc_ok; bool env_is_default = false; bool env_is_fallback = false; if (need_env) { env_location = pcmk__env_option(PCMK__ENV_AUTHKEY_LOCATION); need_env = false; } // Try location in environment variable, if set if (env_location != NULL) { env_rc = get_remote_key(env_location, key); if (env_rc == pcmk_rc_ok) { return pcmk_rc_ok; } env_is_default = !strcmp(env_location, DEFAULT_REMOTE_KEY_LOCATION); env_is_fallback = !strcmp(env_location, ALT_REMOTE_KEY_LOCATION); /* @TODO It would be more secure to fail, rather than fall back to the * default, if an explicitly set key location is not readable, and it * would be better to never use the Corosync location as a fallback. * However, that would break any deployments currently working with the * fallbacks. * * @COMPAT Change at 3.0.0 */ } // Try default location, if environment wasn't explicitly set to it if (env_is_default) { default_rc = env_rc; } else { default_rc = get_remote_key(DEFAULT_REMOTE_KEY_LOCATION, key); } // Try fallback location, if environment wasn't set to it and default failed // @COMPAT Drop at 3.0.0 if (env_is_fallback) { alt_rc = env_rc; } else if (default_rc != pcmk_rc_ok) { alt_rc = get_remote_key(ALT_REMOTE_KEY_LOCATION, key); } // We have all results, so log and return if ((env_rc != pcmk_rc_ok) && (default_rc != pcmk_rc_ok) && (alt_rc != pcmk_rc_ok)) { // Environment set, everything failed crm_warn("Could not read Pacemaker Remote key from %s (%s%s%s%s%s): %s", env_location, env_is_default? "" : "or default location ", env_is_default? "" : DEFAULT_REMOTE_KEY_LOCATION, !env_is_default && !env_is_fallback? " " : "", env_is_fallback? "" : "or fallback location ", env_is_fallback? "" : ALT_REMOTE_KEY_LOCATION, pcmk_rc_str(env_rc)); return ENOKEY; } if (env_rc != pcmk_rc_ok) { // Environment set but failed, using a default crm_warn("Could not read Pacemaker Remote key from %s " "(using %s location %s instead): %s", env_location, (default_rc == pcmk_rc_ok)? "default" : "fallback", (default_rc == pcmk_rc_ok)? DEFAULT_REMOTE_KEY_LOCATION : ALT_REMOTE_KEY_LOCATION, pcmk_rc_str(env_rc)); crm_warn("This undocumented behavior is deprecated and unsafe and will " "be removed in a future release"); return pcmk_rc_ok; } if (default_rc != pcmk_rc_ok) { if (alt_rc == pcmk_rc_ok) { // Environment variable unset, used alternate location // This gets caught by the default return below, but we additionally // warn on this behavior here. crm_warn("Read Pacemaker Remote key from alternate location %s", ALT_REMOTE_KEY_LOCATION); crm_warn("This undocumented behavior is deprecated and unsafe and will " "be removed in a future release"); } else { // Environment unset, defaults failed crm_warn("Could not read Pacemaker Remote key from default location %s" " (or fallback location %s): %s", DEFAULT_REMOTE_KEY_LOCATION, ALT_REMOTE_KEY_LOCATION, pcmk_rc_str(default_rc)); return ENOKEY; } } return pcmk_rc_ok; // Environment variable unset, a default worked } static void lrmd_gnutls_global_init(void) { static int gnutls_init = 0; if (!gnutls_init) { crm_gnutls_global_init(); } gnutls_init = 1; } static void tls_handshake_failed(lrmd_t *lrmd, int tls_rc, int rc) { lrmd_private_t *native = lrmd->lrmd_private; crm_warn("Disconnecting after TLS handshake with " "Pacemaker Remote server %s:%d failed: %s", native->server, native->port, (rc == EPROTO)? gnutls_strerror(tls_rc) : pcmk_rc_str(rc)); report_async_connection_result(lrmd, pcmk_rc2legacy(rc)); gnutls_deinit(*native->remote->tls_session); gnutls_free(native->remote->tls_session); native->remote->tls_session = NULL; lrmd_tls_connection_destroy(lrmd); } static void tls_handshake_succeeded(lrmd_t *lrmd) { int rc = pcmk_rc_ok; lrmd_private_t *native = lrmd->lrmd_private; crm_info("TLS connection to Pacemaker Remote server %s:%d succeeded", native->server, native->port); rc = add_tls_to_mainloop(lrmd, true); report_async_connection_result(lrmd, pcmk_rc2legacy(rc)); } #endif static void report_async_connection_result(lrmd_t * lrmd, int rc) { lrmd_private_t *native = lrmd->lrmd_private; if (native->callback) { lrmd_event_data_t event = { 0, }; event.type = lrmd_event_connect; event.remote_nodename = native->remote_nodename; event.connection_rc = rc; native->callback(&event); } } #ifdef HAVE_GNUTLS_GNUTLS_H /*! * \internal * \brief Perform a TLS client handshake with a Pacemaker Remote server * * \param[in] lrmd Newly established Pacemaker Remote executor connection * * \return Standard Pacemaker return code */ static int tls_client_handshake(lrmd_t *lrmd) { lrmd_private_t *native = lrmd->lrmd_private; int tls_rc = GNUTLS_E_SUCCESS; int rc = pcmk__tls_client_handshake(native->remote, TLS_HANDSHAKE_TIMEOUT, &tls_rc); if (rc != pcmk_rc_ok) { tls_handshake_failed(lrmd, tls_rc, rc); } return rc; } /*! * \internal * \brief Add trigger and file descriptor mainloop sources for TLS * * \param[in,out] lrmd API connection with established TLS session * \param[in] do_handshake Whether to perform executor handshake * * \return Standard Pacemaker return code */ static int add_tls_to_mainloop(lrmd_t *lrmd, bool do_handshake) { lrmd_private_t *native = lrmd->lrmd_private; int rc = pcmk_rc_ok; char *name = crm_strdup_printf("pacemaker-remote-%s:%d", native->server, native->port); struct mainloop_fd_callbacks tls_fd_callbacks = { .dispatch = lrmd_tls_dispatch, .destroy = lrmd_tls_connection_destroy, }; native->process_notify = mainloop_add_trigger(G_PRIORITY_HIGH, lrmd_tls_dispatch, lrmd); native->source = mainloop_add_fd(name, G_PRIORITY_HIGH, native->sock, lrmd, &tls_fd_callbacks); /* Async connections lose the client name provided by the API caller, so we * have to use our generated name here to perform the executor handshake. * * @TODO Keep track of the caller-provided name. Perhaps we should be using * that name in this function instead of generating one anyway. */ if (do_handshake) { rc = lrmd_handshake(lrmd, name); rc = pcmk_legacy2rc(rc); } free(name); return rc; } struct handshake_data_s { lrmd_t *lrmd; time_t start_time; int timeout_sec; }; static gboolean try_handshake_cb(gpointer user_data) { struct handshake_data_s *hs = user_data; lrmd_t *lrmd = hs->lrmd; lrmd_private_t *native = lrmd->lrmd_private; pcmk__remote_t *remote = native->remote; int rc = pcmk_rc_ok; int tls_rc = GNUTLS_E_SUCCESS; if (time(NULL) >= hs->start_time + hs->timeout_sec) { rc = ETIME; tls_handshake_failed(lrmd, GNUTLS_E_TIMEDOUT, rc); free(hs); return 0; } rc = pcmk__tls_client_try_handshake(remote, &tls_rc); if (rc == pcmk_rc_ok) { tls_handshake_succeeded(lrmd); free(hs); return 0; } else if (rc == EAGAIN) { mainloop_set_trigger(native->handshake_trigger); return 1; } else { rc = EKEYREJECTED; tls_handshake_failed(lrmd, tls_rc, rc); free(hs); return 0; } } static void lrmd_tcp_connect_cb(void *userdata, int rc, int sock) { lrmd_t *lrmd = userdata; lrmd_private_t *native = lrmd->lrmd_private; gnutls_datum_t psk_key = { NULL, 0 }; int tls_rc = GNUTLS_E_SUCCESS; native->async_timer = 0; if (rc != pcmk_rc_ok) { lrmd_tls_connection_destroy(lrmd); crm_info("Could not connect to Pacemaker Remote at %s:%d: %s " CRM_XS " rc=%d", native->server, native->port, pcmk_rc_str(rc), rc); report_async_connection_result(lrmd, pcmk_rc2legacy(rc)); return; } /* The TCP connection was successful, so establish the TLS connection. */ native->sock = sock; rc = lrmd__init_remote_key(&psk_key); if (rc != pcmk_rc_ok) { crm_info("Could not connect to Pacemaker Remote at %s:%d: %s " CRM_XS " rc=%d", native->server, native->port, pcmk_rc_str(rc), rc); lrmd_tls_connection_destroy(lrmd); report_async_connection_result(lrmd, pcmk_rc2legacy(rc)); return; } gnutls_psk_allocate_client_credentials(&native->psk_cred_c); gnutls_psk_set_client_credentials(native->psk_cred_c, DEFAULT_REMOTE_USERNAME, &psk_key, GNUTLS_PSK_KEY_RAW); gnutls_free(psk_key.data); native->remote->tls_session = pcmk__new_tls_session(sock, GNUTLS_CLIENT, GNUTLS_CRD_PSK, native->psk_cred_c); if (native->remote->tls_session == NULL) { lrmd_tls_connection_destroy(lrmd); report_async_connection_result(lrmd, -EPROTO); return; } /* If the TLS handshake immediately succeeds or fails, we can handle that * now without having to deal with mainloops and retries. Otherwise, add a * trigger to keep trying until we get a result (or it times out). */ rc = pcmk__tls_client_try_handshake(native->remote, &tls_rc); if (rc == EAGAIN) { struct handshake_data_s *hs = NULL; if (native->handshake_trigger != NULL) { return; } hs = pcmk__assert_alloc(1, sizeof(struct handshake_data_s)); hs->lrmd = lrmd; hs->start_time = time(NULL); hs->timeout_sec = TLS_HANDSHAKE_TIMEOUT; native->handshake_trigger = mainloop_add_trigger(G_PRIORITY_LOW, try_handshake_cb, hs); mainloop_set_trigger(native->handshake_trigger); } else if (rc == pcmk_rc_ok) { tls_handshake_succeeded(lrmd); } else { tls_handshake_failed(lrmd, tls_rc, rc); } } static int lrmd_tls_connect_async(lrmd_t * lrmd, int timeout /*ms */ ) { int rc = pcmk_rc_ok; int timer_id = 0; lrmd_private_t *native = lrmd->lrmd_private; lrmd_gnutls_global_init(); native->sock = -1; rc = pcmk__connect_remote(native->server, native->port, timeout, &timer_id, &(native->sock), lrmd, lrmd_tcp_connect_cb); if (rc != pcmk_rc_ok) { crm_warn("Pacemaker Remote connection to %s:%d failed: %s " CRM_XS " rc=%d", native->server, native->port, pcmk_rc_str(rc), rc); return rc; } native->async_timer = timer_id; return rc; } static int lrmd_tls_connect(lrmd_t * lrmd, int *fd) { int rc = pcmk_rc_ok; lrmd_private_t *native = lrmd->lrmd_private; gnutls_datum_t psk_key = { NULL, 0 }; lrmd_gnutls_global_init(); native->sock = -1; rc = pcmk__connect_remote(native->server, native->port, 0, NULL, &(native->sock), NULL, NULL); if (rc != pcmk_rc_ok) { crm_warn("Pacemaker Remote connection to %s:%d failed: %s " CRM_XS " rc=%d", native->server, native->port, pcmk_rc_str(rc), rc); lrmd_tls_connection_destroy(lrmd); return ENOTCONN; } rc = lrmd__init_remote_key(&psk_key); if (rc != pcmk_rc_ok) { lrmd_tls_connection_destroy(lrmd); return rc; } gnutls_psk_allocate_client_credentials(&native->psk_cred_c); gnutls_psk_set_client_credentials(native->psk_cred_c, DEFAULT_REMOTE_USERNAME, &psk_key, GNUTLS_PSK_KEY_RAW); gnutls_free(psk_key.data); native->remote->tls_session = pcmk__new_tls_session(native->sock, GNUTLS_CLIENT, GNUTLS_CRD_PSK, native->psk_cred_c); if (native->remote->tls_session == NULL) { lrmd_tls_connection_destroy(lrmd); return EPROTO; } if (tls_client_handshake(lrmd) != pcmk_rc_ok) { return EKEYREJECTED; } crm_info("Client TLS connection established with Pacemaker Remote server %s:%d", native->server, native->port); if (fd) { *fd = native->sock; } else { rc = add_tls_to_mainloop(lrmd, false); } return rc; } #endif static int lrmd_api_connect(lrmd_t * lrmd, const char *name, int *fd) { int rc = -ENOTCONN; lrmd_private_t *native = lrmd->lrmd_private; switch (native->type) { case pcmk__client_ipc: rc = lrmd_ipc_connect(lrmd, fd); rc = pcmk_rc2legacy(rc); break; #ifdef HAVE_GNUTLS_GNUTLS_H case pcmk__client_tls: rc = lrmd_tls_connect(lrmd, fd); rc = pcmk_rc2legacy(rc); break; #endif default: crm_err("Unsupported executor connection type (bug?): %d", native->type); rc = -EPROTONOSUPPORT; } if (rc == pcmk_ok) { rc = lrmd_handshake(lrmd, name); } return rc; } static int lrmd_api_connect_async(lrmd_t * lrmd, const char *name, int timeout) { int rc = pcmk_ok; lrmd_private_t *native = lrmd->lrmd_private; CRM_CHECK(native && native->callback, return -EINVAL); switch (native->type) { case pcmk__client_ipc: /* fake async connection with ipc. it should be fast * enough that we gain very little from async */ rc = lrmd_api_connect(lrmd, name, NULL); if (!rc) { report_async_connection_result(lrmd, rc); } break; #ifdef HAVE_GNUTLS_GNUTLS_H case pcmk__client_tls: rc = lrmd_tls_connect_async(lrmd, timeout); break; #endif default: crm_err("Unsupported executor connection type (bug?): %d", native->type); rc = -EPROTONOSUPPORT; } return rc; } static void lrmd_ipc_disconnect(lrmd_t * lrmd) { lrmd_private_t *native = lrmd->lrmd_private; if (native->source != NULL) { /* Attached to mainloop */ mainloop_del_ipc_client(native->source); native->source = NULL; native->ipc = NULL; } else if (native->ipc) { /* Not attached to mainloop */ crm_ipc_t *ipc = native->ipc; native->ipc = NULL; crm_ipc_close(ipc); crm_ipc_destroy(ipc); } } #ifdef HAVE_GNUTLS_GNUTLS_H static void lrmd_tls_disconnect(lrmd_t * lrmd) { lrmd_private_t *native = lrmd->lrmd_private; if (native->remote->tls_session) { gnutls_bye(*native->remote->tls_session, GNUTLS_SHUT_RDWR); gnutls_deinit(*native->remote->tls_session); gnutls_free(native->remote->tls_session); native->remote->tls_session = NULL; } if (native->async_timer) { g_source_remove(native->async_timer); native->async_timer = 0; } if (native->source != NULL) { /* Attached to mainloop */ mainloop_del_ipc_client(native->source); native->source = NULL; } else if (native->sock) { close(native->sock); native->sock = 0; } if (native->pending_notify) { g_list_free_full(native->pending_notify, lrmd_free_xml); native->pending_notify = NULL; } } #endif static int lrmd_api_disconnect(lrmd_t * lrmd) { lrmd_private_t *native = lrmd->lrmd_private; int rc = pcmk_ok; switch (native->type) { case pcmk__client_ipc: crm_debug("Disconnecting from local executor"); lrmd_ipc_disconnect(lrmd); break; #ifdef HAVE_GNUTLS_GNUTLS_H case pcmk__client_tls: crm_debug("Disconnecting from remote executor on %s", native->remote_nodename); lrmd_tls_disconnect(lrmd); break; #endif default: crm_err("Unsupported executor connection type (bug?): %d", native->type); rc = -EPROTONOSUPPORT; } free(native->token); native->token = NULL; free(native->peer_version); native->peer_version = NULL; return rc; } static int lrmd_api_register_rsc(lrmd_t * lrmd, const char *rsc_id, const char *class, const char *provider, const char *type, enum lrmd_call_options options) { int rc = pcmk_ok; xmlNode *data = NULL; if (!class || !type || !rsc_id) { return -EINVAL; } if (pcmk_is_set(pcmk_get_ra_caps(class), pcmk_ra_cap_provider) && (provider == NULL)) { return -EINVAL; } data = pcmk__xe_create(NULL, PCMK__XE_LRMD_RSC); crm_xml_add(data, PCMK__XA_LRMD_ORIGIN, __func__); crm_xml_add(data, PCMK__XA_LRMD_RSC_ID, rsc_id); crm_xml_add(data, PCMK__XA_LRMD_CLASS, class); crm_xml_add(data, PCMK__XA_LRMD_PROVIDER, provider); crm_xml_add(data, PCMK__XA_LRMD_TYPE, type); rc = lrmd_send_command(lrmd, LRMD_OP_RSC_REG, data, NULL, 0, options, TRUE); free_xml(data); return rc; } static int lrmd_api_unregister_rsc(lrmd_t * lrmd, const char *rsc_id, enum lrmd_call_options options) { int rc = pcmk_ok; xmlNode *data = pcmk__xe_create(NULL, PCMK__XE_LRMD_RSC); crm_xml_add(data, PCMK__XA_LRMD_ORIGIN, __func__); crm_xml_add(data, PCMK__XA_LRMD_RSC_ID, rsc_id); rc = lrmd_send_command(lrmd, LRMD_OP_RSC_UNREG, data, NULL, 0, options, TRUE); free_xml(data); return rc; } lrmd_rsc_info_t * lrmd_new_rsc_info(const char *rsc_id, const char *standard, const char *provider, const char *type) { lrmd_rsc_info_t *rsc_info = pcmk__assert_alloc(1, sizeof(lrmd_rsc_info_t)); rsc_info->id = pcmk__str_copy(rsc_id); rsc_info->standard = pcmk__str_copy(standard); rsc_info->provider = pcmk__str_copy(provider); rsc_info->type = pcmk__str_copy(type); return rsc_info; } lrmd_rsc_info_t * lrmd_copy_rsc_info(lrmd_rsc_info_t * rsc_info) { return lrmd_new_rsc_info(rsc_info->id, rsc_info->standard, rsc_info->provider, rsc_info->type); } void lrmd_free_rsc_info(lrmd_rsc_info_t * rsc_info) { if (!rsc_info) { return; } free(rsc_info->id); free(rsc_info->type); free(rsc_info->standard); free(rsc_info->provider); free(rsc_info); } static lrmd_rsc_info_t * lrmd_api_get_rsc_info(lrmd_t * lrmd, const char *rsc_id, enum lrmd_call_options options) { lrmd_rsc_info_t *rsc_info = NULL; xmlNode *data = pcmk__xe_create(NULL, PCMK__XE_LRMD_RSC); xmlNode *output = NULL; const char *class = NULL; const char *provider = NULL; const char *type = NULL; crm_xml_add(data, PCMK__XA_LRMD_ORIGIN, __func__); crm_xml_add(data, PCMK__XA_LRMD_RSC_ID, rsc_id); lrmd_send_command(lrmd, LRMD_OP_RSC_INFO, data, &output, 0, options, TRUE); free_xml(data); if (!output) { return NULL; } class = crm_element_value(output, PCMK__XA_LRMD_CLASS); provider = crm_element_value(output, PCMK__XA_LRMD_PROVIDER); type = crm_element_value(output, PCMK__XA_LRMD_TYPE); if (!class || !type) { free_xml(output); return NULL; } else if (pcmk_is_set(pcmk_get_ra_caps(class), pcmk_ra_cap_provider) && !provider) { free_xml(output); return NULL; } rsc_info = lrmd_new_rsc_info(rsc_id, class, provider, type); free_xml(output); return rsc_info; } void lrmd_free_op_info(lrmd_op_info_t *op_info) { if (op_info) { free(op_info->rsc_id); free(op_info->action); free(op_info->interval_ms_s); free(op_info->timeout_ms_s); free(op_info); } } static int lrmd_api_get_recurring_ops(lrmd_t *lrmd, const char *rsc_id, int timeout_ms, enum lrmd_call_options options, GList **output) { xmlNode *data = NULL; xmlNode *output_xml = NULL; int rc = pcmk_ok; if (output == NULL) { return -EINVAL; } *output = NULL; // Send request if (rsc_id) { data = pcmk__xe_create(NULL, PCMK__XE_LRMD_RSC); crm_xml_add(data, PCMK__XA_LRMD_ORIGIN, __func__); crm_xml_add(data, PCMK__XA_LRMD_RSC_ID, rsc_id); } rc = lrmd_send_command(lrmd, LRMD_OP_GET_RECURRING, data, &output_xml, timeout_ms, options, TRUE); if (data) { free_xml(data); } // Process reply if ((rc != pcmk_ok) || (output_xml == NULL)) { return rc; } for (const xmlNode *rsc_xml = pcmk__xe_first_child(output_xml, PCMK__XE_LRMD_RSC, NULL, NULL); (rsc_xml != NULL) && (rc == pcmk_ok); rsc_xml = pcmk__xe_next_same(rsc_xml)) { rsc_id = crm_element_value(rsc_xml, PCMK__XA_LRMD_RSC_ID); if (rsc_id == NULL) { crm_err("Could not parse recurring operation information from executor"); continue; } for (const xmlNode *op_xml = pcmk__xe_first_child(rsc_xml, PCMK__XE_LRMD_RSC_OP, NULL, NULL); op_xml != NULL; op_xml = pcmk__xe_next_same(op_xml)) { lrmd_op_info_t *op_info = calloc(1, sizeof(lrmd_op_info_t)); if (op_info == NULL) { rc = -ENOMEM; break; } op_info->rsc_id = strdup(rsc_id); op_info->action = crm_element_value_copy(op_xml, PCMK__XA_LRMD_RSC_ACTION); op_info->interval_ms_s = crm_element_value_copy(op_xml, PCMK__XA_LRMD_RSC_INTERVAL); op_info->timeout_ms_s = crm_element_value_copy(op_xml, PCMK__XA_LRMD_TIMEOUT); *output = g_list_prepend(*output, op_info); } } free_xml(output_xml); return rc; } static void lrmd_api_set_callback(lrmd_t * lrmd, lrmd_event_callback callback) { lrmd_private_t *native = lrmd->lrmd_private; native->callback = callback; } void lrmd_internal_set_proxy_callback(lrmd_t * lrmd, void *userdata, void (*callback)(lrmd_t *lrmd, void *userdata, xmlNode *msg)) { lrmd_private_t *native = lrmd->lrmd_private; native->proxy_callback = callback; native->proxy_callback_userdata = userdata; } void lrmd_internal_proxy_dispatch(lrmd_t *lrmd, xmlNode *msg) { lrmd_private_t *native = lrmd->lrmd_private; if (native->proxy_callback) { crm_log_xml_trace(msg, "PROXY_INBOUND"); native->proxy_callback(lrmd, native->proxy_callback_userdata, msg); } } int lrmd_internal_proxy_send(lrmd_t * lrmd, xmlNode *msg) { if (lrmd == NULL) { return -ENOTCONN; } crm_xml_add(msg, PCMK__XA_LRMD_OP, CRM_OP_IPC_FWD); crm_log_xml_trace(msg, "PROXY_OUTBOUND"); return lrmd_send_xml_no_reply(lrmd, msg); } static int stonith_get_metadata(const char *provider, const char *type, char **output) { int rc = pcmk_ok; stonith_t *stonith_api = stonith_api_new(); if (stonith_api == NULL) { crm_err("Could not get fence agent meta-data: API memory allocation failed"); return -ENOMEM; } rc = stonith_api->cmds->metadata(stonith_api, st_opt_sync_call, type, provider, output, 0); if ((rc == pcmk_ok) && (*output == NULL)) { rc = -EIO; } stonith_api->cmds->free(stonith_api); return rc; } static int lrmd_api_get_metadata(lrmd_t *lrmd, const char *standard, const char *provider, const char *type, char **output, enum lrmd_call_options options) { return lrmd->cmds->get_metadata_params(lrmd, standard, provider, type, output, options, NULL); } static int lrmd_api_get_metadata_params(lrmd_t *lrmd, const char *standard, const char *provider, const char *type, char **output, enum lrmd_call_options options, lrmd_key_value_t *params) { svc_action_t *action = NULL; GHashTable *params_table = NULL; if (!standard || !type) { lrmd_key_value_freeall(params); return -EINVAL; } if (pcmk__str_eq(standard, PCMK_RESOURCE_CLASS_STONITH, pcmk__str_casei)) { lrmd_key_value_freeall(params); return stonith_get_metadata(provider, type, output); } params_table = pcmk__strkey_table(free, free); for (const lrmd_key_value_t *param = params; param; param = param->next) { pcmk__insert_dup(params_table, param->key, param->value); } action = services__create_resource_action(type, standard, provider, type, PCMK_ACTION_META_DATA, 0, PCMK_DEFAULT_METADATA_TIMEOUT_MS, params_table, 0); lrmd_key_value_freeall(params); if (action == NULL) { return -ENOMEM; } if (action->rc != PCMK_OCF_UNKNOWN) { services_action_free(action); return -EINVAL; } if (!services_action_sync(action)) { crm_err("Failed to retrieve meta-data for %s:%s:%s", standard, provider, type); services_action_free(action); return -EIO; } if (!action->stdout_data) { crm_err("Failed to receive meta-data for %s:%s:%s", standard, provider, type); services_action_free(action); return -EIO; } + /* This is a complaint about services__get_nagios_metadata not NULL-terminating + * stdout_data when it calls fread(). However, that function allocates with + * pcmk__assert_alloc which uses calloc() to allocate and fill with zeros, so + * this is a false positive. + */ + // coverity[string_null] *output = strdup(action->stdout_data); services_action_free(action); return pcmk_ok; } static int lrmd_api_exec(lrmd_t *lrmd, const char *rsc_id, const char *action, const char *userdata, guint interval_ms, int timeout, /* ms */ int start_delay, /* ms */ enum lrmd_call_options options, lrmd_key_value_t * params) { int rc = pcmk_ok; xmlNode *data = pcmk__xe_create(NULL, PCMK__XE_LRMD_RSC); xmlNode *args = pcmk__xe_create(data, PCMK__XE_ATTRIBUTES); lrmd_key_value_t *tmp = NULL; crm_xml_add(data, PCMK__XA_LRMD_ORIGIN, __func__); crm_xml_add(data, PCMK__XA_LRMD_RSC_ID, rsc_id); crm_xml_add(data, PCMK__XA_LRMD_RSC_ACTION, action); crm_xml_add(data, PCMK__XA_LRMD_RSC_USERDATA_STR, userdata); crm_xml_add_ms(data, PCMK__XA_LRMD_RSC_INTERVAL, interval_ms); crm_xml_add_int(data, PCMK__XA_LRMD_TIMEOUT, timeout); crm_xml_add_int(data, PCMK__XA_LRMD_RSC_START_DELAY, start_delay); for (tmp = params; tmp; tmp = tmp->next) { hash2smartfield((gpointer) tmp->key, (gpointer) tmp->value, args); } rc = lrmd_send_command(lrmd, LRMD_OP_RSC_EXEC, data, NULL, timeout, options, TRUE); free_xml(data); lrmd_key_value_freeall(params); return rc; } /* timeout is in ms */ static int lrmd_api_exec_alert(lrmd_t *lrmd, const char *alert_id, const char *alert_path, int timeout, lrmd_key_value_t *params) { int rc = pcmk_ok; xmlNode *data = pcmk__xe_create(NULL, PCMK__XE_LRMD_ALERT); xmlNode *args = pcmk__xe_create(data, PCMK__XE_ATTRIBUTES); lrmd_key_value_t *tmp = NULL; crm_xml_add(data, PCMK__XA_LRMD_ORIGIN, __func__); crm_xml_add(data, PCMK__XA_LRMD_ALERT_ID, alert_id); crm_xml_add(data, PCMK__XA_LRMD_ALERT_PATH, alert_path); crm_xml_add_int(data, PCMK__XA_LRMD_TIMEOUT, timeout); for (tmp = params; tmp; tmp = tmp->next) { hash2smartfield((gpointer) tmp->key, (gpointer) tmp->value, args); } rc = lrmd_send_command(lrmd, LRMD_OP_ALERT_EXEC, data, NULL, timeout, lrmd_opt_notify_orig_only, TRUE); free_xml(data); lrmd_key_value_freeall(params); return rc; } static int lrmd_api_cancel(lrmd_t *lrmd, const char *rsc_id, const char *action, guint interval_ms) { int rc = pcmk_ok; xmlNode *data = pcmk__xe_create(NULL, PCMK__XE_LRMD_RSC); crm_xml_add(data, PCMK__XA_LRMD_ORIGIN, __func__); crm_xml_add(data, PCMK__XA_LRMD_RSC_ACTION, action); crm_xml_add(data, PCMK__XA_LRMD_RSC_ID, rsc_id); crm_xml_add_ms(data, PCMK__XA_LRMD_RSC_INTERVAL, interval_ms); rc = lrmd_send_command(lrmd, LRMD_OP_RSC_CANCEL, data, NULL, 0, 0, TRUE); free_xml(data); return rc; } static int list_stonith_agents(lrmd_list_t ** resources) { int rc = 0; stonith_t *stonith_api = stonith_api_new(); stonith_key_value_t *stonith_resources = NULL; stonith_key_value_t *dIter = NULL; if (stonith_api == NULL) { crm_err("Could not list fence agents: API memory allocation failed"); return -ENOMEM; } stonith_api->cmds->list_agents(stonith_api, st_opt_sync_call, NULL, &stonith_resources, 0); stonith_api->cmds->free(stonith_api); for (dIter = stonith_resources; dIter; dIter = dIter->next) { rc++; if (resources) { *resources = lrmd_list_add(*resources, dIter->value); } } stonith_key_value_freeall(stonith_resources, 1, 0); return rc; } static int lrmd_api_list_agents(lrmd_t * lrmd, lrmd_list_t ** resources, const char *class, const char *provider) { int rc = 0; int stonith_count = 0; // Initially, whether to include stonith devices if (pcmk__str_eq(class, PCMK_RESOURCE_CLASS_STONITH, pcmk__str_casei)) { stonith_count = 1; } else { GList *gIter = NULL; GList *agents = resources_list_agents(class, provider); for (gIter = agents; gIter != NULL; gIter = gIter->next) { *resources = lrmd_list_add(*resources, (const char *)gIter->data); rc++; } g_list_free_full(agents, free); if (!class) { stonith_count = 1; } } if (stonith_count) { // Now, if stonith devices are included, how many there are stonith_count = list_stonith_agents(resources); if (stonith_count > 0) { rc += stonith_count; } } if (rc == 0) { crm_notice("No agents found for class %s", class); rc = -EPROTONOSUPPORT; } return rc; } static bool does_provider_have_agent(const char *agent, const char *provider, const char *class) { bool found = false; GList *agents = NULL; GList *gIter2 = NULL; agents = resources_list_agents(class, provider); for (gIter2 = agents; gIter2 != NULL; gIter2 = gIter2->next) { if (pcmk__str_eq(agent, gIter2->data, pcmk__str_casei)) { found = true; } } g_list_free_full(agents, free); return found; } static int lrmd_api_list_ocf_providers(lrmd_t * lrmd, const char *agent, lrmd_list_t ** providers) { int rc = pcmk_ok; char *provider = NULL; GList *ocf_providers = NULL; GList *gIter = NULL; ocf_providers = resources_list_providers(PCMK_RESOURCE_CLASS_OCF); for (gIter = ocf_providers; gIter != NULL; gIter = gIter->next) { provider = gIter->data; if (!agent || does_provider_have_agent(agent, provider, PCMK_RESOURCE_CLASS_OCF)) { *providers = lrmd_list_add(*providers, (const char *)gIter->data); rc++; } } g_list_free_full(ocf_providers, free); return rc; } static int lrmd_api_list_standards(lrmd_t * lrmd, lrmd_list_t ** supported) { int rc = 0; GList *standards = NULL; GList *gIter = NULL; standards = resources_list_standards(); for (gIter = standards; gIter != NULL; gIter = gIter->next) { *supported = lrmd_list_add(*supported, (const char *)gIter->data); rc++; } if (list_stonith_agents(NULL) > 0) { *supported = lrmd_list_add(*supported, PCMK_RESOURCE_CLASS_STONITH); rc++; } g_list_free_full(standards, free); return rc; } /*! * \internal * \brief Create an executor API object * * \param[out] api Will be set to newly created API object (it is the * caller's responsibility to free this value with * lrmd_api_delete() if this function succeeds) * \param[in] nodename If the object will be used for a remote connection, * the node name to use in cluster for remote executor * \param[in] server If the object will be used for a remote connection, * the resolvable host name to connect to * \param[in] port If the object will be used for a remote connection, * port number on \p server to connect to * * \return Standard Pacemaker return code * \note If the caller leaves one of \p nodename or \p server NULL, the other's * value will be used for both. If the caller leaves both NULL, an API * object will be created for a local executor connection. */ int lrmd__new(lrmd_t **api, const char *nodename, const char *server, int port) { lrmd_private_t *pvt = NULL; if (api == NULL) { return EINVAL; } *api = NULL; // Allocate all memory needed *api = calloc(1, sizeof(lrmd_t)); if (*api == NULL) { return ENOMEM; } pvt = calloc(1, sizeof(lrmd_private_t)); if (pvt == NULL) { lrmd_api_delete(*api); *api = NULL; return ENOMEM; } (*api)->lrmd_private = pvt; // @TODO Do we need to do this for local connections? pvt->remote = calloc(1, sizeof(pcmk__remote_t)); (*api)->cmds = calloc(1, sizeof(lrmd_api_operations_t)); if ((pvt->remote == NULL) || ((*api)->cmds == NULL)) { lrmd_api_delete(*api); *api = NULL; return ENOMEM; } // Set methods (*api)->cmds->connect = lrmd_api_connect; (*api)->cmds->connect_async = lrmd_api_connect_async; (*api)->cmds->is_connected = lrmd_api_is_connected; (*api)->cmds->poke_connection = lrmd_api_poke_connection; (*api)->cmds->disconnect = lrmd_api_disconnect; (*api)->cmds->register_rsc = lrmd_api_register_rsc; (*api)->cmds->unregister_rsc = lrmd_api_unregister_rsc; (*api)->cmds->get_rsc_info = lrmd_api_get_rsc_info; (*api)->cmds->get_recurring_ops = lrmd_api_get_recurring_ops; (*api)->cmds->set_callback = lrmd_api_set_callback; (*api)->cmds->get_metadata = lrmd_api_get_metadata; (*api)->cmds->exec = lrmd_api_exec; (*api)->cmds->cancel = lrmd_api_cancel; (*api)->cmds->list_agents = lrmd_api_list_agents; (*api)->cmds->list_ocf_providers = lrmd_api_list_ocf_providers; (*api)->cmds->list_standards = lrmd_api_list_standards; (*api)->cmds->exec_alert = lrmd_api_exec_alert; (*api)->cmds->get_metadata_params = lrmd_api_get_metadata_params; if ((nodename == NULL) && (server == NULL)) { pvt->type = pcmk__client_ipc; } else { #ifdef HAVE_GNUTLS_GNUTLS_H if (nodename == NULL) { nodename = server; } else if (server == NULL) { server = nodename; } pvt->type = pcmk__client_tls; pvt->remote_nodename = strdup(nodename); pvt->server = strdup(server); if ((pvt->remote_nodename == NULL) || (pvt->server == NULL)) { lrmd_api_delete(*api); *api = NULL; return ENOMEM; } pvt->port = port; if (pvt->port == 0) { pvt->port = crm_default_remote_port(); } #else crm_err("Cannot communicate with Pacemaker Remote " "because GnuTLS is not enabled for this build"); lrmd_api_delete(*api); *api = NULL; return EOPNOTSUPP; #endif } return pcmk_rc_ok; } lrmd_t * lrmd_api_new(void) { lrmd_t *api = NULL; pcmk__assert(lrmd__new(&api, NULL, NULL, 0) == pcmk_rc_ok); return api; } lrmd_t * lrmd_remote_api_new(const char *nodename, const char *server, int port) { lrmd_t *api = NULL; pcmk__assert(lrmd__new(&api, nodename, server, port) == pcmk_rc_ok); return api; } void lrmd_api_delete(lrmd_t * lrmd) { if (lrmd == NULL) { return; } if (lrmd->cmds != NULL) { // Never NULL, but make static analysis happy if (lrmd->cmds->disconnect != NULL) { // Also never really NULL lrmd->cmds->disconnect(lrmd); // No-op if already disconnected } free(lrmd->cmds); } if (lrmd->lrmd_private != NULL) { lrmd_private_t *native = lrmd->lrmd_private; #ifdef HAVE_GNUTLS_GNUTLS_H free(native->server); #endif free(native->remote_nodename); free(native->remote); free(native->token); free(native->peer_version); free(lrmd->lrmd_private); } free(lrmd); } struct metadata_cb { void (*callback)(int pid, const pcmk__action_result_t *result, void *user_data); void *user_data; }; /*! * \internal * \brief Process asynchronous metadata completion * * \param[in,out] action Metadata action that completed */ static void metadata_complete(svc_action_t *action) { struct metadata_cb *metadata_cb = (struct metadata_cb *) action->cb_data; pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; services__copy_result(action, &result); pcmk__set_result_output(&result, action->stdout_data, action->stderr_data); metadata_cb->callback(0, &result, metadata_cb->user_data); result.action_stdout = NULL; // Prevent free, because action owns it result.action_stderr = NULL; // Prevent free, because action owns it pcmk__reset_result(&result); free(metadata_cb); } /*! * \internal * \brief Retrieve agent metadata asynchronously * * \param[in] rsc Resource agent specification * \param[in] callback Function to call with result (this will always be * called, whether by this function directly or later * via the main loop, and on success the metadata will * be in its result argument's action_stdout) * \param[in,out] user_data User data to pass to callback * * \return Standard Pacemaker return code * \note This function is not a lrmd_api_operations_t method because it does not * need an lrmd_t object and does not go through the executor, but * executes the agent directly. */ int lrmd__metadata_async(const lrmd_rsc_info_t *rsc, void (*callback)(int pid, const pcmk__action_result_t *result, void *user_data), void *user_data) { svc_action_t *action = NULL; struct metadata_cb *metadata_cb = NULL; pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; CRM_CHECK(callback != NULL, return EINVAL); if ((rsc == NULL) || (rsc->standard == NULL) || (rsc->type == NULL)) { pcmk__set_result(&result, PCMK_OCF_NOT_CONFIGURED, PCMK_EXEC_ERROR_FATAL, "Invalid resource specification"); callback(0, &result, user_data); pcmk__reset_result(&result); return EINVAL; } if (strcmp(rsc->standard, PCMK_RESOURCE_CLASS_STONITH) == 0) { return stonith__metadata_async(rsc->type, PCMK_DEFAULT_METADATA_TIMEOUT_MS / 1000, callback, user_data); } action = services__create_resource_action(pcmk__s(rsc->id, rsc->type), rsc->standard, rsc->provider, rsc->type, PCMK_ACTION_META_DATA, 0, PCMK_DEFAULT_METADATA_TIMEOUT_MS, NULL, 0); if (action == NULL) { pcmk__set_result(&result, PCMK_OCF_UNKNOWN_ERROR, PCMK_EXEC_ERROR, "Out of memory"); callback(0, &result, user_data); pcmk__reset_result(&result); return ENOMEM; } if (action->rc != PCMK_OCF_UNKNOWN) { services__copy_result(action, &result); callback(0, &result, user_data); pcmk__reset_result(&result); services_action_free(action); return EINVAL; } action->cb_data = calloc(1, sizeof(struct metadata_cb)); if (action->cb_data == NULL) { services_action_free(action); pcmk__set_result(&result, PCMK_OCF_UNKNOWN_ERROR, PCMK_EXEC_ERROR, "Out of memory"); callback(0, &result, user_data); pcmk__reset_result(&result); return ENOMEM; } metadata_cb = (struct metadata_cb *) action->cb_data; metadata_cb->callback = callback; metadata_cb->user_data = user_data; if (!services_action_async(action, metadata_complete)) { services_action_free(action); return pcmk_rc_error; // @TODO Derive from action->rc and ->status } // The services library has taken responsibility for action return pcmk_rc_ok; } /*! * \internal * \brief Set the result of an executor event * * \param[in,out] event Executor event to set * \param[in] rc OCF exit status of event * \param[in] op_status Executor status of event * \param[in] exit_reason Human-friendly description of event */ void lrmd__set_result(lrmd_event_data_t *event, enum ocf_exitcode rc, int op_status, const char *exit_reason) { if (event == NULL) { return; } event->rc = rc; event->op_status = op_status; // lrmd_event_data_t has (const char *) members that lrmd_free_event() frees pcmk__str_update((char **) &event->exit_reason, exit_reason); } /*! * \internal * \brief Clear an executor event's exit reason, output, and error output * * \param[in,out] event Executor event to reset */ void lrmd__reset_result(lrmd_event_data_t *event) { if (event == NULL) { return; } free((void *) event->exit_reason); event->exit_reason = NULL; free((void *) event->output); event->output = NULL; } /*! * \internal * \brief Get the uptime of a remote resource connection * * When the cluster connects to a remote resource, part of that resource's * handshake includes the uptime of the remote resource's connection. This * uptime is stored in the lrmd_t object. * * \return The connection's uptime, or -1 if unknown */ time_t lrmd__uptime(lrmd_t *lrmd) { lrmd_private_t *native = lrmd->lrmd_private; if (native->remote == NULL) { return -1; } else { return native->remote->uptime; } } const char * lrmd__node_start_state(lrmd_t *lrmd) { lrmd_private_t *native = lrmd->lrmd_private; if (native->remote == NULL) { return NULL; } else { return native->remote->start_state; } } diff --git a/lib/pacemaker/Makefile.am b/lib/pacemaker/Makefile.am index 11aba01d08..0015573d56 100644 --- a/lib/pacemaker/Makefile.am +++ b/lib/pacemaker/Makefile.am @@ -1,75 +1,75 @@ # # Copyright 2004-2024 the Pacemaker project contributors # # The version control history for this file may have further details. # # This source code is licensed under the GNU General Public License version 2 # or later (GPLv2+) WITHOUT ANY WARRANTY. # include $(top_srcdir)/mk/common.mk AM_CPPFLAGS += -I$(top_builddir) -I$(top_srcdir) SUBDIRS = tests noinst_HEADERS = libpacemaker_private.h ## libraries lib_LTLIBRARIES = libpacemaker.la -libpacemaker_la_LDFLAGS = -version-info 9:1:8 +libpacemaker_la_LDFLAGS = -version-info 9:2:8 libpacemaker_la_CFLAGS = $(CFLAGS_HARDENED_LIB) libpacemaker_la_LDFLAGS += $(LDFLAGS_HARDENED_LIB) libpacemaker_la_LIBADD = $(top_builddir)/lib/pengine/libpe_status.la libpacemaker_la_LIBADD += $(top_builddir)/lib/cib/libcib.la libpacemaker_la_LIBADD += $(top_builddir)/lib/lrmd/liblrmd.la libpacemaker_la_LIBADD += $(top_builddir)/lib/fencing/libstonithd.la libpacemaker_la_LIBADD += $(top_builddir)/lib/services/libcrmservice.la libpacemaker_la_LIBADD += $(top_builddir)/lib/common/libcrmcommon.la # -L$(top_builddir)/lib/pils -lpils -export-dynamic -module -avoid-version ## Library sources (*must* use += format for bumplibs) libpacemaker_la_SOURCES = libpacemaker_la_SOURCES += pcmk_acl.c libpacemaker_la_SOURCES += pcmk_agents.c libpacemaker_la_SOURCES += pcmk_cluster_queries.c libpacemaker_la_SOURCES += pcmk_fence.c libpacemaker_la_SOURCES += pcmk_graph_consumer.c libpacemaker_la_SOURCES += pcmk_graph_logging.c libpacemaker_la_SOURCES += pcmk_graph_producer.c libpacemaker_la_SOURCES += pcmk_injections.c libpacemaker_la_SOURCES += pcmk_options.c libpacemaker_la_SOURCES += pcmk_output.c libpacemaker_la_SOURCES += pcmk_resource.c libpacemaker_la_SOURCES += pcmk_result_code.c libpacemaker_la_SOURCES += pcmk_rule.c libpacemaker_la_SOURCES += pcmk_sched_actions.c libpacemaker_la_SOURCES += pcmk_sched_bundle.c libpacemaker_la_SOURCES += pcmk_sched_clone.c libpacemaker_la_SOURCES += pcmk_sched_colocation.c libpacemaker_la_SOURCES += pcmk_sched_constraints.c libpacemaker_la_SOURCES += pcmk_sched_fencing.c libpacemaker_la_SOURCES += pcmk_sched_group.c libpacemaker_la_SOURCES += pcmk_sched_instances.c libpacemaker_la_SOURCES += pcmk_sched_location.c libpacemaker_la_SOURCES += pcmk_sched_migration.c libpacemaker_la_SOURCES += pcmk_sched_nodes.c libpacemaker_la_SOURCES += pcmk_sched_ordering.c libpacemaker_la_SOURCES += pcmk_sched_primitive.c libpacemaker_la_SOURCES += pcmk_sched_probes.c libpacemaker_la_SOURCES += pcmk_sched_promotable.c libpacemaker_la_SOURCES += pcmk_sched_recurring.c libpacemaker_la_SOURCES += pcmk_sched_remote.c libpacemaker_la_SOURCES += pcmk_sched_resource.c libpacemaker_la_SOURCES += pcmk_sched_tickets.c libpacemaker_la_SOURCES += pcmk_sched_utilization.c libpacemaker_la_SOURCES += pcmk_scheduler.c libpacemaker_la_SOURCES += pcmk_setup.c libpacemaker_la_SOURCES += pcmk_simulate.c libpacemaker_la_SOURCES += pcmk_status.c libpacemaker_la_SOURCES += pcmk_ticket.c libpacemaker_la_SOURCES += pcmk_verify.c diff --git a/lib/pacemaker/pcmk_sched_ordering.c b/lib/pacemaker/pcmk_sched_ordering.c index 090732b466..226ca4277a 100644 --- a/lib/pacemaker/pcmk_sched_ordering.c +++ b/lib/pacemaker/pcmk_sched_ordering.c @@ -1,1532 +1,1533 @@ /* - * Copyright 2004-2024 the Pacemaker project contributors + * Copyright 2004-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU General Public License version 2 * or later (GPLv2+) WITHOUT ANY WARRANTY. */ #include #include // PRIx32 #include #include #include #include #include "libpacemaker_private.h" enum pe_order_kind { pe_order_kind_optional, pe_order_kind_mandatory, pe_order_kind_serialize, }; enum ordering_symmetry { ordering_asymmetric, // the only relation in an asymmetric ordering ordering_symmetric, // the normal relation in a symmetric ordering ordering_symmetric_inverse, // the inverse relation in a symmetric ordering }; #define EXPAND_CONSTRAINT_IDREF(__set, __rsc, __name) do { \ __rsc = pcmk__find_constraint_resource(scheduler->resources, \ __name); \ if (__rsc == NULL) { \ pcmk__config_err("%s: No resource found for %s", __set, __name);\ return pcmk_rc_unpack_error; \ } \ } while (0) static const char * invert_action(const char *action) { if (pcmk__str_eq(action, PCMK_ACTION_START, pcmk__str_none)) { return PCMK_ACTION_STOP; } else if (pcmk__str_eq(action, PCMK_ACTION_STOP, pcmk__str_none)) { return PCMK_ACTION_START; } else if (pcmk__str_eq(action, PCMK_ACTION_PROMOTE, pcmk__str_none)) { return PCMK_ACTION_DEMOTE; } else if (pcmk__str_eq(action, PCMK_ACTION_DEMOTE, pcmk__str_none)) { return PCMK_ACTION_PROMOTE; } else if (pcmk__str_eq(action, PCMK_ACTION_PROMOTED, pcmk__str_none)) { return PCMK_ACTION_DEMOTED; } else if (pcmk__str_eq(action, PCMK_ACTION_DEMOTED, pcmk__str_none)) { return PCMK_ACTION_PROMOTED; } else if (pcmk__str_eq(action, PCMK_ACTION_RUNNING, pcmk__str_none)) { return PCMK_ACTION_STOPPED; } else if (pcmk__str_eq(action, PCMK_ACTION_STOPPED, pcmk__str_none)) { return PCMK_ACTION_RUNNING; } pcmk__config_warn("Unknown action '%s' specified in order constraint", action); return NULL; } static enum pe_order_kind get_ordering_type(const xmlNode *xml_obj) { enum pe_order_kind kind_e = pe_order_kind_mandatory; const char *kind = crm_element_value(xml_obj, PCMK_XA_KIND); if (kind == NULL) { const char *score = crm_element_value(xml_obj, PCMK_XA_SCORE); kind_e = pe_order_kind_mandatory; if (score) { // @COMPAT deprecated informally since 1.0.7, formally since 2.0.1 int score_i = 0; (void) pcmk_parse_score(score, &score_i, 0); if (score_i == 0) { kind_e = pe_order_kind_optional; } pcmk__warn_once(pcmk__wo_order_score, "Support for '" PCMK_XA_SCORE "' in " PCMK_XE_RSC_ORDER " is deprecated and will be " "removed in a future release " "(use '" PCMK_XA_KIND "' instead)"); } } else if (pcmk__str_eq(kind, PCMK_VALUE_MANDATORY, pcmk__str_none)) { kind_e = pe_order_kind_mandatory; } else if (pcmk__str_eq(kind, PCMK_VALUE_OPTIONAL, pcmk__str_none)) { kind_e = pe_order_kind_optional; } else if (pcmk__str_eq(kind, PCMK_VALUE_SERIALIZE, pcmk__str_none)) { kind_e = pe_order_kind_serialize; } else { pcmk__config_err("Resetting '" PCMK_XA_KIND "' for constraint %s to " "'" PCMK_VALUE_MANDATORY "' because '%s' is not valid", pcmk__s(pcmk__xe_id(xml_obj), "missing ID"), kind); } return kind_e; } /*! * \internal * \brief Get ordering symmetry from XML * * \param[in] xml_obj Ordering XML * \param[in] parent_kind Default ordering kind * \param[in] parent_symmetrical_s Parent element's \c PCMK_XA_SYMMETRICAL * setting, if any * * \retval ordering_symmetric Ordering is symmetric * \retval ordering_asymmetric Ordering is asymmetric */ static enum ordering_symmetry get_ordering_symmetry(const xmlNode *xml_obj, enum pe_order_kind parent_kind, const char *parent_symmetrical_s) { int rc = pcmk_rc_ok; bool symmetric = false; enum pe_order_kind kind = parent_kind; // Default to parent's kind // Check ordering XML for explicit kind if ((crm_element_value(xml_obj, PCMK_XA_KIND) != NULL) || (crm_element_value(xml_obj, PCMK_XA_SCORE) != NULL)) { kind = get_ordering_type(xml_obj); } // Check ordering XML (and parent) for explicit PCMK_XA_SYMMETRICAL setting rc = pcmk__xe_get_bool_attr(xml_obj, PCMK_XA_SYMMETRICAL, &symmetric); if (rc != pcmk_rc_ok && parent_symmetrical_s != NULL) { symmetric = crm_is_true(parent_symmetrical_s); rc = pcmk_rc_ok; } if (rc == pcmk_rc_ok) { if (symmetric) { if (kind == pe_order_kind_serialize) { pcmk__config_warn("Ignoring " PCMK_XA_SYMMETRICAL " for '%s' because not valid with " PCMK_XA_KIND " of '" PCMK_VALUE_SERIALIZE "'", pcmk__xe_id(xml_obj)); } else { return ordering_symmetric; } } return ordering_asymmetric; } // Use default symmetry if (kind == pe_order_kind_serialize) { return ordering_asymmetric; } return ordering_symmetric; } /*! * \internal * \brief Get ordering flags appropriate to ordering kind * * \param[in] kind Ordering kind * \param[in] first Action name for 'first' action * \param[in] symmetry This ordering's symmetry role * * \return Minimal ordering flags appropriate to \p kind */ static uint32_t ordering_flags_for_kind(enum pe_order_kind kind, const char *first, enum ordering_symmetry symmetry) { uint32_t flags = pcmk__ar_none; // so we trace-log all flags set switch (kind) { case pe_order_kind_optional: pcmk__set_relation_flags(flags, pcmk__ar_ordered); break; case pe_order_kind_serialize: /* This flag is not used anywhere directly but means the relation * will not match an equality comparison against pcmk__ar_none or * pcmk__ar_ordered. */ pcmk__set_relation_flags(flags, pcmk__ar_serialize); break; case pe_order_kind_mandatory: pcmk__set_relation_flags(flags, pcmk__ar_ordered); switch (symmetry) { case ordering_asymmetric: pcmk__set_relation_flags(flags, pcmk__ar_asymmetric); break; case ordering_symmetric: pcmk__set_relation_flags(flags, pcmk__ar_first_implies_then); if (pcmk__strcase_any_of(first, PCMK_ACTION_START, PCMK_ACTION_PROMOTE, NULL)) { pcmk__set_relation_flags(flags, pcmk__ar_unrunnable_first_blocks); } break; case ordering_symmetric_inverse: pcmk__set_relation_flags(flags, pcmk__ar_then_implies_first); break; } break; } return flags; } /*! * \internal * \brief Find resource corresponding to ID specified in ordering * * \param[in] xml Ordering XML * \param[in] resource_attr XML attribute name for resource ID * \param[in] instance_attr XML attribute name for instance number. * This option is deprecated and will be removed in a * future release. * \param[in] scheduler Scheduler data * * \return Resource corresponding to \p id, or NULL if none */ static pcmk_resource_t * get_ordering_resource(const xmlNode *xml, const char *resource_attr, const char *instance_attr, const pcmk_scheduler_t *scheduler) { // @COMPAT: instance_attr and instance_id variables deprecated since 2.1.5 pcmk_resource_t *rsc = NULL; const char *rsc_id = crm_element_value(xml, resource_attr); const char *instance_id = crm_element_value(xml, instance_attr); if (rsc_id == NULL) { pcmk__config_err("Ignoring constraint '%s' without %s", pcmk__xe_id(xml), resource_attr); return NULL; } rsc = pcmk__find_constraint_resource(scheduler->resources, rsc_id); if (rsc == NULL) { pcmk__config_err("Ignoring constraint '%s' because resource '%s' " "does not exist", pcmk__xe_id(xml), rsc_id); return NULL; } if (instance_id != NULL) { pcmk__warn_once(pcmk__wo_order_inst, "Support for " PCMK__XA_FIRST_INSTANCE " and " PCMK__XA_THEN_INSTANCE " is deprecated and will be " "removed in a future release."); if (!pcmk__is_clone(rsc)) { pcmk__config_err("Ignoring constraint '%s' because resource '%s' " "is not a clone but instance '%s' was requested", pcmk__xe_id(xml), rsc_id, instance_id); return NULL; } rsc = find_clone_instance(rsc, instance_id); if (rsc == NULL) { pcmk__config_err("Ignoring constraint '%s' because resource '%s' " "does not have an instance '%s'", pcmk__xe_id(xml), rsc_id, instance_id); return NULL; } } return rsc; } /*! * \internal * \brief Determine minimum number of 'first' instances required in ordering * * \param[in] rsc 'First' resource in ordering * \param[in] xml Ordering XML * * \return Minimum 'first' instances required (or 0 if not applicable) */ static int get_minimum_first_instances(const pcmk_resource_t *rsc, const xmlNode *xml) { const char *clone_min = NULL; bool require_all = false; if (!pcmk__is_clone(rsc)) { return 0; } clone_min = g_hash_table_lookup(rsc->meta, PCMK_META_CLONE_MIN); if (clone_min != NULL) { int clone_min_int = 0; pcmk__scan_min_int(clone_min, &clone_min_int, 0); return clone_min_int; } /* @COMPAT 1.1.13: * PCMK_XA_REQUIRE_ALL=PCMK_VALUE_FALSE is deprecated equivalent of * PCMK_META_CLONE_MIN=1 */ if (pcmk__xe_get_bool_attr(xml, PCMK_XA_REQUIRE_ALL, &require_all) != ENODATA) { pcmk__warn_once(pcmk__wo_require_all, "Support for " PCMK_XA_REQUIRE_ALL " in ordering " "constraints is deprecated and will be removed in a " "future release (use " PCMK_META_CLONE_MIN " clone " "meta-attribute instead)"); if (!require_all) { return 1; } } return 0; } /*! * \internal * \brief Create orderings for a constraint with \c PCMK_META_CLONE_MIN > 0 * * \param[in] id Ordering ID * \param[in,out] rsc_first 'First' resource in ordering (a clone) * \param[in] action_first 'First' action in ordering * \param[in] rsc_then 'Then' resource in ordering * \param[in] action_then 'Then' action in ordering * \param[in] flags Ordering flags * \param[in] clone_min Minimum required instances of 'first' */ static void clone_min_ordering(const char *id, pcmk_resource_t *rsc_first, const char *action_first, pcmk_resource_t *rsc_then, const char *action_then, uint32_t flags, int clone_min) { // Create a pseudo-action for when the minimum instances are active char *task = crm_strdup_printf(PCMK_ACTION_CLONE_ONE_OR_MORE ":%s", id); pcmk_action_t *clone_min_met = get_pseudo_op(task, rsc_first->cluster); free(task); /* Require the pseudo-action to have the required number of actions to be * considered runnable before allowing the pseudo-action to be runnable. */ clone_min_met->required_runnable_before = clone_min; pcmk__set_action_flags(clone_min_met, pcmk_action_min_runnable); // Order the actions for each clone instance before the pseudo-action for (GList *iter = rsc_first->children; iter != NULL; iter = iter->next) { pcmk_resource_t *child = iter->data; pcmk__new_ordering(child, pcmk__op_key(child->id, action_first, 0), NULL, NULL, NULL, clone_min_met, pcmk__ar_min_runnable |pcmk__ar_first_implies_then_graphed, rsc_first->cluster); } // Order "then" action after the pseudo-action (if runnable) pcmk__new_ordering(NULL, NULL, clone_min_met, rsc_then, pcmk__op_key(rsc_then->id, action_then, 0), NULL, flags|pcmk__ar_unrunnable_first_blocks, rsc_first->cluster); } /*! * \internal * \brief Update ordering flags for restart-type=restart * * \param[in] rsc 'Then' resource in ordering * \param[in] kind Ordering kind * \param[in] flag Ordering flag to set (when applicable) * \param[in,out] flags Ordering flag set to update * * \compat The \c PCMK__META_RESTART_TYPE resource meta-attribute is deprecated. * Eventually, it will be removed, and \c pe_restart_ignore will be the * only behavior, at which time this can just be removed entirely. */ #define handle_restart_type(rsc, kind, flag, flags) do { \ if (((kind) == pe_order_kind_optional) \ && ((rsc)->restart_type == pe_restart_restart)) { \ pcmk__set_relation_flags((flags), (flag)); \ } \ } while (0) /*! * \internal * \brief Create new ordering for inverse of symmetric constraint * * \param[in] id Ordering ID (for logging only) * \param[in] kind Ordering kind * \param[in] rsc_first 'First' resource in ordering (a clone) * \param[in] action_first 'First' action in ordering * \param[in,out] rsc_then 'Then' resource in ordering * \param[in] action_then 'Then' action in ordering */ static void inverse_ordering(const char *id, enum pe_order_kind kind, pcmk_resource_t *rsc_first, const char *action_first, pcmk_resource_t *rsc_then, const char *action_then) { action_then = invert_action(action_then); action_first = invert_action(action_first); if ((action_then == NULL) || (action_first == NULL)) { pcmk__config_warn("Cannot invert constraint '%s' " "(please specify inverse manually)", id); } else { uint32_t flags = ordering_flags_for_kind(kind, action_first, ordering_symmetric_inverse); handle_restart_type(rsc_then, kind, pcmk__ar_then_implies_first, flags); pcmk__order_resource_actions(rsc_then, action_then, rsc_first, action_first, flags); } } static void unpack_simple_rsc_order(xmlNode *xml_obj, pcmk_scheduler_t *scheduler) { pcmk_resource_t *rsc_then = NULL; pcmk_resource_t *rsc_first = NULL; int min_required_before = 0; enum pe_order_kind kind = pe_order_kind_mandatory; uint32_t flags = pcmk__ar_none; enum ordering_symmetry symmetry; const char *action_then = NULL; const char *action_first = NULL; const char *id = NULL; CRM_CHECK(xml_obj != NULL, return); id = crm_element_value(xml_obj, PCMK_XA_ID); if (id == NULL) { pcmk__config_err("Ignoring <%s> constraint without " PCMK_XA_ID, xml_obj->name); return; } rsc_first = get_ordering_resource(xml_obj, PCMK_XA_FIRST, PCMK__XA_FIRST_INSTANCE, scheduler); if (rsc_first == NULL) { return; } rsc_then = get_ordering_resource(xml_obj, PCMK_XA_THEN, PCMK__XA_THEN_INSTANCE, scheduler); if (rsc_then == NULL) { return; } action_first = crm_element_value(xml_obj, PCMK_XA_FIRST_ACTION); if (action_first == NULL) { action_first = PCMK_ACTION_START; } action_then = crm_element_value(xml_obj, PCMK_XA_THEN_ACTION); if (action_then == NULL) { action_then = action_first; } kind = get_ordering_type(xml_obj); symmetry = get_ordering_symmetry(xml_obj, kind, NULL); flags = ordering_flags_for_kind(kind, action_first, symmetry); handle_restart_type(rsc_then, kind, pcmk__ar_first_implies_then, flags); /* If there is a minimum number of instances that must be runnable before * the 'then' action is runnable, we use a pseudo-action for convenience: * minimum number of clone instances have runnable actions -> * pseudo-action is runnable -> dependency is runnable. */ min_required_before = get_minimum_first_instances(rsc_first, xml_obj); if (min_required_before > 0) { clone_min_ordering(id, rsc_first, action_first, rsc_then, action_then, flags, min_required_before); } else { pcmk__order_resource_actions(rsc_first, action_first, rsc_then, action_then, flags); } if (symmetry == ordering_symmetric) { inverse_ordering(id, kind, rsc_first, action_first, rsc_then, action_then); } } /*! * \internal * \brief Create a new ordering between two actions * * \param[in,out] first_rsc Resource for 'first' action (if NULL and * \p first_action is a resource action, that * resource will be used) * \param[in,out] first_action_task Action key for 'first' action (if NULL and * \p first_action is not NULL, its UUID will * be used) * \param[in,out] first_action 'first' action (if NULL, \p first_rsc and * \p first_action_task must be set) * * \param[in] then_rsc Resource for 'then' action (if NULL and * \p then_action is a resource action, that * resource will be used) * \param[in,out] then_action_task Action key for 'then' action (if NULL and * \p then_action is not NULL, its UUID will * be used) * \param[in] then_action 'then' action (if NULL, \p then_rsc and * \p then_action_task must be set) * * \param[in] flags Group of enum pcmk__action_relation_flags * \param[in,out] sched Scheduler data to add ordering to * * \note This function takes ownership of first_action_task and * then_action_task, which do not need to be freed by the caller. */ void pcmk__new_ordering(pcmk_resource_t *first_rsc, char *first_action_task, pcmk_action_t *first_action, pcmk_resource_t *then_rsc, char *then_action_task, pcmk_action_t *then_action, uint32_t flags, pcmk_scheduler_t *sched) { pcmk__action_relation_t *order = NULL; // One of action or resource must be specified for each side CRM_CHECK(((first_action != NULL) || (first_rsc != NULL)) && ((then_action != NULL) || (then_rsc != NULL)), free(first_action_task); free(then_action_task); return); if ((first_rsc == NULL) && (first_action != NULL)) { first_rsc = first_action->rsc; } if ((then_rsc == NULL) && (then_action != NULL)) { then_rsc = then_action->rsc; } order = pcmk__assert_alloc(1, sizeof(pcmk__action_relation_t)); order->id = sched->order_id++; order->flags = flags; order->rsc1 = first_rsc; order->rsc2 = then_rsc; order->action1 = first_action; order->action2 = then_action; order->task1 = first_action_task; order->task2 = then_action_task; if ((order->task1 == NULL) && (first_action != NULL)) { order->task1 = strdup(first_action->uuid); } if ((order->task2 == NULL) && (then_action != NULL)) { order->task2 = strdup(then_action->uuid); } if ((order->rsc1 == NULL) && (first_action != NULL)) { order->rsc1 = first_action->rsc; } if ((order->rsc2 == NULL) && (then_action != NULL)) { order->rsc2 = then_action->rsc; } pcmk__rsc_trace(first_rsc, "Created ordering %d for %s then %s", (sched->order_id - 1), pcmk__s(order->task1, "an underspecified action"), pcmk__s(order->task2, "an underspecified action")); sched->ordering_constraints = g_list_prepend(sched->ordering_constraints, order); pcmk__order_migration_equivalents(order); } /*! * \brief Unpack a set in an ordering constraint * * \param[in] set Set XML to unpack * \param[in] parent_kind \c PCMK_XE_RSC_ORDER XML \c PCMK_XA_KIND * attribute * \param[in] parent_symmetrical_s \c PCMK_XE_RSC_ORDER XML * \c PCMK_XA_SYMMETRICAL attribute * \param[in,out] scheduler Scheduler data * * \return Standard Pacemaker return code */ static int unpack_order_set(const xmlNode *set, enum pe_order_kind parent_kind, const char *parent_symmetrical_s, pcmk_scheduler_t *scheduler) { GList *set_iter = NULL; GList *resources = NULL; pcmk_resource_t *last = NULL; pcmk_resource_t *resource = NULL; int local_kind = parent_kind; bool sequential = false; uint32_t flags = pcmk__ar_ordered; enum ordering_symmetry symmetry; char *key = NULL; const char *id = pcmk__xe_id(set); const char *action = crm_element_value(set, PCMK_XA_ACTION); const char *sequential_s = crm_element_value(set, PCMK_XA_SEQUENTIAL); const char *kind_s = crm_element_value(set, PCMK_XA_KIND); if (action == NULL) { action = PCMK_ACTION_START; } if (kind_s) { local_kind = get_ordering_type(set); } if (sequential_s == NULL) { sequential_s = "1"; } sequential = crm_is_true(sequential_s); symmetry = get_ordering_symmetry(set, parent_kind, parent_symmetrical_s); flags = ordering_flags_for_kind(local_kind, action, symmetry); for (const xmlNode *xml_rsc = pcmk__xe_first_child(set, PCMK_XE_RESOURCE_REF, NULL, NULL); xml_rsc != NULL; xml_rsc = pcmk__xe_next_same(xml_rsc)) { EXPAND_CONSTRAINT_IDREF(id, resource, pcmk__xe_id(xml_rsc)); resources = g_list_append(resources, resource); } if (pcmk__list_of_1(resources)) { crm_trace("Single set: %s", id); goto done; } set_iter = resources; while (set_iter != NULL) { resource = (pcmk_resource_t *) set_iter->data; set_iter = set_iter->next; key = pcmk__op_key(resource->id, action, 0); if (local_kind == pe_order_kind_serialize) { /* Serialize before everything that comes after */ for (GList *iter = set_iter; iter != NULL; iter = iter->next) { pcmk_resource_t *then_rsc = iter->data; char *then_key = pcmk__op_key(then_rsc->id, action, 0); pcmk__new_ordering(resource, strdup(key), NULL, then_rsc, then_key, NULL, flags, scheduler); } } else if (sequential) { if (last != NULL) { pcmk__order_resource_actions(last, action, resource, action, flags); } last = resource; } free(key); } if (symmetry == ordering_asymmetric) { goto done; } last = NULL; action = invert_action(action); flags = ordering_flags_for_kind(local_kind, action, ordering_symmetric_inverse); set_iter = resources; while (set_iter != NULL) { resource = (pcmk_resource_t *) set_iter->data; set_iter = set_iter->next; if (sequential) { if (last != NULL) { pcmk__order_resource_actions(resource, action, last, action, flags); } last = resource; } } done: g_list_free(resources); return pcmk_rc_ok; } /*! * \brief Order two resource sets relative to each other * * \param[in] id Ordering ID (for logging) * \param[in] set1 First listed set * \param[in] set2 Second listed set * \param[in] kind Ordering kind * \param[in,out] scheduler Scheduler data * \param[in] symmetry Which ordering symmetry applies to this relation * * \return Standard Pacemaker return code */ static int order_rsc_sets(const char *id, const xmlNode *set1, const xmlNode *set2, enum pe_order_kind kind, pcmk_scheduler_t *scheduler, enum ordering_symmetry symmetry) { const xmlNode *xml_rsc = NULL; const xmlNode *xml_rsc_2 = NULL; pcmk_resource_t *rsc_1 = NULL; pcmk_resource_t *rsc_2 = NULL; const char *action_1 = crm_element_value(set1, PCMK_XA_ACTION); const char *action_2 = crm_element_value(set2, PCMK_XA_ACTION); uint32_t flags = pcmk__ar_none; bool require_all = true; (void) pcmk__xe_get_bool_attr(set1, PCMK_XA_REQUIRE_ALL, &require_all); if (action_1 == NULL) { action_1 = PCMK_ACTION_START; } if (action_2 == NULL) { action_2 = PCMK_ACTION_START; } if (symmetry == ordering_symmetric_inverse) { action_1 = invert_action(action_1); action_2 = invert_action(action_2); } if (pcmk__str_eq(PCMK_ACTION_STOP, action_1, pcmk__str_none) || pcmk__str_eq(PCMK_ACTION_DEMOTE, action_1, pcmk__str_none)) { /* Assuming: A -> ( B || C) -> D * The one-or-more logic only applies during the start/promote phase. * During shutdown neither B nor can shutdown until D is down, so simply * turn require_all back on. */ require_all = true; } flags = ordering_flags_for_kind(kind, action_1, symmetry); /* If we have an unordered set1, whether it is sequential or not is * irrelevant in regards to set2. */ if (!require_all) { char *task = crm_strdup_printf(PCMK_ACTION_ONE_OR_MORE ":%s", pcmk__xe_id(set1)); pcmk_action_t *unordered_action = get_pseudo_op(task, scheduler); free(task); pcmk__set_action_flags(unordered_action, pcmk_action_min_runnable); for (xml_rsc = pcmk__xe_first_child(set1, PCMK_XE_RESOURCE_REF, NULL, NULL); xml_rsc != NULL; xml_rsc = pcmk__xe_next_same(xml_rsc)) { EXPAND_CONSTRAINT_IDREF(id, rsc_1, pcmk__xe_id(xml_rsc)); /* Add an ordering constraint between every element in set1 and the * pseudo action. If any action in set1 is runnable the pseudo * action will be runnable. */ pcmk__new_ordering(rsc_1, pcmk__op_key(rsc_1->id, action_1, 0), NULL, NULL, NULL, unordered_action, pcmk__ar_min_runnable |pcmk__ar_first_implies_then_graphed, scheduler); } for (xml_rsc_2 = pcmk__xe_first_child(set2, PCMK_XE_RESOURCE_REF, NULL, NULL); xml_rsc_2 != NULL; xml_rsc_2 = pcmk__xe_next_same(xml_rsc_2)) { EXPAND_CONSTRAINT_IDREF(id, rsc_2, pcmk__xe_id(xml_rsc_2)); /* Add an ordering constraint between the pseudo-action and every * element in set2. If the pseudo-action is runnable, every action * in set2 will be runnable. */ pcmk__new_ordering(NULL, NULL, unordered_action, rsc_2, pcmk__op_key(rsc_2->id, action_2, 0), NULL, flags|pcmk__ar_unrunnable_first_blocks, scheduler); } return pcmk_rc_ok; } if (pcmk__xe_attr_is_true(set1, PCMK_XA_SEQUENTIAL)) { if (symmetry == ordering_symmetric_inverse) { // Get the first one xml_rsc = pcmk__xe_first_child(set1, PCMK_XE_RESOURCE_REF, NULL, NULL); if (xml_rsc != NULL) { EXPAND_CONSTRAINT_IDREF(id, rsc_1, pcmk__xe_id(xml_rsc)); } } else { // Get the last one const char *rid = NULL; for (xml_rsc = pcmk__xe_first_child(set1, PCMK_XE_RESOURCE_REF, NULL, NULL); xml_rsc != NULL; xml_rsc = pcmk__xe_next_same(xml_rsc)) { rid = pcmk__xe_id(xml_rsc); } EXPAND_CONSTRAINT_IDREF(id, rsc_1, rid); } } if (pcmk__xe_attr_is_true(set2, PCMK_XA_SEQUENTIAL)) { if (symmetry == ordering_symmetric_inverse) { // Get the last one const char *rid = NULL; for (xml_rsc = pcmk__xe_first_child(set2, PCMK_XE_RESOURCE_REF, NULL, NULL); xml_rsc != NULL; xml_rsc = pcmk__xe_next_same(xml_rsc)) { rid = pcmk__xe_id(xml_rsc); } EXPAND_CONSTRAINT_IDREF(id, rsc_2, rid); } else { // Get the first one xml_rsc = pcmk__xe_first_child(set2, PCMK_XE_RESOURCE_REF, NULL, NULL); if (xml_rsc != NULL) { EXPAND_CONSTRAINT_IDREF(id, rsc_2, pcmk__xe_id(xml_rsc)); } } } if ((rsc_1 != NULL) && (rsc_2 != NULL)) { pcmk__order_resource_actions(rsc_1, action_1, rsc_2, action_2, flags); } else if (rsc_1 != NULL) { for (xml_rsc = pcmk__xe_first_child(set2, PCMK_XE_RESOURCE_REF, NULL, NULL); xml_rsc != NULL; xml_rsc = pcmk__xe_next_same(xml_rsc)) { EXPAND_CONSTRAINT_IDREF(id, rsc_2, pcmk__xe_id(xml_rsc)); pcmk__order_resource_actions(rsc_1, action_1, rsc_2, action_2, flags); } } else if (rsc_2 != NULL) { for (xml_rsc = pcmk__xe_first_child(set1, PCMK_XE_RESOURCE_REF, NULL, NULL); xml_rsc != NULL; xml_rsc = pcmk__xe_next_same(xml_rsc)) { EXPAND_CONSTRAINT_IDREF(id, rsc_1, pcmk__xe_id(xml_rsc)); pcmk__order_resource_actions(rsc_1, action_1, rsc_2, action_2, flags); } } else { for (xml_rsc = pcmk__xe_first_child(set1, PCMK_XE_RESOURCE_REF, NULL, NULL); xml_rsc != NULL; xml_rsc = pcmk__xe_next_same(xml_rsc)) { EXPAND_CONSTRAINT_IDREF(id, rsc_1, pcmk__xe_id(xml_rsc)); for (xmlNode *xml_rsc_2 = pcmk__xe_first_child(set2, PCMK_XE_RESOURCE_REF, NULL, NULL); xml_rsc_2 != NULL; xml_rsc_2 = pcmk__xe_next_same(xml_rsc_2)) { EXPAND_CONSTRAINT_IDREF(id, rsc_2, pcmk__xe_id(xml_rsc_2)); pcmk__order_resource_actions(rsc_1, action_1, rsc_2, action_2, flags); } } } return pcmk_rc_ok; } /*! * \internal * \brief If an ordering constraint uses resource tags, expand them * * \param[in,out] xml_obj Ordering constraint XML * \param[out] expanded_xml Equivalent XML with tags expanded * \param[in] scheduler Scheduler data * * \return Standard Pacemaker return code (specifically, pcmk_rc_ok on success, * and pcmk_rc_unpack_error on invalid configuration) */ static int unpack_order_tags(xmlNode *xml_obj, xmlNode **expanded_xml, const pcmk_scheduler_t *scheduler) { const char *id_first = NULL; const char *id_then = NULL; const char *action_first = NULL; const char *action_then = NULL; pcmk_resource_t *rsc_first = NULL; pcmk_resource_t *rsc_then = NULL; pcmk_tag_t *tag_first = NULL; pcmk_tag_t *tag_then = NULL; xmlNode *rsc_set_first = NULL; xmlNode *rsc_set_then = NULL; bool any_sets = false; // Check whether there are any resource sets with template or tag references *expanded_xml = pcmk__expand_tags_in_sets(xml_obj, scheduler); if (*expanded_xml != NULL) { crm_log_xml_trace(*expanded_xml, "Expanded " PCMK_XE_RSC_ORDER); return pcmk_rc_ok; } id_first = crm_element_value(xml_obj, PCMK_XA_FIRST); id_then = crm_element_value(xml_obj, PCMK_XA_THEN); if ((id_first == NULL) || (id_then == NULL)) { return pcmk_rc_ok; } if (!pcmk__valid_resource_or_tag(scheduler, id_first, &rsc_first, &tag_first)) { pcmk__config_err("Ignoring constraint '%s' because '%s' is not a " "valid resource or tag", pcmk__xe_id(xml_obj), id_first); return pcmk_rc_unpack_error; } if (!pcmk__valid_resource_or_tag(scheduler, id_then, &rsc_then, &tag_then)) { pcmk__config_err("Ignoring constraint '%s' because '%s' is not a " "valid resource or tag", pcmk__xe_id(xml_obj), id_then); return pcmk_rc_unpack_error; } if ((rsc_first != NULL) && (rsc_then != NULL)) { // Neither side references a template or tag return pcmk_rc_ok; } action_first = crm_element_value(xml_obj, PCMK_XA_FIRST_ACTION); action_then = crm_element_value(xml_obj, PCMK_XA_THEN_ACTION); *expanded_xml = pcmk__xml_copy(NULL, xml_obj); /* Convert template/tag reference in PCMK_XA_FIRST into constraint * PCMK_XE_RESOURCE_SET */ if (!pcmk__tag_to_set(*expanded_xml, &rsc_set_first, PCMK_XA_FIRST, true, scheduler)) { free_xml(*expanded_xml); *expanded_xml = NULL; return pcmk_rc_unpack_error; } if (rsc_set_first != NULL) { if (action_first != NULL) { /* Move PCMK_XA_FIRST_ACTION into converted PCMK_XE_RESOURCE_SET as * PCMK_XA_ACTION */ crm_xml_add(rsc_set_first, PCMK_XA_ACTION, action_first); pcmk__xe_remove_attr(*expanded_xml, PCMK_XA_FIRST_ACTION); } any_sets = true; } /* Convert template/tag reference in PCMK_XA_THEN into constraint * PCMK_XE_RESOURCE_SET */ if (!pcmk__tag_to_set(*expanded_xml, &rsc_set_then, PCMK_XA_THEN, true, scheduler)) { free_xml(*expanded_xml); *expanded_xml = NULL; return pcmk_rc_unpack_error; } if (rsc_set_then != NULL) { if (action_then != NULL) { /* Move PCMK_XA_THEN_ACTION into converted PCMK_XE_RESOURCE_SET as * PCMK_XA_ACTION */ crm_xml_add(rsc_set_then, PCMK_XA_ACTION, action_then); pcmk__xe_remove_attr(*expanded_xml, PCMK_XA_THEN_ACTION); } any_sets = true; } if (any_sets) { crm_log_xml_trace(*expanded_xml, "Expanded " PCMK_XE_RSC_ORDER); } else { free_xml(*expanded_xml); *expanded_xml = NULL; } return pcmk_rc_ok; } /*! * \internal * \brief Unpack ordering constraint XML * * \param[in,out] xml_obj Ordering constraint XML to unpack * \param[in,out] scheduler Scheduler data */ void pcmk__unpack_ordering(xmlNode *xml_obj, pcmk_scheduler_t *scheduler) { xmlNode *set = NULL; xmlNode *last = NULL; xmlNode *orig_xml = NULL; xmlNode *expanded_xml = NULL; const char *id = crm_element_value(xml_obj, PCMK_XA_ID); const char *invert = crm_element_value(xml_obj, PCMK_XA_SYMMETRICAL); enum pe_order_kind kind = get_ordering_type(xml_obj); enum ordering_symmetry symmetry = get_ordering_symmetry(xml_obj, kind, NULL); // Expand any resource tags in the constraint XML if (unpack_order_tags(xml_obj, &expanded_xml, scheduler) != pcmk_rc_ok) { return; } if (expanded_xml != NULL) { orig_xml = xml_obj; xml_obj = expanded_xml; } // If the constraint has resource sets, unpack them for (set = pcmk__xe_first_child(xml_obj, PCMK_XE_RESOURCE_SET, NULL, NULL); set != NULL; set = pcmk__xe_next_same(set)) { set = expand_idref(set, scheduler->input); if ((set == NULL) // Configuration error, message already logged || (unpack_order_set(set, kind, invert, scheduler) != pcmk_rc_ok)) { if (expanded_xml != NULL) { free_xml(expanded_xml); } return; } if (last != NULL) { if (order_rsc_sets(id, last, set, kind, scheduler, symmetry) != pcmk_rc_ok) { if (expanded_xml != NULL) { free_xml(expanded_xml); } return; } if ((symmetry == ordering_symmetric) && (order_rsc_sets(id, set, last, kind, scheduler, ordering_symmetric_inverse) != pcmk_rc_ok)) { if (expanded_xml != NULL) { free_xml(expanded_xml); } return; } } last = set; } if (expanded_xml) { free_xml(expanded_xml); xml_obj = orig_xml; } // If the constraint has no resource sets, unpack it as a simple ordering if (last == NULL) { return unpack_simple_rsc_order(xml_obj, scheduler); } } static bool ordering_is_invalid(pcmk_action_t *action, pcmk__related_action_t *input) { /* Prevent user-defined ordering constraints between resources * running in a guest node and the resource that defines that node. */ if (!pcmk_is_set(input->type, pcmk__ar_guest_allowed) && (input->action->rsc != NULL) && pcmk__rsc_corresponds_to_guest(action->rsc, input->action->node)) { pcmk__config_warn("Invalid ordering constraint between %s and %s", input->action->rsc->id, action->rsc->id); return true; } /* If there's an order like * "rscB_stop node2"-> "load_stopped_node2" -> "rscA_migrate_to node1" * * then rscA is being migrated from node1 to node2, while rscB is being * migrated from node2 to node1. If there would be a graph loop, * break the order "load_stopped_node2" -> "rscA_migrate_to node1". */ if (((uint32_t) input->type == pcmk__ar_if_on_same_node_or_target) && (action->rsc != NULL) && pcmk__str_eq(action->task, PCMK_ACTION_MIGRATE_TO, pcmk__str_none) && pcmk__graph_has_loop(action, action, input)) { return true; } return false; } void pcmk__disable_invalid_orderings(pcmk_scheduler_t *scheduler) { for (GList *iter = scheduler->actions; iter != NULL; iter = iter->next) { pcmk_action_t *action = (pcmk_action_t *) iter->data; pcmk__related_action_t *input = NULL; for (GList *input_iter = action->actions_before; input_iter != NULL; input_iter = input_iter->next) { input = input_iter->data; if (ordering_is_invalid(action, input)) { input->type = (enum pe_ordering) pcmk__ar_none; } } } } /*! * \internal * \brief Order stops on a node before the node's shutdown * * \param[in,out] node Node being shut down * \param[in] shutdown_op Shutdown action for node */ void pcmk__order_stops_before_shutdown(pcmk_node_t *node, pcmk_action_t *shutdown_op) { for (GList *iter = node->details->data_set->actions; iter != NULL; iter = iter->next) { pcmk_action_t *action = (pcmk_action_t *) iter->data; // Only stops on the node shutting down are relevant if (!pcmk__same_node(action->node, node) || !pcmk__str_eq(action->task, PCMK_ACTION_STOP, pcmk__str_none)) { continue; } // Resources and nodes in maintenance mode won't be touched if (pcmk_is_set(action->rsc->flags, pcmk_rsc_maintenance)) { pcmk__rsc_trace(action->rsc, "Not ordering %s before shutdown of %s because " "resource in maintenance mode", action->uuid, pcmk__node_name(node)); continue; } else if (node->details->maintenance) { pcmk__rsc_trace(action->rsc, "Not ordering %s before shutdown of %s because " "node in maintenance mode", action->uuid, pcmk__node_name(node)); continue; } /* Don't touch a resource that is unmanaged or blocked, to avoid * blocking the shutdown (though if another action depends on this one, * we may still end up blocking) */ if (!pcmk_any_flags_set(action->rsc->flags, pcmk_rsc_managed|pcmk_rsc_blocked)) { pcmk__rsc_trace(action->rsc, "Not ordering %s before shutdown of %s because " "resource is unmanaged or blocked", action->uuid, pcmk__node_name(node)); continue; } pcmk__rsc_trace(action->rsc, "Ordering %s before shutdown of %s", action->uuid, pcmk__node_name(node)); pcmk__clear_action_flags(action, pcmk_action_optional); pcmk__new_ordering(action->rsc, NULL, action, NULL, strdup(PCMK_ACTION_DO_SHUTDOWN), shutdown_op, pcmk__ar_ordered|pcmk__ar_unrunnable_first_blocks, node->details->data_set); } } /*! * \brief Find resource actions matching directly or as child * * \param[in] rsc Resource to check * \param[in] original_key Action key to search for (possibly referencing * parent of \rsc) * * \return Newly allocated list of matching actions * \note It is the caller's responsibility to free the result with g_list_free() */ static GList * find_actions_by_task(const pcmk_resource_t *rsc, const char *original_key) { // Search under given task key directly GList *list = find_actions(rsc->actions, original_key, NULL); if (list == NULL) { // Search again using this resource's ID char *key = NULL; char *task = NULL; guint interval_ms = 0; CRM_CHECK(parse_op_key(original_key, NULL, &task, &interval_ms), return NULL); key = pcmk__op_key(rsc->id, task, interval_ms); list = find_actions(rsc->actions, key, NULL); free(key); free(task); } return list; } /*! * \internal * \brief Order relevant resource actions after a given action * * \param[in,out] first_action Action to order after (or NULL if none runnable) * \param[in] rsc Resource whose actions should be ordered * \param[in,out] order Ordering constraint being applied */ static void order_resource_actions_after(pcmk_action_t *first_action, const pcmk_resource_t *rsc, pcmk__action_relation_t *order) { GList *then_actions = NULL; uint32_t flags = pcmk__ar_none; CRM_CHECK((rsc != NULL) && (order != NULL), return); flags = order->flags; pcmk__rsc_trace(rsc, "Applying ordering %d for 'then' resource %s", order->id, rsc->id); if (order->action2 != NULL) { then_actions = g_list_prepend(NULL, order->action2); } else { then_actions = find_actions_by_task(rsc, order->task2); } if (then_actions == NULL) { pcmk__rsc_trace(rsc, "Ignoring ordering %d: no %s actions found for %s", order->id, order->task2, rsc->id); return; } if ((first_action != NULL) && (first_action->rsc == rsc) && pcmk_is_set(first_action->flags, pcmk_action_migration_abort)) { pcmk__rsc_trace(rsc, "Detected dangling migration ordering (%s then %s %s)", first_action->uuid, order->task2, rsc->id); pcmk__clear_relation_flags(flags, pcmk__ar_first_implies_then); } if ((first_action == NULL) && !pcmk_is_set(flags, pcmk__ar_first_implies_then)) { pcmk__rsc_debug(rsc, "Ignoring ordering %d for %s: No first action found", order->id, rsc->id); g_list_free(then_actions); return; } for (GList *iter = then_actions; iter != NULL; iter = iter->next) { pcmk_action_t *then_action_iter = (pcmk_action_t *) iter->data; if (first_action != NULL) { order_actions(first_action, then_action_iter, flags); } else { pcmk__clear_action_flags(then_action_iter, pcmk_action_runnable); + // coverity[null_field] order->rsc1 can't be NULL here crm_warn("%s of %s is unrunnable because there is no %s of %s " "to order it after", then_action_iter->task, rsc->id, order->task1, order->rsc1->id); } } g_list_free(then_actions); } static void rsc_order_first(pcmk_resource_t *first_rsc, pcmk__action_relation_t *order) { GList *first_actions = NULL; pcmk_action_t *first_action = order->action1; pcmk_resource_t *then_rsc = order->rsc2; pcmk__assert(first_rsc != NULL); pcmk__rsc_trace(first_rsc, "Applying ordering constraint %d (first: %s)", order->id, first_rsc->id); if (first_action != NULL) { first_actions = g_list_prepend(NULL, first_action); } else { first_actions = find_actions_by_task(first_rsc, order->task1); } if ((first_actions == NULL) && (first_rsc == then_rsc)) { pcmk__rsc_trace(first_rsc, "Ignoring constraint %d: first (%s for %s) not found", order->id, order->task1, first_rsc->id); } else if (first_actions == NULL) { char *key = NULL; char *op_type = NULL; guint interval_ms = 0; parse_op_key(order->task1, NULL, &op_type, &interval_ms); key = pcmk__op_key(first_rsc->id, op_type, interval_ms); if ((first_rsc->fns->state(first_rsc, TRUE) == pcmk_role_stopped) && pcmk__str_eq(op_type, PCMK_ACTION_STOP, pcmk__str_none)) { free(key); pcmk__rsc_trace(first_rsc, "Ignoring constraint %d: first (%s for %s) " "not found", order->id, order->task1, first_rsc->id); } else if ((first_rsc->fns->state(first_rsc, TRUE) == pcmk_role_unpromoted) && pcmk__str_eq(op_type, PCMK_ACTION_DEMOTE, pcmk__str_none)) { free(key); pcmk__rsc_trace(first_rsc, "Ignoring constraint %d: first (%s for %s) " "not found", order->id, order->task1, first_rsc->id); } else { pcmk__rsc_trace(first_rsc, "Creating first (%s for %s) for constraint %d ", order->task1, first_rsc->id, order->id); first_action = custom_action(first_rsc, key, op_type, NULL, TRUE, first_rsc->cluster); first_actions = g_list_prepend(NULL, first_action); } free(op_type); } if (then_rsc == NULL) { if (order->action2 == NULL) { pcmk__rsc_trace(first_rsc, "Ignoring constraint %d: then not found", order->id); return; } then_rsc = order->action2->rsc; } for (GList *iter = first_actions; iter != NULL; iter = iter->next) { first_action = iter->data; if (then_rsc == NULL) { order_actions(first_action, order->action2, order->flags); } else { order_resource_actions_after(first_action, then_rsc, order); } } g_list_free(first_actions); } // GFunc to call pcmk__block_colocation_dependents() static void block_colocation_dependents(gpointer data, gpointer user_data) { pcmk__block_colocation_dependents(data); } // GFunc to call pcmk__update_action_for_orderings() static void update_action_for_orderings(gpointer data, gpointer user_data) { pcmk__update_action_for_orderings((pcmk_action_t *) data, (pcmk_scheduler_t *) user_data); } /*! * \internal * \brief Apply all ordering constraints * * \param[in,out] sched Scheduler data */ void pcmk__apply_orderings(pcmk_scheduler_t *sched) { crm_trace("Applying ordering constraints"); /* Ordering constraints need to be processed in the order they were created. * rsc_order_first() and order_resource_actions_after() require the relevant * actions to already exist in some cases, but rsc_order_first() will create * the 'first' action in certain cases. Thus calling rsc_order_first() can * change the behavior of later-created orderings. * * Also, g_list_append() should be avoided for performance reasons, so we * prepend orderings when creating them and reverse the list here. * * @TODO This is brittle and should be carefully redesigned so that the * order of creation doesn't matter, and the reverse becomes unneeded. */ sched->ordering_constraints = g_list_reverse(sched->ordering_constraints); for (GList *iter = sched->ordering_constraints; iter != NULL; iter = iter->next) { pcmk__action_relation_t *order = iter->data; pcmk_resource_t *rsc = order->rsc1; if (rsc != NULL) { rsc_order_first(rsc, order); continue; } rsc = order->rsc2; if (rsc != NULL) { order_resource_actions_after(order->action1, rsc, order); } else { crm_trace("Applying ordering constraint %d (non-resource actions)", order->id); order_actions(order->action1, order->action2, order->flags); } } g_list_foreach(sched->actions, block_colocation_dependents, NULL); crm_trace("Ordering probes"); pcmk__order_probes(sched); crm_trace("Updating %d actions", g_list_length(sched->actions)); g_list_foreach(sched->actions, update_action_for_orderings, sched); pcmk__disable_invalid_orderings(sched); } /*! * \internal * \brief Order a given action after each action in a given list * * \param[in,out] after "After" action * \param[in,out] list List of "before" actions */ void pcmk__order_after_each(pcmk_action_t *after, GList *list) { const char *after_desc = (after->task == NULL)? after->uuid : after->task; for (GList *iter = list; iter != NULL; iter = iter->next) { pcmk_action_t *before = (pcmk_action_t *) iter->data; const char *before_desc = before->task? before->task : before->uuid; crm_debug("Ordering %s on %s before %s on %s", before_desc, pcmk__node_name(before->node), after_desc, pcmk__node_name(after->node)); order_actions(before, after, pcmk__ar_ordered); } } /*! * \internal * \brief Order promotions and demotions for restarts of a clone or bundle * * \param[in,out] rsc Clone or bundle to order */ void pcmk__promotable_restart_ordering(pcmk_resource_t *rsc) { // Order start and promote after all instances are stopped pcmk__order_resource_actions(rsc, PCMK_ACTION_STOPPED, rsc, PCMK_ACTION_START, pcmk__ar_ordered); pcmk__order_resource_actions(rsc, PCMK_ACTION_STOPPED, rsc, PCMK_ACTION_PROMOTE, pcmk__ar_ordered); // Order stop, start, and promote after all instances are demoted pcmk__order_resource_actions(rsc, PCMK_ACTION_DEMOTED, rsc, PCMK_ACTION_STOP, pcmk__ar_ordered); pcmk__order_resource_actions(rsc, PCMK_ACTION_DEMOTED, rsc, PCMK_ACTION_START, pcmk__ar_ordered); pcmk__order_resource_actions(rsc, PCMK_ACTION_DEMOTED, rsc, PCMK_ACTION_PROMOTE, pcmk__ar_ordered); // Order promote after all instances are started pcmk__order_resource_actions(rsc, PCMK_ACTION_RUNNING, rsc, PCMK_ACTION_PROMOTE, pcmk__ar_ordered); // Order demote after all instances are demoted pcmk__order_resource_actions(rsc, PCMK_ACTION_DEMOTE, rsc, PCMK_ACTION_DEMOTED, pcmk__ar_ordered); } diff --git a/lib/pacemaker/pcmk_simulate.c b/lib/pacemaker/pcmk_simulate.c index 3c19bcf847..30144d826c 100644 --- a/lib/pacemaker/pcmk_simulate.c +++ b/lib/pacemaker/pcmk_simulate.c @@ -1,1024 +1,1024 @@ /* - * Copyright 2021-2024 the Pacemaker project contributors + * Copyright 2021-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include #include #include "libpacemaker_private.h" static pcmk__output_t *out = NULL; static cib_t *fake_cib = NULL; static GList *fake_resource_list = NULL; static const GList *fake_op_fail_list = NULL; static void set_effective_date(pcmk_scheduler_t *scheduler, bool print_original, const char *use_date); /*! * \internal * \brief Create an action name for use in a dot graph * * \param[in] action Action to create name for * \param[in] verbose If true, add action ID to name * * \return Newly allocated string with action name * \note It is the caller's responsibility to free the result. */ static char * create_action_name(const pcmk_action_t *action, bool verbose) { char *action_name = NULL; const char *prefix = ""; const char *action_host = NULL; const char *clone_name = NULL; const char *task = action->task; if (action->node != NULL) { action_host = action->node->details->uname; } else if (!pcmk_is_set(action->flags, pcmk_action_pseudo)) { action_host = ""; } if (pcmk__str_eq(action->task, PCMK_ACTION_CANCEL, pcmk__str_none)) { prefix = "Cancel "; task = action->cancel_task; } if (action->rsc != NULL) { clone_name = action->rsc->clone_name; } if (clone_name != NULL) { char *key = NULL; guint interval_ms = 0; if (pcmk__guint_from_hash(action->meta, PCMK_META_INTERVAL, 0, &interval_ms) != pcmk_rc_ok) { interval_ms = 0; } if (pcmk__strcase_any_of(action->task, PCMK_ACTION_NOTIFY, PCMK_ACTION_NOTIFIED, NULL)) { const char *n_type = g_hash_table_lookup(action->meta, "notify_key_type"); const char *n_task = g_hash_table_lookup(action->meta, "notify_key_operation"); pcmk__assert(n_type != NULL); pcmk__assert(n_task != NULL); key = pcmk__notify_key(clone_name, n_type, n_task); } else { key = pcmk__op_key(clone_name, task, interval_ms); } if (action_host != NULL) { action_name = crm_strdup_printf("%s%s %s", prefix, key, action_host); } else { action_name = crm_strdup_printf("%s%s", prefix, key); } free(key); } else if (pcmk__str_eq(action->task, PCMK_ACTION_STONITH, pcmk__str_none)) { const char *op = g_hash_table_lookup(action->meta, PCMK__META_STONITH_ACTION); action_name = crm_strdup_printf("%s%s '%s' %s", prefix, action->task, op, action_host); } else if (action->rsc && action_host) { action_name = crm_strdup_printf("%s%s %s", prefix, action->uuid, action_host); } else if (action_host) { action_name = crm_strdup_printf("%s%s %s", prefix, action->task, action_host); } else { action_name = crm_strdup_printf("%s", action->uuid); } if (verbose) { char *with_id = crm_strdup_printf("%s (%d)", action_name, action->id); free(action_name); action_name = with_id; } return action_name; } /*! * \internal * \brief Display the status of a cluster * * \param[in,out] scheduler Scheduler data * \param[in] show_opts How to modify display (as pcmk_show_opt_e flags) * \param[in] section_opts Sections to display (as pcmk_section_e flags) * \param[in] title What to use as list title * \param[in] print_spacer Whether to display a spacer first */ static void print_cluster_status(pcmk_scheduler_t *scheduler, uint32_t show_opts, uint32_t section_opts, const char *title, bool print_spacer) { pcmk__output_t *out = scheduler->priv; GList *all = NULL; crm_exit_t stonith_rc = 0; enum pcmk_pacemakerd_state state = pcmk_pacemakerd_state_invalid; section_opts |= pcmk_section_nodes | pcmk_section_resources; show_opts |= pcmk_show_inactive_rscs | pcmk_show_failed_detail; all = g_list_prepend(all, (gpointer) "*"); PCMK__OUTPUT_SPACER_IF(out, print_spacer); out->begin_list(out, NULL, NULL, "%s", title); out->message(out, "cluster-status", scheduler, state, stonith_rc, NULL, pcmk__fence_history_none, section_opts, show_opts, NULL, all, all); out->end_list(out); g_list_free(all); } /*! * \internal * \brief Display a summary of all actions scheduled in a transition * * \param[in,out] scheduler Scheduler data (fully scheduled) * \param[in] print_spacer Whether to display a spacer first */ static void print_transition_summary(pcmk_scheduler_t *scheduler, bool print_spacer) { pcmk__output_t *out = scheduler->priv; PCMK__OUTPUT_SPACER_IF(out, print_spacer); out->begin_list(out, NULL, NULL, "Transition Summary"); pcmk__output_actions(scheduler); out->end_list(out); } /*! * \internal * \brief Reset scheduler input, output, date, and flags * * \param[in,out] scheduler Scheduler data * \param[in] input What to set as cluster input * \param[in] out What to set as cluster output object * \param[in] use_date What to set as cluster's current timestamp * \param[in] flags Group of enum pcmk_scheduler_flags to set */ static void reset(pcmk_scheduler_t *scheduler, xmlNodePtr input, pcmk__output_t *out, const char *use_date, unsigned int flags) { scheduler->input = input; scheduler->priv = out; set_effective_date(scheduler, true, use_date); if (pcmk_is_set(flags, pcmk_sim_sanitized)) { pcmk__set_scheduler_flags(scheduler, pcmk_sched_sanitized); } if (pcmk_is_set(flags, pcmk_sim_show_scores)) { pcmk__set_scheduler_flags(scheduler, pcmk_sched_output_scores); } if (pcmk_is_set(flags, pcmk_sim_show_utilization)) { pcmk__set_scheduler_flags(scheduler, pcmk_sched_show_utilization); } } /*! * \brief Write out a file in dot(1) format describing the actions that will * be taken by the scheduler in response to an input CIB file. * * \param[in,out] scheduler Scheduler data * \param[in] dot_file The filename to write * \param[in] all_actions Write all actions, even those that are optional * or are on unmanaged resources * \param[in] verbose Add extra information, such as action IDs, to the * output * * \return Standard Pacemaker return code */ static int write_sim_dotfile(pcmk_scheduler_t *scheduler, const char *dot_file, bool all_actions, bool verbose) { GList *iter = NULL; FILE *dot_strm = fopen(dot_file, "w"); if (dot_strm == NULL) { return errno; } fprintf(dot_strm, " digraph \"g\" {\n"); for (iter = scheduler->actions; iter != NULL; iter = iter->next) { pcmk_action_t *action = (pcmk_action_t *) iter->data; const char *style = "dashed"; const char *font = "black"; - const char *color = "black"; + const char *color = NULL; char *action_name = create_action_name(action, verbose); if (pcmk_is_set(action->flags, pcmk_action_pseudo)) { font = "orange"; } if (pcmk_is_set(action->flags, pcmk_action_added_to_graph)) { style = PCMK__VALUE_BOLD; color = "green"; } else if ((action->rsc != NULL) && !pcmk_is_set(action->rsc->flags, pcmk_rsc_managed)) { color = "red"; font = "purple"; if (!all_actions) { goto do_not_write; } } else if (pcmk_is_set(action->flags, pcmk_action_optional)) { color = "blue"; if (!all_actions) { goto do_not_write; } } else { color = "red"; CRM_LOG_ASSERT(!pcmk_is_set(action->flags, pcmk_action_runnable)); } pcmk__set_action_flags(action, pcmk_action_added_to_graph); fprintf(dot_strm, "\"%s\" [ style=%s color=\"%s\" fontcolor=\"%s\"]\n", action_name, style, color, font); do_not_write: free(action_name); } for (iter = scheduler->actions; iter != NULL; iter = iter->next) { pcmk_action_t *action = (pcmk_action_t *) iter->data; for (GList *before_iter = action->actions_before; before_iter != NULL; before_iter = before_iter->next) { pcmk__related_action_t *before = before_iter->data; char *before_name = NULL; char *after_name = NULL; const char *style = "dashed"; bool optional = true; if (before->state == pe_link_dumped) { optional = false; style = PCMK__VALUE_BOLD; } else if ((uint32_t) before->type == pcmk__ar_none) { continue; } else if (pcmk_is_set(before->action->flags, pcmk_action_added_to_graph) && pcmk_is_set(action->flags, pcmk_action_added_to_graph) && (uint32_t) before->type != pcmk__ar_if_on_same_node_or_target) { optional = false; } if (all_actions || !optional) { before_name = create_action_name(before->action, verbose); after_name = create_action_name(action, verbose); fprintf(dot_strm, "\"%s\" -> \"%s\" [ style = %s]\n", before_name, after_name, style); free(before_name); free(after_name); } } } fprintf(dot_strm, "}\n"); fflush(dot_strm); fclose(dot_strm); return pcmk_rc_ok; } /*! * \brief Profile the configuration updates and scheduler actions in a single * CIB file, printing the profiling timings. * * \note \p scheduler->priv must have been set to a valid \p pcmk__output_t * object before this function is called. * * \param[in] xml_file The CIB file to profile * \param[in] repeat Number of times to run * \param[in,out] scheduler Scheduler data * \param[in] use_date The date to set the cluster's time to (may be NULL) */ static void profile_file(const char *xml_file, long long repeat, pcmk_scheduler_t *scheduler, const char *use_date) { pcmk__output_t *out = scheduler->priv; xmlNode *cib_object = NULL; clock_t start = 0; clock_t end; unsigned long long scheduler_flags = pcmk_sched_no_compat; pcmk__assert(out != NULL); cib_object = pcmk__xml_read(xml_file); start = clock(); if (pcmk_find_cib_element(cib_object, PCMK_XE_STATUS) == NULL) { pcmk__xe_create(cib_object, PCMK_XE_STATUS); } if (pcmk__update_configured_schema(&cib_object, false) != pcmk_rc_ok) { free_xml(cib_object); return; } if (!pcmk__validate_xml(cib_object, NULL, NULL, NULL)) { free_xml(cib_object); return; } if (pcmk_is_set(scheduler->flags, pcmk_sched_output_scores)) { scheduler_flags |= pcmk_sched_output_scores; } if (pcmk_is_set(scheduler->flags, pcmk_sched_show_utilization)) { scheduler_flags |= pcmk_sched_show_utilization; } for (int i = 0; i < repeat; ++i) { xmlNode *input = cib_object; if (repeat > 1) { input = pcmk__xml_copy(NULL, cib_object); } scheduler->input = input; set_effective_date(scheduler, false, use_date); pcmk__schedule_actions(input, scheduler_flags, scheduler); pe_reset_working_set(scheduler); } end = clock(); out->message(out, "profile", xml_file, start, end); } void pcmk__profile_dir(const char *dir, long long repeat, pcmk_scheduler_t *scheduler, const char *use_date) { pcmk__output_t *out = scheduler->priv; - struct dirent **namelist; + struct dirent **namelist = NULL; int file_num = scandir(dir, &namelist, 0, alphasort); pcmk__assert(out != NULL); if (file_num > 0) { struct stat prop; char buffer[FILENAME_MAX]; out->begin_list(out, NULL, NULL, "Timings"); while (file_num--) { if ('.' == namelist[file_num]->d_name[0]) { free(namelist[file_num]); continue; } else if (!pcmk__ends_with_ext(namelist[file_num]->d_name, ".xml")) { free(namelist[file_num]); continue; } snprintf(buffer, sizeof(buffer), "%s/%s", dir, namelist[file_num]->d_name); if (stat(buffer, &prop) == 0 && S_ISREG(prop.st_mode)) { profile_file(buffer, repeat, scheduler, use_date); } free(namelist[file_num]); } free(namelist); out->end_list(out); } } /*! * \brief Set the date of the cluster, either to the value given by * \p use_date, or to the \c PCMK_XA_EXECUTION_DATE value in the CIB. * * \note \p scheduler->priv must have been set to a valid \p pcmk__output_t * object before this function is called. * * \param[in,out] scheduler Scheduler data * \param[in] print_original If \p true, the \c PCMK_XA_EXECUTION_DATE * should also be printed * \param[in] use_date The date to set the cluster's time to * (may be NULL) */ static void set_effective_date(pcmk_scheduler_t *scheduler, bool print_original, const char *use_date) { pcmk__output_t *out = scheduler->priv; time_t original_date = 0; pcmk__assert(out != NULL); crm_element_value_epoch(scheduler->input, PCMK_XA_EXECUTION_DATE, &original_date); if (use_date) { scheduler->now = crm_time_new(use_date); out->info(out, "Setting effective cluster time: %s", use_date); crm_time_log(LOG_NOTICE, "Pretending 'now' is", scheduler->now, crm_time_log_date | crm_time_log_timeofday); } else if (original_date != 0) { scheduler->now = pcmk__copy_timet(original_date); if (print_original) { char *when = crm_time_as_string(scheduler->now, crm_time_log_date|crm_time_log_timeofday); out->info(out, "Using the original execution date of: %s", when); free(when); } } } /*! * \internal * \brief Simulate successfully executing a pseudo-action in a graph * * \param[in,out] graph Graph to update with pseudo-action result * \param[in,out] action Pseudo-action to simulate executing * * \return Standard Pacemaker return code */ static int simulate_pseudo_action(pcmk__graph_t *graph, pcmk__graph_action_t *action) { const char *node = crm_element_value(action->xml, PCMK__META_ON_NODE); const char *task = crm_element_value(action->xml, PCMK__XA_OPERATION_KEY); pcmk__set_graph_action_flags(action, pcmk__graph_action_confirmed); out->message(out, "inject-pseudo-action", node, task); pcmk__update_graph(graph, action); return pcmk_rc_ok; } /*! * \internal * \brief Simulate executing a resource action in a graph * * \param[in,out] graph Graph to update with resource action result * \param[in,out] action Resource action to simulate executing * * \return Standard Pacemaker return code */ static int simulate_resource_action(pcmk__graph_t *graph, pcmk__graph_action_t *action) { int rc; lrmd_event_data_t *op = NULL; int target_outcome = PCMK_OCF_OK; const char *rtype = NULL; const char *rclass = NULL; const char *resource = NULL; const char *rprovider = NULL; const char *resource_config_name = NULL; const char *operation = crm_element_value(action->xml, PCMK_XA_OPERATION); const char *target_rc_s = crm_meta_value(action->params, PCMK__META_OP_TARGET_RC); xmlNode *cib_node = NULL; xmlNode *cib_resource = NULL; xmlNode *action_rsc = pcmk__xe_first_child(action->xml, PCMK_XE_PRIMITIVE, NULL, NULL); char *node = crm_element_value_copy(action->xml, PCMK__META_ON_NODE); char *uuid = NULL; const char *router_node = crm_element_value(action->xml, PCMK__XA_ROUTER_NODE); // Certain actions don't need to be displayed or history entries if (pcmk__str_eq(operation, CRM_OP_REPROBE, pcmk__str_none)) { crm_debug("No history injection for %s op on %s", operation, node); goto done; // Confirm action and update graph } if (action_rsc == NULL) { // Shouldn't be possible crm_log_xml_err(action->xml, "Bad"); free(node); return EPROTO; } /* A resource might be known by different names in the configuration and in * the action (for example, a clone instance). Grab the configuration name * (which is preferred when writing history), and if necessary, the instance * name. */ resource_config_name = crm_element_value(action_rsc, PCMK_XA_ID); if (resource_config_name == NULL) { // Shouldn't be possible crm_log_xml_err(action->xml, "No ID"); free(node); return EPROTO; } resource = resource_config_name; if (pe_find_resource(fake_resource_list, resource) == NULL) { const char *longname = crm_element_value(action_rsc, PCMK__XA_LONG_ID); if ((longname != NULL) && (pe_find_resource(fake_resource_list, longname) != NULL)) { resource = longname; } } // Certain actions need to be displayed but don't need history entries if (pcmk__strcase_any_of(operation, PCMK_ACTION_DELETE, PCMK_ACTION_META_DATA, NULL)) { out->message(out, "inject-rsc-action", resource, operation, node, (guint) 0); goto done; // Confirm action and update graph } rclass = crm_element_value(action_rsc, PCMK_XA_CLASS); rtype = crm_element_value(action_rsc, PCMK_XA_TYPE); rprovider = crm_element_value(action_rsc, PCMK_XA_PROVIDER); pcmk__scan_min_int(target_rc_s, &target_outcome, 0); pcmk__assert(fake_cib->cmds->query(fake_cib, NULL, NULL, cib_sync_call|cib_scope_local) == pcmk_ok); // Ensure the action node is in the CIB uuid = crm_element_value_copy(action->xml, PCMK__META_ON_NODE_UUID); cib_node = pcmk__inject_node(fake_cib, node, ((router_node == NULL)? uuid: node)); free(uuid); pcmk__assert(cib_node != NULL); // Add a history entry for the action cib_resource = pcmk__inject_resource_history(out, cib_node, resource, resource_config_name, rclass, rtype, rprovider); if (cib_resource == NULL) { crm_err("Could not simulate action %d history for resource %s", action->id, resource); free(node); free_xml(cib_node); return EINVAL; } // Simulate and display an executor event for the action result op = pcmk__event_from_graph_action(cib_resource, action, PCMK_EXEC_DONE, target_outcome, "User-injected result"); out->message(out, "inject-rsc-action", resource, op->op_type, node, op->interval_ms); // Check whether action is in a list of desired simulated failures for (const GList *iter = fake_op_fail_list; iter != NULL; iter = iter->next) { const char *spec = (const char *) iter->data; char *key = NULL; const char *match_name = NULL; const char *offset = NULL; // Allow user to specify anonymous clone with or without instance number key = crm_strdup_printf(PCMK__OP_FMT "@%s=", resource, op->op_type, op->interval_ms, node); if (strncasecmp(key, spec, strlen(key)) == 0) { match_name = resource; } free(key); // If not found, try the resource's name in the configuration if ((match_name == NULL) && (strcmp(resource, resource_config_name) != 0)) { key = crm_strdup_printf(PCMK__OP_FMT "@%s=", resource_config_name, op->op_type, op->interval_ms, node); if (strncasecmp(key, spec, strlen(key)) == 0) { match_name = resource_config_name; } free(key); } if (match_name == NULL) { continue; // This failed action entry doesn't match } // ${match_name}_${task}_${interval_in_ms}@${node}=${rc} rc = sscanf(spec, "%*[^=]=%d", (int *) &op->rc); if (rc != 1) { out->err(out, "Invalid failed operation '%s' " "(result code must be integer)", spec); continue; // Keep checking other list entries } out->info(out, "Pretending action %d failed with rc=%d", action->id, op->rc); pcmk__set_graph_action_flags(action, pcmk__graph_action_failed); graph->abort_priority = PCMK_SCORE_INFINITY; if (pcmk__str_eq(op->op_type, PCMK_ACTION_START, pcmk__str_none)) { offset = pcmk__s(graph->failed_start_offset, PCMK_VALUE_INFINITY); } else if (pcmk__str_eq(op->op_type, PCMK_ACTION_STOP, pcmk__str_none)) { offset = pcmk__s(graph->failed_stop_offset, PCMK_VALUE_INFINITY); } pcmk__inject_failcount(out, fake_cib, cib_node, match_name, op->op_type, op->interval_ms, op->rc, pcmk_str_is_infinity(offset)); break; } pcmk__inject_action_result(cib_resource, op, target_outcome); lrmd_free_event(op); rc = fake_cib->cmds->modify(fake_cib, PCMK_XE_STATUS, cib_node, cib_sync_call|cib_scope_local); pcmk__assert(rc == pcmk_ok); done: free(node); free_xml(cib_node); pcmk__set_graph_action_flags(action, pcmk__graph_action_confirmed); pcmk__update_graph(graph, action); return pcmk_rc_ok; } /*! * \internal * \brief Simulate successfully executing a cluster action * * \param[in,out] graph Graph to update with action result * \param[in,out] action Cluster action to simulate * * \return Standard Pacemaker return code */ static int simulate_cluster_action(pcmk__graph_t *graph, pcmk__graph_action_t *action) { const char *node = crm_element_value(action->xml, PCMK__META_ON_NODE); const char *task = crm_element_value(action->xml, PCMK_XA_OPERATION); xmlNode *rsc = pcmk__xe_first_child(action->xml, PCMK_XE_PRIMITIVE, NULL, NULL); pcmk__set_graph_action_flags(action, pcmk__graph_action_confirmed); out->message(out, "inject-cluster-action", node, task, rsc); pcmk__update_graph(graph, action); return pcmk_rc_ok; } /*! * \internal * \brief Simulate successfully executing a fencing action * * \param[in,out] graph Graph to update with action result * \param[in,out] action Fencing action to simulate * * \return Standard Pacemaker return code */ static int simulate_fencing_action(pcmk__graph_t *graph, pcmk__graph_action_t *action) { const char *op = crm_meta_value(action->params, PCMK__META_STONITH_ACTION); char *target = crm_element_value_copy(action->xml, PCMK__META_ON_NODE); out->message(out, "inject-fencing-action", target, op); if (!pcmk__str_eq(op, PCMK_ACTION_ON, pcmk__str_casei)) { int rc = pcmk_ok; GString *xpath = g_string_sized_new(512); // Set node state to offline xmlNode *cib_node = pcmk__inject_node_state_change(fake_cib, target, false); pcmk__assert(cib_node != NULL); crm_xml_add(cib_node, PCMK_XA_CRM_DEBUG_ORIGIN, __func__); rc = fake_cib->cmds->replace(fake_cib, PCMK_XE_STATUS, cib_node, cib_sync_call|cib_scope_local); pcmk__assert(rc == pcmk_ok); // Simulate controller clearing node's resource history and attributes pcmk__g_strcat(xpath, "//" PCMK__XE_NODE_STATE "[@" PCMK_XA_UNAME "='", target, "']/" PCMK__XE_LRM, NULL); fake_cib->cmds->remove(fake_cib, (const char *) xpath->str, NULL, cib_xpath|cib_sync_call|cib_scope_local); g_string_truncate(xpath, 0); pcmk__g_strcat(xpath, "//" PCMK__XE_NODE_STATE "[@" PCMK_XA_UNAME "='", target, "']" "/" PCMK__XE_TRANSIENT_ATTRIBUTES, NULL); fake_cib->cmds->remove(fake_cib, (const char *) xpath->str, NULL, cib_xpath|cib_sync_call|cib_scope_local); free_xml(cib_node); g_string_free(xpath, TRUE); } pcmk__set_graph_action_flags(action, pcmk__graph_action_confirmed); pcmk__update_graph(graph, action); free(target); return pcmk_rc_ok; } enum pcmk__graph_status pcmk__simulate_transition(pcmk_scheduler_t *scheduler, cib_t *cib, const GList *op_fail_list) { pcmk__graph_t *transition = NULL; enum pcmk__graph_status graph_rc; pcmk__graph_functions_t simulation_fns = { simulate_pseudo_action, simulate_resource_action, simulate_cluster_action, simulate_fencing_action, }; out = scheduler->priv; fake_cib = cib; fake_op_fail_list = op_fail_list; if (!out->is_quiet(out)) { out->begin_list(out, NULL, NULL, "Executing Cluster Transition"); } pcmk__set_graph_functions(&simulation_fns); transition = pcmk__unpack_graph(scheduler->graph, crm_system_name); pcmk__log_graph(LOG_DEBUG, transition); fake_resource_list = scheduler->resources; do { graph_rc = pcmk__execute_graph(transition); } while (graph_rc == pcmk__graph_active); fake_resource_list = NULL; if (graph_rc != pcmk__graph_complete) { out->err(out, "Transition failed: %s", pcmk__graph_status2text(graph_rc)); pcmk__log_graph(LOG_ERR, transition); out->err(out, "An invalid transition was produced"); } pcmk__free_graph(transition); if (!out->is_quiet(out)) { // If not quiet, we'll need the resulting CIB for later display xmlNode *cib_object = NULL; int rc = fake_cib->cmds->query(fake_cib, NULL, &cib_object, cib_sync_call|cib_scope_local); pcmk__assert(rc == pcmk_ok); pe_reset_working_set(scheduler); scheduler->input = cib_object; out->end_list(out); } return graph_rc; } int pcmk__simulate(pcmk_scheduler_t *scheduler, pcmk__output_t *out, const pcmk_injections_t *injections, unsigned int flags, uint32_t section_opts, const char *use_date, const char *input_file, const char *graph_file, const char *dot_file) { int printed = pcmk_rc_no_output; int rc = pcmk_rc_ok; xmlNodePtr input = NULL; cib_t *cib = NULL; rc = cib__signon_query(out, &cib, &input); if (rc != pcmk_rc_ok) { goto simulate_done; } reset(scheduler, input, out, use_date, flags); cluster_status(scheduler); if ((cib->variant == cib_native) && pcmk_is_set(section_opts, pcmk_section_times)) { if (pcmk__our_nodename == NULL) { // Currently used only in the times section pcmk__query_node_name(out, 0, &pcmk__our_nodename, 0); } scheduler->localhost = pcmk__our_nodename; } if (!out->is_quiet(out)) { const bool show_pending = pcmk_is_set(flags, pcmk_sim_show_pending); if (pcmk_is_set(scheduler->flags, pcmk_sched_in_maintenance)) { printed = out->message(out, "maint-mode", scheduler->flags); } if (scheduler->disabled_resources || scheduler->blocked_resources) { PCMK__OUTPUT_SPACER_IF(out, printed == pcmk_rc_ok); printed = out->info(out, "%d of %d resource instances DISABLED and " "%d BLOCKED from further action due to failure", scheduler->disabled_resources, scheduler->ninstances, scheduler->blocked_resources); } /* Most formatted output headers use caps for each word, but this one * only has the first word capitalized for compatibility with pcs. */ print_cluster_status(scheduler, (show_pending? pcmk_show_pending : 0), section_opts, "Current cluster status", (printed == pcmk_rc_ok)); printed = pcmk_rc_ok; } // If the user requested any injections, handle them if ((injections->node_down != NULL) || (injections->node_fail != NULL) || (injections->node_up != NULL) || (injections->op_inject != NULL) || (injections->ticket_activate != NULL) || (injections->ticket_grant != NULL) || (injections->ticket_revoke != NULL) || (injections->ticket_standby != NULL) || (injections->watchdog != NULL)) { PCMK__OUTPUT_SPACER_IF(out, printed == pcmk_rc_ok); pcmk__inject_scheduler_input(scheduler, cib, injections); printed = pcmk_rc_ok; rc = cib->cmds->query(cib, NULL, &input, cib_sync_call); if (rc != pcmk_rc_ok) { rc = pcmk_legacy2rc(rc); goto simulate_done; } cleanup_calculations(scheduler); reset(scheduler, input, out, use_date, flags); cluster_status(scheduler); } if (input_file != NULL) { rc = pcmk__xml_write_file(input, input_file, false, NULL); if (rc != pcmk_rc_ok) { goto simulate_done; } } if (pcmk_any_flags_set(flags, pcmk_sim_process | pcmk_sim_simulate)) { pcmk__output_t *logger_out = NULL; unsigned long long scheduler_flags = pcmk_sched_no_compat; if (pcmk_is_set(scheduler->flags, pcmk_sched_output_scores)) { scheduler_flags |= pcmk_sched_output_scores; } if (pcmk_is_set(scheduler->flags, pcmk_sched_show_utilization)) { scheduler_flags |= pcmk_sched_show_utilization; } if (pcmk_all_flags_set(scheduler->flags, pcmk_sched_output_scores |pcmk_sched_show_utilization)) { PCMK__OUTPUT_SPACER_IF(out, printed == pcmk_rc_ok); out->begin_list(out, NULL, NULL, "Assignment Scores and Utilization Information"); printed = pcmk_rc_ok; } else if (pcmk_is_set(scheduler->flags, pcmk_sched_output_scores)) { PCMK__OUTPUT_SPACER_IF(out, printed == pcmk_rc_ok); out->begin_list(out, NULL, NULL, "Assignment Scores"); printed = pcmk_rc_ok; } else if (pcmk_is_set(scheduler->flags, pcmk_sched_show_utilization)) { PCMK__OUTPUT_SPACER_IF(out, printed == pcmk_rc_ok); out->begin_list(out, NULL, NULL, "Utilization Information"); printed = pcmk_rc_ok; } else { rc = pcmk__log_output_new(&logger_out); if (rc != pcmk_rc_ok) { goto simulate_done; } pe__register_messages(logger_out); pcmk__register_lib_messages(logger_out); scheduler->priv = logger_out; } pcmk__schedule_actions(input, scheduler_flags, scheduler); if (logger_out == NULL) { out->end_list(out); } else { logger_out->finish(logger_out, CRM_EX_OK, true, NULL); pcmk__output_free(logger_out); scheduler->priv = out; } input = NULL; /* Don't try and free it twice */ if (graph_file != NULL) { rc = pcmk__xml_write_file(scheduler->graph, graph_file, false, NULL); if (rc != pcmk_rc_ok) { rc = pcmk_rc_graph_error; goto simulate_done; } } if (dot_file != NULL) { rc = write_sim_dotfile(scheduler, dot_file, pcmk_is_set(flags, pcmk_sim_all_actions), pcmk_is_set(flags, pcmk_sim_verbose)); if (rc != pcmk_rc_ok) { rc = pcmk_rc_dot_error; goto simulate_done; } } if (!out->is_quiet(out)) { print_transition_summary(scheduler, printed == pcmk_rc_ok); } } rc = pcmk_rc_ok; if (!pcmk_is_set(flags, pcmk_sim_simulate)) { goto simulate_done; } PCMK__OUTPUT_SPACER_IF(out, printed == pcmk_rc_ok); if (pcmk__simulate_transition(scheduler, cib, injections->op_fail) != pcmk__graph_complete) { rc = pcmk_rc_invalid_transition; } if (out->is_quiet(out)) { goto simulate_done; } set_effective_date(scheduler, true, use_date); if (pcmk_is_set(flags, pcmk_sim_show_scores)) { pcmk__set_scheduler_flags(scheduler, pcmk_sched_output_scores); } if (pcmk_is_set(flags, pcmk_sim_show_utilization)) { pcmk__set_scheduler_flags(scheduler, pcmk_sched_show_utilization); } cluster_status(scheduler); print_cluster_status(scheduler, 0, section_opts, "Revised Cluster Status", true); simulate_done: cib__clean_up_connection(&cib); return rc; } int pcmk_simulate(xmlNodePtr *xml, pcmk_scheduler_t *scheduler, const pcmk_injections_t *injections, unsigned int flags, unsigned int section_opts, const char *use_date, const char *input_file, const char *graph_file, const char *dot_file) { pcmk__output_t *out = NULL; int rc = pcmk_rc_ok; rc = pcmk__xml_output_new(&out, xml); if (rc != pcmk_rc_ok) { return rc; } pe__register_messages(out); pcmk__register_lib_messages(out); rc = pcmk__simulate(scheduler, out, injections, flags, section_opts, use_date, input_file, graph_file, dot_file); pcmk__xml_output_finish(out, pcmk_rc2exitc(rc), xml); return rc; } diff --git a/lib/pengine/Makefile.am b/lib/pengine/Makefile.am index 67d45b8fd4..7d487ac7c0 100644 --- a/lib/pengine/Makefile.am +++ b/lib/pengine/Makefile.am @@ -1,82 +1,82 @@ # # Copyright 2004-2024 the Pacemaker project contributors # # The version control history for this file may have further details. # # This source code is licensed under the GNU General Public License version 2 # or later (GPLv2+) WITHOUT ANY WARRANTY. # include $(top_srcdir)/mk/common.mk # Without "." here, check-recursive will run through the subdirectories first # and then run "make check" here. This will fail, because there's things in # the subdirectories that need check_LTLIBRARIES built first. Adding "." here # changes the order so the subdirectories are processed afterwards. SUBDIRS = . tests ## libraries lib_LTLIBRARIES = libpe_rules.la \ libpe_status.la check_LTLIBRARIES = libpe_status_test.la noinst_HEADERS = pe_status_private.h libpe_rules_la_LDFLAGS = -version-info 30:3:4 libpe_rules_la_CFLAGS = $(CFLAGS_HARDENED_LIB) libpe_rules_la_LDFLAGS += $(LDFLAGS_HARDENED_LIB) libpe_rules_la_LIBADD = $(top_builddir)/lib/common/libcrmcommon.la ## Library sources (*must* use += format for bumplibs) libpe_rules_la_SOURCES = common.c libpe_rules_la_SOURCES += rules.c libpe_rules_la_SOURCES += rules_alerts.c -libpe_status_la_LDFLAGS = -version-info 35:2:7 +libpe_status_la_LDFLAGS = -version-info 35:3:7 libpe_status_la_CFLAGS = $(CFLAGS_HARDENED_LIB) libpe_status_la_LDFLAGS += $(LDFLAGS_HARDENED_LIB) libpe_status_la_LIBADD = $(top_builddir)/lib/common/libcrmcommon.la ## Library sources (*must* use += format for bumplibs) libpe_status_la_SOURCES = libpe_status_la_SOURCES += bundle.c libpe_status_la_SOURCES += clone.c libpe_status_la_SOURCES += common.c libpe_status_la_SOURCES += complex.c libpe_status_la_SOURCES += failcounts.c libpe_status_la_SOURCES += group.c libpe_status_la_SOURCES += native.c libpe_status_la_SOURCES += pe_actions.c libpe_status_la_SOURCES += pe_health.c libpe_status_la_SOURCES += pe_digest.c libpe_status_la_SOURCES += pe_notif.c libpe_status_la_SOURCES += pe_output.c libpe_status_la_SOURCES += remote.c libpe_status_la_SOURCES += rules.c libpe_status_la_SOURCES += status.c libpe_status_la_SOURCES += tags.c libpe_status_la_SOURCES += unpack.c libpe_status_la_SOURCES += utils.c # # libpe_status_test is only used with unit tests, so we can # mock system calls. See lib/common/mock.c for details. # include $(top_srcdir)/mk/tap.mk libpe_status_test_la_SOURCES = $(libpe_status_la_SOURCES) libpe_status_test_la_LDFLAGS = $(libpe_status_la_LDFLAGS) \ -rpath $(libdir) \ $(LDFLAGS_WRAP) # See comments on libcrmcommon_test_la in lib/common/Makefile.am regarding these flags. libpe_status_test_la_CFLAGS = $(libpe_status_la_CFLAGS) \ -DPCMK__UNIT_TESTING \ -fno-builtin \ -fno-inline libpe_status_test_la_LIBADD = $(top_builddir)/lib/common/libcrmcommon_test.la \ -lcmocka \ -lm diff --git a/lib/pengine/bundle.c b/lib/pengine/bundle.c index f7b0755dc1..eaed40f56a 100644 --- a/lib/pengine/bundle.c +++ b/lib/pengine/bundle.c @@ -1,2275 +1,2276 @@ /* * Copyright 2004-2024 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include enum pe__bundle_mount_flags { pe__bundle_mount_none = 0x00, // mount instance-specific subdirectory rather than source directly pe__bundle_mount_subdir = 0x01 }; typedef struct { char *source; char *target; char *options; uint32_t flags; // bitmask of pe__bundle_mount_flags } pe__bundle_mount_t; typedef struct { char *source; char *target; } pe__bundle_port_t; enum pe__container_agent { PE__CONTAINER_AGENT_UNKNOWN, PE__CONTAINER_AGENT_DOCKER, PE__CONTAINER_AGENT_RKT, PE__CONTAINER_AGENT_PODMAN, }; #define PE__CONTAINER_AGENT_UNKNOWN_S "unknown" #define PE__CONTAINER_AGENT_DOCKER_S "docker" #define PE__CONTAINER_AGENT_RKT_S "rkt" #define PE__CONTAINER_AGENT_PODMAN_S "podman" typedef struct pe__bundle_variant_data_s { int promoted_max; int nreplicas; int nreplicas_per_host; char *prefix; char *image; const char *ip_last; char *host_network; char *host_netmask; char *control_port; char *container_network; char *ip_range_start; gboolean add_host; gchar *container_host_options; char *container_command; char *launcher_options; const char *attribute_target; pcmk_resource_t *child; GList *replicas; // pcmk__bundle_replica_t * GList *ports; // pe__bundle_port_t * GList *mounts; // pe__bundle_mount_t * enum pe__container_agent agent_type; } pe__bundle_variant_data_t; #define get_bundle_variant_data(data, rsc) \ pcmk__assert(pcmk__is_bundle(rsc) && (rsc->variant_opaque != NULL)); \ data = (pe__bundle_variant_data_t *) rsc->variant_opaque; /*! * \internal * \brief Get maximum number of bundle replicas allowed to run * * \param[in] rsc Bundle or bundled resource to check * * \return Maximum replicas for bundle corresponding to \p rsc */ int pe__bundle_max(const pcmk_resource_t *rsc) { const pe__bundle_variant_data_t *bundle_data = NULL; get_bundle_variant_data(bundle_data, pe__const_top_resource(rsc, true)); return bundle_data->nreplicas; } /*! * \internal * \brief Get the resource inside a bundle * * \param[in] bundle Bundle to check * * \return Resource inside \p bundle if any, otherwise NULL */ pcmk_resource_t * pe__bundled_resource(const pcmk_resource_t *rsc) { const pe__bundle_variant_data_t *bundle_data = NULL; get_bundle_variant_data(bundle_data, pe__const_top_resource(rsc, true)); return bundle_data->child; } /*! * \internal * \brief Get containerized resource corresponding to a given bundle container * * \param[in] instance Collective instance that might be a bundle container * * \return Bundled resource instance inside \p instance if it is a bundle * container instance, otherwise NULL */ const pcmk_resource_t * pe__get_rsc_in_container(const pcmk_resource_t *instance) { const pe__bundle_variant_data_t *data = NULL; const pcmk_resource_t *top = pe__const_top_resource(instance, true); if (!pcmk__is_bundle(top)) { return NULL; } get_bundle_variant_data(data, top); for (const GList *iter = data->replicas; iter != NULL; iter = iter->next) { const pcmk__bundle_replica_t *replica = iter->data; if (instance == replica->container) { return replica->child; } } return NULL; } /*! * \internal * \brief Check whether a given node is created by a bundle * * \param[in] bundle Bundle resource to check * \param[in] node Node to check * * \return true if \p node is an instance of \p bundle, otherwise false */ bool pe__node_is_bundle_instance(const pcmk_resource_t *bundle, const pcmk_node_t *node) { pe__bundle_variant_data_t *bundle_data = NULL; get_bundle_variant_data(bundle_data, bundle); for (GList *iter = bundle_data->replicas; iter != NULL; iter = iter->next) { pcmk__bundle_replica_t *replica = iter->data; if (pcmk__same_node(node, replica->node)) { return true; } } return false; } /*! * \internal * \brief Get the container of a bundle's first replica * * \param[in] bundle Bundle resource to get container for * * \return Container resource from first replica of \p bundle if any, * otherwise NULL */ pcmk_resource_t * pe__first_container(const pcmk_resource_t *bundle) { const pe__bundle_variant_data_t *bundle_data = NULL; const pcmk__bundle_replica_t *replica = NULL; get_bundle_variant_data(bundle_data, bundle); if (bundle_data->replicas == NULL) { return NULL; } replica = bundle_data->replicas->data; return replica->container; } /*! * \internal * \brief Iterate over bundle replicas * * \param[in,out] bundle Bundle to iterate over * \param[in] fn Function to call for each replica (its return value * indicates whether to continue iterating) * \param[in,out] user_data Pointer to pass to \p fn */ void pe__foreach_bundle_replica(pcmk_resource_t *bundle, bool (*fn)(pcmk__bundle_replica_t *, void *), void *user_data) { const pe__bundle_variant_data_t *bundle_data = NULL; get_bundle_variant_data(bundle_data, bundle); for (GList *iter = bundle_data->replicas; iter != NULL; iter = iter->next) { if (!fn((pcmk__bundle_replica_t *) iter->data, user_data)) { break; } } } /*! * \internal * \brief Iterate over const bundle replicas * * \param[in] bundle Bundle to iterate over * \param[in] fn Function to call for each replica (its return value * indicates whether to continue iterating) * \param[in,out] user_data Pointer to pass to \p fn */ void pe__foreach_const_bundle_replica(const pcmk_resource_t *bundle, bool (*fn)(const pcmk__bundle_replica_t *, void *), void *user_data) { const pe__bundle_variant_data_t *bundle_data = NULL; get_bundle_variant_data(bundle_data, bundle); for (const GList *iter = bundle_data->replicas; iter != NULL; iter = iter->next) { if (!fn((const pcmk__bundle_replica_t *) iter->data, user_data)) { break; } } } static char * next_ip(const char *last_ip) { unsigned int oct1 = 0; unsigned int oct2 = 0; unsigned int oct3 = 0; unsigned int oct4 = 0; int rc = sscanf(last_ip, "%u.%u.%u.%u", &oct1, &oct2, &oct3, &oct4); if (rc != 4) { /*@ TODO check for IPv6 */ return NULL; } else if (oct3 > 253) { return NULL; } else if (oct4 > 253) { ++oct3; oct4 = 1; } else { ++oct4; } return crm_strdup_printf("%u.%u.%u.%u", oct1, oct2, oct3, oct4); } static void allocate_ip(pe__bundle_variant_data_t *data, pcmk__bundle_replica_t *replica, GString *buffer) { if(data->ip_range_start == NULL) { return; } else if(data->ip_last) { replica->ipaddr = next_ip(data->ip_last); } else { replica->ipaddr = strdup(data->ip_range_start); } data->ip_last = replica->ipaddr; switch (data->agent_type) { case PE__CONTAINER_AGENT_DOCKER: case PE__CONTAINER_AGENT_PODMAN: if (data->add_host) { g_string_append_printf(buffer, " --add-host=%s-%d:%s", data->prefix, replica->offset, replica->ipaddr); } else { g_string_append_printf(buffer, " --hosts-entry=%s=%s-%d", replica->ipaddr, data->prefix, replica->offset); } break; case PE__CONTAINER_AGENT_RKT: g_string_append_printf(buffer, " --hosts-entry=%s=%s-%d", replica->ipaddr, data->prefix, replica->offset); break; default: // PE__CONTAINER_AGENT_UNKNOWN break; } } static xmlNode * create_resource(const char *name, const char *provider, const char *kind) { xmlNode *rsc = pcmk__xe_create(NULL, PCMK_XE_PRIMITIVE); crm_xml_add(rsc, PCMK_XA_ID, name); crm_xml_add(rsc, PCMK_XA_CLASS, PCMK_RESOURCE_CLASS_OCF); crm_xml_add(rsc, PCMK_XA_PROVIDER, provider); crm_xml_add(rsc, PCMK_XA_TYPE, kind); return rsc; } /*! * \internal * \brief Check whether cluster can manage resource inside container * * \param[in,out] data Container variant data * * \return TRUE if networking configuration is acceptable, FALSE otherwise * * \note The resource is manageable if an IP range or control port has been * specified. If a control port is used without an IP range, replicas per * host must be 1. */ static bool valid_network(pe__bundle_variant_data_t *data) { if(data->ip_range_start) { return TRUE; } if(data->control_port) { if(data->nreplicas_per_host > 1) { pcmk__config_err("Specifying the '" PCMK_XA_CONTROL_PORT "' for %s " "requires '" PCMK_XA_REPLICAS_PER_HOST "=1'", data->prefix); data->nreplicas_per_host = 1; // @TODO to be sure: // pcmk__clear_rsc_flags(rsc, pcmk_rsc_unique); } return TRUE; } return FALSE; } static int create_ip_resource(pcmk_resource_t *parent, pe__bundle_variant_data_t *data, pcmk__bundle_replica_t *replica) { if(data->ip_range_start) { char *id = NULL; xmlNode *xml_ip = NULL; xmlNode *xml_obj = NULL; id = crm_strdup_printf("%s-ip-%s", data->prefix, replica->ipaddr); crm_xml_sanitize_id(id); xml_ip = create_resource(id, "heartbeat", "IPaddr2"); free(id); xml_obj = pcmk__xe_create(xml_ip, PCMK_XE_INSTANCE_ATTRIBUTES); crm_xml_set_id(xml_obj, "%s-attributes-%d", data->prefix, replica->offset); crm_create_nvpair_xml(xml_obj, NULL, "ip", replica->ipaddr); if(data->host_network) { crm_create_nvpair_xml(xml_obj, NULL, "nic", data->host_network); } if(data->host_netmask) { crm_create_nvpair_xml(xml_obj, NULL, "cidr_netmask", data->host_netmask); } else { crm_create_nvpair_xml(xml_obj, NULL, "cidr_netmask", "32"); } xml_obj = pcmk__xe_create(xml_ip, PCMK_XE_OPERATIONS); crm_create_op_xml(xml_obj, pcmk__xe_id(xml_ip), PCMK_ACTION_MONITOR, "60s", NULL); // TODO: Other ops? Timeouts and intervals from underlying resource? if (pe__unpack_resource(xml_ip, &replica->ip, parent, parent->cluster) != pcmk_rc_ok) { return pcmk_rc_unpack_error; } parent->children = g_list_append(parent->children, replica->ip); } return pcmk_rc_ok; } static const char* container_agent_str(enum pe__container_agent t) { switch (t) { case PE__CONTAINER_AGENT_DOCKER: return PE__CONTAINER_AGENT_DOCKER_S; case PE__CONTAINER_AGENT_RKT: return PE__CONTAINER_AGENT_RKT_S; case PE__CONTAINER_AGENT_PODMAN: return PE__CONTAINER_AGENT_PODMAN_S; default: // PE__CONTAINER_AGENT_UNKNOWN break; } return PE__CONTAINER_AGENT_UNKNOWN_S; } static int create_container_resource(pcmk_resource_t *parent, const pe__bundle_variant_data_t *data, pcmk__bundle_replica_t *replica) { char *id = NULL; xmlNode *xml_container = NULL; xmlNode *xml_obj = NULL; // Agent-specific const char *hostname_opt = NULL; const char *env_opt = NULL; const char *agent_str = NULL; int volid = 0; // rkt-only GString *buffer = NULL; GString *dbuffer = NULL; // Where syntax differences are drop-in replacements, set them now switch (data->agent_type) { case PE__CONTAINER_AGENT_DOCKER: case PE__CONTAINER_AGENT_PODMAN: hostname_opt = "-h "; env_opt = "-e "; break; case PE__CONTAINER_AGENT_RKT: hostname_opt = "--hostname="; env_opt = "--environment="; break; default: // PE__CONTAINER_AGENT_UNKNOWN return pcmk_rc_unpack_error; } agent_str = container_agent_str(data->agent_type); buffer = g_string_sized_new(4096); id = crm_strdup_printf("%s-%s-%d", data->prefix, agent_str, replica->offset); crm_xml_sanitize_id(id); xml_container = create_resource(id, "heartbeat", agent_str); free(id); xml_obj = pcmk__xe_create(xml_container, PCMK_XE_INSTANCE_ATTRIBUTES); crm_xml_set_id(xml_obj, "%s-attributes-%d", data->prefix, replica->offset); crm_create_nvpair_xml(xml_obj, NULL, "image", data->image); crm_create_nvpair_xml(xml_obj, NULL, "allow_pull", PCMK_VALUE_TRUE); crm_create_nvpair_xml(xml_obj, NULL, "force_kill", PCMK_VALUE_FALSE); crm_create_nvpair_xml(xml_obj, NULL, "reuse", PCMK_VALUE_FALSE); if (data->agent_type == PE__CONTAINER_AGENT_DOCKER) { g_string_append(buffer, " --restart=no"); } /* Set a container hostname only if we have an IP to map it to. The user can * set -h or --uts=host themselves if they want a nicer name for logs, but * this makes applications happy who need their hostname to match the IP * they bind to. */ if (data->ip_range_start != NULL) { g_string_append_printf(buffer, " %s%s-%d", hostname_opt, data->prefix, replica->offset); } pcmk__g_strcat(buffer, " ", env_opt, "PCMK_stderr=1", NULL); if (data->container_network != NULL) { pcmk__g_strcat(buffer, " --net=", data->container_network, NULL); } if (data->control_port != NULL) { pcmk__g_strcat(buffer, " ", env_opt, "PCMK_" PCMK__ENV_REMOTE_PORT "=", data->control_port, NULL); } else { g_string_append_printf(buffer, " %sPCMK_" PCMK__ENV_REMOTE_PORT "=%d", env_opt, DEFAULT_REMOTE_PORT); } for (GList *iter = data->mounts; iter != NULL; iter = iter->next) { pe__bundle_mount_t *mount = (pe__bundle_mount_t *) iter->data; char *source = NULL; if (pcmk_is_set(mount->flags, pe__bundle_mount_subdir)) { source = crm_strdup_printf("%s/%s-%d", mount->source, data->prefix, replica->offset); pcmk__add_separated_word(&dbuffer, 1024, source, ","); } switch (data->agent_type) { case PE__CONTAINER_AGENT_DOCKER: case PE__CONTAINER_AGENT_PODMAN: pcmk__g_strcat(buffer, " -v ", pcmk__s(source, mount->source), ":", mount->target, NULL); if (mount->options != NULL) { pcmk__g_strcat(buffer, ":", mount->options, NULL); } break; case PE__CONTAINER_AGENT_RKT: g_string_append_printf(buffer, " --volume vol%d,kind=host," "source=%s%s%s " "--mount volume=vol%d,target=%s", volid, pcmk__s(source, mount->source), (mount->options != NULL)? "," : "", pcmk__s(mount->options, ""), volid, mount->target); volid++; break; default: break; } free(source); } for (GList *iter = data->ports; iter != NULL; iter = iter->next) { pe__bundle_port_t *port = (pe__bundle_port_t *) iter->data; switch (data->agent_type) { case PE__CONTAINER_AGENT_DOCKER: case PE__CONTAINER_AGENT_PODMAN: if (replica->ipaddr != NULL) { pcmk__g_strcat(buffer, " -p ", replica->ipaddr, ":", port->source, ":", port->target, NULL); } else if (!pcmk__str_eq(data->container_network, PCMK_VALUE_HOST, pcmk__str_none)) { // No need to do port mapping if net == host pcmk__g_strcat(buffer, " -p ", port->source, ":", port->target, NULL); } break; case PE__CONTAINER_AGENT_RKT: if (replica->ipaddr != NULL) { pcmk__g_strcat(buffer, " --port=", port->target, ":", replica->ipaddr, ":", port->source, NULL); } else { pcmk__g_strcat(buffer, " --port=", port->target, ":", port->source, NULL); } break; default: break; } } /* @COMPAT: We should use pcmk__add_word() here, but we can't yet, because * it would cause restarts during rolling upgrades. * * In a previous version of the container resource creation logic, if * data->launcher_options is not NULL, we append * (" %s", data->launcher_options) even if data->launcher_options is an * empty string. Likewise for data->container_host_options. Using * * pcmk__add_word(buffer, 0, data->launcher_options) * * removes that extra trailing space, causing a resource definition change. */ if (data->launcher_options != NULL) { pcmk__g_strcat(buffer, " ", data->launcher_options, NULL); } if (data->container_host_options != NULL) { pcmk__g_strcat(buffer, " ", data->container_host_options, NULL); } crm_create_nvpair_xml(xml_obj, NULL, "run_opts", (const char *) buffer->str); g_string_free(buffer, TRUE); crm_create_nvpair_xml(xml_obj, NULL, "mount_points", (dbuffer != NULL)? (const char *) dbuffer->str : ""); if (dbuffer != NULL) { g_string_free(dbuffer, TRUE); } if (replica->child != NULL) { if (data->container_command != NULL) { crm_create_nvpair_xml(xml_obj, NULL, "run_cmd", data->container_command); } else { crm_create_nvpair_xml(xml_obj, NULL, "run_cmd", SBIN_DIR "/pacemaker-remoted"); } /* TODO: Allow users to specify their own? * * We just want to know if the container is alive; we'll monitor the * child independently. */ crm_create_nvpair_xml(xml_obj, NULL, "monitor_cmd", "/bin/true"); #if 0 /* @TODO Consider supporting the use case where we can start and stop * resources, but not proxy local commands (such as setting node * attributes), by running the local executor in stand-alone mode. * However, this would probably be better done via ACLs as with other * Pacemaker Remote nodes. */ } else if ((child != NULL) && data->untrusted) { crm_create_nvpair_xml(xml_obj, NULL, "run_cmd", CRM_DAEMON_DIR "/pacemaker-execd"); crm_create_nvpair_xml(xml_obj, NULL, "monitor_cmd", CRM_DAEMON_DIR "/pacemaker/cts-exec-helper -c poke"); #endif } else { if (data->container_command != NULL) { crm_create_nvpair_xml(xml_obj, NULL, "run_cmd", data->container_command); } /* TODO: Allow users to specify their own? * * We don't know what's in the container, so we just want to know if it * is alive. */ crm_create_nvpair_xml(xml_obj, NULL, "monitor_cmd", "/bin/true"); } xml_obj = pcmk__xe_create(xml_container, PCMK_XE_OPERATIONS); crm_create_op_xml(xml_obj, pcmk__xe_id(xml_container), PCMK_ACTION_MONITOR, "60s", NULL); // TODO: Other ops? Timeouts and intervals from underlying resource? if (pe__unpack_resource(xml_container, &replica->container, parent, parent->cluster) != pcmk_rc_ok) { return pcmk_rc_unpack_error; } pcmk__set_rsc_flags(replica->container, pcmk_rsc_replica_container); parent->children = g_list_append(parent->children, replica->container); return pcmk_rc_ok; } /*! * \brief Ban a node from a resource's (and its children's) allowed nodes list * * \param[in,out] rsc Resource to modify * \param[in] uname Name of node to ban */ static void disallow_node(pcmk_resource_t *rsc, const char *uname) { gpointer match = g_hash_table_lookup(rsc->allowed_nodes, uname); if (match) { ((pcmk_node_t *) match)->weight = -PCMK_SCORE_INFINITY; ((pcmk_node_t *) match)->rsc_discover_mode = pcmk_probe_never; } if (rsc->children) { g_list_foreach(rsc->children, (GFunc) disallow_node, (gpointer) uname); } } static int create_remote_resource(pcmk_resource_t *parent, pe__bundle_variant_data_t *data, pcmk__bundle_replica_t *replica) { if (replica->child && valid_network(data)) { GHashTableIter gIter; pcmk_node_t *node = NULL; xmlNode *xml_remote = NULL; char *id = crm_strdup_printf("%s-%d", data->prefix, replica->offset); char *port_s = NULL; const char *uname = NULL; const char *connect_name = NULL; if (pe_find_resource(parent->cluster->resources, id) != NULL) { free(id); // The biggest hammer we have id = crm_strdup_printf("pcmk-internal-%s-remote-%d", replica->child->id, replica->offset); //@TODO return error instead of asserting? pcmk__assert(pe_find_resource(parent->cluster->resources, id) == NULL); } /* REMOTE_CONTAINER_HACK: Using "#uname" as the server name when the * connection does not have its own IP is a magic string that we use to * support nested remotes (i.e. a bundle running on a remote node). */ connect_name = (replica->ipaddr? replica->ipaddr : "#uname"); if (data->control_port == NULL) { port_s = pcmk__itoa(DEFAULT_REMOTE_PORT); } /* This sets replica->container as replica->remote's container, which is * similar to what happens with guest nodes. This is how the scheduler * knows that the bundle node is fenced by recovering the container, and * that remote should be ordered relative to the container. */ xml_remote = pe_create_remote_xml(NULL, id, replica->container->id, NULL, NULL, NULL, connect_name, (data->control_port? data->control_port : port_s)); free(port_s); /* Abandon our created ID, and pull the copy from the XML, because we * need something that will get freed during scheduler data cleanup to * use as the node ID and uname. */ free(id); id = NULL; uname = pcmk__xe_id(xml_remote); /* Ensure a node has been created for the guest (it may have already * been, if it has a permanent node attribute), and ensure its weight is * -INFINITY so no other resources can run on it. */ node = pcmk_find_node(parent->cluster, uname); if (node == NULL) { node = pe_create_node(uname, uname, PCMK_VALUE_REMOTE, -PCMK_SCORE_INFINITY, parent->cluster); } else { node->weight = -PCMK_SCORE_INFINITY; } node->rsc_discover_mode = pcmk_probe_never; /* unpack_remote_nodes() ensures that each remote node and guest node * has a pcmk_node_t entry. Ideally, it would do the same for bundle * nodes. Unfortunately, a bundle has to be mostly unpacked before it's * obvious what nodes will be needed, so we do it just above. * * Worse, that means that the node may have been utilized while * unpacking other resources, without our weight correction. The most * likely place for this to happen is when pe__unpack_resource() calls * resource_location() to set a default score in symmetric clusters. * This adds a node *copy* to each resource's allowed nodes, and these * copies will have the wrong weight. * * As a hacky workaround, fix those copies here. * * @TODO Possible alternative: ensure bundles are unpacked before other * resources, so the weight is correct before any copies are made. */ g_list_foreach(parent->cluster->resources, (GFunc) disallow_node, (gpointer) uname); replica->node = pe__copy_node(node); replica->node->weight = 500; replica->node->rsc_discover_mode = pcmk_probe_exclusive; /* Ensure the node shows up as allowed and with the correct discovery set */ if (replica->child->allowed_nodes != NULL) { g_hash_table_destroy(replica->child->allowed_nodes); } replica->child->allowed_nodes = pcmk__strkey_table(NULL, free); g_hash_table_insert(replica->child->allowed_nodes, (gpointer) replica->node->details->id, pe__copy_node(replica->node)); { pcmk_node_t *copy = pe__copy_node(replica->node); copy->weight = -PCMK_SCORE_INFINITY; g_hash_table_insert(replica->child->parent->allowed_nodes, (gpointer) replica->node->details->id, copy); } if (pe__unpack_resource(xml_remote, &replica->remote, parent, parent->cluster) != pcmk_rc_ok) { return pcmk_rc_unpack_error; } g_hash_table_iter_init(&gIter, replica->remote->allowed_nodes); while (g_hash_table_iter_next(&gIter, NULL, (void **)&node)) { if (pcmk__is_pacemaker_remote_node(node)) { /* Remote resources can only run on 'normal' cluster node */ node->weight = -PCMK_SCORE_INFINITY; } } replica->node->details->remote_rsc = replica->remote; // Ensure pcmk__is_guest_or_bundle_node() functions correctly replica->remote->container = replica->container; /* A bundle's #kind is closer to "container" (guest node) than the * "remote" set by pe_create_node(). */ pcmk__insert_dup(replica->node->details->attrs, CRM_ATTR_KIND, "container"); /* One effect of this is that setup_container() will add * replica->remote to replica->container's fillers, which will make * pe__resource_contains_guest_node() true for replica->container. * * replica->child does NOT get added to replica->container's fillers. * The only noticeable effect if it did would be for its fail count to * be taken into account when checking replica->container's migration * threshold. */ parent->children = g_list_append(parent->children, replica->remote); } return pcmk_rc_ok; } static int create_replica_resources(pcmk_resource_t *parent, pe__bundle_variant_data_t *data, pcmk__bundle_replica_t *replica) { int rc = pcmk_rc_ok; rc = create_container_resource(parent, data, replica); if (rc != pcmk_rc_ok) { return rc; } rc = create_ip_resource(parent, data, replica); if (rc != pcmk_rc_ok) { return rc; } rc = create_remote_resource(parent, data, replica); if (rc != pcmk_rc_ok) { return rc; } if ((replica->child != NULL) && (replica->ipaddr != NULL)) { pcmk__insert_meta(replica->child, "external-ip", replica->ipaddr); } if (replica->remote != NULL) { /* * Allow the remote connection resource to be allocated to a * different node than the one on which the container is active. * * This makes it possible to have Pacemaker Remote nodes running * containers with pacemaker-remoted inside in order to start * services inside those containers. */ pcmk__set_rsc_flags(replica->remote, pcmk_rsc_remote_nesting_allowed); } return rc; } static void mount_add(pe__bundle_variant_data_t *bundle_data, const char *source, const char *target, const char *options, uint32_t flags) { pe__bundle_mount_t *mount = pcmk__assert_alloc(1, sizeof(pe__bundle_mount_t)); mount->source = pcmk__str_copy(source); mount->target = pcmk__str_copy(target); mount->options = pcmk__str_copy(options); mount->flags = flags; bundle_data->mounts = g_list_append(bundle_data->mounts, mount); } static void mount_free(pe__bundle_mount_t *mount) { free(mount->source); free(mount->target); free(mount->options); free(mount); } static void port_free(pe__bundle_port_t *port) { free(port->source); free(port->target); free(port); } static pcmk__bundle_replica_t * replica_for_remote(pcmk_resource_t *remote) { pcmk_resource_t *top = remote; pe__bundle_variant_data_t *bundle_data = NULL; if (top == NULL) { return NULL; } while (top->parent != NULL) { top = top->parent; } get_bundle_variant_data(bundle_data, top); for (GList *gIter = bundle_data->replicas; gIter != NULL; gIter = gIter->next) { pcmk__bundle_replica_t *replica = gIter->data; if (replica->remote == remote) { return replica; } } CRM_LOG_ASSERT(FALSE); return NULL; } bool pe__bundle_needs_remote_name(pcmk_resource_t *rsc) { const char *value; GHashTable *params = NULL; if (rsc == NULL) { return false; } // Use NULL node since pcmk__bundle_expand() uses that to set value params = pe_rsc_params(rsc, NULL, rsc->cluster); value = g_hash_table_lookup(params, PCMK_REMOTE_RA_ADDR); return pcmk__str_eq(value, "#uname", pcmk__str_casei) && xml_contains_remote_node(rsc->xml); } const char * pe__add_bundle_remote_name(pcmk_resource_t *rsc, xmlNode *xml, const char *field) { // REMOTE_CONTAINER_HACK: Allow remote nodes that start containers with pacemaker remote inside pcmk_node_t *node = NULL; pcmk__bundle_replica_t *replica = NULL; if (!pe__bundle_needs_remote_name(rsc)) { return NULL; } replica = replica_for_remote(rsc); if (replica == NULL) { return NULL; } node = replica->container->allocated_to; if (node == NULL) { /* If it won't be running anywhere after the * transition, go with where it's running now. */ node = pcmk__current_node(replica->container); } if(node == NULL) { crm_trace("Cannot determine address for bundle connection %s", rsc->id); return NULL; } crm_trace("Setting address for bundle connection %s to bundle host %s", rsc->id, pcmk__node_name(node)); if(xml != NULL && field != NULL) { crm_xml_add(xml, field, node->details->uname); } return node->details->uname; } #define pe__set_bundle_mount_flags(mount_xml, flags, flags_to_set) do { \ flags = pcmk__set_flags_as(__func__, __LINE__, LOG_TRACE, \ "Bundle mount", pcmk__xe_id(mount_xml), \ flags, (flags_to_set), #flags_to_set); \ } while (0) gboolean pe__unpack_bundle(pcmk_resource_t *rsc, pcmk_scheduler_t *scheduler) { const char *value = NULL; xmlNode *xml_obj = NULL; const xmlNode *xml_child = NULL; xmlNode *xml_resource = NULL; pe__bundle_variant_data_t *bundle_data = NULL; bool need_log_mount = TRUE; pcmk__assert(rsc != NULL); pcmk__rsc_trace(rsc, "Processing resource %s...", rsc->id); bundle_data = pcmk__assert_alloc(1, sizeof(pe__bundle_variant_data_t)); rsc->variant_opaque = bundle_data; bundle_data->prefix = strdup(rsc->id); xml_obj = pcmk__xe_first_child(rsc->xml, PCMK_XE_DOCKER, NULL, NULL); if (xml_obj != NULL) { bundle_data->agent_type = PE__CONTAINER_AGENT_DOCKER; } else { xml_obj = pcmk__xe_first_child(rsc->xml, PCMK__XE_RKT, NULL, NULL); if (xml_obj != NULL) { pcmk__warn_once(pcmk__wo_rkt, "Support for " PCMK__XE_RKT " in bundles " "(such as %s) is deprecated and will be " "removed in a future release", rsc->id); bundle_data->agent_type = PE__CONTAINER_AGENT_RKT; } else { xml_obj = pcmk__xe_first_child(rsc->xml, PCMK_XE_PODMAN, NULL, NULL); if (xml_obj != NULL) { bundle_data->agent_type = PE__CONTAINER_AGENT_PODMAN; } else { return FALSE; } } } // Use 0 for default, minimum, and invalid PCMK_XA_PROMOTED_MAX value = crm_element_value(xml_obj, PCMK_XA_PROMOTED_MAX); if (value == NULL) { // @COMPAT deprecated since 2.0.0 value = crm_element_value(xml_obj, PCMK__XA_PROMOTED_MAX_LEGACY); if (value != NULL) { pcmk__warn_once(pcmk__wo_bundle_master, "Support for the " PCMK__XA_PROMOTED_MAX_LEGACY " attribute (such as in %s) is deprecated and " "will be removed in a future release. Use " PCMK_XA_PROMOTED_MAX " instead.", rsc->id); } } pcmk__scan_min_int(value, &bundle_data->promoted_max, 0); /* Default replicas to PCMK_XA_PROMOTED_MAX if it was specified and 1 * otherwise */ value = crm_element_value(xml_obj, PCMK_XA_REPLICAS); if ((value == NULL) && (bundle_data->promoted_max > 0)) { bundle_data->nreplicas = bundle_data->promoted_max; } else { pcmk__scan_min_int(value, &bundle_data->nreplicas, 1); } /* * Communication between containers on the same host via the * floating IPs only works if the container is started with: * --userland-proxy=false --ip-masq=false */ value = crm_element_value(xml_obj, PCMK_XA_REPLICAS_PER_HOST); pcmk__scan_min_int(value, &bundle_data->nreplicas_per_host, 1); if (bundle_data->nreplicas_per_host == 1) { pcmk__clear_rsc_flags(rsc, pcmk_rsc_unique); } bundle_data->container_command = crm_element_value_copy(xml_obj, PCMK_XA_RUN_COMMAND); bundle_data->launcher_options = crm_element_value_copy(xml_obj, PCMK_XA_OPTIONS); bundle_data->image = crm_element_value_copy(xml_obj, PCMK_XA_IMAGE); bundle_data->container_network = crm_element_value_copy(xml_obj, PCMK_XA_NETWORK); xml_obj = pcmk__xe_first_child(rsc->xml, PCMK_XE_NETWORK, NULL, NULL); if(xml_obj) { bundle_data->ip_range_start = crm_element_value_copy(xml_obj, PCMK_XA_IP_RANGE_START); bundle_data->host_netmask = crm_element_value_copy(xml_obj, PCMK_XA_HOST_NETMASK); bundle_data->host_network = crm_element_value_copy(xml_obj, PCMK_XA_HOST_INTERFACE); bundle_data->control_port = crm_element_value_copy(xml_obj, PCMK_XA_CONTROL_PORT); value = crm_element_value(xml_obj, PCMK_XA_ADD_HOST); if (crm_str_to_boolean(value, &bundle_data->add_host) != 1) { bundle_data->add_host = TRUE; } for (xml_child = pcmk__xe_first_child(xml_obj, PCMK_XE_PORT_MAPPING, NULL, NULL); xml_child != NULL; xml_child = pcmk__xe_next_same(xml_child)) { pe__bundle_port_t *port = pcmk__assert_alloc(1, sizeof(pe__bundle_port_t)); port->source = crm_element_value_copy(xml_child, PCMK_XA_PORT); if(port->source == NULL) { port->source = crm_element_value_copy(xml_child, PCMK_XA_RANGE); } else { port->target = crm_element_value_copy(xml_child, PCMK_XA_INTERNAL_PORT); } if(port->source != NULL && strlen(port->source) > 0) { if(port->target == NULL) { port->target = strdup(port->source); } bundle_data->ports = g_list_append(bundle_data->ports, port); } else { pcmk__config_err("Invalid " PCMK_XA_PORT " directive %s", pcmk__xe_id(xml_child)); port_free(port); } } } xml_obj = pcmk__xe_first_child(rsc->xml, PCMK_XE_STORAGE, NULL, NULL); for (xml_child = pcmk__xe_first_child(xml_obj, PCMK_XE_STORAGE_MAPPING, NULL, NULL); xml_child != NULL; xml_child = pcmk__xe_next_same(xml_child)) { const char *source = crm_element_value(xml_child, PCMK_XA_SOURCE_DIR); const char *target = crm_element_value(xml_child, PCMK_XA_TARGET_DIR); const char *options = crm_element_value(xml_child, PCMK_XA_OPTIONS); int flags = pe__bundle_mount_none; if (source == NULL) { source = crm_element_value(xml_child, PCMK_XA_SOURCE_DIR_ROOT); pe__set_bundle_mount_flags(xml_child, flags, pe__bundle_mount_subdir); } if (source && target) { mount_add(bundle_data, source, target, options, flags); if (strcmp(target, "/var/log") == 0) { need_log_mount = FALSE; } } else { pcmk__config_err("Invalid mount directive %s", pcmk__xe_id(xml_child)); } } xml_obj = pcmk__xe_first_child(rsc->xml, PCMK_XE_PRIMITIVE, NULL, NULL); if (xml_obj && valid_network(bundle_data)) { char *value = NULL; xmlNode *xml_set = NULL; xml_resource = pcmk__xe_create(NULL, PCMK_XE_CLONE); /* @COMPAT We no longer use the tag, but we need to keep it as * part of the resource name, so that bundles don't restart in a rolling * upgrade. (It also avoids needing to change regression tests.) */ crm_xml_set_id(xml_resource, "%s-%s", bundle_data->prefix, (bundle_data->promoted_max? "master" : (const char *)xml_resource->name)); xml_set = pcmk__xe_create(xml_resource, PCMK_XE_META_ATTRIBUTES); crm_xml_set_id(xml_set, "%s-%s-meta", bundle_data->prefix, xml_resource->name); crm_create_nvpair_xml(xml_set, NULL, PCMK_META_ORDERED, PCMK_VALUE_TRUE); value = pcmk__itoa(bundle_data->nreplicas); crm_create_nvpair_xml(xml_set, NULL, PCMK_META_CLONE_MAX, value); free(value); value = pcmk__itoa(bundle_data->nreplicas_per_host); crm_create_nvpair_xml(xml_set, NULL, PCMK_META_CLONE_NODE_MAX, value); free(value); crm_create_nvpair_xml(xml_set, NULL, PCMK_META_GLOBALLY_UNIQUE, pcmk__btoa(bundle_data->nreplicas_per_host > 1)); if (bundle_data->promoted_max) { crm_create_nvpair_xml(xml_set, NULL, PCMK_META_PROMOTABLE, PCMK_VALUE_TRUE); value = pcmk__itoa(bundle_data->promoted_max); crm_create_nvpair_xml(xml_set, NULL, PCMK_META_PROMOTED_MAX, value); free(value); } //crm_xml_add(xml_obj, PCMK_XA_ID, bundle_data->prefix); pcmk__xml_copy(xml_resource, xml_obj); } else if(xml_obj) { pcmk__config_err("Cannot control %s inside %s without either " PCMK_XA_IP_RANGE_START " or " PCMK_XA_CONTROL_PORT, rsc->id, pcmk__xe_id(xml_obj)); return FALSE; } if(xml_resource) { int lpc = 0; GList *childIter = NULL; pe__bundle_port_t *port = NULL; GString *buffer = NULL; if (pe__unpack_resource(xml_resource, &(bundle_data->child), rsc, scheduler) != pcmk_rc_ok) { return FALSE; } /* Currently, we always map the default authentication key location * into the same location inside the container. * * Ideally, we would respect the host's PCMK_authkey_location, but: * - it may be different on different nodes; * - the actual connection will do extra checking to make sure the key * file exists and is readable, that we can't do here on the DC * - tools such as crm_resource and crm_simulate may not have the same * environment variables as the cluster, causing operation digests to * differ * * Always using the default location inside the container is fine, * because we control the pacemaker_remote environment, and it avoids * having to pass another environment variable to the container. * * @TODO A better solution may be to have only pacemaker_remote use the * environment variable, and have the cluster nodes use a new * cluster option for key location. This would introduce the limitation * of the location being the same on all cluster nodes, but that's * reasonable. */ mount_add(bundle_data, DEFAULT_REMOTE_KEY_LOCATION, DEFAULT_REMOTE_KEY_LOCATION, NULL, pe__bundle_mount_none); if (need_log_mount) { mount_add(bundle_data, CRM_BUNDLE_DIR, "/var/log", NULL, pe__bundle_mount_subdir); } port = pcmk__assert_alloc(1, sizeof(pe__bundle_port_t)); if(bundle_data->control_port) { port->source = strdup(bundle_data->control_port); } else { /* If we wanted to respect PCMK_remote_port, we could use * crm_default_remote_port() here and elsewhere in this file instead * of DEFAULT_REMOTE_PORT. * * However, it gains nothing, since we control both the container * environment and the connection resource parameters, and the user * can use a different port if desired by setting * PCMK_XA_CONTROL_PORT. */ port->source = pcmk__itoa(DEFAULT_REMOTE_PORT); } port->target = strdup(port->source); bundle_data->ports = g_list_append(bundle_data->ports, port); buffer = g_string_sized_new(1024); for (childIter = bundle_data->child->children; childIter != NULL; childIter = childIter->next) { pcmk__bundle_replica_t *replica = NULL; replica = pcmk__assert_alloc(1, sizeof(pcmk__bundle_replica_t)); replica->child = childIter->data; replica->child->exclusive_discover = TRUE; replica->offset = lpc++; // Ensure the child's notify gets set based on the underlying primitive's value if (pcmk_is_set(replica->child->flags, pcmk_rsc_notify)) { pcmk__set_rsc_flags(bundle_data->child, pcmk_rsc_notify); } allocate_ip(bundle_data, replica, buffer); bundle_data->replicas = g_list_append(bundle_data->replicas, replica); + // coverity[null_field] replica->child can't be NULL here bundle_data->attribute_target = g_hash_table_lookup(replica->child->meta, PCMK_META_CONTAINER_ATTRIBUTE_TARGET); } bundle_data->container_host_options = g_string_free(buffer, FALSE); if (bundle_data->attribute_target) { pcmk__insert_dup(rsc->meta, PCMK_META_CONTAINER_ATTRIBUTE_TARGET, bundle_data->attribute_target); pcmk__insert_dup(bundle_data->child->meta, PCMK_META_CONTAINER_ATTRIBUTE_TARGET, bundle_data->attribute_target); } } else { // Just a naked container, no pacemaker-remote GString *buffer = g_string_sized_new(1024); for (int lpc = 0; lpc < bundle_data->nreplicas; lpc++) { pcmk__bundle_replica_t *replica = NULL; replica = pcmk__assert_alloc(1, sizeof(pcmk__bundle_replica_t)); replica->offset = lpc; allocate_ip(bundle_data, replica, buffer); bundle_data->replicas = g_list_append(bundle_data->replicas, replica); } bundle_data->container_host_options = g_string_free(buffer, FALSE); } for (GList *gIter = bundle_data->replicas; gIter != NULL; gIter = gIter->next) { pcmk__bundle_replica_t *replica = gIter->data; if (create_replica_resources(rsc, bundle_data, replica) != pcmk_rc_ok) { pcmk__config_err("Failed unpacking resource %s", rsc->id); rsc->fns->free(rsc); return FALSE; } /* Utilization needs special handling for bundles. It makes no sense for * the inner primitive to have utilization, because it is tied * one-to-one to the guest node created by the container resource -- and * there's no way to set capacities for that guest node anyway. * * What the user really wants is to configure utilization for the * container. However, the schema only allows utilization for * primitives, and the container resource is implicit anyway, so the * user can *only* configure utilization for the inner primitive. If * they do, move the primitive's utilization values to the container. * * @TODO This means that bundles without an inner primitive can't have * utilization. An alternative might be to allow utilization values in * the top-level bundle XML in the schema, and copy those to each * container. */ if (replica->child != NULL) { GHashTable *empty = replica->container->utilization; replica->container->utilization = replica->child->utilization; replica->child->utilization = empty; } } if (bundle_data->child) { rsc->children = g_list_append(rsc->children, bundle_data->child); } return TRUE; } static int replica_resource_active(pcmk_resource_t *rsc, gboolean all) { if (rsc) { gboolean child_active = rsc->fns->active(rsc, all); if (child_active && !all) { return TRUE; } else if (!child_active && all) { return FALSE; } } return -1; } gboolean pe__bundle_active(pcmk_resource_t *rsc, gboolean all) { pe__bundle_variant_data_t *bundle_data = NULL; GList *iter = NULL; get_bundle_variant_data(bundle_data, rsc); for (iter = bundle_data->replicas; iter != NULL; iter = iter->next) { pcmk__bundle_replica_t *replica = iter->data; int rsc_active; rsc_active = replica_resource_active(replica->ip, all); if (rsc_active >= 0) { return (gboolean) rsc_active; } rsc_active = replica_resource_active(replica->child, all); if (rsc_active >= 0) { return (gboolean) rsc_active; } rsc_active = replica_resource_active(replica->container, all); if (rsc_active >= 0) { return (gboolean) rsc_active; } rsc_active = replica_resource_active(replica->remote, all); if (rsc_active >= 0) { return (gboolean) rsc_active; } } /* If "all" is TRUE, we've already checked that no resources were inactive, * so return TRUE; if "all" is FALSE, we didn't find any active resources, * so return FALSE. */ return all; } /*! * \internal * \brief Find the bundle replica corresponding to a given node * * \param[in] bundle Top-level bundle resource * \param[in] node Node to search for * * \return Bundle replica if found, NULL otherwise */ pcmk_resource_t * pe__find_bundle_replica(const pcmk_resource_t *bundle, const pcmk_node_t *node) { pe__bundle_variant_data_t *bundle_data = NULL; pcmk__assert((bundle != NULL) && (node != NULL)); get_bundle_variant_data(bundle_data, bundle); for (GList *gIter = bundle_data->replicas; gIter != NULL; gIter = gIter->next) { pcmk__bundle_replica_t *replica = gIter->data; pcmk__assert((replica != NULL) && (replica->node != NULL)); if (pcmk__same_node(replica->node, node)) { return replica->child; } } return NULL; } /*! * \internal * \deprecated This function will be removed in a future release */ static void print_rsc_in_list(pcmk_resource_t *rsc, const char *pre_text, long options, void *print_data) { if (rsc != NULL) { if (options & pe_print_html) { status_print("
  • "); } rsc->fns->print(rsc, pre_text, options, print_data); if (options & pe_print_html) { status_print("
  • \n"); } } } /*! * \internal * \deprecated This function will be removed in a future release */ static void bundle_print_xml(pcmk_resource_t *rsc, const char *pre_text, long options, void *print_data) { pe__bundle_variant_data_t *bundle_data = NULL; char *child_text = NULL; CRM_CHECK(rsc != NULL, return); if (pre_text == NULL) { pre_text = ""; } child_text = crm_strdup_printf("%s ", pre_text); get_bundle_variant_data(bundle_data, rsc); status_print("%sid); status_print("type=\"%s\" ", container_agent_str(bundle_data->agent_type)); status_print("image=\"%s\" ", bundle_data->image); status_print("unique=\"%s\" ", pcmk__flag_text(rsc->flags, pcmk_rsc_unique)); status_print("managed=\"%s\" ", pcmk__flag_text(rsc->flags, pcmk_rsc_managed)); status_print("failed=\"%s\" ", pcmk__flag_text(rsc->flags, pcmk_rsc_failed)); status_print(">\n"); for (GList *gIter = bundle_data->replicas; gIter != NULL; gIter = gIter->next) { pcmk__bundle_replica_t *replica = gIter->data; pcmk__assert(replica); status_print("%s \n", pre_text, replica->offset); print_rsc_in_list(replica->ip, child_text, options, print_data); print_rsc_in_list(replica->child, child_text, options, print_data); print_rsc_in_list(replica->container, child_text, options, print_data); print_rsc_in_list(replica->remote, child_text, options, print_data); status_print("%s \n", pre_text); } status_print("%s\n", pre_text); free(child_text); } PCMK__OUTPUT_ARGS("bundle", "uint32_t", "pcmk_resource_t *", "GList *", "GList *") int pe__bundle_xml(pcmk__output_t *out, va_list args) { uint32_t show_opts = va_arg(args, uint32_t); pcmk_resource_t *rsc = va_arg(args, pcmk_resource_t *); GList *only_node = va_arg(args, GList *); GList *only_rsc = va_arg(args, GList *); pe__bundle_variant_data_t *bundle_data = NULL; int rc = pcmk_rc_no_output; gboolean printed_header = FALSE; gboolean print_everything = TRUE; const char *desc = NULL; pcmk__assert(rsc != NULL); get_bundle_variant_data(bundle_data, rsc); if (rsc->fns->is_filtered(rsc, only_rsc, TRUE)) { return rc; } print_everything = pcmk__str_in_list(rsc->id, only_rsc, pcmk__str_star_matches); for (GList *gIter = bundle_data->replicas; gIter != NULL; gIter = gIter->next) { pcmk__bundle_replica_t *replica = gIter->data; char *id = NULL; gboolean print_ip, print_child, print_ctnr, print_remote; pcmk__assert(replica); if (pcmk__rsc_filtered_by_node(replica->container, only_node)) { continue; } print_ip = replica->ip != NULL && !replica->ip->fns->is_filtered(replica->ip, only_rsc, print_everything); print_child = replica->child != NULL && !replica->child->fns->is_filtered(replica->child, only_rsc, print_everything); print_ctnr = !replica->container->fns->is_filtered(replica->container, only_rsc, print_everything); print_remote = replica->remote != NULL && !replica->remote->fns->is_filtered(replica->remote, only_rsc, print_everything); if (!print_everything && !print_ip && !print_child && !print_ctnr && !print_remote) { continue; } if (!printed_header) { const char *type = container_agent_str(bundle_data->agent_type); const char *unique = pcmk__flag_text(rsc->flags, pcmk_rsc_unique); const char *maintenance = pcmk__flag_text(rsc->flags, pcmk_rsc_maintenance); const char *managed = pcmk__flag_text(rsc->flags, pcmk_rsc_managed); const char *failed = pcmk__flag_text(rsc->flags, pcmk_rsc_failed); printed_header = TRUE; desc = pe__resource_description(rsc, show_opts); rc = pe__name_and_nvpairs_xml(out, true, PCMK_XE_BUNDLE, PCMK_XA_ID, rsc->id, PCMK_XA_TYPE, type, PCMK_XA_IMAGE, bundle_data->image, PCMK_XA_UNIQUE, unique, PCMK_XA_MAINTENANCE, maintenance, PCMK_XA_MANAGED, managed, PCMK_XA_FAILED, failed, PCMK_XA_DESCRIPTION, desc, NULL); pcmk__assert(rc == pcmk_rc_ok); } id = pcmk__itoa(replica->offset); rc = pe__name_and_nvpairs_xml(out, true, PCMK_XE_REPLICA, PCMK_XA_ID, id, NULL); free(id); pcmk__assert(rc == pcmk_rc_ok); if (print_ip) { out->message(out, (const char *) replica->ip->xml->name, show_opts, replica->ip, only_node, only_rsc); } if (print_child) { out->message(out, (const char *) replica->child->xml->name, show_opts, replica->child, only_node, only_rsc); } if (print_ctnr) { out->message(out, (const char *) replica->container->xml->name, show_opts, replica->container, only_node, only_rsc); } if (print_remote) { out->message(out, (const char *) replica->remote->xml->name, show_opts, replica->remote, only_node, only_rsc); } pcmk__output_xml_pop_parent(out); // replica } if (printed_header) { pcmk__output_xml_pop_parent(out); // bundle } return rc; } static void pe__bundle_replica_output_html(pcmk__output_t *out, pcmk__bundle_replica_t *replica, pcmk_node_t *node, uint32_t show_opts) { pcmk_resource_t *rsc = replica->child; int offset = 0; char buffer[LINE_MAX]; if(rsc == NULL) { rsc = replica->container; } if (replica->remote) { offset += snprintf(buffer + offset, LINE_MAX - offset, "%s", rsc_printable_id(replica->remote)); } else { offset += snprintf(buffer + offset, LINE_MAX - offset, "%s", rsc_printable_id(replica->container)); } if (replica->ipaddr) { offset += snprintf(buffer + offset, LINE_MAX - offset, " (%s)", replica->ipaddr); } pe__common_output_html(out, rsc, buffer, node, show_opts); } /*! * \internal * \brief Get a string describing a resource's unmanaged state or lack thereof * * \param[in] rsc Resource to describe * * \return A string indicating that a resource is in maintenance mode or * otherwise unmanaged, or an empty string otherwise */ static const char * get_unmanaged_str(const pcmk_resource_t *rsc) { if (pcmk_is_set(rsc->flags, pcmk_rsc_maintenance)) { return " (maintenance)"; } if (!pcmk_is_set(rsc->flags, pcmk_rsc_managed)) { return " (unmanaged)"; } return ""; } PCMK__OUTPUT_ARGS("bundle", "uint32_t", "pcmk_resource_t *", "GList *", "GList *") int pe__bundle_html(pcmk__output_t *out, va_list args) { uint32_t show_opts = va_arg(args, uint32_t); pcmk_resource_t *rsc = va_arg(args, pcmk_resource_t *); GList *only_node = va_arg(args, GList *); GList *only_rsc = va_arg(args, GList *); const char *desc = NULL; pe__bundle_variant_data_t *bundle_data = NULL; int rc = pcmk_rc_no_output; gboolean print_everything = TRUE; pcmk__assert(rsc != NULL); get_bundle_variant_data(bundle_data, rsc); desc = pe__resource_description(rsc, show_opts); if (rsc->fns->is_filtered(rsc, only_rsc, TRUE)) { return rc; } print_everything = pcmk__str_in_list(rsc->id, only_rsc, pcmk__str_star_matches); for (GList *gIter = bundle_data->replicas; gIter != NULL; gIter = gIter->next) { pcmk__bundle_replica_t *replica = gIter->data; gboolean print_ip, print_child, print_ctnr, print_remote; pcmk__assert(replica); if (pcmk__rsc_filtered_by_node(replica->container, only_node)) { continue; } print_ip = replica->ip != NULL && !replica->ip->fns->is_filtered(replica->ip, only_rsc, print_everything); print_child = replica->child != NULL && !replica->child->fns->is_filtered(replica->child, only_rsc, print_everything); print_ctnr = !replica->container->fns->is_filtered(replica->container, only_rsc, print_everything); print_remote = replica->remote != NULL && !replica->remote->fns->is_filtered(replica->remote, only_rsc, print_everything); if (pcmk_is_set(show_opts, pcmk_show_implicit_rscs) || (print_everything == FALSE && (print_ip || print_child || print_ctnr || print_remote))) { /* The text output messages used below require pe_print_implicit to * be set to do anything. */ uint32_t new_show_opts = show_opts | pcmk_show_implicit_rscs; PCMK__OUTPUT_LIST_HEADER(out, FALSE, rc, "Container bundle%s: %s [%s]%s%s%s%s%s", (bundle_data->nreplicas > 1)? " set" : "", rsc->id, bundle_data->image, pcmk_is_set(rsc->flags, pcmk_rsc_unique)? " (unique)" : "", desc ? " (" : "", desc ? desc : "", desc ? ")" : "", get_unmanaged_str(rsc)); if (pcmk__list_of_multiple(bundle_data->replicas)) { out->begin_list(out, NULL, NULL, "Replica[%d]", replica->offset); } if (print_ip) { out->message(out, (const char *) replica->ip->xml->name, new_show_opts, replica->ip, only_node, only_rsc); } if (print_child) { out->message(out, (const char *) replica->child->xml->name, new_show_opts, replica->child, only_node, only_rsc); } if (print_ctnr) { out->message(out, (const char *) replica->container->xml->name, new_show_opts, replica->container, only_node, only_rsc); } if (print_remote) { out->message(out, (const char *) replica->remote->xml->name, new_show_opts, replica->remote, only_node, only_rsc); } if (pcmk__list_of_multiple(bundle_data->replicas)) { out->end_list(out); } } else if (print_everything == FALSE && !(print_ip || print_child || print_ctnr || print_remote)) { continue; } else { PCMK__OUTPUT_LIST_HEADER(out, FALSE, rc, "Container bundle%s: %s [%s]%s%s%s%s%s", (bundle_data->nreplicas > 1)? " set" : "", rsc->id, bundle_data->image, pcmk_is_set(rsc->flags, pcmk_rsc_unique)? " (unique)" : "", desc ? " (" : "", desc ? desc : "", desc ? ")" : "", get_unmanaged_str(rsc)); pe__bundle_replica_output_html(out, replica, pcmk__current_node(replica->container), show_opts); } } PCMK__OUTPUT_LIST_FOOTER(out, rc); return rc; } static void pe__bundle_replica_output_text(pcmk__output_t *out, pcmk__bundle_replica_t *replica, pcmk_node_t *node, uint32_t show_opts) { const pcmk_resource_t *rsc = replica->child; int offset = 0; char buffer[LINE_MAX]; if(rsc == NULL) { rsc = replica->container; } if (replica->remote) { offset += snprintf(buffer + offset, LINE_MAX - offset, "%s", rsc_printable_id(replica->remote)); } else { offset += snprintf(buffer + offset, LINE_MAX - offset, "%s", rsc_printable_id(replica->container)); } if (replica->ipaddr) { offset += snprintf(buffer + offset, LINE_MAX - offset, " (%s)", replica->ipaddr); } pe__common_output_text(out, rsc, buffer, node, show_opts); } PCMK__OUTPUT_ARGS("bundle", "uint32_t", "pcmk_resource_t *", "GList *", "GList *") int pe__bundle_text(pcmk__output_t *out, va_list args) { uint32_t show_opts = va_arg(args, uint32_t); pcmk_resource_t *rsc = va_arg(args, pcmk_resource_t *); GList *only_node = va_arg(args, GList *); GList *only_rsc = va_arg(args, GList *); const char *desc = NULL; pe__bundle_variant_data_t *bundle_data = NULL; int rc = pcmk_rc_no_output; gboolean print_everything = TRUE; desc = pe__resource_description(rsc, show_opts); get_bundle_variant_data(bundle_data, rsc); pcmk__assert(rsc != NULL); if (rsc->fns->is_filtered(rsc, only_rsc, TRUE)) { return rc; } print_everything = pcmk__str_in_list(rsc->id, only_rsc, pcmk__str_star_matches); for (GList *gIter = bundle_data->replicas; gIter != NULL; gIter = gIter->next) { pcmk__bundle_replica_t *replica = gIter->data; gboolean print_ip, print_child, print_ctnr, print_remote; pcmk__assert(replica); if (pcmk__rsc_filtered_by_node(replica->container, only_node)) { continue; } print_ip = replica->ip != NULL && !replica->ip->fns->is_filtered(replica->ip, only_rsc, print_everything); print_child = replica->child != NULL && !replica->child->fns->is_filtered(replica->child, only_rsc, print_everything); print_ctnr = !replica->container->fns->is_filtered(replica->container, only_rsc, print_everything); print_remote = replica->remote != NULL && !replica->remote->fns->is_filtered(replica->remote, only_rsc, print_everything); if (pcmk_is_set(show_opts, pcmk_show_implicit_rscs) || (print_everything == FALSE && (print_ip || print_child || print_ctnr || print_remote))) { /* The text output messages used below require pe_print_implicit to * be set to do anything. */ uint32_t new_show_opts = show_opts | pcmk_show_implicit_rscs; PCMK__OUTPUT_LIST_HEADER(out, FALSE, rc, "Container bundle%s: %s [%s]%s%s%s%s%s", (bundle_data->nreplicas > 1)? " set" : "", rsc->id, bundle_data->image, pcmk_is_set(rsc->flags, pcmk_rsc_unique)? " (unique)" : "", desc ? " (" : "", desc ? desc : "", desc ? ")" : "", get_unmanaged_str(rsc)); if (pcmk__list_of_multiple(bundle_data->replicas)) { out->list_item(out, NULL, "Replica[%d]", replica->offset); } out->begin_list(out, NULL, NULL, NULL); if (print_ip) { out->message(out, (const char *) replica->ip->xml->name, new_show_opts, replica->ip, only_node, only_rsc); } if (print_child) { out->message(out, (const char *) replica->child->xml->name, new_show_opts, replica->child, only_node, only_rsc); } if (print_ctnr) { out->message(out, (const char *) replica->container->xml->name, new_show_opts, replica->container, only_node, only_rsc); } if (print_remote) { out->message(out, (const char *) replica->remote->xml->name, new_show_opts, replica->remote, only_node, only_rsc); } out->end_list(out); } else if (print_everything == FALSE && !(print_ip || print_child || print_ctnr || print_remote)) { continue; } else { PCMK__OUTPUT_LIST_HEADER(out, FALSE, rc, "Container bundle%s: %s [%s]%s%s%s%s%s", (bundle_data->nreplicas > 1)? " set" : "", rsc->id, bundle_data->image, pcmk_is_set(rsc->flags, pcmk_rsc_unique)? " (unique)" : "", desc ? " (" : "", desc ? desc : "", desc ? ")" : "", get_unmanaged_str(rsc)); pe__bundle_replica_output_text(out, replica, pcmk__current_node(replica->container), show_opts); } } PCMK__OUTPUT_LIST_FOOTER(out, rc); return rc; } /*! * \internal * \deprecated This function will be removed in a future release */ static void print_bundle_replica(pcmk__bundle_replica_t *replica, const char *pre_text, long options, void *print_data) { pcmk_node_t *node = NULL; pcmk_resource_t *rsc = replica->child; int offset = 0; char buffer[LINE_MAX]; if(rsc == NULL) { rsc = replica->container; } if (replica->remote) { offset += snprintf(buffer + offset, LINE_MAX - offset, "%s", rsc_printable_id(replica->remote)); } else { offset += snprintf(buffer + offset, LINE_MAX - offset, "%s", rsc_printable_id(replica->container)); } if (replica->ipaddr) { offset += snprintf(buffer + offset, LINE_MAX - offset, " (%s)", replica->ipaddr); } node = pcmk__current_node(replica->container); common_print(rsc, pre_text, buffer, node, options, print_data); } /*! * \internal * \deprecated This function will be removed in a future release */ void pe__print_bundle(pcmk_resource_t *rsc, const char *pre_text, long options, void *print_data) { pe__bundle_variant_data_t *bundle_data = NULL; char *child_text = NULL; CRM_CHECK(rsc != NULL, return); if (options & pe_print_xml) { bundle_print_xml(rsc, pre_text, options, print_data); return; } get_bundle_variant_data(bundle_data, rsc); if (pre_text == NULL) { pre_text = " "; } status_print("%sContainer bundle%s: %s [%s]%s%s\n", pre_text, ((bundle_data->nreplicas > 1)? " set" : ""), rsc->id, bundle_data->image, pcmk_is_set(rsc->flags, pcmk_rsc_unique)? " (unique)" : "", pcmk_is_set(rsc->flags, pcmk_rsc_managed)? "" : " (unmanaged)"); if (options & pe_print_html) { status_print("
    \n
      \n"); } for (GList *gIter = bundle_data->replicas; gIter != NULL; gIter = gIter->next) { pcmk__bundle_replica_t *replica = gIter->data; pcmk__assert(replica); if (options & pe_print_html) { status_print("
    • "); } if (pcmk_is_set(options, pe_print_implicit)) { child_text = crm_strdup_printf(" %s", pre_text); if (pcmk__list_of_multiple(bundle_data->replicas)) { status_print(" %sReplica[%d]\n", pre_text, replica->offset); } if (options & pe_print_html) { status_print("
      \n
        \n"); } print_rsc_in_list(replica->ip, child_text, options, print_data); print_rsc_in_list(replica->container, child_text, options, print_data); print_rsc_in_list(replica->remote, child_text, options, print_data); print_rsc_in_list(replica->child, child_text, options, print_data); if (options & pe_print_html) { status_print("
      \n"); } } else { child_text = crm_strdup_printf("%s ", pre_text); print_bundle_replica(replica, child_text, options, print_data); } free(child_text); if (options & pe_print_html) { status_print("
    • \n"); } } if (options & pe_print_html) { status_print("
    \n"); } } static void free_bundle_replica(pcmk__bundle_replica_t *replica) { if (replica == NULL) { return; } if (replica->node) { free(replica->node); replica->node = NULL; } if (replica->ip) { free_xml(replica->ip->xml); replica->ip->xml = NULL; replica->ip->fns->free(replica->ip); replica->ip = NULL; } if (replica->container) { free_xml(replica->container->xml); replica->container->xml = NULL; replica->container->fns->free(replica->container); replica->container = NULL; } if (replica->remote) { free_xml(replica->remote->xml); replica->remote->xml = NULL; replica->remote->fns->free(replica->remote); replica->remote = NULL; } free(replica->ipaddr); free(replica); } void pe__free_bundle(pcmk_resource_t *rsc) { pe__bundle_variant_data_t *bundle_data = NULL; CRM_CHECK(rsc != NULL, return); get_bundle_variant_data(bundle_data, rsc); pcmk__rsc_trace(rsc, "Freeing %s", rsc->id); free(bundle_data->prefix); free(bundle_data->image); free(bundle_data->control_port); free(bundle_data->host_network); free(bundle_data->host_netmask); free(bundle_data->ip_range_start); free(bundle_data->container_network); free(bundle_data->launcher_options); free(bundle_data->container_command); g_free(bundle_data->container_host_options); g_list_free_full(bundle_data->replicas, (GDestroyNotify) free_bundle_replica); g_list_free_full(bundle_data->mounts, (GDestroyNotify)mount_free); g_list_free_full(bundle_data->ports, (GDestroyNotify)port_free); g_list_free(rsc->children); if(bundle_data->child) { free_xml(bundle_data->child->xml); bundle_data->child->xml = NULL; bundle_data->child->fns->free(bundle_data->child); } common_free(rsc); } enum rsc_role_e pe__bundle_resource_state(const pcmk_resource_t *rsc, gboolean current) { enum rsc_role_e container_role = pcmk_role_unknown; return container_role; } /*! * \brief Get the number of configured replicas in a bundle * * \param[in] rsc Bundle resource * * \return Number of configured replicas, or 0 on error */ int pe_bundle_replicas(const pcmk_resource_t *rsc) { if (pcmk__is_bundle(rsc)) { pe__bundle_variant_data_t *bundle_data = NULL; get_bundle_variant_data(bundle_data, rsc); return bundle_data->nreplicas; } return 0; } void pe__count_bundle(pcmk_resource_t *rsc) { pe__bundle_variant_data_t *bundle_data = NULL; get_bundle_variant_data(bundle_data, rsc); for (GList *item = bundle_data->replicas; item != NULL; item = item->next) { pcmk__bundle_replica_t *replica = item->data; if (replica->ip) { replica->ip->fns->count(replica->ip); } if (replica->child) { replica->child->fns->count(replica->child); } if (replica->container) { replica->container->fns->count(replica->container); } if (replica->remote) { replica->remote->fns->count(replica->remote); } } } gboolean pe__bundle_is_filtered(const pcmk_resource_t *rsc, GList *only_rsc, gboolean check_parent) { gboolean passes = FALSE; pe__bundle_variant_data_t *bundle_data = NULL; if (pcmk__str_in_list(rsc_printable_id(rsc), only_rsc, pcmk__str_star_matches)) { passes = TRUE; } else { get_bundle_variant_data(bundle_data, rsc); for (GList *gIter = bundle_data->replicas; gIter != NULL; gIter = gIter->next) { pcmk__bundle_replica_t *replica = gIter->data; if (replica->ip != NULL && !replica->ip->fns->is_filtered(replica->ip, only_rsc, FALSE)) { passes = TRUE; break; } else if (replica->child != NULL && !replica->child->fns->is_filtered(replica->child, only_rsc, FALSE)) { passes = TRUE; break; } else if (!replica->container->fns->is_filtered(replica->container, only_rsc, FALSE)) { passes = TRUE; break; } else if (replica->remote != NULL && !replica->remote->fns->is_filtered(replica->remote, only_rsc, FALSE)) { passes = TRUE; break; } } } return !passes; } /*! * \internal * \brief Get a list of a bundle's containers * * \param[in] bundle Bundle resource * * \return Newly created list of \p bundle's containers * \note It is the caller's responsibility to free the result with * g_list_free(). */ GList * pe__bundle_containers(const pcmk_resource_t *bundle) { GList *containers = NULL; const pe__bundle_variant_data_t *data = NULL; get_bundle_variant_data(data, bundle); for (GList *iter = data->replicas; iter != NULL; iter = iter->next) { pcmk__bundle_replica_t *replica = iter->data; containers = g_list_append(containers, replica->container); } return containers; } // Bundle implementation of pcmk_rsc_methods_t:active_node() pcmk_node_t * pe__bundle_active_node(const pcmk_resource_t *rsc, unsigned int *count_all, unsigned int *count_clean) { pcmk_node_t *active = NULL; pcmk_node_t *node = NULL; pcmk_resource_t *container = NULL; GList *containers = NULL; GList *iter = NULL; GHashTable *nodes = NULL; const pe__bundle_variant_data_t *data = NULL; if (count_all != NULL) { *count_all = 0; } if (count_clean != NULL) { *count_clean = 0; } if (rsc == NULL) { return NULL; } /* For the purposes of this method, we only care about where the bundle's * containers are active, so build a list of active containers. */ get_bundle_variant_data(data, rsc); for (iter = data->replicas; iter != NULL; iter = iter->next) { pcmk__bundle_replica_t *replica = iter->data; if (replica->container->running_on != NULL) { containers = g_list_append(containers, replica->container); } } if (containers == NULL) { return NULL; } /* If the bundle has only a single active container, just use that * container's method. If live migration is ever supported for bundle * containers, this will allow us to prefer the migration source when there * is only one container and it is migrating. For now, this just lets us * avoid creating the nodes table. */ if (pcmk__list_of_1(containers)) { container = containers->data; node = container->fns->active_node(container, count_all, count_clean); g_list_free(containers); return node; } // Add all containers' active nodes to a hash table (for uniqueness) nodes = g_hash_table_new(NULL, NULL); for (iter = containers; iter != NULL; iter = iter->next) { container = iter->data; for (GList *node_iter = container->running_on; node_iter != NULL; node_iter = node_iter->next) { node = node_iter->data; // If insert returns true, we haven't counted this node yet if (g_hash_table_insert(nodes, (gpointer) node->details, (gpointer) node) && !pe__count_active_node(rsc, node, &active, count_all, count_clean)) { goto done; } } } done: g_list_free(containers); g_hash_table_destroy(nodes); return active; } /*! * \internal * \brief Get maximum bundle resource instances per node * * \param[in] rsc Bundle resource to check * * \return Maximum number of \p rsc instances that can be active on one node */ unsigned int pe__bundle_max_per_node(const pcmk_resource_t *rsc) { pe__bundle_variant_data_t *bundle_data = NULL; get_bundle_variant_data(bundle_data, rsc); pcmk__assert(bundle_data->nreplicas_per_host >= 0); return (unsigned int) bundle_data->nreplicas_per_host; } diff --git a/lib/services/Makefile.am b/lib/services/Makefile.am index fbd959162b..905afa6a86 100644 --- a/lib/services/Makefile.am +++ b/lib/services/Makefile.am @@ -1,44 +1,44 @@ # # Copyright 2012-2024 the Pacemaker project contributors # # The version control history for this file may have further details. # # This source code is licensed under the GNU Lesser General Public License # version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. # MAINTAINERCLEANFILES = Makefile.in AM_CPPFLAGS = -I$(top_srcdir)/include lib_LTLIBRARIES = libcrmservice.la noinst_HEADERS = $(wildcard *.h) -libcrmservice_la_LDFLAGS = -version-info 32:2:4 +libcrmservice_la_LDFLAGS = -version-info 32:3:4 libcrmservice_la_CFLAGS = libcrmservice_la_CFLAGS += $(CFLAGS_HARDENED_LIB) libcrmservice_la_LDFLAGS += $(LDFLAGS_HARDENED_LIB) libcrmservice_la_LIBADD = $(top_builddir)/lib/common/libcrmcommon.la \ $(DBUS_LIBS) ## Library sources (*must* use += format for bumplibs) libcrmservice_la_SOURCES = services.c libcrmservice_la_SOURCES += services_linux.c libcrmservice_la_SOURCES += services_ocf.c if BUILD_LSB libcrmservice_la_SOURCES += services_lsb.c endif if BUILD_DBUS libcrmservice_la_SOURCES += dbus.c endif if BUILD_UPSTART libcrmservice_la_SOURCES += upstart.c endif if BUILD_SYSTEMD libcrmservice_la_SOURCES += systemd.c endif if BUILD_NAGIOS libcrmservice_la_SOURCES += services_nagios.c endif diff --git a/lib/services/services_linux.c b/lib/services/services_linux.c index f5928905fc..b0b79ca40f 100644 --- a/lib/services/services_linux.c +++ b/lib/services/services_linux.c @@ -1,1507 +1,1507 @@ /* - * Copyright 2010-2024 the Pacemaker project contributors + * Copyright 2010-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #ifndef _GNU_SOURCE # define _GNU_SOURCE #endif #include #include #include #include #include #include #include #include #include #include #include "crm/crm.h" #include "crm/common/mainloop.h" #include "crm/services.h" #include "crm/services_internal.h" #include "services_private.h" static void close_pipe(int fildes[]); /* We have two alternative ways of handling SIGCHLD when synchronously waiting * for spawned processes to complete. Both rely on polling a file descriptor to * discover SIGCHLD events. * * If sys/signalfd.h is available (e.g. on Linux), we call signalfd() to * generate the file descriptor. Otherwise, we use the "self-pipe trick" * (opening a pipe and writing a byte to it when SIGCHLD is received). */ #ifdef HAVE_SYS_SIGNALFD_H // signalfd() implementation #include // Everything needed to manage SIGCHLD handling struct sigchld_data_s { sigset_t mask; // Signals to block now (including SIGCHLD) sigset_t old_mask; // Previous set of blocked signals bool ignored; // If SIGCHLD for another child has been ignored }; // Initialize SIGCHLD data and prepare for use static bool sigchld_setup(struct sigchld_data_s *data) { sigemptyset(&(data->mask)); sigaddset(&(data->mask), SIGCHLD); sigemptyset(&(data->old_mask)); // Block SIGCHLD (saving previous set of blocked signals to restore later) if (sigprocmask(SIG_BLOCK, &(data->mask), &(data->old_mask)) < 0) { crm_info("Wait for child process completion failed: %s " CRM_XS " source=sigprocmask", pcmk_rc_str(errno)); return false; } data->ignored = false; return true; } // Get a file descriptor suitable for polling for SIGCHLD events static int sigchld_open(struct sigchld_data_s *data) { int fd; CRM_CHECK(data != NULL, return -1); fd = signalfd(-1, &(data->mask), SFD_NONBLOCK); if (fd < 0) { crm_info("Wait for child process completion failed: %s " CRM_XS " source=signalfd", pcmk_rc_str(errno)); } return fd; } // Close a file descriptor returned by sigchld_open() static void sigchld_close(int fd) { if (fd > 0) { close(fd); } } // Return true if SIGCHLD was received from polled fd static bool sigchld_received(int fd, int pid, struct sigchld_data_s *data) { struct signalfd_siginfo fdsi; ssize_t s; if (fd < 0) { return false; } s = read(fd, &fdsi, sizeof(struct signalfd_siginfo)); if (s != sizeof(struct signalfd_siginfo)) { crm_info("Wait for child process completion failed: %s " CRM_XS " source=read", pcmk_rc_str(errno)); } else if (fdsi.ssi_signo == SIGCHLD) { if (fdsi.ssi_pid == pid) { return true; } else { /* This SIGCHLD is for another child. We have to ignore it here but * will still need to resend it after this synchronous action has * completed and SIGCHLD has been restored to be handled by the * previous SIGCHLD handler, so that it will be handled. */ data->ignored = true; return false; } } return false; } // Do anything needed after done waiting for SIGCHLD static void sigchld_cleanup(struct sigchld_data_s *data) { // Restore the original set of blocked signals if ((sigismember(&(data->old_mask), SIGCHLD) == 0) && (sigprocmask(SIG_UNBLOCK, &(data->mask), NULL) < 0)) { crm_warn("Could not clean up after child process completion: %s", pcmk_rc_str(errno)); } // Resend any ignored SIGCHLD for other children so that they'll be handled. if (data->ignored && kill(getpid(), SIGCHLD) != 0) { crm_warn("Could not resend ignored SIGCHLD to ourselves: %s", pcmk_rc_str(errno)); } } #else // HAVE_SYS_SIGNALFD_H not defined // Self-pipe implementation (see above for function descriptions) struct sigchld_data_s { int pipe_fd[2]; // Pipe file descriptors struct sigaction sa; // Signal handling info (with SIGCHLD) struct sigaction old_sa; // Previous signal handling info bool ignored; // If SIGCHLD for another child has been ignored }; // We need a global to use in the signal handler volatile struct sigchld_data_s *last_sigchld_data = NULL; static void sigchld_handler(void) { // We received a SIGCHLD, so trigger pipe polling if ((last_sigchld_data != NULL) && (last_sigchld_data->pipe_fd[1] >= 0) && (write(last_sigchld_data->pipe_fd[1], "", 1) == -1)) { crm_info("Wait for child process completion failed: %s " CRM_XS " source=write", pcmk_rc_str(errno)); } } static bool sigchld_setup(struct sigchld_data_s *data) { int rc; data->pipe_fd[0] = data->pipe_fd[1] = -1; if (pipe(data->pipe_fd) == -1) { crm_info("Wait for child process completion failed: %s " CRM_XS " source=pipe", pcmk_rc_str(errno)); return false; } rc = pcmk__set_nonblocking(data->pipe_fd[0]); if (rc != pcmk_rc_ok) { crm_info("Could not set pipe input non-blocking: %s " CRM_XS " rc=%d", pcmk_rc_str(rc), rc); } rc = pcmk__set_nonblocking(data->pipe_fd[1]); if (rc != pcmk_rc_ok) { crm_info("Could not set pipe output non-blocking: %s " CRM_XS " rc=%d", pcmk_rc_str(rc), rc); } // Set SIGCHLD handler data->sa.sa_handler = (sighandler_t) sigchld_handler; data->sa.sa_flags = 0; sigemptyset(&(data->sa.sa_mask)); if (sigaction(SIGCHLD, &(data->sa), &(data->old_sa)) < 0) { crm_info("Wait for child process completion failed: %s " CRM_XS " source=sigaction", pcmk_rc_str(errno)); } data->ignored = false; // Remember data for use in signal handler last_sigchld_data = data; return true; } static int sigchld_open(struct sigchld_data_s *data) { CRM_CHECK(data != NULL, return -1); return data->pipe_fd[0]; } static void sigchld_close(int fd) { // Pipe will be closed in sigchld_cleanup() return; } static bool sigchld_received(int fd, int pid, struct sigchld_data_s *data) { char ch; if (fd < 0) { return false; } // Clear out the self-pipe while (read(fd, &ch, 1) == 1) /*omit*/; return true; } static void sigchld_cleanup(struct sigchld_data_s *data) { // Restore the previous SIGCHLD handler if (sigaction(SIGCHLD, &(data->old_sa), NULL) < 0) { crm_warn("Could not clean up after child process completion: %s", pcmk_rc_str(errno)); } close_pipe(data->pipe_fd); // Resend any ignored SIGCHLD for other children so that they'll be handled. if (data->ignored && kill(getpid(), SIGCHLD) != 0) { crm_warn("Could not resend ignored SIGCHLD to ourselves: %s", pcmk_rc_str(errno)); } } #endif /*! * \internal * \brief Close the two file descriptors of a pipe * * \param[in,out] fildes Array of file descriptors opened by pipe() */ static void close_pipe(int fildes[]) { if (fildes[0] >= 0) { close(fildes[0]); fildes[0] = -1; } if (fildes[1] >= 0) { close(fildes[1]); fildes[1] = -1; } } #define out_type(is_stderr) ((is_stderr)? "stderr" : "stdout") // Maximum number of bytes of stdout or stderr we'll accept #define MAX_OUTPUT (10 * 1024 * 1024) static gboolean svc_read_output(int fd, svc_action_t * op, bool is_stderr) { char *data = NULL; ssize_t rc = 0; size_t len = 0; size_t discarded = 0; char buf[500]; static const size_t buf_read_len = sizeof(buf) - 1; if (fd < 0) { crm_trace("No fd for %s", op->id); return FALSE; } if (is_stderr && op->stderr_data) { len = strlen(op->stderr_data); data = op->stderr_data; crm_trace("Reading %s stderr into offset %lld", op->id, (long long) len); } else if (is_stderr == FALSE && op->stdout_data) { len = strlen(op->stdout_data); data = op->stdout_data; crm_trace("Reading %s stdout into offset %lld", op->id, (long long) len); } else { crm_trace("Reading %s %s", op->id, out_type(is_stderr)); } do { errno = 0; rc = read(fd, buf, buf_read_len); if (rc > 0) { if (len < MAX_OUTPUT) { buf[rc] = 0; crm_trace("Received %lld bytes of %s %s: %.80s", (long long) rc, op->id, out_type(is_stderr), buf); data = pcmk__realloc(data, len + rc + 1); strcpy(data + len, buf); len += rc; } else { discarded += rc; } } else if (errno != EINTR) { // Fatal error or EOF rc = 0; break; } } while ((rc == buf_read_len) || (rc < 0)); if (discarded > 0) { crm_warn("Truncated %s %s to %lld bytes (discarded %lld)", op->id, out_type(is_stderr), (long long) len, (long long) discarded); } if (is_stderr) { op->stderr_data = data; } else { op->stdout_data = data; } return rc != 0; } static int dispatch_stdout(gpointer userdata) { svc_action_t *op = (svc_action_t *) userdata; return svc_read_output(op->opaque->stdout_fd, op, FALSE); } static int dispatch_stderr(gpointer userdata) { svc_action_t *op = (svc_action_t *) userdata; return svc_read_output(op->opaque->stderr_fd, op, TRUE); } static void pipe_out_done(gpointer user_data) { svc_action_t *op = (svc_action_t *) user_data; crm_trace("%p", op); op->opaque->stdout_gsource = NULL; if (op->opaque->stdout_fd > STDOUT_FILENO) { close(op->opaque->stdout_fd); } op->opaque->stdout_fd = -1; } static void pipe_err_done(gpointer user_data) { svc_action_t *op = (svc_action_t *) user_data; op->opaque->stderr_gsource = NULL; if (op->opaque->stderr_fd > STDERR_FILENO) { close(op->opaque->stderr_fd); } op->opaque->stderr_fd = -1; } static struct mainloop_fd_callbacks stdout_callbacks = { .dispatch = dispatch_stdout, .destroy = pipe_out_done, }; static struct mainloop_fd_callbacks stderr_callbacks = { .dispatch = dispatch_stderr, .destroy = pipe_err_done, }; static void set_ocf_env(const char *key, const char *value, gpointer user_data) { if (setenv(key, value, 1) != 0) { crm_perror(LOG_ERR, "setenv failed for key:%s and value:%s", key, value); } } static void set_ocf_env_with_prefix(gpointer key, gpointer value, gpointer user_data) { char buffer[500]; snprintf(buffer, sizeof(buffer), strcmp(key, "OCF_CHECK_LEVEL") != 0 ? "OCF_RESKEY_%s" : "%s", (char *)key); set_ocf_env(buffer, value, user_data); } static void set_alert_env(gpointer key, gpointer value, gpointer user_data) { int rc; if (value != NULL) { rc = setenv(key, value, 1); } else { rc = unsetenv(key); } if (rc < 0) { crm_perror(LOG_ERR, "setenv %s=%s", (char*)key, (value? (char*)value : "")); } else { crm_trace("setenv %s=%s", (char*)key, (value? (char*)value : "")); } } /*! * \internal * \brief Add environment variables suitable for an action * * \param[in] op Action to use */ static void add_action_env_vars(const svc_action_t *op) { void (*env_setter)(gpointer, gpointer, gpointer) = NULL; if (op->agent == NULL) { env_setter = set_alert_env; /* we deal with alert handler */ } else if (pcmk__str_eq(op->standard, PCMK_RESOURCE_CLASS_OCF, pcmk__str_casei)) { env_setter = set_ocf_env_with_prefix; } if (env_setter != NULL && op->params != NULL) { g_hash_table_foreach(op->params, env_setter, NULL); } if (env_setter == NULL || env_setter == set_alert_env) { return; } set_ocf_env("OCF_RA_VERSION_MAJOR", PCMK_OCF_MAJOR_VERSION, NULL); set_ocf_env("OCF_RA_VERSION_MINOR", PCMK_OCF_MINOR_VERSION, NULL); set_ocf_env("OCF_ROOT", OCF_ROOT_DIR, NULL); set_ocf_env("OCF_EXIT_REASON_PREFIX", PCMK_OCF_REASON_PREFIX, NULL); if (op->rsc) { set_ocf_env("OCF_RESOURCE_INSTANCE", op->rsc, NULL); } if (op->agent != NULL) { set_ocf_env("OCF_RESOURCE_TYPE", op->agent, NULL); } /* Notes: this is not added to specification yet. Sept 10,2004 */ if (op->provider != NULL) { set_ocf_env("OCF_RESOURCE_PROVIDER", op->provider, NULL); } } static void pipe_in_single_parameter(gpointer key, gpointer value, gpointer user_data) { svc_action_t *op = user_data; char *buffer = crm_strdup_printf("%s=%s\n", (char *)key, (char *) value); size_t len = strlen(buffer); size_t total = 0; ssize_t ret = 0; do { errno = 0; ret = write(op->opaque->stdin_fd, buffer + total, len - total); if (ret > 0) { total += ret; } } while ((errno == EINTR) && (total < len)); free(buffer); } /*! * \internal * \brief Pipe parameters in via stdin for action * * \param[in] op Action to use */ static void pipe_in_action_stdin_parameters(const svc_action_t *op) { if (op->params) { g_hash_table_foreach(op->params, pipe_in_single_parameter, (gpointer) op); } } gboolean recurring_action_timer(gpointer data) { svc_action_t *op = data; crm_debug("Scheduling another invocation of %s", op->id); /* Clean out the old result */ free(op->stdout_data); op->stdout_data = NULL; free(op->stderr_data); op->stderr_data = NULL; op->opaque->repeat_timer = 0; services_action_async(op, NULL); return FALSE; } /*! * \internal * \brief Finalize handling of an asynchronous operation * * Given a completed asynchronous operation, cancel or reschedule it as * appropriate if recurring, call its callback if registered, stop tracking it, * and clean it up. * * \param[in,out] op Operation to finalize * * \return Standard Pacemaker return code * \retval EINVAL Caller supplied NULL or invalid \p op * \retval EBUSY Uncanceled recurring action has only been cleaned up * \retval pcmk_rc_ok Action has been freed * * \note If the return value is not pcmk_rc_ok, the caller is responsible for * freeing the action. */ int services__finalize_async_op(svc_action_t *op) { CRM_CHECK((op != NULL) && !(op->synchronous), return EINVAL); if (op->interval_ms != 0) { // Recurring operations must be either cancelled or rescheduled if (op->cancel) { services__set_cancelled(op); cancel_recurring_action(op); } else { op->opaque->repeat_timer = g_timeout_add(op->interval_ms, recurring_action_timer, (void *) op); } } if (op->opaque->callback != NULL) { op->opaque->callback(op); } // Stop tracking the operation (as in-flight or blocked) op->pid = 0; services_untrack_op(op); if ((op->interval_ms != 0) && !(op->cancel)) { // Do not free recurring actions (they will get freed when cancelled) services_action_cleanup(op); return EBUSY; } services_action_free(op); return pcmk_rc_ok; } static void close_op_input(svc_action_t *op) { if (op->opaque->stdin_fd >= 0) { close(op->opaque->stdin_fd); } } static void finish_op_output(svc_action_t *op, bool is_stderr) { mainloop_io_t **source; int fd; if (is_stderr) { source = &(op->opaque->stderr_gsource); fd = op->opaque->stderr_fd; } else { source = &(op->opaque->stdout_gsource); fd = op->opaque->stdout_fd; } if (op->synchronous || *source) { crm_trace("Finish reading %s[%d] %s", op->id, op->pid, (is_stderr? "stderr" : "stdout")); svc_read_output(fd, op, is_stderr); if (op->synchronous) { close(fd); } else { mainloop_del_fd(*source); *source = NULL; } } } // Log an operation's stdout and stderr static void log_op_output(svc_action_t *op) { char *prefix = crm_strdup_printf("%s[%d] error output", op->id, op->pid); /* The library caller has better context to know how important the output * is, so log it at info and debug severity here. They can log it again at * higher severity if appropriate. */ crm_log_output(LOG_INFO, prefix, op->stderr_data); strcpy(prefix + strlen(prefix) - strlen("error output"), "output"); crm_log_output(LOG_DEBUG, prefix, op->stdout_data); free(prefix); } // Truncate exit reasons at this many characters #define EXIT_REASON_MAX_LEN 128 static void parse_exit_reason_from_stderr(svc_action_t *op) { const char *reason_start = NULL; const char *reason_end = NULL; const int prefix_len = strlen(PCMK_OCF_REASON_PREFIX); if ((op->stderr_data == NULL) || // Only OCF agents have exit reasons in stderr !pcmk__str_eq(op->standard, PCMK_RESOURCE_CLASS_OCF, pcmk__str_none)) { return; } // Find the last occurrence of the magic string indicating an exit reason for (const char *cur = strstr(op->stderr_data, PCMK_OCF_REASON_PREFIX); cur != NULL; cur = strstr(cur, PCMK_OCF_REASON_PREFIX)) { cur += prefix_len; // Skip over magic string reason_start = cur; } if ((reason_start == NULL) || (reason_start[0] == '\n') || (reason_start[0] == '\0')) { return; // No or empty exit reason } // Exit reason goes to end of line (or end of output) reason_end = strchr(reason_start, '\n'); if (reason_end == NULL) { reason_end = reason_start + strlen(reason_start); } // Limit size of exit reason to something reasonable if (reason_end > (reason_start + EXIT_REASON_MAX_LEN)) { reason_end = reason_start + EXIT_REASON_MAX_LEN; } free(op->opaque->exit_reason); op->opaque->exit_reason = strndup(reason_start, reason_end - reason_start); } /*! * \internal * \brief Process the completion of an asynchronous child process * * \param[in,out] p Child process that completed * \param[in] pid Process ID of child * \param[in] core (Unused) * \param[in] signo Signal that interrupted child, if any * \param[in] exitcode Exit status of child process */ static void async_action_complete(mainloop_child_t *p, pid_t pid, int core, int signo, int exitcode) { svc_action_t *op = mainloop_child_userdata(p); mainloop_clear_child_userdata(p); CRM_CHECK(op->pid == pid, services__set_result(op, services__generic_error(op), PCMK_EXEC_ERROR, "Bug in mainloop handling"); return); /* Depending on the priority the mainloop gives the stdout and stderr * file descriptors, this function could be called before everything has * been read from them, so force a final read now. */ finish_op_output(op, true); finish_op_output(op, false); close_op_input(op); if (signo == 0) { crm_debug("%s[%d] exited with status %d", op->id, op->pid, exitcode); services__set_result(op, exitcode, PCMK_EXEC_DONE, NULL); log_op_output(op); parse_exit_reason_from_stderr(op); } else if (mainloop_child_timeout(p)) { const char *kind = services__action_kind(op); crm_info("%s %s[%d] timed out after %s", kind, op->id, op->pid, pcmk__readable_interval(op->timeout)); services__format_result(op, services__generic_error(op), PCMK_EXEC_TIMEOUT, "%s did not complete within %s", kind, pcmk__readable_interval(op->timeout)); } else if (op->cancel) { /* If an in-flight recurring operation was killed because it was * cancelled, don't treat that as a failure. */ crm_info("%s[%d] terminated with signal %d (%s)", op->id, op->pid, signo, strsignal(signo)); services__set_result(op, PCMK_OCF_OK, PCMK_EXEC_CANCELLED, NULL); } else { crm_info("%s[%d] terminated with signal %d (%s)", op->id, op->pid, signo, strsignal(signo)); services__format_result(op, PCMK_OCF_UNKNOWN_ERROR, PCMK_EXEC_ERROR, "%s interrupted by %s signal", services__action_kind(op), strsignal(signo)); } services__finalize_async_op(op); } /*! * \internal * \brief Return agent standard's exit status for "generic error" * * When returning an internal error for an action, a value that is appropriate * to the action's agent standard must be used. This function returns a value * appropriate for errors in general. * * \param[in] op Action that error is for * * \return Exit status appropriate to agent standard * \note Actions without a standard will get PCMK_OCF_UNKNOWN_ERROR. */ int services__generic_error(const svc_action_t *op) { if ((op == NULL) || (op->standard == NULL)) { return PCMK_OCF_UNKNOWN_ERROR; } #if PCMK__ENABLE_LSB if (pcmk__str_eq(op->standard, PCMK_RESOURCE_CLASS_LSB, pcmk__str_casei) && pcmk__str_eq(op->action, PCMK_ACTION_STATUS, pcmk__str_casei)) { return PCMK_LSB_STATUS_UNKNOWN; } #endif #if SUPPORT_NAGIOS if (pcmk__str_eq(op->standard, PCMK_RESOURCE_CLASS_NAGIOS, pcmk__str_casei)) { return NAGIOS_STATE_UNKNOWN; } #endif return PCMK_OCF_UNKNOWN_ERROR; } /*! * \internal * \brief Return agent standard's exit status for "not installed" * * When returning an internal error for an action, a value that is appropriate * to the action's agent standard must be used. This function returns a value * appropriate for "not installed" errors. * * \param[in] op Action that error is for * * \return Exit status appropriate to agent standard * \note Actions without a standard will get PCMK_OCF_UNKNOWN_ERROR. */ int services__not_installed_error(const svc_action_t *op) { if ((op == NULL) || (op->standard == NULL)) { return PCMK_OCF_UNKNOWN_ERROR; } #if PCMK__ENABLE_LSB if (pcmk__str_eq(op->standard, PCMK_RESOURCE_CLASS_LSB, pcmk__str_casei) && pcmk__str_eq(op->action, PCMK_ACTION_STATUS, pcmk__str_casei)) { return PCMK_LSB_STATUS_NOT_INSTALLED; } #endif #if SUPPORT_NAGIOS if (pcmk__str_eq(op->standard, PCMK_RESOURCE_CLASS_NAGIOS, pcmk__str_casei)) { return NAGIOS_STATE_UNKNOWN; } #endif return PCMK_OCF_NOT_INSTALLED; } /*! * \internal * \brief Return agent standard's exit status for "insufficient privileges" * * When returning an internal error for an action, a value that is appropriate * to the action's agent standard must be used. This function returns a value * appropriate for "insufficient privileges" errors. * * \param[in] op Action that error is for * * \return Exit status appropriate to agent standard * \note Actions without a standard will get PCMK_OCF_UNKNOWN_ERROR. */ int services__authorization_error(const svc_action_t *op) { if ((op == NULL) || (op->standard == NULL)) { return PCMK_OCF_UNKNOWN_ERROR; } #if PCMK__ENABLE_LSB if (pcmk__str_eq(op->standard, PCMK_RESOURCE_CLASS_LSB, pcmk__str_casei) && pcmk__str_eq(op->action, PCMK_ACTION_STATUS, pcmk__str_casei)) { return PCMK_LSB_STATUS_INSUFFICIENT_PRIV; } #endif #if SUPPORT_NAGIOS if (pcmk__str_eq(op->standard, PCMK_RESOURCE_CLASS_NAGIOS, pcmk__str_casei)) { return NAGIOS_INSUFFICIENT_PRIV; } #endif return PCMK_OCF_INSUFFICIENT_PRIV; } /*! * \internal * \brief Return agent standard's exit status for "not configured" * * When returning an internal error for an action, a value that is appropriate * to the action's agent standard must be used. This function returns a value * appropriate for "not configured" errors. * * \param[in] op Action that error is for * \param[in] is_fatal Whether problem is cluster-wide instead of only local * * \return Exit status appropriate to agent standard * \note Actions without a standard will get PCMK_OCF_UNKNOWN_ERROR. */ int services__configuration_error(const svc_action_t *op, bool is_fatal) { if ((op == NULL) || (op->standard == NULL)) { return PCMK_OCF_UNKNOWN_ERROR; } #if PCMK__ENABLE_LSB if (pcmk__str_eq(op->standard, PCMK_RESOURCE_CLASS_LSB, pcmk__str_casei) && pcmk__str_eq(op->action, PCMK_ACTION_STATUS, pcmk__str_casei)) { return PCMK_LSB_NOT_CONFIGURED; } #endif #if SUPPORT_NAGIOS if (pcmk__str_eq(op->standard, PCMK_RESOURCE_CLASS_NAGIOS, pcmk__str_casei)) { return NAGIOS_STATE_UNKNOWN; } #endif return is_fatal? PCMK_OCF_NOT_CONFIGURED : PCMK_OCF_INVALID_PARAM; } /*! * \internal * \brief Set operation rc and status per errno from stat(), fork() or execvp() * * \param[in,out] op Operation to set rc and status for * \param[in] error Value of errno after system call * * \return void */ void services__handle_exec_error(svc_action_t * op, int error) { const char *name = op->opaque->exec; if (name == NULL) { name = op->agent; if (name == NULL) { name = op->id; } } switch (error) { /* see execve(2), stat(2) and fork(2) */ case ENOENT: /* No such file or directory */ case EISDIR: /* Is a directory */ case ENOTDIR: /* Path component is not a directory */ case EINVAL: /* Invalid executable format */ case ENOEXEC: /* Invalid executable format */ services__format_result(op, services__not_installed_error(op), PCMK_EXEC_NOT_INSTALLED, "%s: %s", name, pcmk_rc_str(error)); break; case EACCES: /* permission denied (various errors) */ case EPERM: /* permission denied (various errors) */ services__format_result(op, services__authorization_error(op), PCMK_EXEC_ERROR, "%s: %s", name, pcmk_rc_str(error)); break; default: services__set_result(op, services__generic_error(op), PCMK_EXEC_ERROR, pcmk_rc_str(error)); } } /*! * \internal * \brief Exit a child process that failed before executing agent * * \param[in] op Action that failed * \param[in] exit_status Exit status code to use * \param[in] exit_reason Exit reason to output if for OCF agent */ static void exit_child(const svc_action_t *op, int exit_status, const char *exit_reason) { if ((op != NULL) && (exit_reason != NULL) && pcmk__str_eq(op->standard, PCMK_RESOURCE_CLASS_OCF, pcmk__str_none)) { fprintf(stderr, PCMK_OCF_REASON_PREFIX "%s\n", exit_reason); } pcmk_common_cleanup(); _exit(exit_status); } static void action_launch_child(svc_action_t *op) { int rc; /* SIGPIPE is ignored (which is different from signal blocking) by the gnutls library. * Depending on the libqb version in use, libqb may set SIGPIPE to be ignored as well. * We do not want this to be inherited by the child process. By resetting this the signal * to the default behavior, we avoid some potential odd problems that occur during OCF * scripts when SIGPIPE is ignored by the environment. */ signal(SIGPIPE, SIG_DFL); #if defined(HAVE_SCHED_SETSCHEDULER) if (sched_getscheduler(0) != SCHED_OTHER) { struct sched_param sp; memset(&sp, 0, sizeof(sp)); sp.sched_priority = 0; if (sched_setscheduler(0, SCHED_OTHER, &sp) == -1) { crm_info("Could not reset scheduling policy for %s", op->id); } } #endif if (setpriority(PRIO_PROCESS, 0, 0) == -1) { crm_info("Could not reset process priority for %s", op->id); } /* Man: The call setpgrp() is equivalent to setpgid(0,0) * _and_ compiles on BSD variants too * need to investigate if it works the same too. */ setpgid(0, 0); pcmk__close_fds_in_child(false); /* It would be nice if errors in this function could be reported as * execution status (for example, PCMK_EXEC_NO_SECRETS for the secrets error * below) instead of exit status. However, we've already forked, so * exit status is all we have. At least for OCF actions, we can output an * exit reason for the parent to parse. */ #if SUPPORT_CIBSECRETS rc = pcmk__substitute_secrets(op->rsc, op->params); if (rc != pcmk_rc_ok) { if (pcmk__str_eq(op->action, PCMK_ACTION_STOP, pcmk__str_casei)) { crm_info("Proceeding with stop operation for %s " "despite being unable to load CIB secrets (%s)", op->rsc, pcmk_rc_str(rc)); } else { crm_err("Considering %s unconfigured " "because unable to load CIB secrets: %s", op->rsc, pcmk_rc_str(rc)); exit_child(op, services__configuration_error(op, false), "Unable to load CIB secrets"); } } #endif add_action_env_vars(op); /* Become the desired user */ if (op->opaque->uid && (geteuid() == 0)) { // If requested, set effective group if (op->opaque->gid && (setgid(op->opaque->gid) < 0)) { crm_err("Considering %s unauthorized because could not set " "child group to %d: %s", op->id, op->opaque->gid, strerror(errno)); exit_child(op, services__authorization_error(op), "Could not set group for child process"); } // Erase supplementary group list // (We could do initgroups() if we kept a copy of the username) if (setgroups(0, NULL) < 0) { crm_err("Considering %s unauthorized because could not " "clear supplementary groups: %s", op->id, strerror(errno)); exit_child(op, services__authorization_error(op), "Could not clear supplementary groups for child process"); } // Set effective user if (setuid(op->opaque->uid) < 0) { crm_err("Considering %s unauthorized because could not set user " "to %d: %s", op->id, op->opaque->uid, strerror(errno)); exit_child(op, services__authorization_error(op), "Could not set user for child process"); } } // Execute the agent (doesn't return if successful) execvp(op->opaque->exec, op->opaque->args); // An earlier stat() should have avoided most possible errors rc = errno; services__handle_exec_error(op, rc); crm_err("Unable to execute %s: %s", op->id, strerror(rc)); exit_child(op, op->rc, "Child process was unable to execute file"); } /*! * \internal * \brief Wait for synchronous action to complete, and set its result * * \param[in,out] op Action to wait for * \param[in,out] data Child signal data */ static void wait_for_sync_result(svc_action_t *op, struct sigchld_data_s *data) { int status = 0; int timeout = op->timeout; time_t start = time(NULL); struct pollfd fds[3]; int wait_rc = 0; const char *wait_reason = NULL; fds[0].fd = op->opaque->stdout_fd; fds[0].events = POLLIN; fds[0].revents = 0; fds[1].fd = op->opaque->stderr_fd; fds[1].events = POLLIN; fds[1].revents = 0; fds[2].fd = sigchld_open(data); fds[2].events = POLLIN; fds[2].revents = 0; crm_trace("Waiting for %s[%d]", op->id, op->pid); do { int poll_rc = poll(fds, 3, timeout); wait_reason = NULL; if (poll_rc > 0) { if (fds[0].revents & POLLIN) { svc_read_output(op->opaque->stdout_fd, op, FALSE); } if (fds[1].revents & POLLIN) { svc_read_output(op->opaque->stderr_fd, op, TRUE); } if ((fds[2].revents & POLLIN) && sigchld_received(fds[2].fd, op->pid, data)) { wait_rc = waitpid(op->pid, &status, WNOHANG); if ((wait_rc > 0) || ((wait_rc < 0) && (errno == ECHILD))) { // Child process exited or doesn't exist break; } else if (wait_rc < 0) { wait_reason = pcmk_rc_str(errno); crm_info("Wait for completion of %s[%d] failed: %s " CRM_XS " source=waitpid", op->id, op->pid, wait_reason); wait_rc = 0; // Act as if process is still running #ifndef HAVE_SYS_SIGNALFD_H } else { /* The child hasn't exited, so this SIGCHLD could be for * another child. We have to ignore it here but will still * need to resend it after this synchronous action has * completed and SIGCHLD has been restored to be handled by * the previous handler, so that it will be handled. */ data->ignored = true; #endif } } } else if (poll_rc == 0) { // Poll timed out with no descriptors ready timeout = 0; break; } else if ((poll_rc < 0) && (errno != EINTR)) { wait_reason = pcmk_rc_str(errno); crm_info("Wait for completion of %s[%d] failed: %s " CRM_XS " source=poll", op->id, op->pid, wait_reason); break; } timeout = op->timeout - (time(NULL) - start) * 1000; } while ((op->timeout < 0 || timeout > 0)); crm_trace("Stopped waiting for %s[%d]", op->id, op->pid); finish_op_output(op, true); finish_op_output(op, false); close_op_input(op); sigchld_close(fds[2].fd); if (wait_rc <= 0) { if ((op->timeout > 0) && (timeout <= 0)) { services__format_result(op, services__generic_error(op), PCMK_EXEC_TIMEOUT, "%s did not exit within specified timeout", services__action_kind(op)); crm_info("%s[%d] timed out after %dms", op->id, op->pid, op->timeout); } else { services__set_result(op, services__generic_error(op), PCMK_EXEC_ERROR, wait_reason); } /* If only child hasn't been successfully waited for, yet. This is to limit killing wrong target a bit more. */ if ((wait_rc == 0) && (waitpid(op->pid, &status, WNOHANG) == 0)) { if (kill(op->pid, SIGKILL)) { crm_warn("Could not kill rogue child %s[%d]: %s", op->id, op->pid, pcmk_rc_str(errno)); } /* Safe to skip WNOHANG here as we sent non-ignorable signal. */ while ((waitpid(op->pid, &status, 0) == (pid_t) -1) && (errno == EINTR)) { /* keep waiting */; } } } else if (WIFEXITED(status)) { services__set_result(op, WEXITSTATUS(status), PCMK_EXEC_DONE, NULL); parse_exit_reason_from_stderr(op); crm_info("%s[%d] exited with status %d", op->id, op->pid, op->rc); } else if (WIFSIGNALED(status)) { int signo = WTERMSIG(status); services__format_result(op, services__generic_error(op), PCMK_EXEC_ERROR, "%s interrupted by %s signal", services__action_kind(op), strsignal(signo)); crm_info("%s[%d] terminated with signal %d (%s)", op->id, op->pid, signo, strsignal(signo)); #ifdef WCOREDUMP if (WCOREDUMP(status)) { crm_warn("%s[%d] dumped core", op->id, op->pid); } #endif } else { // Shouldn't be possible to get here services__set_result(op, services__generic_error(op), PCMK_EXEC_ERROR, "Unable to wait for child to complete"); } } /*! * \internal * \brief Execute an action whose standard uses executable files * * \param[in,out] op Action to execute * * \return Standard Pacemaker return value * \retval EBUSY Recurring operation could not be initiated * \retval pcmk_rc_error Synchronous action failed * \retval pcmk_rc_ok Synchronous action succeeded, or asynchronous action * should not be freed (because it's pending or because * it failed to execute and was already freed) * * \note If the return value for an asynchronous action is not pcmk_rc_ok, the * caller is responsible for freeing the action. */ int services__execute_file(svc_action_t *op) { int stdout_fd[2]; int stderr_fd[2]; int stdin_fd[2] = {-1, -1}; int rc; struct stat st; struct sigchld_data_s data = { .ignored = false }; // Catch common failure conditions early if (stat(op->opaque->exec, &st) != 0) { rc = errno; crm_info("Cannot execute '%s': %s " CRM_XS " stat rc=%d", op->opaque->exec, pcmk_rc_str(rc), rc); services__handle_exec_error(op, rc); goto done; } if (pipe(stdout_fd) < 0) { rc = errno; crm_info("Cannot execute '%s': %s " CRM_XS " pipe(stdout) rc=%d", op->opaque->exec, pcmk_rc_str(rc), rc); services__handle_exec_error(op, rc); goto done; } if (pipe(stderr_fd) < 0) { rc = errno; close_pipe(stdout_fd); crm_info("Cannot execute '%s': %s " CRM_XS " pipe(stderr) rc=%d", op->opaque->exec, pcmk_rc_str(rc), rc); services__handle_exec_error(op, rc); goto done; } if (pcmk_is_set(pcmk_get_ra_caps(op->standard), pcmk_ra_cap_stdin)) { if (pipe(stdin_fd) < 0) { rc = errno; close_pipe(stdout_fd); close_pipe(stderr_fd); crm_info("Cannot execute '%s': %s " CRM_XS " pipe(stdin) rc=%d", op->opaque->exec, pcmk_rc_str(rc), rc); services__handle_exec_error(op, rc); goto done; } } if (op->synchronous && !sigchld_setup(&data)) { close_pipe(stdin_fd); close_pipe(stdout_fd); close_pipe(stderr_fd); sigchld_cleanup(&data); services__set_result(op, services__generic_error(op), PCMK_EXEC_ERROR, "Could not manage signals for child process"); goto done; } op->pid = fork(); switch (op->pid) { case -1: rc = errno; close_pipe(stdin_fd); close_pipe(stdout_fd); close_pipe(stderr_fd); crm_info("Cannot execute '%s': %s " CRM_XS " fork rc=%d", op->opaque->exec, pcmk_rc_str(rc), rc); services__handle_exec_error(op, rc); if (op->synchronous) { sigchld_cleanup(&data); } goto done; break; case 0: /* Child */ close(stdout_fd[0]); close(stderr_fd[0]); if (stdin_fd[1] >= 0) { close(stdin_fd[1]); } if (STDOUT_FILENO != stdout_fd[1]) { if (dup2(stdout_fd[1], STDOUT_FILENO) != STDOUT_FILENO) { crm_warn("Can't redirect output from '%s': %s " CRM_XS " errno=%d", op->opaque->exec, pcmk_rc_str(errno), errno); } close(stdout_fd[1]); } if (STDERR_FILENO != stderr_fd[1]) { if (dup2(stderr_fd[1], STDERR_FILENO) != STDERR_FILENO) { crm_warn("Can't redirect error output from '%s': %s " CRM_XS " errno=%d", op->opaque->exec, pcmk_rc_str(errno), errno); } close(stderr_fd[1]); } if ((stdin_fd[0] >= 0) && (STDIN_FILENO != stdin_fd[0])) { if (dup2(stdin_fd[0], STDIN_FILENO) != STDIN_FILENO) { crm_warn("Can't redirect input to '%s': %s " CRM_XS " errno=%d", op->opaque->exec, pcmk_rc_str(errno), errno); } close(stdin_fd[0]); } if (op->synchronous) { sigchld_cleanup(&data); } action_launch_child(op); pcmk__assert(false); // action_launch_child() should not return } /* Only the parent reaches here */ close(stdout_fd[1]); close(stderr_fd[1]); if (stdin_fd[0] >= 0) { close(stdin_fd[0]); } op->opaque->stdout_fd = stdout_fd[0]; rc = pcmk__set_nonblocking(op->opaque->stdout_fd); if (rc != pcmk_rc_ok) { crm_info("Could not set '%s' output non-blocking: %s " CRM_XS " rc=%d", op->opaque->exec, pcmk_rc_str(rc), rc); } op->opaque->stderr_fd = stderr_fd[0]; rc = pcmk__set_nonblocking(op->opaque->stderr_fd); if (rc != pcmk_rc_ok) { crm_info("Could not set '%s' error output non-blocking: %s " CRM_XS " rc=%d", op->opaque->exec, pcmk_rc_str(rc), rc); } op->opaque->stdin_fd = stdin_fd[1]; if (op->opaque->stdin_fd >= 0) { // using buffer behind non-blocking-fd here - that could be improved // as long as no other standard uses stdin_fd assume stonith rc = pcmk__set_nonblocking(op->opaque->stdin_fd); if (rc != pcmk_rc_ok) { crm_info("Could not set '%s' input non-blocking: %s " CRM_XS " fd=%d,rc=%d", op->opaque->exec, pcmk_rc_str(rc), op->opaque->stdin_fd, rc); } pipe_in_action_stdin_parameters(op); // as long as we are handling parameters directly in here just close close(op->opaque->stdin_fd); op->opaque->stdin_fd = -1; } // after fds are setup properly and before we plug anything into mainloop if (op->opaque->fork_callback) { op->opaque->fork_callback(op); } if (op->synchronous) { wait_for_sync_result(op, &data); sigchld_cleanup(&data); goto done; } crm_trace("Waiting async for '%s'[%d]", op->opaque->exec, op->pid); mainloop_child_add_with_flags(op->pid, op->timeout, op->id, op, pcmk_is_set(op->flags, SVC_ACTION_LEAVE_GROUP)? mainloop_leave_pid_group : 0, async_action_complete); op->opaque->stdout_gsource = mainloop_add_fd(op->id, G_PRIORITY_LOW, op->opaque->stdout_fd, op, &stdout_callbacks); op->opaque->stderr_gsource = mainloop_add_fd(op->id, G_PRIORITY_LOW, op->opaque->stderr_fd, op, &stderr_callbacks); services_add_inflight_op(op); return pcmk_rc_ok; done: if (op->synchronous) { return (op->rc == PCMK_OCF_OK)? pcmk_rc_ok : pcmk_rc_error; } else { return services__finalize_async_op(op); } } GList * services_os_get_single_directory_list(const char *root, gboolean files, gboolean executable) { GList *list = NULL; - struct dirent **namelist; + struct dirent **namelist = NULL; int entries = 0, lpc = 0; char buffer[PATH_MAX]; entries = scandir(root, &namelist, NULL, alphasort); if (entries <= 0) { return list; } for (lpc = 0; lpc < entries; lpc++) { struct stat sb; if ('.' == namelist[lpc]->d_name[0]) { free(namelist[lpc]); continue; } snprintf(buffer, sizeof(buffer), "%s/%s", root, namelist[lpc]->d_name); if (stat(buffer, &sb)) { continue; } if (S_ISDIR(sb.st_mode)) { if (files) { free(namelist[lpc]); continue; } } else if (S_ISREG(sb.st_mode)) { if (files == FALSE) { free(namelist[lpc]); continue; } else if (executable && (sb.st_mode & S_IXUSR) == 0 && (sb.st_mode & S_IXGRP) == 0 && (sb.st_mode & S_IXOTH) == 0) { free(namelist[lpc]); continue; } } list = g_list_append(list, strdup(namelist[lpc]->d_name)); free(namelist[lpc]); } free(namelist); return list; } GList * services_os_get_directory_list(const char *root, gboolean files, gboolean executable) { GList *result = NULL; char *dirs = strdup(root); char *dir = NULL; if (pcmk__str_empty(dirs)) { free(dirs); return result; } for (dir = strtok(dirs, ":"); dir != NULL; dir = strtok(NULL, ":")) { GList *tmp = services_os_get_single_directory_list(dir, files, executable); if (tmp) { result = g_list_concat(result, tmp); } } free(dirs); return result; } diff --git a/lib/services/services_nagios.c b/lib/services/services_nagios.c index 69b1611139..5b4dbe7db3 100644 --- a/lib/services/services_nagios.c +++ b/lib/services/services_nagios.c @@ -1,220 +1,221 @@ /* - * Copyright 2010-2024 the Pacemaker project contributors + * Copyright 2010-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #ifndef _GNU_SOURCE # define _GNU_SOURCE #endif #include #include #include #include #include #include #include #include #include #include #include "crm/crm.h" #include #include "crm/common/mainloop.h" #include "crm/services.h" #include "services_private.h" #include "services_nagios.h" /*! * \internal * \brief Prepare a Nagios action * * \param[in,out] op Action to prepare * * \return Standard Pacemaker return code */ int services__nagios_prepare(svc_action_t *op) { op->opaque->exec = pcmk__full_path(op->agent, NAGIOS_PLUGIN_DIR); op->opaque->args[0] = strdup(op->opaque->exec); if (op->opaque->args[0] == NULL) { return ENOMEM; } if (pcmk__str_eq(op->action, PCMK_ACTION_MONITOR, pcmk__str_casei) && (op->interval_ms == 0)) { // Invoke --version for a nagios probe op->opaque->args[1] = strdup("--version"); if (op->opaque->args[1] == NULL) { return ENOMEM; } } else if (op->params != NULL) { GHashTableIter iter; char *key = NULL; char *value = NULL; int index = 1; // 0 is already set to executable name g_hash_table_iter_init(&iter, op->params); while (g_hash_table_iter_next(&iter, (gpointer *) & key, (gpointer *) & value)) { if (index > (PCMK__NELEM(op->opaque->args) - 2)) { return E2BIG; } if (pcmk__str_eq(key, PCMK_XA_CRM_FEATURE_SET, pcmk__str_casei) || strstr(key, CRM_META "_")) { continue; } op->opaque->args[index++] = crm_strdup_printf("--%s", key); op->opaque->args[index++] = strdup(value); if (op->opaque->args[index - 1] == NULL) { return ENOMEM; } } } // Nagios actions don't need to keep the parameters if (op->params != NULL) { g_hash_table_destroy(op->params); op->params = NULL; } return pcmk_rc_ok; } /*! * \internal * \brief Map a Nagios result to a standard OCF result * * \param[in] exit_status Nagios exit status * * \return Standard OCF result */ enum ocf_exitcode services__nagios2ocf(int exit_status) { switch (exit_status) { case NAGIOS_STATE_OK: return PCMK_OCF_OK; case NAGIOS_INSUFFICIENT_PRIV: return PCMK_OCF_INSUFFICIENT_PRIV; case NAGIOS_STATE_WARNING: return PCMK_OCF_DEGRADED; case NAGIOS_STATE_CRITICAL: case NAGIOS_STATE_UNKNOWN: default: return PCMK_OCF_UNKNOWN_ERROR; } } static inline char * nagios_metadata_name(const char *plugin) { return crm_strdup_printf(NAGIOS_METADATA_DIR "/%s.xml", plugin); } GList * services__list_nagios_agents(void) { GList *plugin_list = NULL; GList *result = NULL; plugin_list = services_os_get_directory_list(NAGIOS_PLUGIN_DIR, TRUE, TRUE); // Return only the plugins that have metadata for (GList *gIter = plugin_list; gIter != NULL; gIter = gIter->next) { struct stat st; const char *plugin = gIter->data; char *metadata = nagios_metadata_name(plugin); if (stat(metadata, &st) == 0) { result = g_list_append(result, strdup(plugin)); } free(metadata); } g_list_free_full(plugin_list, free); return result; } gboolean services__nagios_agent_exists(const char *name) { char *buf = NULL; gboolean rc = FALSE; struct stat st; if (name == NULL) { return rc; } buf = crm_strdup_printf(NAGIOS_PLUGIN_DIR "/%s", name); if (stat(buf, &st) == 0) { rc = TRUE; } free(buf); return rc; } int services__get_nagios_metadata(const char *type, char **output) { int rc = pcmk_ok; FILE *file_strm = NULL; int start = 0, length = 0, read_len = 0; char *metadata_file = nagios_metadata_name(type); file_strm = fopen(metadata_file, "r"); if (file_strm == NULL) { crm_err("Metadata file %s does not exist", metadata_file); free(metadata_file); return -EIO; } /* see how big the file is */ start = ftell(file_strm); fseek(file_strm, 0L, SEEK_END); - length = ftell(file_strm); - fseek(file_strm, 0L, start); + length = ftell(file_strm); pcmk__assert(length >= 0); + fseek(file_strm, 0L, SEEK_SET); + pcmk__assert(start == ftell(file_strm)); - if (length <= 0) { + if (length == 0) { crm_info("%s was not valid", metadata_file); free(*output); *output = NULL; rc = -EIO; } else { crm_trace("Reading %d bytes from file", length); *output = pcmk__assert_alloc(1, (length + 1)); read_len = fread(*output, 1, length, file_strm); if (read_len != length) { crm_err("Calculated and read bytes differ: %d vs. %d", length, read_len); free(*output); *output = NULL; rc = -EIO; } } fclose(file_strm); free(metadata_file); return rc; } diff --git a/m4/version.m4 b/m4/version.m4 index f49a9f0d47..d6bb9178ac 100644 --- a/m4/version.m4 +++ b/m4/version.m4 @@ -1,2 +1,2 @@ -m4_define([VERSION_NUMBER], [2.1.9]) +m4_define([VERSION_NUMBER], [2.1.10]) m4_define([PCMK_URL], [https://ClusterLabs.org/projects/pacemaker/]) diff --git a/po/zh_CN.po b/po/zh_CN.po index 6be10d9fff..f42f3b9143 100644 --- a/po/zh_CN.po +++ b/po/zh_CN.po @@ -1,1452 +1,1466 @@ # # Copyright 2003-2024 the Pacemaker project contributors # # The version control history for this file may have further details. # # This source code is licensed under the GNU Lesser General Public License # version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. # #, fuzzy msgid "" msgstr "" "Project-Id-Version: Pacemaker 2\n" "Report-Msgid-Bugs-To: developers@clusterlabs.org\n" -"POT-Creation-Date: 2024-10-02 16:48-0500\n" +"POT-Creation-Date: 2025-06-16 17:33-0400\n" "PO-Revision-Date: 2021-11-08 11:04+0800\n" "Last-Translator: Vivi \n" "Language-Team: CHINESE \n" "Language: zh_CN\n" "MIME-Version: 1.0\n" "Content-Type: text/plain; charset=UTF-8\n" "Content-Transfer-Encoding: 8bit\n" #: daemons/fenced/pacemaker-fenced.c:505 msgid "Instance attributes available for all \"stonith\"-class resources" msgstr " 可用于所有stonith类资源的实例属性" #: daemons/fenced/pacemaker-fenced.c:507 msgid "" "Instance attributes available for all \"stonith\"-class resources and used " "by Pacemaker's fence daemon, formerly known as stonithd" msgstr "" " 可用于所有stonith类资源的实例属性,并由Pacemaker的fence守护程序使用(以前称" "为stonithd)" #: daemons/fenced/pacemaker-fenced.c:518 msgid "Deprecated (will be removed in a future release)" msgstr "已弃用(将在未来版本中删除)" #: daemons/fenced/pacemaker-fenced.c:521 msgid "Intended for use in regression testing only" msgstr "仅适用于回归测试" #: daemons/fenced/pacemaker-fenced.c:524 msgid "Send logs to the additional named logfile" msgstr "将日志发送到其他命名日志文件" #: lib/common/options.c:57 msgid "Pacemaker version on cluster node elected Designated Controller (DC)" msgstr "集群选定的控制器节点(DC)的 Pacemaker 版本" #: lib/common/options.c:59 #, fuzzy msgid "" "Includes a hash which identifies the exact revision the code was built from. " "Used for diagnostic purposes." msgstr "它包含一个标识所构建代码修订版本的哈希值. 其可用于诊断." #: lib/common/options.c:66 #, fuzzy msgid "The messaging layer on which Pacemaker is currently running" msgstr "Pacemaker 当前运行的消息传递层" #: lib/common/options.c:67 msgid "Used for informational and diagnostic purposes." msgstr "用于提供信息和诊断." #: lib/common/options.c:73 msgid "An arbitrary name for the cluster" msgstr "任意的集群名称" #: lib/common/options.c:74 msgid "" "This optional value is mostly for users' convenience as desired in " "administration, but may also be used in Pacemaker configuration rules via " "the #cluster-name node attribute, and by higher-level tools and resource " "agents." msgstr "" "该可选值主要是为了方便用户根据管理的需要使用, 可以通过 #cluster-name 节点属性" "在 Pacemaker 配置规则中使用, 以及被更高级的工具和资源代理使用." #: lib/common/options.c:83 msgid "How long to wait for a response from other nodes during start-up" msgstr "启动过程中等待其他节点响应的时间" #: lib/common/options.c:84 msgid "" "The optimal value will depend on the speed and load of your network and the " "type of switches used." msgstr "其最佳值将取决于您的网络速度和负载以及使用的交换机类型." #: lib/common/options.c:91 msgid "" "Polling interval to recheck cluster state and evaluate rules with date " "specifications" msgstr "重新检查集群状态及评估日期规范规则的轮询间隔" #: lib/common/options.c:93 #, fuzzy msgid "" "Pacemaker is primarily event-driven, and looks ahead to know when to recheck " "cluster state for failure-timeout settings and most time-based rules. " "However, it will also recheck the cluster after this amount of inactivity, " "to evaluate rules with date specifications and serve as a fail-safe for " "certain types of scheduler bugs. A value of 0 disables polling. A positive " "value sets an interval in seconds, unless other units are specified (for " "example, \"5min\")." msgstr "" "Pacemaker 主要是通过事件驱动的, 并会提前预测何时重新检查集群状态以评估大多数" "基于时间的规则以及 failure-timeout 配置, 然而无论如何, 经过指定的时间后如果没" "有活动, 它将重新检查集群, 以评估具有日期规范的规则, 并为某些类型的调度程序缺" "陷提供故障保护. 如果值为0, 将禁用轮询. 如果值为正数, 则设置以秒为单位的时间间" "隔, 除非指定了其它单位 (例如, \"5min\")." #: lib/common/options.c:107 msgid "How a cluster node should react if notified of its own fencing" msgstr "集群节点在收到针对自己的 fence 操作结果通知时应如何反应" #: lib/common/options.c:108 #, fuzzy msgid "" "A cluster node may receive notification of a \"succeeded\" fencing that " "targeted it if fencing is misconfigured, or if fabric fencing is in use that " "doesn't cut cluster communication. Use \"stop\" to attempt to immediately " "stop Pacemaker and stay stopped, or \"panic\" to attempt to immediately " "reboot the local node, falling back to stop on failure." msgstr "" "如果有错误的 fence 配置, 或者在使用 fabric fence 机制 (并不会切断集群通信), " "则集群节点可能会收到针对自己的 \"succeeded\" fence 结果通知. 使用 \"stop\" 尝" "试立即停止 pacemaker 并保持停止状态,或者使用 \"panic\" 尝试立即重新启动本地节" "点,如果失败则返回执行 stop." #: lib/common/options.c:119 msgid "" "Declare an election failed if it is not decided within this much time. If " "you need to adjust this value, it probably indicates the presence of a bug." msgstr "" "如果集群在本项设置时间内没有作出决定则宣布选举失败. 这可能表明当前存在错误, " "您需要调整该值." #: lib/common/options.c:128 msgid "" "Exit immediately if shutdown does not complete within this much time. If you " "need to adjust this value, it probably indicates the presence of a bug." msgstr "" "如果在这段时间内关机仍未完成, pacemaker 将立即退出. 这可能表明当前存在错误, " "您需要调整该值." #: lib/common/options.c:138 lib/common/options.c:147 msgid "" "If you need to adjust this value, it probably indicates the presence of a " "bug." msgstr "这可能表明当前存在错误, 您需要调整该值." #: lib/common/options.c:156 #, fuzzy msgid "" "Enabling this option will slow down cluster recovery under all conditions" msgstr "启用此选项将在所有情况下减慢集群恢复的速度" #: lib/common/options.c:158 msgid "" "Delay cluster recovery for this much time to allow for additional events to " "occur. Useful if your configuration is sensitive to the order in which ping " "updates arrive." msgstr "" "集群恢复将被推迟指定的时间间隔, 以等待更多事件发生. 如果您的配置对 ping 更新" "到达的顺序很敏感, 则可以使用此选项." #: lib/common/options.c:169 msgid "What to do when the cluster does not have quorum" msgstr "当集群没有达到必需票数时该如何做" #: lib/common/options.c:176 msgid "Whether to lock resources to a cleanly shut down node" msgstr "是否锁定资源到完全关闭的节点" #: lib/common/options.c:177 msgid "" "When true, resources active on a node when it is cleanly shut down are kept " "\"locked\" to that node (not allowed to run elsewhere) until they start " "again on that node after it rejoins (or for at most shutdown-lock-limit, if " "set). Stonith resources and Pacemaker Remote connections are never locked. " "Clone and bundle instances and the promoted role of promotable clones are " "currently never locked, though support could be added in a future release." msgstr "" "设置为 true 时, 在完全关闭的节点上活动的资源将被 \"locked\" 到该节点 (不允许" "在其它方运行), 直到该节点重新加入后它们再次在该节点上启动 (最长为 shutdown-" "lock-limit,如果已设置). Stonith 资源和 Pacemaker Remote 连接永远不会被锁定. " "克隆和捆绑实例以及可提升克隆的提升角色目前不会被锁定, 尽管可能在未来的发行版" "中添加支持. " #: lib/common/options.c:190 msgid "Do not lock resources to a cleanly shut down node longer than this" msgstr "资源会被锁定到完全关闭的节点的最长时间" #: lib/common/options.c:192 msgid "" "If shutdown-lock is true and this is set to a nonzero time duration, " "shutdown locks will expire after this much time has passed since the " "shutdown was initiated, even if the node has not rejoined." msgstr "" "如果 shutdown-lock 为 true, 并且将此选项设置为非零时间间隔, 则自关闭操作执行" "经过此时间后,shutdown lock 将过期, 即使该节点尚未重新加入也是如此. " #: lib/common/options.c:201 msgid "Enable Access Control Lists (ACLs) for the CIB" msgstr "为 CIB 启用访问控制列表 (ACL) " #: lib/common/options.c:208 msgid "Whether resources can run on any node by default" msgstr "默认情况下资源是否可以在任何节点上运行" #: lib/common/options.c:215 msgid "" "Whether the cluster should refrain from monitoring, starting, and stopping " "resources" msgstr "集群是否应避免监视, 启动和停止资源" #: lib/common/options.c:223 msgid "" "Whether a start failure should prevent a resource from being recovered on " "the same node" msgstr "资源启动失败是否应阻止在同一节点上恢复该资源" #: lib/common/options.c:225 msgid "" "When true, the cluster will immediately ban a resource from a node if it " "fails to start there. When false, the cluster will instead check the " "resource's fail count against its migration-threshold." msgstr "" "当为true, 如果资源启动失败, 集群将立即禁止节点启动该资源, 当为false, 集群将检" "查资源的失败次数是否超过了其 migration-threshold. " #: lib/common/options.c:233 msgid "Whether the cluster should check for active resources during start-up" msgstr "集群是否在启动期间检查活动的资源" +#: lib/common/options.c:242 +#, fuzzy +msgid "Whether remote nodes can be fenced without quorum" +msgstr "节点是否可以被 fence 作为集群恢复的一部分" + #: lib/common/options.c:243 +msgid "" +"By default, inquorate nodes can fence Pacemaker Remote nodes that are part " +"of its partition regardless of whether the resource was successfully " +"restarted elsewhere. If false, an additional check will be added to only " +"fence remote nodes if the cluster thinks they were unable to be restarted." +msgstr "" + +#: lib/common/options.c:253 #, fuzzy msgid "Whether nodes may be fenced as part of recovery" msgstr "节点是否可以被 fence 作为集群恢复的一部分" -#: lib/common/options.c:244 +#: lib/common/options.c:254 msgid "" "If false, unresponsive nodes are immediately assumed to be harmless, and " "resources that were active on them may be recovered elsewhere. This can " "result in a \"split-brain\" situation, potentially leading to data loss and/" "or service unavailability." msgstr "" "如果为 false, 则立即假定无响应的节点是无害的, 并且可以在其它位置恢复在其上活" "动的资源. 这可能会导致 \"split-brain\" 情况, 从而可能导致数据丢失和(或)服务不" "可用. " -#: lib/common/options.c:254 +#: lib/common/options.c:264 msgid "" "Action to send to fence device when a node needs to be fenced (\"poweroff\" " "is a deprecated alias for \"off\")" msgstr "" "当节点需要被 fence 时, 向 fence 设备发送的操作 (\"poweroff\" 作为 \"off\" 的" "别名已被弃用)" -#: lib/common/options.c:262 +#: lib/common/options.c:272 msgid "" "How long to wait for on, off, and reboot fence actions to complete by default" msgstr "默认情况下, 等待 on, off, 和 reboot fence 操作完成的时间" -#: lib/common/options.c:270 +#: lib/common/options.c:280 msgid "Whether watchdog integration is enabled" msgstr "是否启用 watchdog 集成设置" -#: lib/common/options.c:271 +#: lib/common/options.c:281 msgid "" "This is set automatically by the cluster according to whether SBD is " "detected to be in use. User-configured values are ignored. The value `true` " "is meaningful if diskless SBD is used and `stonith-watchdog-timeout` is " "nonzero. In that case, if fencing is required, watchdog-based self-fencing " "will be performed via SBD without requiring a fencing resource explicitly " "configured." msgstr "" "集群会根据是否检测到 SBD 正在使用来自动设置此值. 用户配置的值将被忽略. 如果使" "用了无盘 SBD 并且 `stonith-watchdog-timeout` 不为零, 则值 `true` 才有实际意" "义. 在这种情况下, 如果需要fence, 将通过 SBD 执行基于 watchdog 的自我 fence, " "而不需要明确配置 fence 资源." -#: lib/common/options.c:292 +#: lib/common/options.c:302 #, fuzzy msgid "" "How long before nodes can be assumed to be safely down when watchdog-based " "self-fencing via SBD is in use" msgstr "" "当基于 watchdog 的自我 fence 机制通过SBD 被执行时, 节点被认为安全下线的等待时" "间有多长" -#: lib/common/options.c:294 +#: lib/common/options.c:304 #, fuzzy msgid "" "If this is set to a positive value, lost nodes are assumed to achieve self-" "fencing using watchdog-based SBD within this much time. This does not " "require a fencing resource to be explicitly configured, though a " "fence_watchdog resource can be configured, to limit use to specific nodes. " "If this is set to 0 (the default), the cluster will never assume watchdog-" "based self-fencing. If this is set to a negative value, the cluster will use " "twice the local value of the `SBD_WATCHDOG_TIMEOUT` environment variable if " "that is positive, or otherwise treat this as 0. WARNING: When used, this " "timeout must be larger than `SBD_WATCHDOG_TIMEOUT` on all nodes that use " "watchdog-based SBD, and Pacemaker will refuse to start on any of those nodes " "where this is not true for the local value or SBD is not active. When this " "is set to a negative value, `SBD_WATCHDOG_TIMEOUT` must be set to the same " "value on all nodes that use SBD, otherwise data corruption or loss could " "occur." msgstr "" "如果设为正值, 丢失的节点将在设定的时间内被认定使用基于 watchdog 的 SBD 完成自" "我 fence. 这不需要明确配置一个 fence 资源, 但可以配置一个 fence_watchdog 资源" "来限制对特定节点使用. 如果设为0 (默认值), 集群将永远不会认定节点使用基于 " "watchdog 的自我 fence. 如果设为负值, 集群将使用本地 `SBD_WATCHDOG_TIMEOUT` 环" "境变量的两倍值(如果该值为正), 否则会将该值视为0. 警告: 在所有使用基于 " "watchdog 的 SBD 的节点上, 此超时值需大于 `SBD_WATCHDOG_TIMEOUT` 的值, 否则 " "Pacemaker 不会在任何不符合此条件的节点上启动, 也不会在任何未启用 SBD 的节点上" "启动. 当设为负值时所有使用 SBD 的节点上 `SBD_WATCHDOG_TIMEOUT` 的值必须设置为" "相同的值, 否则可能导致数据损坏或丢失." -#: lib/common/options.c:314 +#: lib/common/options.c:324 msgid "" "How many times fencing can fail before it will no longer be immediately re-" "attempted on a target" msgstr "fence 操作失败多少次会停止立即尝试" -#: lib/common/options.c:327 +#: lib/common/options.c:337 msgid "Allow performing fencing operations in parallel" msgstr "允许并行执行 fencing 操作" -#: lib/common/options.c:334 +#: lib/common/options.c:344 #, fuzzy msgid "Whether to fence unseen nodes at start-up" msgstr "*** 仅高级使用 *** 是否在启动时fence不可见节点" -#: lib/common/options.c:335 +#: lib/common/options.c:345 #, fuzzy msgid "" "Setting this to false may lead to a \"split-brain\" situation, potentially " "leading to data loss and/or service unavailability." msgstr "" "将此设置为 false 可能会导致 \"split-brain\" 的情况,可能导致数据丢失和(或)服" "务不可用。" -#: lib/common/options.c:342 +#: lib/common/options.c:352 msgid "" "Apply fencing delay targeting the lost nodes with the highest total resource " "priority" msgstr "针对具有最高总资源优先级的丢失节点应用fencing延迟" -#: lib/common/options.c:344 +#: lib/common/options.c:354 msgid "" "Apply specified delay for the fencings that are targeting the lost nodes " "with the highest total resource priority in case we don't have the majority " "of the nodes in our cluster partition, so that the more significant nodes " "potentially win any fencing match, which is especially meaningful under " "split-brain of 2-node cluster. A promoted resource instance takes the base " "priority + 1 on calculation if the base priority is not 0. Any static/random " "delays that are introduced by `pcmk_delay_base/max` configured for the " "corresponding fencing resources will be added to this delay. This delay " "should be significantly greater than, safely twice, the maximum " "`pcmk_delay_base/max`. By default, priority fencing delay is disabled." msgstr "" "如果我们所在的集群分区并不拥有大多数集群节点,则针对丢失节点的fence操作应用指" "定的延迟,这样更重要的节点就能够赢得fence竞赛。这对于双节点集群在split-brain" "状况下尤其有意义。如果基本优先级不为0,在计算时主资源实例获得基本优先级+1。任" "何对于相应的 fence 资源由 pcmk_delay_base/max 配置所引入的静态/随机延迟会被添" "加到此延迟。为了安全, 这个延迟应该明显大于 pcmk_delay_base/max 的最大设置值," "例如两倍。默认情况下,优先级fencing延迟已禁用。" -#: lib/common/options.c:361 +#: lib/common/options.c:371 msgid "" "How long to wait for a node that has joined the cluster to join the " "controller process group" msgstr "等待已加入集群的节点加入控制器进程组的时间" -#: lib/common/options.c:363 +#: lib/common/options.c:373 msgid "" "Fence nodes that do not join the controller process group within this much " "time after joining the cluster, to allow the cluster to continue managing " "resources. A value of 0 means never fence pending nodes. Setting the value " "to 2h means fence nodes after 2 hours." msgstr "" "如果节点加入集群后在此时间内不加入控制器进程组,Fence该节点,以便群集继续管理" "资源。值为0表示永远不 fence 待定节点。将值设置为2h表示2小时后 fence 待定节" "点。" -#: lib/common/options.c:373 +#: lib/common/options.c:383 msgid "Maximum time for node-to-node communication" msgstr "最大节点间通信时间" -#: lib/common/options.c:374 +#: lib/common/options.c:384 msgid "" "The node elected Designated Controller (DC) will consider an action failed " "if it does not get a response from the node executing the action within this " "time (after considering the action's own timeout). The \"correct\" value " "will depend on the speed and load of your network and cluster nodes." msgstr "" "如果一个操作未在该时间内(并且考虑操作本身的超时时长)从执行该操作的节点获得" "响应,则会被选为指定控制器(DC)的节点认定为失败。\"正确\" 值将取决于速度和您" "的网络和集群节点的负载。" -#: lib/common/options.c:386 +#: lib/common/options.c:396 msgid "Maximum amount of system load that should be used by cluster nodes" msgstr "集群节点应该使用的最大系统负载量" -#: lib/common/options.c:388 +#: lib/common/options.c:398 msgid "" "The cluster will slow down its recovery process when the amount of system " "resources used (currently CPU) approaches this limit" msgstr "当使用的系统资源量(当前指 CPU)接近此限制时, 集群将减慢其恢复过程" -#: lib/common/options.c:395 +#: lib/common/options.c:405 msgid "" "Maximum number of jobs that can be scheduled per node (defaults to 2x cores)" msgstr "每个节点可以调度的最大作业数(默认为2x内核数)" -#: lib/common/options.c:403 +#: lib/common/options.c:413 #, fuzzy msgid "" "Maximum number of jobs that the cluster may execute in parallel across all " "nodes" msgstr "集群可以在所有节点上并发执行的最大作业数" -#: lib/common/options.c:405 +#: lib/common/options.c:415 msgid "" "The \"correct\" value will depend on the speed and load of your network and " "cluster nodes. If set to 0, the cluster will impose a dynamically calculated " "limit when any node has a high load." msgstr "" "\"正确\" 值将取决于速度和您的网络与集群节点的负载。如果设置为0,当任何节点具" "有高负载时,集群将施加一个动态计算的限制。" -#: lib/common/options.c:414 +#: lib/common/options.c:424 msgid "" "The number of live migration actions that the cluster is allowed to execute " "in parallel on a node (-1 means no limit)" msgstr "允许集群在一个节点上并行执行的实时迁移操作的数量(-1表示没有限制)" -#: lib/common/options.c:433 +#: lib/common/options.c:443 msgid "Maximum IPC message backlog before disconnecting a cluster daemon" msgstr "断开集群守护程序之前的最大IPC消息积压" -#: lib/common/options.c:434 +#: lib/common/options.c:444 msgid "" "Raise this if log has \"Evicting client\" messages for cluster daemon PIDs " "(a good value is the number of resources in the cluster multiplied by the " "number of nodes)." msgstr "" "如果日志中有针对集群守护程序PID的消息“Evicting client”,(则建议将值设为集群" "中的资源数量乘以节点数量)" -#: lib/common/options.c:444 +#: lib/common/options.c:454 #, fuzzy msgid "Whether the cluster should stop all active resources" msgstr "集群是否在启动期间检查运行资源" -#: lib/common/options.c:451 +#: lib/common/options.c:461 msgid "Whether to stop resources that were removed from the configuration" msgstr "是否停止配置已被删除的资源" -#: lib/common/options.c:459 +#: lib/common/options.c:469 msgid "Whether to cancel recurring actions removed from the configuration" msgstr "是否取消配置已被删除的的重复操作" -#: lib/common/options.c:467 +#: lib/common/options.c:477 #, fuzzy msgid "Whether to remove stopped resources from the executor" msgstr "是否从pacemaker-execd 守护进程中清除已停止的资源" -#: lib/common/options.c:468 +#: lib/common/options.c:478 #, fuzzy msgid "Values other than default are poorly tested and potentially dangerous." msgstr "非默认值未经过充分的测试,有潜在的风险。该选项将在未来的版本中删除。" -#: lib/common/options.c:477 +#: lib/common/options.c:487 msgid "The number of scheduler inputs resulting in errors to save" msgstr "保存导致错误的调度程序输入的数量" -#: lib/common/options.c:478 lib/common/options.c:485 lib/common/options.c:492 +#: lib/common/options.c:488 lib/common/options.c:495 lib/common/options.c:502 msgid "Zero to disable, -1 to store unlimited." msgstr "零表示禁用,-1表示存储不受限制。" -#: lib/common/options.c:484 +#: lib/common/options.c:494 msgid "The number of scheduler inputs resulting in warnings to save" msgstr "保存导致警告的调度程序输入的数量" -#: lib/common/options.c:491 +#: lib/common/options.c:501 msgid "The number of scheduler inputs without errors or warnings to save" msgstr "保存没有错误或警告的调度程序输入的数量" -#: lib/common/options.c:503 +#: lib/common/options.c:513 #, fuzzy msgid "How cluster should react to node health attributes" msgstr "集群节点对节点健康属性如何反应" -#: lib/common/options.c:504 +#: lib/common/options.c:514 msgid "" "Requires external entities to create node attributes (named with the prefix " "\"#health\") with values \"red\", \"yellow\", or \"green\"." msgstr "" "需要外部实体创建具有“red”,“yellow”或“green”值的节点属性(前缀为“#health”)" -#: lib/common/options.c:512 +#: lib/common/options.c:522 msgid "Base health score assigned to a node" msgstr "分配给节点的基本健康分数" -#: lib/common/options.c:513 +#: lib/common/options.c:523 msgid "Only used when \"node-health-strategy\" is set to \"progressive\"." msgstr "仅在“node-health-strategy”设置为“progressive”时使用。" -#: lib/common/options.c:520 +#: lib/common/options.c:530 msgid "The score to use for a node health attribute whose value is \"green\"" msgstr "为节点健康属性值为“green”所使用的分数" -#: lib/common/options.c:522 lib/common/options.c:531 lib/common/options.c:540 +#: lib/common/options.c:532 lib/common/options.c:541 lib/common/options.c:550 msgid "" -"Only used when \"node-health-strategy\" is set to \"custom\" or \"progressive" -"\"." +"Only used when \"node-health-strategy\" is set to \"custom\" or " +"\"progressive\"." msgstr "仅在“node-health-strategy”设置为“custom”或“progressive”时使用。" -#: lib/common/options.c:529 +#: lib/common/options.c:539 msgid "The score to use for a node health attribute whose value is \"yellow\"" msgstr "为节点健康属性值为“yellow”所使用的分数" -#: lib/common/options.c:538 +#: lib/common/options.c:548 msgid "The score to use for a node health attribute whose value is \"red\"" msgstr "为节点健康属性值为“red”所使用的分数" -#: lib/common/options.c:551 +#: lib/common/options.c:561 #, fuzzy msgid "How the cluster should allocate resources to nodes" msgstr "集群应该如何分配资源到节点" -#: lib/common/options.c:569 +#: lib/common/options.c:579 #, fuzzy msgid "An alternate parameter to supply instead of 'port'" msgstr "用于替代 'port' 的其它参数" -#: lib/common/options.c:570 +#: lib/common/options.c:580 #, fuzzy msgid "" "Some devices do not support the standard 'port' parameter or may provide " "additional ones. Use this to specify an alternate, device-specific, " "parameter that should indicate the machine to be fenced. A value of \"none\" " "can be used to tell the cluster not to supply any additional parameters." msgstr "" "一些设备不支持标准的 'port' 参数, 或者可能会提供其它的参数. 使用此选项可指定" "一个替代的, 该设备专用的参数, 该参数应该指出需要 fence 的机器. 可以使用 " "\"none\" 值用于告诉集群不要提供任何其它的参数. " -#: lib/common/options.c:580 +#: lib/common/options.c:590 #, fuzzy msgid "" "A mapping of node names to port numbers for devices that do not support node " "names." msgstr "为不支持主机名的设备提供主机名到端口号的映射. " -#: lib/common/options.c:582 +#: lib/common/options.c:592 #, fuzzy msgid "" "For example, \"node1:1;node2:2,3\" would tell the cluster to use port 1 for " "node1 and ports 2 and 3 for node2." msgstr "" "例如, \"node1:1;node2:2,3\" 将会告诉集群对node1使用端口1, 对node2使用端口2和" "3." -#: lib/common/options.c:589 +#: lib/common/options.c:599 msgid "Nodes targeted by this device" msgstr "此设备针对的节点" -#: lib/common/options.c:590 +#: lib/common/options.c:600 #, fuzzy msgid "" "Comma-separated list of nodes that can be targeted by this device (for " "example, \"node1,node2,node3\"). If pcmk_host_check is \"static-list\", " "either this or pcmk_host_map must be set." msgstr "" "此设备可以针对的节点列表,节点之间用逗号分隔(例如,node1,node2, node3).如果" "pcmk_host_list=\"static-list\")" -#: lib/common/options.c:600 +#: lib/common/options.c:610 #, fuzzy msgid "How to determine which nodes can be targeted by the device" msgstr "如何确定设备可以针对哪些节点" -#: lib/common/options.c:601 +#: lib/common/options.c:611 #, fuzzy msgid "" "Use \"dynamic-list\" to query the device via the 'list' command; \"static-" "list\" to check the pcmk_host_list attribute; \"status\" to query the device " "via the 'status' command; or \"none\" to assume every device can fence every " "node. The default value is \"static-list\" if pcmk_host_map or " "pcmk_host_list is set; otherwise \"dynamic-list\" if the device supports the " "list operation; otherwise \"status\" if the device supports the status " "operation; otherwise \"none\"" msgstr "" -"选项值 \"dynamic-list\" 表示通过 'list' 命令查询设备; 选项值 \"static-list" -"\"表示检查 pcmk_host_list 属性; 选项值 \"status\" 表示通过 'status' 命令查询" -"设备; 或使用选项值 \"none\" 假设每个设备都可以 fence 所有节点. 如果" +"选项值 \"dynamic-list\" 表示通过 'list' 命令查询设备; 选项值 \"static-" +"list\"表示检查 pcmk_host_list 属性; 选项值 \"status\" 表示通过 'status' 命令" +"查询设备; 或使用选项值 \"none\" 假设每个设备都可以 fence 所有节点. 如果" "\"pcmk_host_map\"或\"pcmk_host_list\"被设置,默认值为\"static-list\";否则," "如果设备支持列表操作,则为\"dynamic-list\";如果设备支持状态操作,则为" "\"status\";否则为\"none\"" -#: lib/common/options.c:614 +#: lib/common/options.c:624 msgid "" "Enable a delay of no more than the time specified before executing fencing " "actions." msgstr "在执行 fence 操作前启用不超过指定时间的延迟" -#: lib/common/options.c:616 +#: lib/common/options.c:626 msgid "" "Enable a delay of no more than the time specified before executing fencing " "actions. Pacemaker derives the overall delay by taking the value of " "pcmk_delay_base and adding a random delay value such that the sum is kept " "below this maximum." msgstr "" "在执行 fence 操作前启用不超过指定时间的延迟. Pacemaker通过获取 " "pcmk_delay_base 的值并添加随机延迟值来得出总延迟, 并且确保总和不超过此最大值." -#: lib/common/options.c:625 +#: lib/common/options.c:635 msgid "Enable a base delay for fencing actions and specify base delay value." msgstr "为 fence 操作启用一个指定的基础延迟. " -#: lib/common/options.c:627 +#: lib/common/options.c:637 #, fuzzy msgid "" "This enables a static delay for fencing actions, which can help avoid " "\"death matches\" where two nodes try to fence each other at the same time. " "If pcmk_delay_max is also used, a random delay will be added such that the " "total delay is kept below that value. This can be set to a single time value " "to apply to any node targeted by this device (useful if a separate device is " "configured for each target), or to a node map (for example, \"node1:1s;" "node2:5\") to set a different value for each target." msgstr "" "这为 fence 操作启用一个静态延迟, 这有助于避免 \"death matches\" 即两个节点同" "时尝试互相 fence. 如果还同时使用了pcmk_delay_max, 则会添加一个随机延迟, 并确" "保总延迟保持在该值以下. 可以将其设置为单个时间值, 以应用于该设备的所有目标节" "点 (如果为每个目标节点都配置了单独的设备的情况下, 这很有用) 或设置成一个节点" "映射形式 (例如,\"node1:1s;node2:5\") 从而为每个目标节点设置不同值. " -#: lib/common/options.c:640 +#: lib/common/options.c:650 msgid "" "The maximum number of actions can be performed in parallel on this device" msgstr "可以在该设备上并发执行的最多操作数量" -#: lib/common/options.c:642 +#: lib/common/options.c:652 #, fuzzy msgid "" "Cluster property concurrent-fencing=\"true\" needs to be configured first. " "Then use this to specify the maximum number of actions can be performed in " "parallel on this device. A value of -1 means an unlimited number of actions " "can be performed in parallel." msgstr "" "需要先配置集群属性 concurrent-fencing=\"true\". 然后使用此参数指定可以在该设" "备上并发执行的最多操作数量. -1 表示可以并行执行无限数量的操作. " -#: lib/common/options.c:652 +#: lib/common/options.c:662 #, fuzzy msgid "An alternate command to run instead of 'reboot'" msgstr "运行替代命令,而不是'reboot'" -#: lib/common/options.c:653 +#: lib/common/options.c:663 #, fuzzy msgid "" "Some devices do not support the standard commands or may provide additional " "ones. Use this to specify an alternate, device-specific, command that " "implements the 'reboot' action." msgstr "" "一些设备不支持标准命令或可能提供其他命令,使用此选项可以指定一个该设备特定的" "替代命令,用来实现'reboot'操作。" -#: lib/common/options.c:661 +#: lib/common/options.c:671 #, fuzzy msgid "" "Specify an alternate timeout to use for 'reboot' actions instead of stonith-" "timeout" msgstr "指定用于'reboot' 操作的替代超时,而不是stonith-timeout" -#: lib/common/options.c:663 +#: lib/common/options.c:673 #, fuzzy msgid "" "Some devices need much more/less time to complete than normal. Use this to " "specify an alternate, device-specific, timeout for 'reboot' actions." msgstr "" "一些设备需要比正常情况下更多或更少的时间来完成操作,使用此选项指定一个用" "于'reboot'操作的该设备特定的替代超时。" -#: lib/common/options.c:671 +#: lib/common/options.c:681 #, fuzzy msgid "" "The maximum number of times to try the 'reboot' command within the timeout " "period" msgstr "在超时前重试'reboot'命令的最大次数" -#: lib/common/options.c:673 +#: lib/common/options.c:683 #, fuzzy msgid "" "Some devices do not support multiple connections. Operations may \"fail\" if " "the device is busy with another task. In that case, Pacemaker will " "automatically retry the operation if there is time remaining. Use this " "option to alter the number of times Pacemaker tries a 'reboot' action before " "giving up." msgstr "" "一些设备不支持多个连接。 如果设备忙于另一个任务,则操作可能会'失败' ,因此" "Pacemaker将自动重试(如果时间允许)。 使用此选项更改Pacemaker在放弃之前重" "试'reboot' 操作的次数." -#: lib/common/options.c:683 +#: lib/common/options.c:693 #, fuzzy msgid "An alternate command to run instead of 'off'" msgstr "运行替代命令,而不是'off'" -#: lib/common/options.c:684 +#: lib/common/options.c:694 #, fuzzy msgid "" "Some devices do not support the standard commands or may provide additional " "ones. Use this to specify an alternate, device-specific, command that " "implements the 'off' action." msgstr "" "一些设备不支持标准命令或可能提供其他命令,使用此选项可指定一个该设备专用的替代" "命令,用来实现'off'操作。" -#: lib/common/options.c:692 +#: lib/common/options.c:702 #, fuzzy msgid "" "Specify an alternate timeout to use for 'off' actions instead of stonith-" "timeout" msgstr "指定用于off 操作的替代超时,而不是stonith-timeout" -#: lib/common/options.c:694 +#: lib/common/options.c:704 #, fuzzy msgid "" "Some devices need much more/less time to complete than normal. Use this to " "specify an alternate, device-specific, timeout for 'off' actions." msgstr "" "一些设备需要比正常情况下更多或更少的时间来完成操作,使用此选项指定一个用" "于'off'操作的该设备特定的替代超时。" -#: lib/common/options.c:702 +#: lib/common/options.c:712 #, fuzzy msgid "" "The maximum number of times to try the 'off' command within the timeout " "period" msgstr "在超时前重试'off'命令的最大次数" -#: lib/common/options.c:704 +#: lib/common/options.c:714 #, fuzzy msgid "" "Some devices do not support multiple connections. Operations may \"fail\" if " "the device is busy with another task. In that case, Pacemaker will " "automatically retry the operation if there is time remaining. Use this " "option to alter the number of times Pacemaker tries a 'off' action before " "giving up." msgstr "" " 一些设备不支持多个连接。 如果设备忙于另一个任务,则操作可能会'失败' , 因此" "Pacemaker将自动重试(如果时间允许)。 使用此选项更改Pacemaker在放弃之前重" "试'off' 操作的次数." -#: lib/common/options.c:714 +#: lib/common/options.c:724 #, fuzzy msgid "An alternate command to run instead of 'on'" msgstr "仅高级使用:运行替代命令,而不是'on'" -#: lib/common/options.c:715 +#: lib/common/options.c:725 #, fuzzy msgid "" "Some devices do not support the standard commands or may provide additional " "ones. Use this to specify an alternate, device-specific, command that " "implements the 'on' action." msgstr "" "一些设备不支持标准命令或可能提供其他命令,使用此选项可指定一个该设备特定的替" "代命令,用来实现'on'操作。" -#: lib/common/options.c:723 +#: lib/common/options.c:733 #, fuzzy msgid "" "Specify an alternate timeout to use for 'on' actions instead of stonith-" "timeout" msgstr "指定用于on 操作的替代超时,而不是stonith-timeout" -#: lib/common/options.c:725 +#: lib/common/options.c:735 #, fuzzy msgid "" "Some devices need much more/less time to complete than normal. Use this to " "specify an alternate, device-specific, timeout for 'on' actions." msgstr "" "一些设备需要比正常情况下更多或更少的时间来完成操作,使用此选项指定一个用" "于'on'操作的该设备特定的替代超时。" -#: lib/common/options.c:733 +#: lib/common/options.c:743 #, fuzzy msgid "" "The maximum number of times to try the 'on' command within the timeout period" msgstr "在超时前重试'on'命令的最大次数" -#: lib/common/options.c:735 +#: lib/common/options.c:745 #, fuzzy msgid "" "Some devices do not support multiple connections. Operations may \"fail\" if " "the device is busy with another task. In that case, Pacemaker will " "automatically retry the operation if there is time remaining. Use this " "option to alter the number of times Pacemaker tries a 'on' action before " "giving up." msgstr "" " 一些设备不支持多个连接。 如果设备忙于另一个任务,则操作可能会'失败' , 因此" "Pacemaker将自动重试(如果时间允许)。 使用此选项更改Pacemaker在放弃之前重" "试'on' 操作的次数." -#: lib/common/options.c:745 +#: lib/common/options.c:755 #, fuzzy msgid "An alternate command to run instead of 'list'" msgstr "运行替代命令,而不是'list'" -#: lib/common/options.c:746 +#: lib/common/options.c:756 #, fuzzy msgid "" "Some devices do not support the standard commands or may provide additional " "ones. Use this to specify an alternate, device-specific, command that " "implements the 'list' action." msgstr "" "一些设备不支持标准命令或可能提供其他命令,使用此选项可指定一个该设备特定的替" "代命令,用来实现'list'操作。" -#: lib/common/options.c:754 +#: lib/common/options.c:764 #, fuzzy msgid "" "Specify an alternate timeout to use for 'list' actions instead of stonith-" "timeout" msgstr "指定用于list 操作的替代超时,而不是stonith-timeout" -#: lib/common/options.c:756 +#: lib/common/options.c:766 #, fuzzy msgid "" "Some devices need much more/less time to complete than normal. Use this to " "specify an alternate, device-specific, timeout for 'list' actions." msgstr "" "一些设备需要比正常情况下更多或更少的时间来完成操作,使用此选项指定一个用" "于'list'操作的该设备特定的替代超时。" -#: lib/common/options.c:764 +#: lib/common/options.c:774 #, fuzzy msgid "" "The maximum number of times to try the 'list' command within the timeout " "period" msgstr "在超时前重试'list'命令的最大次数" -#: lib/common/options.c:766 +#: lib/common/options.c:776 #, fuzzy msgid "" "Some devices do not support multiple connections. Operations may \"fail\" if " "the device is busy with another task. In that case, Pacemaker will " "automatically retry the operation if there is time remaining. Use this " "option to alter the number of times Pacemaker tries a 'list' action before " "giving up." msgstr "" " 一些设备不支持多个连接。 如果设备忙于另一个任务,则操作可能会'失败' , 因此" "Pacemaker将自动重试(如果时间允许)。 使用此选项更改Pacemaker在放弃之前重" "试'list' 操作的次数." -#: lib/common/options.c:776 +#: lib/common/options.c:786 #, fuzzy msgid "An alternate command to run instead of 'monitor'" msgstr "运行替代命令,而不是'monitor'" -#: lib/common/options.c:777 +#: lib/common/options.c:787 #, fuzzy msgid "" "Some devices do not support the standard commands or may provide additional " "ones. Use this to specify an alternate, device-specific, command that " "implements the 'monitor' action." msgstr "" "一些设备不支持标准命令或可能提供其他命令,使用此选项可指定一个该设备特定的替" "代命令,用来实现'monitor'操作。" -#: lib/common/options.c:785 +#: lib/common/options.c:795 #, fuzzy msgid "" "Specify an alternate timeout to use for 'monitor' actions instead of stonith-" "timeout" msgstr "指定用于monitor 操作的替代超时,而不是stonith-timeout" -#: lib/common/options.c:787 +#: lib/common/options.c:797 #, fuzzy msgid "" "Some devices need much more/less time to complete than normal. Use this to " "specify an alternate, device-specific, timeout for 'monitor' actions." msgstr "" "一些设备需要比正常情况下更多或更少的时间来完成操作,使用此选项指定一个用" "于'monitor'操作的该设备特定的替代超时。" -#: lib/common/options.c:795 +#: lib/common/options.c:805 #, fuzzy msgid "" "The maximum number of times to try the 'monitor' command within the timeout " "period" msgstr "在超时前重试'monitor'命令的最大次数" -#: lib/common/options.c:797 +#: lib/common/options.c:807 #, fuzzy msgid "" "Some devices do not support multiple connections. Operations may \"fail\" if " "the device is busy with another task. In that case, Pacemaker will " "automatically retry the operation if there is time remaining. Use this " "option to alter the number of times Pacemaker tries a 'monitor' action " "before giving up." msgstr "" " 一些设备不支持多个连接。 如果设备忙于另一个任务,则操作可能会'失败' , 因此" "Pacemaker将自动重试(如果时间允许)。 使用此选项更改Pacemaker在放弃之前重" "试'monitor' 操作的次数." -#: lib/common/options.c:807 +#: lib/common/options.c:817 #, fuzzy msgid "An alternate command to run instead of 'status'" msgstr "运行替代命令,而不是'status'" -#: lib/common/options.c:808 +#: lib/common/options.c:818 #, fuzzy msgid "" "Some devices do not support the standard commands or may provide additional " "ones. Use this to specify an alternate, device-specific, command that " "implements the 'status' action." msgstr "" "一些设备不支持标准命令或可能提供其他命令,使用此选项可指定一个该设备特定的替" "代命令,用来实现'status'操作。" -#: lib/common/options.c:816 +#: lib/common/options.c:826 #, fuzzy msgid "" "Specify an alternate timeout to use for 'status' actions instead of stonith-" "timeout" msgstr "指定用于status 操作的替代超时,而不是stonith-timeout" -#: lib/common/options.c:818 +#: lib/common/options.c:828 #, fuzzy msgid "" "Some devices need much more/less time to complete than normal. Use this to " "specify an alternate, device-specific, timeout for 'status' actions." msgstr "" "一些设备需要比正常情况下更多或更少的时间来完成操作,使用此选项指定一个用" "于'status'操作的该设备特定的替代超时" -#: lib/common/options.c:826 +#: lib/common/options.c:836 #, fuzzy msgid "" "The maximum number of times to try the 'status' command within the timeout " "period" msgstr "仅高级使用:在超时前重试'status'命令的最大次数" -#: lib/common/options.c:828 +#: lib/common/options.c:838 #, fuzzy msgid "" "Some devices do not support multiple connections. Operations may \"fail\" if " "the device is busy with another task. In that case, Pacemaker will " "automatically retry the operation if there is time remaining. Use this " "option to alter the number of times Pacemaker tries a 'status' action before " "giving up." msgstr "" " 一些设备不支持多个连接。 如果设备忙于另一个任务,则操作可能会'失败' , 因此" "Pacemaker将自动重试(如果时间允许)。 使用此选项更改Pacemaker在放弃之前重" "试'status' 操作的次数." -#: lib/common/options.c:849 +#: lib/common/options.c:859 msgid "Resource assignment priority" msgstr "" -#: lib/common/options.c:850 +#: lib/common/options.c:860 msgid "" "If not all resources can be active, the cluster will stop lower-priority " "resources in order to keep higher-priority ones active." msgstr "" -#: lib/common/options.c:858 +#: lib/common/options.c:868 msgid "Default value for influence in colocation constraints" msgstr "" -#: lib/common/options.c:859 +#: lib/common/options.c:869 msgid "" "Use this value as the default for influence in all colocation constraints " "involving this resource, as well as in the implicit colocation constraints " "created if this resource is in a group." msgstr "" -#: lib/common/options.c:869 +#: lib/common/options.c:879 #, fuzzy msgid "State the cluster should attempt to keep this resource in" msgstr "集群是否在启动期间检查运行资源" -#: lib/common/options.c:870 +#: lib/common/options.c:880 msgid "" "\"Stopped\" forces the resource to be stopped. \"Started\" allows the " "resource to be started (and in the case of promotable clone resources, " "promoted if appropriate). \"Unpromoted\" allows the resource to be started, " "but only in the unpromoted role if the resource is promotable. \"Promoted\" " "is equivalent to \"Started\"." msgstr "" -#: lib/common/options.c:881 +#: lib/common/options.c:891 #, fuzzy msgid "Whether the cluster is allowed to actively change the resource's state" msgstr "集群是否在启动期间检查运行资源" -#: lib/common/options.c:883 +#: lib/common/options.c:893 msgid "" "If false, the cluster will not start, stop, promote, or demote the resource " "on any node. Recurring actions for the resource are unaffected. If true, a " "true value for the maintenance-mode cluster option, the maintenance node " "attribute, or the maintenance resource meta-attribute overrides this." msgstr "" -#: lib/common/options.c:893 +#: lib/common/options.c:903 msgid "" "If true, the cluster will not schedule any actions involving the resource" msgstr "" -#: lib/common/options.c:895 +#: lib/common/options.c:905 msgid "" "If true, the cluster will not start, stop, promote, or demote the resource " "on any node, and will pause any recurring monitors (except those specifying " "role as \"Stopped\"). If false, a true value for the maintenance-mode " "cluster option or maintenance node attribute overrides this." msgstr "" -#: lib/common/options.c:905 +#: lib/common/options.c:915 msgid "Score to add to the current node when a resource is already active" msgstr "" -#: lib/common/options.c:907 +#: lib/common/options.c:917 msgid "" "Score to add to the current node when a resource is already active. This " "allows running resources to stay where they are, even if they would be " "placed elsewhere if they were being started from a stopped state. The " "default is 1 for individual clone instances, and 0 for all other resources." msgstr "" -#: lib/common/options.c:920 +#: lib/common/options.c:930 msgid "Conditions under which the resource can be started" msgstr "" -#: lib/common/options.c:921 +#: lib/common/options.c:931 msgid "" "Conditions under which the resource can be started. \"nothing\" means the " "cluster can always start this resource. \"quorum\" means the cluster can " "start this resource only if a majority of the configured nodes are active. " "\"fencing\" means the cluster can start this resource only if a majority of " "the configured nodes are active and any failed or unknown nodes have been " "fenced. \"unfencing\" means the cluster can start this resource only if a " "majority of the configured nodes are active and any failed or unknown nodes " "have been fenced, and only on nodes that have been unfenced. The default is " "\"quorum\" for resources with a class of stonith; otherwise, \"unfencing\" " "if unfencing is active in the cluster; otherwise, \"fencing\" if the stonith-" "enabled cluster option is true; otherwise, \"quorum\"." msgstr "" -#: lib/common/options.c:942 +#: lib/common/options.c:952 msgid "" "Number of failures on a node before the resource becomes ineligible to run " "there." msgstr "" -#: lib/common/options.c:944 +#: lib/common/options.c:954 msgid "" "Number of failures that may occur for this resource on a node, before that " "node is marked ineligible to host this resource. A value of 0 indicates that " "this feature is disabled (the node will never be marked ineligible). By " "contrast, the cluster treats \"INFINITY\" (the default) as a very large but " "finite number. This option has an effect only if the failed operation " "specifies its on-fail attribute as \"restart\" (the default), and " "additionally for failed start operations, if the start-failure-is-fatal " "cluster property is set to false." msgstr "" -#: lib/common/options.c:958 +#: lib/common/options.c:968 msgid "Number of seconds before acting as if a failure had not occurred" msgstr "" -#: lib/common/options.c:959 +#: lib/common/options.c:969 msgid "" "Number of seconds after a failed action for this resource before acting as " "if the failure had not occurred, and potentially allowing the resource back " "to the node on which it failed. A value of 0 indicates that this feature is " "disabled." msgstr "" -#: lib/common/options.c:970 +#: lib/common/options.c:980 msgid "" "What to do if the cluster finds the resource active on more than one node" msgstr "" -#: lib/common/options.c:972 +#: lib/common/options.c:982 msgid "" "What to do if the cluster finds the resource active on more than one node. " "\"block\" means to mark the resource as unmanaged. \"stop_only\" means to " "stop all active instances of this resource and leave them stopped. " "\"stop_start\" means to stop all active instances of this resource and start " "the resource in one location only. \"stop_unexpected\" means to stop all " "active instances of this resource except where the resource should be " "active. (This should be used only when extra instances are not expected to " "disrupt existing instances, and the resource agent's monitor of an existing " "instance is capable of detecting any problems that could be caused. Note " "that any resources ordered after this one will still need to be restarted.)" msgstr "" -#: lib/common/options.c:991 +#: lib/common/options.c:1001 #, fuzzy msgid "" "Whether the cluster should try to \"live migrate\" this resource when it " "needs to be moved" msgstr "集群是否在启动期间检查运行资源" -#: lib/common/options.c:993 +#: lib/common/options.c:1003 msgid "" "Whether the cluster should try to \"live migrate\" this resource when it " "needs to be moved. The default is true for ocf:pacemaker:remote resources, " "and false otherwise." msgstr "" -#: lib/common/options.c:1002 +#: lib/common/options.c:1012 msgid "" "Whether the resource should be allowed to run on a node even if the node's " "health score would otherwise prevent it" msgstr "" -#: lib/common/options.c:1010 +#: lib/common/options.c:1020 #, fuzzy msgid "Where to check user-defined node attributes" msgstr "*** 仅高级使用 *** 是否在启动时fence不可见节点" -#: lib/common/options.c:1011 +#: lib/common/options.c:1021 msgid "" "Whether to check user-defined node attributes on the physical host where a " "container is running or on the local node. This is usually set for a bundle " -"resource and inherited by the bundle's primitive resource. A value of \"host" -"\" means to check user-defined node attributes on the underlying physical " -"host. Any other value means to check user-defined node attributes on the " -"local node (for a bundled primitive resource, this is the bundle node)." +"resource and inherited by the bundle's primitive resource. A value of " +"\"host\" means to check user-defined node attributes on the underlying " +"physical host. Any other value means to check user-defined node attributes " +"on the local node (for a bundled primitive resource, this is the bundle " +"node)." msgstr "" -#: lib/common/options.c:1024 +#: lib/common/options.c:1034 msgid "" "Name of the Pacemaker Remote guest node this resource is associated with, if " "any" msgstr "" -#: lib/common/options.c:1026 +#: lib/common/options.c:1036 msgid "" "Name of the Pacemaker Remote guest node this resource is associated with, if " "any. If specified, this both enables the resource as a guest node and " "defines the unique name used to identify the guest node. The guest must be " "configured to run the Pacemaker Remote daemon when it is started. WARNING: " "This value cannot overlap with any resource or node IDs." msgstr "" -#: lib/common/options.c:1038 +#: lib/common/options.c:1048 msgid "" "If remote-node is specified, the IP address or hostname used to connect to " "the guest via Pacemaker Remote" msgstr "" -#: lib/common/options.c:1040 +#: lib/common/options.c:1050 msgid "" "If remote-node is specified, the IP address or hostname used to connect to " "the guest via Pacemaker Remote. The Pacemaker Remote daemon on the guest " "must be configured to accept connections on this address. The default is the " "value of the remote-node meta-attribute." msgstr "" -#: lib/common/options.c:1050 +#: lib/common/options.c:1060 msgid "" "If remote-node is specified, port on the guest used for its Pacemaker Remote " "connection" msgstr "" -#: lib/common/options.c:1052 +#: lib/common/options.c:1062 msgid "" "If remote-node is specified, the port on the guest used for its Pacemaker " "Remote connection. The Pacemaker Remote daemon on the guest must be " "configured to listen on this port." msgstr "" -#: lib/common/options.c:1060 +#: lib/common/options.c:1070 msgid "" "If remote-node is specified, how long before a pending Pacemaker Remote " "guest connection times out." msgstr "" -#: lib/common/options.c:1068 +#: lib/common/options.c:1078 msgid "" "If remote-node is specified, this acts as the allow-migrate meta-attribute " "for the implicit remote connection resource (ocf:pacemaker:remote)." msgstr "" -#: lib/common/cmdline.c:70 +#: lib/common/cmdline.c:72 msgid "Display software version and exit" msgstr "显示软件版本信息" -#: lib/common/cmdline.c:73 +#: lib/common/cmdline.c:75 msgid "Increase debug output (may be specified multiple times)" msgstr "显示更多调试信息(可多次指定)" -#: lib/common/cmdline.c:92 +#: lib/common/cmdline.c:94 msgid "FORMAT" msgstr "格式" -#: lib/common/cmdline.c:94 +#: lib/common/cmdline.c:96 msgid "Specify file name for output (or \"-\" for stdout)" msgstr "指定输出的文件名 或指定'-' 表示标准输出" -#: lib/common/cmdline.c:94 +#: lib/common/cmdline.c:96 msgid "DEST" msgstr "目标" -#: lib/common/cmdline.c:100 +#: lib/common/cmdline.c:102 msgid "Output Options:" msgstr "输出选项" -#: lib/common/cmdline.c:100 +#: lib/common/cmdline.c:102 msgid "Show output help" msgstr "显示输出帮助" #: tools/crm_resource.c:204 #, c-format msgid "Aborting because no messages received in %d seconds" msgstr "中止,因为在%d秒内没有接收到消息" #: tools/crm_resource.c:374 #, c-format msgid "Invalid check level setting: %s" msgstr "无效的检查级别设置:%s" #: tools/crm_resource.c:891 #, c-format msgid "" "Resource '%s' not moved: active in %d locations (promoted in %d).\n" "To prevent '%s' from running on a specific location, specify a node.To " "prevent '%s' from being promoted at a specific location, specify a node and " "the --promoted option." msgstr "" "资源'%s'未移动:在%d个位置运行(其中在%d个位置为主实例)\n" "若要阻止'%s'在特定位置运行,请指定一个节点。若要防止'%s'在指定位置升级,指定" "一个节点并使用--promoted选项" #: tools/crm_resource.c:902 #, c-format msgid "" "Resource '%s' not moved: active in %d locations.\n" "To prevent '%s' from running on a specific location, specify a node." msgstr "" "资源%s未移动:在%d个位置运行\n" "若要防止'%s'运行在特定位置,指定一个节点" #: tools/crm_resource.c:979 #, c-format msgid "Could not get modified CIB: %s\n" msgstr "无法获得修改的CIB:%s\n" #: tools/crm_resource.c:1077 #, c-format msgid "No cluster connection to Pacemaker Remote node %s detected" msgstr "未检测到至pacemaker远程节点%s的集群连接" #: tools/crm_resource.c:1138 msgid "Must specify -t with resource type" msgstr "需要使用-t指定资源类型" #: tools/crm_resource.c:1144 msgid "Must supply -v with new value" msgstr "必须使用-v指定新值" #: tools/crm_resource.c:1176 msgid "Could not create executor connection" msgstr "无法创建到pacemaker-execd守护进程的连接" #: tools/crm_resource.c:1201 #, fuzzy, c-format msgid "Metadata query for %s failed: %s" msgstr ",查询%s的元数据失败: %s\n" #: tools/crm_resource.c:1207 #, c-format msgid "'%s' is not a valid agent specification" msgstr "'%s' 是一个无效的代理" #: tools/crm_resource.c:1220 msgid "--resource cannot be used with --class, --agent, and --provider" msgstr "--resource 不能与 --class, --agent, --provider一起使用" #: tools/crm_resource.c:1225 msgid "" "--class, --agent, and --provider can only be used with --validate and --" "force-*" msgstr "--class, --agent和--provider只能被用于--validate和--force-*" #: tools/crm_resource.c:1234 msgid "stonith does not support providers" msgstr "stonith 不支持提供者" #: tools/crm_resource.c:1238 #, c-format msgid "%s is not a known stonith agent" msgstr "%s 不是一个已知stonith代理" #: tools/crm_resource.c:1243 #, c-format msgid "%s:%s:%s is not a known resource" msgstr "%s:%s:%s 不是一个已知资源" #: tools/crm_resource.c:1551 #, c-format msgid "Error creating output format %s: %s" msgstr "创建输出格式错误 %s:%s" #: tools/crm_resource.c:1572 msgid "--expired requires --clear or -U" msgstr "--expired需要和--clear或-U一起使用" #: tools/crm_resource.c:1589 #, c-format msgid "Error parsing '%s' as a name=value pair" msgstr "'%s'解析错误,格式为name=value" #: tools/crm_resource.c:1688 msgid "Must supply a resource id with -r" msgstr "必须使用-r指定资源id" #: tools/crm_resource.c:1694 msgid "Must supply a node name with -N" msgstr "必须使用-N指定节点名称" #: tools/crm_resource.c:1708 msgid "Could not create CIB connection" msgstr "无法创建到CIB的连接" #: tools/crm_resource.c:1716 #, c-format msgid "Could not connect to the CIB: %s" msgstr "不能连接到CIB:%s" #: tools/crm_resource.c:1739 #, c-format msgid "Resource '%s' not found" msgstr "没有发现'%s'资源" #: tools/crm_resource.c:1751 #, c-format msgid "Cannot operate on clone resource instance '%s'" msgstr "不能操作克隆资源实例'%s'" #: tools/crm_resource.c:1763 #, c-format msgid "Node '%s' not found" msgstr "没有发现%s节点" #: tools/crm_resource.c:1774 #, c-format msgid "Error connecting to the controller: %s" msgstr "连接到控制器错误:%s" #: tools/crm_resource.c:1783 #, fuzzy, c-format msgid "Error connecting to %s: %s" msgstr "连接到控制器错误:%s" -#: tools/crm_resource.c:2052 +#: tools/crm_resource.c:2055 msgid "You need to supply a value with the -v option" msgstr "需要使用-v选项提供一个值" -#: tools/crm_resource.c:2106 +#: tools/crm_resource.c:2109 msgid "You need to specify a resource type with -t" msgstr "需要使用-t指定资源类型" -#: tools/crm_resource.c:2113 +#: tools/crm_resource.c:2116 #, fuzzy, c-format msgid "Could not delete resource %s: %s" msgstr "无法删除资源:%s:%s" -#: tools/crm_resource.c:2123 +#: tools/crm_resource.c:2126 #, c-format msgid "Unimplemented command: %d" msgstr "无效的命令:%d" -#: tools/crm_resource.c:2153 +#: tools/crm_resource.c:2156 #, c-format msgid "Error performing operation: %s" msgstr "执行操作错误:%s" #, fuzzy #~ msgid "For example, \"node1,node2,node3\"." #~ msgstr "例如, \"node1,node2,node3\"." #, fuzzy #~ msgid "*** Advanced Use Only ***" #~ msgstr "*** Advanced Use Only(仅限高级用户使用) ***" #, fuzzy #~ msgid "" #~ "Zero disables polling, while positive values are an interval in seconds " #~ "(unless other units are specified, for example \"5min\")" #~ msgstr "" #~ "0 表示禁用轮询,而正值表示以秒为单位的时间间隔(除非指定了其他单位, 例如 " #~ "\"5min\" 表示5分钟)" #~ msgid " Allowed values: " #~ msgstr " 允许的值: " #~ msgid "" #~ "This value is not used by Pacemaker, but is kept for backward " #~ "compatibility, and certain legacy fence agents might use it." #~ msgstr "" #~ "Pacemaker不使用此值,但保留此值是为了向后兼容,某些传统的fence 代理可能会" #~ "使用它。" #~ msgid "No agents found for standard '%s'" #~ msgstr "没有发现指定的'%s'标准代理" #, fuzzy #~ msgid "No agents found for standard '%s' and provider '%s'" #~ msgstr "没有发现指定的标准%s和提供者%S的资源代理" #~ msgid "No %s found for %s" #~ msgstr "没有发现%s符合%s" #~ msgid "No %s found" #~ msgstr "没有发现%s" #~ msgid "" #~ "If nonzero, along with `have-watchdog=true` automatically set by the " #~ "cluster, when fencing is required, watchdog-based self-fencing will be " #~ "performed via SBD without requiring a fencing resource explicitly " #~ "configured. If `stonith-watchdog-timeout` is set to a positive value, " #~ "unseen nodes are assumed to self-fence within this much time. +WARNING:+ " #~ "It must be ensured that this value is larger than the " #~ "`SBD_WATCHDOG_TIMEOUT` environment variable on all nodes. Pacemaker " #~ "verifies the settings individually on all nodes and prevents startup or " #~ "shuts down if configured wrongly on the fly. It's strongly recommended " #~ "that `SBD_WATCHDOG_TIMEOUT` is set to the same value on all nodes. If " #~ "`stonith-watchdog-timeout` is set to a negative value, and " #~ "`SBD_WATCHDOG_TIMEOUT` is set, twice that value will be used. +WARNING:+ " #~ "In this case, it's essential (currently not verified by Pacemaker) that " #~ "`SBD_WATCHDOG_TIMEOUT` is set to the same value on all nodes." #~ msgstr "" #~ "如果值非零,且集群设置了 `have-watchdog=true` ,当需要 fence 操作时,基于 " #~ "watchdog 的自我 fence 机制将通过SBD执行,而不需要显式配置 fence 资源。如" #~ "果 `stonith-watchdog-timeout` 被设为正值,则假定不可见的节点在这段时间内自" #~ "我fence。 +WARNING:+ 必须确保该值大于所有节点上的`SBD_WATCHDOG_TIMEOUT` 环" #~ "境变量。Pacemaker将在所有节点上单独验证设置,如发现有错误的动态配置,将防" #~ "止节点启动或关闭。强烈建议在所有节点上将 `SBD_WATCHDOG_TIMEOUT` 设置为相同" #~ "的值。如果 `stonith-watchdog-timeout` 设置为负值。并且设置了 " #~ "`SBD_WATCHDOG_TIMEOUT` ,则将使用该值的两倍, +WARNING:+ 在这种情况下,必" #~ "须将所有节点上 `SBD_WATCHDOG_TIMEOUT` 设置为相同的值(目前没有通过pacemaker" #~ "验证)。" diff --git a/tools/cibadmin.c b/tools/cibadmin.c index 74a142c9f3..c6004ccdce 100644 --- a/tools/cibadmin.c +++ b/tools/cibadmin.c @@ -1,997 +1,1003 @@ /* - * Copyright 2004-2024 the Pacemaker project contributors + * Copyright 2004-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU General Public License version 2 * or later (GPLv2+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #define SUMMARY "query and edit the Pacemaker configuration" #define INDENT " " enum cibadmin_section_type { cibadmin_section_all = 0, cibadmin_section_scope, cibadmin_section_xpath, }; static int request_id = 0; static cib_t *the_cib = NULL; static GMainLoop *mainloop = NULL; static crm_exit_t exit_code = CRM_EX_OK; static struct { const char *cib_action; int cmd_options; enum cibadmin_section_type section_type; char *cib_section; char *validate_with; gint message_timeout_sec; enum pcmk__acl_render_how acl_render_mode; gchar *cib_user; gchar *dest_node; gchar *input_file; gchar *input_xml; gboolean input_stdin; bool delete_all; gboolean allow_create; gboolean force; gboolean get_node_path; gboolean local; gboolean no_children; gboolean score_update; gboolean sync_call; /* @COMPAT: For "-!" version option. Not advertised nor marked as * deprecated, but accepted. */ gboolean extended_version; //! \deprecated gboolean no_bcast; } options; int do_init(void); static int do_work(xmlNode *input, xmlNode **output); void cibadmin_op_callback(xmlNode *msg, int call_id, int rc, xmlNode *output, void *user_data); static void print_xml_output(xmlNode * xml) { if (!xml) { return; } else if (xml->type != XML_ELEMENT_NODE) { return; } if (pcmk_is_set(options.cmd_options, cib_xpath_address)) { const char *id = crm_element_value(xml, PCMK_XA_ID); if (pcmk__xe_is(xml, PCMK__XE_XPATH_QUERY)) { xmlNode *child = NULL; for (child = xml->children; child; child = child->next) { print_xml_output(child); } } else if (id) { printf("%s\n", id); } } else { GString *buf = g_string_sized_new(1024); pcmk__xml_string(xml, pcmk__xml_fmt_pretty, buf, 0); fprintf(stdout, "%s", buf->str); g_string_free(buf, TRUE); } } // Upgrade requested but already at latest schema static void report_schema_unchanged(void) { const char *err = pcmk_rc_str(pcmk_rc_schema_unchanged); crm_info("Upgrade unnecessary: %s\n", err); printf("Upgrade unnecessary: %s\n", err); exit_code = CRM_EX_OK; } /*! * \internal * \brief Check whether the current CIB action is dangerous * \return true if \p options.cib_action is dangerous, or false otherwise */ static inline bool cib_action_is_dangerous(void) { return options.no_bcast || options.delete_all || pcmk__str_any_of(options.cib_action, PCMK__CIB_REQUEST_UPGRADE, PCMK__CIB_REQUEST_ERASE, NULL); } /*! * \internal * \brief Determine whether the given CIB scope is valid for \p cibadmin * * \param[in] scope Scope to validate * * \return true if \p scope is valid, or false otherwise * \note An invalid scope applies the operation to the entire CIB. */ static inline bool scope_is_valid(const char *scope) { return pcmk__str_any_of(scope, PCMK_XE_CONFIGURATION, PCMK_XE_NODES, PCMK_XE_RESOURCES, PCMK_XE_CONSTRAINTS, PCMK_XE_CRM_CONFIG, PCMK_XE_RSC_DEFAULTS, PCMK_XE_OP_DEFAULTS, PCMK_XE_ACLS, PCMK_XE_FENCING_TOPOLOGY, PCMK_XE_TAGS, PCMK_XE_ALERTS, PCMK_XE_STATUS, NULL); } static gboolean command_cb(const gchar *option_name, const gchar *optarg, gpointer data, GError **error) { options.delete_all = false; if (pcmk__str_any_of(option_name, "-u", "--upgrade", NULL)) { options.cib_action = PCMK__CIB_REQUEST_UPGRADE; } else if (pcmk__str_any_of(option_name, "-Q", "--query", NULL)) { options.cib_action = PCMK__CIB_REQUEST_QUERY; } else if (pcmk__str_any_of(option_name, "-E", "--erase", NULL)) { options.cib_action = PCMK__CIB_REQUEST_ERASE; } else if (pcmk__str_any_of(option_name, "-B", "--bump", NULL)) { options.cib_action = PCMK__CIB_REQUEST_BUMP; } else if (pcmk__str_any_of(option_name, "-C", "--create", NULL)) { options.cib_action = PCMK__CIB_REQUEST_CREATE; } else if (pcmk__str_any_of(option_name, "-M", "--modify", NULL)) { options.cib_action = PCMK__CIB_REQUEST_MODIFY; } else if (pcmk__str_any_of(option_name, "-P", "--patch", NULL)) { options.cib_action = PCMK__CIB_REQUEST_APPLY_PATCH; } else if (pcmk__str_any_of(option_name, "-R", "--replace", NULL)) { options.cib_action = PCMK__CIB_REQUEST_REPLACE; } else if (pcmk__str_any_of(option_name, "-D", "--delete", NULL)) { options.cib_action = PCMK__CIB_REQUEST_DELETE; } else if (pcmk__str_any_of(option_name, "-d", "--delete-all", NULL)) { options.cib_action = PCMK__CIB_REQUEST_DELETE; options.delete_all = true; } else if (pcmk__str_any_of(option_name, "-a", "--empty", NULL)) { options.cib_action = "empty"; pcmk__str_update(&options.validate_with, optarg); } else if (pcmk__str_any_of(option_name, "-5", "--md5-sum", NULL)) { options.cib_action = "md5-sum"; } else if (pcmk__str_any_of(option_name, "-6", "--md5-sum-versioned", NULL)) { options.cib_action = "md5-sum-versioned"; } else { // Should be impossible return FALSE; } return TRUE; } static gboolean show_access_cb(const gchar *option_name, const gchar *optarg, gpointer data, GError **error) { if (pcmk__str_eq(optarg, "auto", pcmk__str_null_matches)) { options.acl_render_mode = pcmk__acl_render_default; } else if (g_strcmp0(optarg, "namespace") == 0) { options.acl_render_mode = pcmk__acl_render_namespace; } else if (g_strcmp0(optarg, "text") == 0) { options.acl_render_mode = pcmk__acl_render_text; } else if (g_strcmp0(optarg, "color") == 0) { options.acl_render_mode = pcmk__acl_render_color; } else { g_set_error(error, PCMK__EXITC_ERROR, CRM_EX_USAGE, "Invalid value '%s' for option '%s'", optarg, option_name); return FALSE; } return TRUE; } static gboolean section_cb(const gchar *option_name, const gchar *optarg, gpointer data, GError **error) { if (pcmk__str_any_of(option_name, "-o", "--scope", NULL)) { options.section_type = cibadmin_section_scope; } else if (pcmk__str_any_of(option_name, "-A", "--xpath", NULL)) { options.section_type = cibadmin_section_xpath; } else { // Should be impossible return FALSE; } pcmk__str_update(&options.cib_section, optarg); return TRUE; } static GOptionEntry command_entries[] = { { "upgrade", 'u', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Upgrade the configuration to the latest syntax", NULL }, { "query", 'Q', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Query the contents of the CIB", NULL }, { "erase", 'E', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Erase the contents of the whole CIB", NULL }, { "bump", 'B', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Increase the CIB's epoch value by 1", NULL }, { "create", 'C', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Create an object in the CIB (will fail if object already exists)", NULL }, { "modify", 'M', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Find object somewhere in CIB's XML tree and update it (fails if object " "does not exist unless -c is also specified)", NULL }, { "patch", 'P', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Supply an update in the form of an XML diff (see crm_diff(8))", NULL }, { "replace", 'R', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Recursively replace an object in the CIB", NULL }, { "delete", 'D', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Delete first object matching supplied criteria (for example, " "<" PCMK_XE_OP " " PCMK_XA_ID "=\"rsc1_op1\" " PCMK_XA_NAME "=\"monitor\"/>).\n" INDENT "The XML element name and all attributes must match in order for " "the element to be deleted.", NULL }, { "delete-all", 'd', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "When used with --xpath, remove all matching objects in the " "configuration instead of just the first one", NULL }, { "empty", 'a', G_OPTION_FLAG_OPTIONAL_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Output an empty CIB. Accepts an optional schema name argument to use as " "the " PCMK_XA_VALIDATE_WITH " value.\n" INDENT "If no schema is given, the latest will be used.", "[schema]" }, { "md5-sum", '5', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Calculate the on-disk CIB digest", NULL }, { "md5-sum-versioned", '6', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Calculate an on-the-wire versioned CIB digest", NULL }, { NULL } }; static GOptionEntry data_entries[] = { /* @COMPAT: These arguments should be last-wins. We can have an enum option * that stores the input type, along with a single string option that stores * the XML string for --xml-text, filename for --xml-file, or NULL for * --xml-pipe. */ { "xml-text", 'X', G_OPTION_FLAG_NONE, G_OPTION_ARG_STRING, &options.input_xml, "Retrieve XML from the supplied string", "value" }, { "xml-file", 'x', G_OPTION_FLAG_NONE, G_OPTION_ARG_FILENAME, &options.input_file, "Retrieve XML from the named file", "value" }, { "xml-pipe", 'p', G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &options.input_stdin, "Retrieve XML from stdin", NULL }, { NULL } }; static GOptionEntry addl_entries[] = { { "force", 'f', G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &options.force, "Force the action to be performed", NULL }, { "timeout", 't', G_OPTION_FLAG_NONE, G_OPTION_ARG_INT, &options.message_timeout_sec, "Time (in seconds) to wait before declaring the operation failed", "value" }, { "user", 'U', G_OPTION_FLAG_NONE, G_OPTION_ARG_STRING, &options.cib_user, "Run the command with permissions of the named user (valid only for the " "root and " CRM_DAEMON_USER " accounts)", "value" }, { "sync-call", 's', G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &options.sync_call, "Wait for call to complete before returning", NULL }, { "local", 'l', G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &options.local, "Command takes effect locally (should be used only for queries)", NULL }, { "scope", 'o', G_OPTION_FLAG_NONE, G_OPTION_ARG_CALLBACK, section_cb, "Limit scope of operation to specific section of CIB\n" INDENT "Valid values: " PCMK_XE_CONFIGURATION ", " PCMK_XE_NODES ", " PCMK_XE_RESOURCES ", " PCMK_XE_CONSTRAINTS ", " PCMK_XE_CRM_CONFIG ", " PCMK_XE_RSC_DEFAULTS ",\n" INDENT " " PCMK_XE_OP_DEFAULTS ", " PCMK_XE_ACLS ", " PCMK_XE_FENCING_TOPOLOGY ", " PCMK_XE_TAGS ", " PCMK_XE_ALERTS ", " PCMK_XE_STATUS "\n" INDENT "If both --scope/-o and --xpath/-a are specified, the last one to " "appear takes effect", "value" }, { "xpath", 'A', G_OPTION_FLAG_NONE, G_OPTION_ARG_CALLBACK, section_cb, "A valid XPath to use instead of --scope/-o\n" INDENT "If both --scope/-o and --xpath/-a are specified, the last one to " "appear takes effect", "value" }, { "node-path", 'e', G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &options.get_node_path, "When performing XPath queries, return paths of any matches found\n" INDENT "(for example, " "\"/" PCMK_XE_CIB "/" PCMK_XE_CONFIGURATION "/" PCMK_XE_RESOURCES "/" PCMK_XE_CLONE "[@" PCMK_XA_ID "='dummy-clone']" "/" PCMK_XE_PRIMITIVE "[@" PCMK_XA_ID "='dummy']\")", NULL }, { "show-access", 'S', G_OPTION_FLAG_OPTIONAL_ARG, G_OPTION_ARG_CALLBACK, show_access_cb, "Whether to use syntax highlighting for ACLs (with -Q/--query and " "-U/--user)\n" INDENT "Allowed values: 'color' (default for terminal), 'text' (plain text, " "default for non-terminal),\n" INDENT " 'namespace', or 'auto' (use default value)\n" INDENT "Default value: 'auto'", "[value]" }, { "score", 0, G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &options.score_update, "Treat new attribute values as atomic score updates where possible " "(with --modify/-M).\n" INDENT "This currently happens by default and cannot be disabled, but\n" INDENT "this default behavior is deprecated and will be removed in a\n" INDENT "future release. Set this flag if this behavior is desired.\n" INDENT "This option takes effect when updating XML attributes. For an\n" INDENT "attribute named \"name\", if the new value is \"name++\" or\n" INDENT "\"name+=X\" for some score X, the new value is set as follows:\n" INDENT "If attribute \"name\" is not already set to some value in\n" INDENT "the element being updated, the new value is set as a literal\n" INDENT "string.\n" INDENT "If the new value is \"name++\", then the attribute is set to \n" INDENT "its existing value (parsed as a score) plus 1.\n" INDENT "If the new value is \"name+=X\" for some score X, then the\n" INDENT "attribute is set to its existing value plus X, where the\n" INDENT "existing value and X are parsed and added as scores.\n" INDENT "Scores are integer values capped at INFINITY and -INFINITY.\n" INDENT "Refer to Pacemaker Explained and to the char2score() function\n" INDENT "for more details on scores, including how they're parsed and\n" INDENT "added.", NULL }, { "allow-create", 'c', G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &options.allow_create, "(Advanced) Allow target of --modify/-M to be created if it does not " "exist", NULL }, { "no-children", 'n', G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &options.no_children, "(Advanced) When querying an object, do not include its children in the " "result", NULL }, { "node", 'N', G_OPTION_FLAG_NONE, G_OPTION_ARG_STRING, &options.dest_node, "(Advanced) Send command to the specified host", "value" }, // @COMPAT: Deprecated { "no-bcast", 'b', G_OPTION_FLAG_HIDDEN, G_OPTION_ARG_NONE, &options.no_bcast, "deprecated", NULL }, // @COMPAT: Deprecated { "host", 'h', G_OPTION_FLAG_HIDDEN, G_OPTION_ARG_STRING, &options.dest_node, "deprecated", NULL }, { NULL } }; static GOptionContext * build_arg_context(pcmk__common_args_t *args) { const char *desc = NULL; GOptionContext *context = NULL; GOptionEntry extra_prog_entries[] = { // @COMPAT: Deprecated { "extended-version", '!', G_OPTION_FLAG_HIDDEN, G_OPTION_ARG_NONE, &options.extended_version, "deprecated", NULL }, { NULL } }; desc = "Examples:\n\n" "Query the configuration from the local node:\n\n" "\t# cibadmin --query --local\n\n" "Query just the cluster options configuration:\n\n" "\t# cibadmin --query --scope " PCMK_XE_CRM_CONFIG "\n\n" "Query all '" PCMK_META_TARGET_ROLE "' settings:\n\n" "\t# cibadmin --query --xpath " "\"//" PCMK_XE_NVPAIR "[@" PCMK_XA_NAME "='" PCMK_META_TARGET_ROLE"']\"\n\n" "Remove all '" PCMK_META_IS_MANAGED "' settings:\n\n" "\t# cibadmin --delete-all --xpath " "\"//" PCMK_XE_NVPAIR "[@" PCMK_XA_NAME "='" PCMK_META_IS_MANAGED "']\"\n\n" "Remove the resource named 'old':\n\n" "\t# cibadmin --delete --xml-text " "'<" PCMK_XE_PRIMITIVE " " PCMK_XA_ID "=\"old\"/>'\n\n" "Remove all resources from the configuration:\n\n" "\t# cibadmin --replace --scope " PCMK_XE_RESOURCES " --xml-text '<" PCMK_XE_RESOURCES "/>'\n\n" "Replace complete configuration with contents of " "$HOME/pacemaker.xml:\n\n" "\t# cibadmin --replace --xml-file $HOME/pacemaker.xml\n\n" "Replace " PCMK_XE_CONSTRAINTS " section of configuration with " "contents of $HOME/constraints.xml:\n\n" "\t# cibadmin --replace --scope " PCMK_XE_CONSTRAINTS " --xml-file $HOME/constraints.xml\n\n" "Increase configuration version to prevent old configurations from " "being loaded accidentally:\n\n" "\t# cibadmin --modify --score --xml-text " "'<" PCMK_XE_CIB " " PCMK_XA_ADMIN_EPOCH "=\"" PCMK_XA_ADMIN_EPOCH "++\"/>'\n\n" "Edit the configuration with your favorite $EDITOR:\n\n" "\t# cibadmin --query > $HOME/local.xml\n\n" "\t# $EDITOR $HOME/local.xml\n\n" "\t# cibadmin --replace --xml-file $HOME/local.xml\n\n" "Assuming terminal, render configuration in color (green for " "writable, blue for readable, red for\n" "denied) to visualize permissions for user tony:\n\n" "\t# cibadmin --show-access=color --query --user tony | less -r\n\n" "SEE ALSO:\n" " crm(8), pcs(8), crm_shadow(8), crm_diff(8)\n"; context = pcmk__build_arg_context(args, NULL, NULL, ""); g_option_context_set_description(context, desc); pcmk__add_main_args(context, extra_prog_entries); pcmk__add_arg_group(context, "commands", "Commands:", "Show command help", command_entries); pcmk__add_arg_group(context, "data", "Data:", "Show data help", data_entries); pcmk__add_arg_group(context, "additional", "Additional Options:", "Show additional options", addl_entries); return context; } int main(int argc, char **argv) { int rc = pcmk_rc_ok; const char *source = NULL; xmlNode *output = NULL; xmlNode *input = NULL; gchar *acl_cred = NULL; GError *error = NULL; pcmk__common_args_t *args = pcmk__new_common_args(SUMMARY); gchar **processed_args = pcmk__cmdline_preproc(argv, "ANSUXhotx"); GOptionContext *context = build_arg_context(args); if (!g_option_context_parse_strv(context, &processed_args, &error)) { exit_code = CRM_EX_USAGE; goto done; } if (g_strv_length(processed_args) > 1) { gchar *help = g_option_context_get_help(context, TRUE, NULL); GString *extra = g_string_sized_new(128); for (int lpc = 1; processed_args[lpc] != NULL; lpc++) { if (extra->len > 0) { g_string_append_c(extra, ' '); } g_string_append(extra, processed_args[lpc]); } exit_code = CRM_EX_USAGE; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "non-option ARGV-elements: %s\n\n%s", extra->str, help); g_free(help); g_string_free(extra, TRUE); goto done; } if (args->version || options.extended_version) { g_strfreev(processed_args); pcmk__free_arg_context(context); /* FIXME: When cibadmin is converted to use formatted output, this can * be replaced by out->version with the appropriate boolean flag. * * options.extended_version is deprecated and will be removed in a * future release. */ pcmk__cli_help(options.extended_version? '!' : 'v'); } /* At LOG_ERR, stderr for CIB calls is rather verbose. Several lines like * * (func@file:line) error: CIB failures * * In cibadmin we explicitly output the XML portion without the prefixes. So * we default to LOG_CRIT. */ pcmk__cli_init_logging("cibadmin", 0); set_crm_log_level(LOG_CRIT); if (args->verbosity > 0) { cib__set_call_options(options.cmd_options, crm_system_name, cib_verbose); for (int i = 0; i < args->verbosity; i++) { crm_bump_log_level(argc, argv); } } if (options.cib_action == NULL) { // @COMPAT: Create a default command if other tools have one gchar *help = g_option_context_get_help(context, TRUE, NULL); exit_code = CRM_EX_USAGE; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Must specify a command option\n\n%s", help); g_free(help); goto done; } if (strcmp(options.cib_action, "empty") == 0) { // Output an empty CIB GString *buf = g_string_sized_new(1024); output = createEmptyCib(1); crm_xml_add(output, PCMK_XA_VALIDATE_WITH, options.validate_with); pcmk__xml_string(output, pcmk__xml_fmt_pretty, buf, 0); fprintf(stdout, "%s", buf->str); g_string_free(buf, TRUE); goto done; } if (cib_action_is_dangerous() && !options.force) { exit_code = CRM_EX_UNSAFE; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "The supplied command is considered dangerous. To prevent " "accidental destruction of the cluster, the --force flag " "is required in order to proceed."); goto done; } if (options.message_timeout_sec < 1) { // Set default timeout options.message_timeout_sec = 30; } if (options.section_type == cibadmin_section_xpath) { // Enable getting section by XPath cib__set_call_options(options.cmd_options, crm_system_name, cib_xpath); } else if (options.section_type == cibadmin_section_scope) { if (!scope_is_valid(options.cib_section)) { // @COMPAT: Consider requiring --force to proceed fprintf(stderr, "Invalid value '%s' for '--scope'. Operation will apply " "to the entire CIB.\n", options.cib_section); } } if (options.allow_create) { // Allow target of --modify/-M to be created if it does not exist cib__set_call_options(options.cmd_options, crm_system_name, cib_can_create); } if (options.delete_all) { // With cibadmin_section_xpath, remove all matching objects cib__set_call_options(options.cmd_options, crm_system_name, cib_multiple); } if (options.get_node_path) { /* Enable getting node path of XPath query matches. * Meaningful only if options.section_type == cibadmin_section_xpath. */ cib__set_call_options(options.cmd_options, crm_system_name, cib_xpath_address); } if (options.local) { // Configure command to take effect only locally cib__set_call_options(options.cmd_options, crm_system_name, cib_scope_local); } // @COMPAT: Deprecated option if (options.no_bcast) { // Configure command to take effect only locally and not to broadcast cib__set_call_options(options.cmd_options, crm_system_name, cib_inhibit_bcast|cib_scope_local); } if (options.no_children) { // When querying an object, don't include its children in the result cib__set_call_options(options.cmd_options, crm_system_name, cib_no_children); } if (options.sync_call || (options.acl_render_mode != pcmk__acl_render_none)) { /* Wait for call to complete before returning. * * The ACL render modes work only with sync calls due to differences in * output handling between sync/async. It shouldn't matter to the user * whether the call is synchronous; for a CIB query, we have to wait for * the result in order to display it in any case. */ cib__set_call_options(options.cmd_options, crm_system_name, cib_sync_call); } if (options.input_file != NULL) { input = pcmk__xml_read(options.input_file); source = options.input_file; } else if (options.input_xml != NULL) { input = pcmk__xml_parse(options.input_xml); source = "input string"; } else if (options.input_stdin) { input = pcmk__xml_read(NULL); source = "STDIN"; } else if (options.acl_render_mode != pcmk__acl_render_none) { char *username = pcmk__uid2username(geteuid()); bool required = pcmk_acl_required(username); free(username); if (required) { if (options.force) { fprintf(stderr, "The supplied command can provide skewed" " result since it is run under user that also" " gets guarded per ACLs on their own right." " Continuing since --force flag was" " provided.\n"); } else { exit_code = CRM_EX_UNSAFE; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "The supplied command can provide skewed result " "since it is run under user that also gets guarded " "per ACLs in their own right. To accept the risk " "of such a possible distortion (without even " "knowing it at this time), use the --force flag."); goto done; } } if (options.cib_user == NULL) { exit_code = CRM_EX_USAGE; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "The supplied command requires -U user specified."); goto done; } /* We already stopped/warned ACL-controlled users about consequences. * * Note: acl_cred takes ownership of options.cib_user here. * options.cib_user is set to NULL so that the CIB is obtained as the * user running the cibadmin command. The CIB must be obtained as a user * with full permissions in order to show the CIB correctly annotated * for the options.cib_user's permissions. */ acl_cred = options.cib_user; options.cib_user = NULL; } if (input != NULL) { crm_log_xml_debug(input, "[admin input]"); } else if (source != NULL) { exit_code = CRM_EX_CONFIG; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Couldn't parse input from %s.", source); goto done; } if (pcmk__str_eq(options.cib_action, "md5-sum", pcmk__str_casei)) { char *digest = NULL; if (input == NULL) { exit_code = CRM_EX_USAGE; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Please supply XML to process with -X, -x, or -p"); goto done; } digest = calculate_on_disk_digest(input); fprintf(stderr, "Digest: "); fprintf(stdout, "%s\n", pcmk__s(digest, "")); free(digest); goto done; } else if (strcmp(options.cib_action, "md5-sum-versioned") == 0) { char *digest = NULL; const char *version = NULL; if (input == NULL) { exit_code = CRM_EX_USAGE; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Please supply XML to process with -X, -x, or -p"); goto done; } version = crm_element_value(input, PCMK_XA_CRM_FEATURE_SET); digest = calculate_xml_versioned_digest(input, FALSE, TRUE, version); fprintf(stderr, "Versioned (%s) digest: ", version); fprintf(stdout, "%s\n", pcmk__s(digest, "")); free(digest); goto done; } else if (pcmk__str_eq(options.cib_action, PCMK__CIB_REQUEST_MODIFY, pcmk__str_none)) { /* @COMPAT When we drop default support for expansion in cibadmin, guard * with `if (options.score_update)` */ cib__set_call_options(options.cmd_options, crm_system_name, cib_score_update); } rc = do_init(); if (rc != pcmk_ok) { rc = pcmk_legacy2rc(rc); exit_code = pcmk_rc2exitc(rc); crm_err("Init failed, could not perform requested operations: %s", pcmk_rc_str(rc)); g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Init failed, could not perform requested operations: %s", pcmk_rc_str(rc)); goto done; } rc = do_work(input, &output); if (!pcmk_is_set(options.cmd_options, cib_sync_call) && (the_cib->variant != cib_file) && (rc >= 0)) { /* For async call, positive rc is the call ID (file always synchronous). * * Wait for the reply by creating a mainloop and running it until the * callbacks are invoked. */ request_id = rc; the_cib->cmds->register_callback(the_cib, request_id, options.message_timeout_sec, FALSE, NULL, "cibadmin_op_callback", cibadmin_op_callback); mainloop = g_main_loop_new(NULL, FALSE); crm_trace("%s waiting for reply from the local CIB", crm_system_name); crm_info("Starting mainloop"); g_main_loop_run(mainloop); } else { rc = pcmk_legacy2rc(rc); if ((rc == pcmk_rc_schema_unchanged) && (strcmp(options.cib_action, PCMK__CIB_REQUEST_UPGRADE) == 0)) { report_schema_unchanged(); } else if (rc != pcmk_rc_ok) { crm_err("Call failed: %s", pcmk_rc_str(rc)); fprintf(stderr, "Call failed: %s\n", pcmk_rc_str(rc)); exit_code = pcmk_rc2exitc(rc); if (rc == pcmk_rc_schema_validation) { if (strcmp(options.cib_action, PCMK__CIB_REQUEST_UPGRADE) == 0) { xmlNode *obj = NULL; if (the_cib->cmds->query(the_cib, NULL, &obj, options.cmd_options) == pcmk_ok) { pcmk__update_schema(&obj, NULL, true, false); } free_xml(obj); } else if (output != NULL) { // Show validation errors to stderr pcmk__validate_xml(output, NULL, NULL, NULL); } } } } if ((output != NULL) && (options.acl_render_mode != pcmk__acl_render_none)) { xmlDoc *acl_evaled_doc; rc = pcmk__acl_annotate_permissions(acl_cred, output->doc, &acl_evaled_doc); if (rc == pcmk_rc_ok) { xmlChar *rendered = NULL; rc = pcmk__acl_evaled_render(acl_evaled_doc, options.acl_render_mode, &rendered); if (rc != pcmk_rc_ok) { exit_code = CRM_EX_CONFIG; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Could not render evaluated access: %s", pcmk_rc_str(rc)); goto done; } printf("%s\n", (char *) rendered); free(rendered); } else { exit_code = CRM_EX_CONFIG; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Could not evaluate access per request (%s, error: %s)", acl_cred, pcmk_rc_str(rc)); goto done; } } else if (output != NULL) { print_xml_output(output); } crm_trace("%s exiting normally", crm_system_name); done: g_strfreev(processed_args); pcmk__free_arg_context(context); g_free(options.cib_user); g_free(options.dest_node); g_free(options.input_file); g_free(options.input_xml); free(options.cib_section); free(options.validate_with); g_free(acl_cred); free_xml(input); free_xml(output); rc = cib__clean_up_connection(&the_cib); if (exit_code == CRM_EX_OK) { exit_code = pcmk_rc2exitc(rc); } pcmk__output_and_clear_error(&error, NULL); crm_exit(exit_code); } static int do_work(xmlNode *input, xmlNode **output) { /* construct the request */ the_cib->call_timeout = options.message_timeout_sec; if ((strcmp(options.cib_action, PCMK__CIB_REQUEST_REPLACE) == 0) && pcmk__xe_is(input, PCMK_XE_CIB)) { xmlNode *status = pcmk_find_cib_element(input, PCMK_XE_STATUS); if (status == NULL) { pcmk__xe_create(input, PCMK_XE_STATUS); } } crm_trace("Passing \"%s\" to variant_op...", options.cib_action); return cib_internal_op(the_cib, options.cib_action, options.dest_node, options.cib_section, input, output, options.cmd_options, options.cib_user); } int do_init(void) { int rc = pcmk_ok; the_cib = cib_new(); - rc = cib__signon_attempts(the_cib, crm_system_name, cib_command, 5); + + if (the_cib == NULL) { + rc = ENOTCONN; + } else { + rc = cib__signon_attempts(the_cib, crm_system_name, cib_command, 5); + } + if (rc != pcmk_ok) { crm_err("Could not connect to the CIB: %s", pcmk_strerror(rc)); fprintf(stderr, "Could not connect to the CIB: %s\n", pcmk_strerror(rc)); } return rc; } void cibadmin_op_callback(xmlNode * msg, int call_id, int rc, xmlNode * output, void *user_data) { rc = pcmk_legacy2rc(rc); exit_code = pcmk_rc2exitc(rc); if (rc == pcmk_rc_schema_unchanged) { report_schema_unchanged(); } else if (rc != pcmk_rc_ok) { crm_warn("Call %s failed: %s " CRM_XS " rc=%d", options.cib_action, pcmk_rc_str(rc), rc); fprintf(stderr, "Call %s failed: %s\n", options.cib_action, pcmk_rc_str(rc)); print_xml_output(output); } else if ((strcmp(options.cib_action, PCMK__CIB_REQUEST_QUERY) == 0) && (output == NULL)) { crm_err("Query returned no output"); crm_log_xml_err(msg, "no output"); } else if (output == NULL) { crm_info("Call passed"); } else { crm_info("Call passed"); print_xml_output(output); } if (call_id == request_id) { g_main_loop_quit(mainloop); } else { crm_info("Message was not the response we were looking for (%d vs. %d)", call_id, request_id); } } diff --git a/tools/crm_resource.c b/tools/crm_resource.c index f3116172b0..898c21ece6 100644 --- a/tools/crm_resource.c +++ b/tools/crm_resource.c @@ -1,2189 +1,2191 @@ /* - * Copyright 2004-2024 the Pacemaker project contributors + * Copyright 2004-2025 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU General Public License version 2 * or later (GPLv2+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include // uint32_t #include #include #include #include #include #include #include #include #include #include #include #include #define SUMMARY "crm_resource - perform tasks related to Pacemaker cluster resources" enum rsc_command { cmd_none = 0, // No command option given (yet) cmd_ban, cmd_cleanup, cmd_clear, cmd_colocations, cmd_cts, cmd_delete, cmd_delete_param, cmd_digests, cmd_execute_agent, cmd_fail, cmd_get_param, cmd_get_property, cmd_list_active_ops, cmd_list_agents, cmd_list_all_ops, cmd_list_alternatives, cmd_list_instances, cmd_list_options, cmd_list_providers, cmd_list_resources, cmd_list_standards, cmd_locate, cmd_metadata, cmd_move, cmd_query_xml, cmd_query_xml_raw, cmd_refresh, cmd_restart, cmd_set_param, cmd_set_property, cmd_wait, cmd_why, }; struct { enum rsc_command rsc_cmd; // crm_resource command to perform // Command-line option values gchar *rsc_id; // Value of --resource gchar *rsc_type; // Value of --resource-type gboolean all; // --all was given gboolean force; // --force was given gboolean clear_expired; // --expired was given gboolean recursive; // --recursive was given gboolean promoted_role_only; // --promoted was given gchar *host_uname; // Value of --node gchar *interval_spec; // Value of --interval gchar *move_lifetime; // Value of --lifetime gchar *operation; // Value of --operation enum pcmk__opt_flags opt_list; // Parsed from --list-options const char *attr_set_type; // Instance, meta, utilization, or element attribute gchar *prop_id; // --nvpair (attribute XML ID) char *prop_name; // Attribute name gchar *prop_set; // --set-name (attribute block XML ID) gchar *prop_value; // --parameter-value (attribute value) guint timeout_ms; // Parsed from --timeout value char *agent_spec; // Standard and/or provider and/or agent gchar *xml_file; // Value of (deprecated) --xml-file int check_level; // Optional value of --validate or --force-check // Resource configuration specified via command-line arguments bool cmdline_config; // Resource configuration was via arguments char *v_agent; // Value of --agent char *v_class; // Value of --class char *v_provider; // Value of --provider GHashTable *cmdline_params; // Resource parameters specified // Positional command-line arguments gchar **remainder; // Positional arguments as given GHashTable *override_params; // Resource parameter values that override config } options = { .attr_set_type = PCMK_XE_INSTANCE_ATTRIBUTES, .check_level = -1, .rsc_cmd = cmd_list_resources, // List all resources if no command given }; gboolean attr_set_type_cb(const gchar *option_name, const gchar *optarg, gpointer data, GError **error); gboolean cmdline_config_cb(const gchar *option_name, const gchar *optarg, gpointer data, GError **error); gboolean option_cb(const gchar *option_name, const gchar *optarg, gpointer data, GError **error); gboolean timeout_cb(const gchar *option_name, const gchar *optarg, gpointer data, GError **error); static crm_exit_t exit_code = CRM_EX_OK; static pcmk__output_t *out = NULL; static pcmk__common_args_t *args = NULL; // Things that should be cleaned up on exit static GError *error = NULL; static GMainLoop *mainloop = NULL; static cib_t *cib_conn = NULL; static pcmk_ipc_api_t *controld_api = NULL; static pcmk_scheduler_t *scheduler = NULL; #define MESSAGE_TIMEOUT_S 60 #define INDENT " " static pcmk__supported_format_t formats[] = { PCMK__SUPPORTED_FORMAT_NONE, PCMK__SUPPORTED_FORMAT_TEXT, PCMK__SUPPORTED_FORMAT_XML, { NULL, NULL, NULL } }; // Clean up and exit static crm_exit_t bye(crm_exit_t ec) { pcmk__output_and_clear_error(&error, out); if (out != NULL) { out->finish(out, ec, true, NULL); pcmk__output_free(out); } pcmk__unregister_formats(); if (cib_conn != NULL) { cib_t *save_cib_conn = cib_conn; cib_conn = NULL; // Ensure we can't free this twice cib__clean_up_connection(&save_cib_conn); } if (controld_api != NULL) { pcmk_ipc_api_t *save_controld_api = controld_api; controld_api = NULL; // Ensure we can't free this twice pcmk_free_ipc_api(save_controld_api); } if (mainloop != NULL) { g_main_loop_unref(mainloop); mainloop = NULL; } pe_free_working_set(scheduler); scheduler = NULL; crm_exit(ec); return ec; } static void quit_main_loop(crm_exit_t ec) { exit_code = ec; if (mainloop != NULL) { GMainLoop *mloop = mainloop; mainloop = NULL; // Don't re-enter this block pcmk_quit_main_loop(mloop, 10); g_main_loop_unref(mloop); } } static gboolean resource_ipc_timeout(gpointer data) { // Start with newline because "Waiting for ..." message doesn't have one if (error != NULL) { g_clear_error(&error); } g_set_error(&error, PCMK__EXITC_ERROR, CRM_EX_TIMEOUT, _("Aborting because no messages received in %d seconds"), MESSAGE_TIMEOUT_S); quit_main_loop(CRM_EX_TIMEOUT); return FALSE; } static void controller_event_callback(pcmk_ipc_api_t *api, enum pcmk_ipc_event event_type, crm_exit_t status, void *event_data, void *user_data) { switch (event_type) { case pcmk_ipc_event_disconnect: if (exit_code == CRM_EX_DISCONNECT) { // Unexpected crm_info("Connection to controller was terminated"); } quit_main_loop(exit_code); break; case pcmk_ipc_event_reply: if (status != CRM_EX_OK) { out->err(out, "Error: bad reply from controller: %s", crm_exit_str(status)); pcmk_disconnect_ipc(api); quit_main_loop(status); } else { if ((pcmk_controld_api_replies_expected(api) == 0) && mainloop && g_main_loop_is_running(mainloop)) { out->info(out, "... got reply (done)"); crm_debug("Got all the replies we expected"); pcmk_disconnect_ipc(api); quit_main_loop(CRM_EX_OK); } else { out->info(out, "... got reply"); } } break; default: break; } } static void start_mainloop(pcmk_ipc_api_t *capi) { unsigned int count = pcmk_controld_api_replies_expected(capi); if (count > 0) { out->info(out, "Waiting for %u %s from the controller", count, pcmk__plural_alt(count, "reply", "replies")); exit_code = CRM_EX_DISCONNECT; // For unexpected disconnects mainloop = g_main_loop_new(NULL, FALSE); g_timeout_add(MESSAGE_TIMEOUT_S * 1000, resource_ipc_timeout, NULL); g_main_loop_run(mainloop); } } static int compare_id(gconstpointer a, gconstpointer b) { return strcmp((const char *)a, (const char *)b); } static GList * build_constraint_list(xmlNode *root) { GList *retval = NULL; xmlNode *cib_constraints = NULL; xmlXPathObjectPtr xpathObj = NULL; int ndx = 0; cib_constraints = pcmk_find_cib_element(root, PCMK_XE_CONSTRAINTS); xpathObj = xpath_search(cib_constraints, "//" PCMK_XE_RSC_LOCATION); for (ndx = 0; ndx < numXpathResults(xpathObj); ndx++) { xmlNode *match = getXpathResult(xpathObj, ndx); retval = g_list_insert_sorted(retval, (gpointer) pcmk__xe_id(match), compare_id); } freeXpathObject(xpathObj); return retval; } static gboolean validate_opt_list(const gchar *optarg) { if (pcmk__str_eq(optarg, PCMK_VALUE_FENCING, pcmk__str_none)) { options.opt_list = pcmk__opt_fencing; } else if (pcmk__str_eq(optarg, PCMK__VALUE_PRIMITIVE, pcmk__str_none)) { options.opt_list = pcmk__opt_primitive; } else { return FALSE; } return TRUE; } /*! * \internal * \brief Process options that set the command * * Nothing else should set \c options.rsc_cmd. * * \param[in] option_name Name of the option being parsed * \param[in] optarg Value to be parsed * \param[in] data Ignored * \param[out] error Where to store recoverable error, if any * * \return \c TRUE if the option was successfully parsed, or \c FALSE if an * error occurred, in which case \p *error is set */ static gboolean command_cb(const gchar *option_name, const gchar *optarg, gpointer data, GError **error) { // Sorted by enum rsc_command name if (pcmk__str_any_of(option_name, "-B", "--ban", NULL)) { options.rsc_cmd = cmd_ban; } else if (pcmk__str_any_of(option_name, "-C", "--cleanup", NULL)) { options.rsc_cmd = cmd_cleanup; } else if (pcmk__str_any_of(option_name, "-U", "--clear", NULL)) { options.rsc_cmd = cmd_clear; } else if (pcmk__str_any_of(option_name, "-a", "--constraints", NULL)) { options.rsc_cmd = cmd_colocations; } else if (pcmk__str_any_of(option_name, "-A", "--stack", NULL)) { options.rsc_cmd = cmd_colocations; options.recursive = TRUE; } else if (pcmk__str_any_of(option_name, "-c", "--list-cts", NULL)) { options.rsc_cmd = cmd_cts; } else if (pcmk__str_any_of(option_name, "-D", "--delete", NULL)) { options.rsc_cmd = cmd_delete; } else if (pcmk__str_any_of(option_name, "-d", "--delete-parameter", NULL)) { options.rsc_cmd = cmd_delete_param; pcmk__str_update(&options.prop_name, optarg); } else if (pcmk__str_eq(option_name, "--digests", pcmk__str_none)) { options.rsc_cmd = cmd_digests; if (options.override_params == NULL) { options.override_params = pcmk__strkey_table(free, free); } } else if (pcmk__str_any_of(option_name, "--force-demote", "--force-promote", "--force-start", "--force-stop", "--force-check", "--validate", NULL)) { options.rsc_cmd = cmd_execute_agent; g_free(options.operation); options.operation = g_strdup(option_name + 2); // skip "--" if (options.override_params == NULL) { options.override_params = pcmk__strkey_table(free, free); } if (optarg != NULL) { if (pcmk__scan_min_int(optarg, &options.check_level, 0) != pcmk_rc_ok) { g_set_error(error, G_OPTION_ERROR, CRM_EX_INVALID_PARAM, _("Invalid check level setting: %s"), optarg); return FALSE; } } } else if (pcmk__str_any_of(option_name, "-F", "--fail", NULL)) { options.rsc_cmd = cmd_fail; } else if (pcmk__str_any_of(option_name, "-g", "--get-parameter", NULL)) { options.rsc_cmd = cmd_get_param; pcmk__str_update(&options.prop_name, optarg); } else if (pcmk__str_any_of(option_name, "-G", "--get-property", NULL)) { options.rsc_cmd = cmd_get_property; pcmk__str_update(&options.prop_name, optarg); } else if (pcmk__str_any_of(option_name, "-O", "--list-operations", NULL)) { options.rsc_cmd = cmd_list_active_ops; } else if (pcmk__str_eq(option_name, "--list-agents", pcmk__str_none)) { options.rsc_cmd = cmd_list_agents; pcmk__str_update(&options.agent_spec, optarg); } else if (pcmk__str_any_of(option_name, "-o", "--list-all-operations", NULL)) { options.rsc_cmd = cmd_list_all_ops; } else if (pcmk__str_eq(option_name, "--list-ocf-alternatives", pcmk__str_none)) { options.rsc_cmd = cmd_list_alternatives; pcmk__str_update(&options.agent_spec, optarg); } else if (pcmk__str_eq(option_name, "--list-options", pcmk__str_none)) { options.rsc_cmd = cmd_list_options; return validate_opt_list(optarg); } else if (pcmk__str_any_of(option_name, "-l", "--list-raw", NULL)) { options.rsc_cmd = cmd_list_instances; } else if (pcmk__str_eq(option_name, "--list-ocf-providers", pcmk__str_none)) { options.rsc_cmd = cmd_list_providers; pcmk__str_update(&options.agent_spec, optarg); } else if (pcmk__str_any_of(option_name, "-L", "--list", NULL)) { options.rsc_cmd = cmd_list_resources; } else if (pcmk__str_eq(option_name, "--list-standards", pcmk__str_none)) { options.rsc_cmd = cmd_list_standards; } else if (pcmk__str_any_of(option_name, "-W", "--locate", NULL)) { options.rsc_cmd = cmd_locate; } else if (pcmk__str_eq(option_name, "--show-metadata", pcmk__str_none)) { options.rsc_cmd = cmd_metadata; pcmk__str_update(&options.agent_spec, optarg); } else if (pcmk__str_any_of(option_name, "-M", "--move", NULL)) { options.rsc_cmd = cmd_move; } else if (pcmk__str_any_of(option_name, "-q", "--query-xml", NULL)) { options.rsc_cmd = cmd_query_xml; } else if (pcmk__str_any_of(option_name, "-w", "--query-xml-raw", NULL)) { options.rsc_cmd = cmd_query_xml_raw; } else if (pcmk__str_any_of(option_name, "-R", "--refresh", NULL)) { options.rsc_cmd = cmd_refresh; } else if (pcmk__str_eq(option_name, "--restart", pcmk__str_none)) { options.rsc_cmd = cmd_restart; } else if (pcmk__str_any_of(option_name, "-p", "--set-parameter", NULL)) { options.rsc_cmd = cmd_set_param; pcmk__str_update(&options.prop_name, optarg); } else if (pcmk__str_any_of(option_name, "-S", "--set-property", NULL)) { options.rsc_cmd = cmd_set_property; pcmk__str_update(&options.prop_name, optarg); } else if (pcmk__str_eq(option_name, "--wait", pcmk__str_none)) { options.rsc_cmd = cmd_wait; } else if (pcmk__str_any_of(option_name, "-Y", "--why", NULL)) { options.rsc_cmd = cmd_why; } return TRUE; } /* short option letters still available: eEJkKXyYZ */ static GOptionEntry query_entries[] = { { "list", 'L', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "List all cluster resources with status", NULL }, { "list-raw", 'l', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "List IDs of all instantiated resources (individual members\n" INDENT "rather than groups etc.)", NULL }, { "list-cts", 'c', G_OPTION_FLAG_HIDDEN|G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, NULL, NULL }, { "list-operations", 'O', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "List active resource operations, optionally filtered by\n" INDENT "--resource and/or --node", NULL }, { "list-all-operations", 'o', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "List all resource operations, optionally filtered by\n" INDENT "--resource and/or --node", NULL }, { "list-options", 0, G_OPTION_FLAG_NONE, G_OPTION_ARG_CALLBACK, command_cb, "List all available options of the given type\n" INDENT "Allowed values:\n" INDENT PCMK__VALUE_PRIMITIVE "(primitive resource meta-attributes), " INDENT PCMK_VALUE_FENCING " (parameters common to all fencing resources)", "TYPE" }, { "list-standards", 0, G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "List supported standards", NULL }, { "list-ocf-providers", 0, G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "List all available OCF providers", NULL }, { "list-agents", 0, G_OPTION_FLAG_NONE, G_OPTION_ARG_CALLBACK, command_cb, "List all agents available for the named standard and/or provider", "STD:PROV" }, { "list-ocf-alternatives", 0, G_OPTION_FLAG_NONE, G_OPTION_ARG_CALLBACK, command_cb, "List all available providers for the named OCF agent", "AGENT" }, { "show-metadata", 0, G_OPTION_FLAG_NONE, G_OPTION_ARG_CALLBACK, command_cb, "Show the metadata for the named class:provider:agent", "SPEC" }, { "query-xml", 'q', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Show XML configuration of resource (after any template expansion)", NULL }, { "query-xml-raw", 'w', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Show XML configuration of resource (before any template expansion)", NULL }, { "get-parameter", 'g', G_OPTION_FLAG_NONE, G_OPTION_ARG_CALLBACK, command_cb, "Display named parameter for resource (use instance attribute\n" INDENT "unless --element, --meta, or --utilization is specified)", "PARAM" }, { "get-property", 'G', G_OPTION_FLAG_HIDDEN, G_OPTION_ARG_CALLBACK, command_cb, "Display named property of resource ('class', 'type', or 'provider') " "(requires --resource)", "PROPERTY" }, { "locate", 'W', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Show node(s) currently running resource", NULL }, { "constraints", 'a', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Display the location and colocation constraints that apply to a\n" INDENT "resource, and if --recursive is specified, to the resources\n" INDENT "directly or indirectly involved in those colocations.\n" INDENT "If the named resource is part of a group, or a clone or\n" INDENT "bundle instance, constraints for the collective resource\n" INDENT "will be shown unless --force is given.", NULL }, { "stack", 'A', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Equivalent to --constraints --recursive", NULL }, { "why", 'Y', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Show why resources are not running, optionally filtered by\n" INDENT "--resource and/or --node", NULL }, { NULL } }; static GOptionEntry command_entries[] = { { "validate", 0, G_OPTION_FLAG_OPTIONAL_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Validate resource configuration by calling agent's validate-all\n" INDENT "action. The configuration may be specified either by giving an\n" INDENT "existing resource name with -r, or by specifying --class,\n" INDENT "--agent, and --provider arguments, along with any number of\n" INDENT "--option arguments. An optional LEVEL argument can be given\n" INDENT "to control the level of checking performed.", "LEVEL" }, { "cleanup", 'C', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "If resource has any past failures, clear its history and fail\n" INDENT "count. Optionally filtered by --resource, --node, --operation\n" INDENT "and --interval (otherwise all). --operation and --interval\n" INDENT "apply to fail counts, but entire history is always clear, to\n" INDENT "allow current state to be rechecked. If the named resource is\n" INDENT "part of a group, or one numbered instance of a clone or bundled\n" INDENT "resource, the clean-up applies to the whole collective resource\n" INDENT "unless --force is given.", NULL }, { "refresh", 'R', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Delete resource's history (including failures) so its current state\n" INDENT "is rechecked. Optionally filtered by --resource and --node\n" INDENT "(otherwise all). If the named resource is part of a group, or one\n" INDENT "numbered instance of a clone or bundled resource, the refresh\n" INDENT "applies to the whole collective resource unless --force is given.", NULL }, { "set-parameter", 'p', G_OPTION_FLAG_NONE, G_OPTION_ARG_CALLBACK, command_cb, "Set named parameter for resource (requires -v). Use instance\n" INDENT "attribute unless --element, --meta, or --utilization is " "specified.", "PARAM" }, { "delete-parameter", 'd', G_OPTION_FLAG_NONE, G_OPTION_ARG_CALLBACK, command_cb, "Delete named parameter for resource. Use instance attribute\n" INDENT "unless --element, --meta or, --utilization is specified.", "PARAM" }, { "set-property", 'S', G_OPTION_FLAG_HIDDEN, G_OPTION_ARG_CALLBACK, command_cb, "Set named property of resource ('class', 'type', or 'provider') " "(requires -r, -t, -v)", "PROPERTY" }, { NULL } }; static GOptionEntry location_entries[] = { { "move", 'M', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Create a constraint to move resource. If --node is specified,\n" INDENT "the constraint will be to move to that node, otherwise it\n" INDENT "will be to ban the current node. Unless --force is specified\n" INDENT "this will return an error if the resource is already running\n" INDENT "on the specified node. If --force is specified, this will\n" INDENT "always ban the current node.\n" INDENT "Optional: --lifetime, --promoted. NOTE: This may prevent the\n" INDENT "resource from running on its previous location until the\n" INDENT "implicit constraint expires or is removed with --clear.", NULL }, { "ban", 'B', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Create a constraint to keep resource off a node.\n" INDENT "Optional: --node, --lifetime, --promoted.\n" INDENT "NOTE: This will prevent the resource from running on the\n" INDENT "affected node until the implicit constraint expires or is\n" INDENT "removed with --clear. If --node is not specified, it defaults\n" INDENT "to the node currently running the resource for primitives\n" INDENT "and groups, or the promoted instance of promotable clones with\n" INDENT PCMK_META_PROMOTED_MAX "=1 (all other situations result in an\n" INDENT "error as there is no sane default).", NULL }, { "clear", 'U', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "Remove all constraints created by the --ban and/or --move\n" INDENT "commands. Requires: --resource. Optional: --node, --promoted,\n" INDENT "--expired. If --node is not specified, all constraints created\n" INDENT "by --ban and --move will be removed for the named resource. If\n" INDENT "--node and --force are specified, any constraint created by\n" INDENT "--move will be cleared, even if it is not for the specified\n" INDENT "node. If --expired is specified, only those constraints whose\n" INDENT "lifetimes have expired will be removed.", NULL }, { "expired", 'e', G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &options.clear_expired, "Modifies the --clear argument to remove constraints with\n" INDENT "expired lifetimes.", NULL }, { "lifetime", 'u', G_OPTION_FLAG_NONE, G_OPTION_ARG_STRING, &options.move_lifetime, "Lifespan (as ISO 8601 duration) of created constraints (with\n" INDENT "-B, -M) see https://en.wikipedia.org/wiki/ISO_8601#Durations)", "TIMESPEC" }, { "promoted", 0, G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &options.promoted_role_only, "Limit scope of command to promoted role (with -B, -M, -U). For\n" INDENT "-B and -M, previously promoted instances may remain\n" INDENT "active in the unpromoted role.", NULL }, // Deprecated since 2.1.0 { "master", 0, G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &options.promoted_role_only, "Deprecated: Use --promoted instead", NULL }, { NULL } }; static GOptionEntry advanced_entries[] = { { "delete", 'D', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "(Advanced) Delete a resource from the CIB. Required: -t", NULL }, { "fail", 'F', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "(Advanced) Tell the cluster this resource has failed", NULL }, { "restart", 0, G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "(Advanced) Tell the cluster to restart this resource and\n" INDENT "anything that depends on it", NULL }, { "wait", 0, G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "(Advanced) Wait until the cluster settles into a stable state", NULL }, { "digests", 0, G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "(Advanced) Show parameter hashes that Pacemaker uses to detect\n" INDENT "configuration changes (only accurate if there is resource\n" INDENT "history on the specified node). Required: --resource, --node.\n" INDENT "Optional: any NAME=VALUE parameters will be used to override\n" INDENT "the configuration (to see what the hash would be with those\n" INDENT "changes).", NULL }, { "force-demote", 0, G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "(Advanced) Bypass the cluster and demote a resource on the local\n" INDENT "node. Unless --force is specified, this will refuse to do so if\n" INDENT "the cluster believes the resource is a clone instance already\n" INDENT "running on the local node.", NULL }, { "force-stop", 0, G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "(Advanced) Bypass the cluster and stop a resource on the local node", NULL }, { "force-start", 0, G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "(Advanced) Bypass the cluster and start a resource on the local\n" INDENT "node. Unless --force is specified, this will refuse to do so if\n" INDENT "the cluster believes the resource is a clone instance already\n" INDENT "running on the local node.", NULL }, { "force-promote", 0, G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, command_cb, "(Advanced) Bypass the cluster and promote a resource on the local\n" INDENT "node. Unless --force is specified, this will refuse to do so if\n" INDENT "the cluster believes the resource is a clone instance already\n" INDENT "running on the local node.", NULL }, { "force-check", 0, G_OPTION_FLAG_OPTIONAL_ARG, G_OPTION_ARG_CALLBACK, command_cb, "(Advanced) Bypass the cluster and check the state of a resource on\n" INDENT "the local node. An optional LEVEL argument can be given\n" INDENT "to control the level of checking performed.", "LEVEL" }, { NULL } }; static GOptionEntry addl_entries[] = { { "node", 'N', G_OPTION_FLAG_NONE, G_OPTION_ARG_STRING, &options.host_uname, "Node name", "NAME" }, { "recursive", 0, G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &options.recursive, "Follow colocation chains when using --set-parameter or --constraints", NULL }, { "resource-type", 't', G_OPTION_FLAG_NONE, G_OPTION_ARG_STRING, &options.rsc_type, "Resource XML element (primitive, group, etc.) (with -D)", "ELEMENT" }, { "parameter-value", 'v', G_OPTION_FLAG_NONE, G_OPTION_ARG_STRING, &options.prop_value, "Value to use with -p", "PARAM" }, { "meta", 'm', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, attr_set_type_cb, "Use resource meta-attribute instead of instance attribute\n" INDENT "(with -p, -g, -d)", NULL }, { "utilization", 'z', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, attr_set_type_cb, "Use resource utilization attribute instead of instance attribute\n" INDENT "(with -p, -g, -d)", NULL }, { "element", 0, G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, attr_set_type_cb, "Use resource element attribute instead of instance attribute\n" INDENT "(with -p, -g, -d)", NULL }, { "operation", 'n', G_OPTION_FLAG_NONE, G_OPTION_ARG_STRING, &options.operation, "Operation to clear instead of all (with -C -r)", "OPERATION" }, { "interval", 'I', G_OPTION_FLAG_NONE, G_OPTION_ARG_STRING, &options.interval_spec, "Interval of operation to clear (default 0) (with -C -r -n)", "N" }, { "class", 0, G_OPTION_FLAG_NONE, G_OPTION_ARG_CALLBACK, cmdline_config_cb, "The standard the resource agent conforms to (for example, ocf).\n" INDENT "Use with --agent, --provider, --option, and --validate.", "CLASS" }, { "agent", 0, G_OPTION_FLAG_NONE, G_OPTION_ARG_CALLBACK, cmdline_config_cb, "The agent to use (for example, IPaddr). Use with --class,\n" INDENT "--provider, --option, and --validate.", "AGENT" }, { "provider", 0, G_OPTION_FLAG_NONE, G_OPTION_ARG_CALLBACK, cmdline_config_cb, "The vendor that supplies the resource agent (for example,\n" INDENT "heartbeat). Use with --class, --agent, --option, and --validate.", "PROVIDER" }, { "option", 0, G_OPTION_FLAG_NONE, G_OPTION_ARG_CALLBACK, option_cb, "Specify a device configuration parameter as NAME=VALUE (may be\n" INDENT "specified multiple times). Use with --validate and without the\n" INDENT "-r option.", "PARAM" }, { "set-name", 's', G_OPTION_FLAG_NONE, G_OPTION_ARG_STRING, &options.prop_set, "(Advanced) XML ID of attributes element to use (with -p, -d)", "ID" }, { "nvpair", 'i', G_OPTION_FLAG_NONE, G_OPTION_ARG_STRING, &options.prop_id, "(Advanced) XML ID of nvpair element to use (with -p, -d)", "ID" }, { "timeout", 'T', G_OPTION_FLAG_NONE, G_OPTION_ARG_CALLBACK, timeout_cb, "(Advanced) Abort if command does not finish in this time (with\n" INDENT "--restart, --wait, --force-*)", "N" }, { "all", 0, G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &options.all, "List all options, including advanced and deprecated (with\n" INDENT "--list-options)", NULL }, { "force", 'f', G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &options.force, "Force the action to be performed. See help for individual commands for\n" INDENT "additional behavior.", NULL }, { "xml-file", 'x', G_OPTION_FLAG_HIDDEN, G_OPTION_ARG_FILENAME, &options.xml_file, NULL, "FILE" }, { "host-uname", 'H', G_OPTION_FLAG_HIDDEN, G_OPTION_ARG_STRING, &options.host_uname, NULL, "HOST" }, { NULL } }; gboolean attr_set_type_cb(const gchar *option_name, const gchar *optarg, gpointer data, GError **error) { if (pcmk__str_any_of(option_name, "-m", "--meta", NULL)) { options.attr_set_type = PCMK_XE_META_ATTRIBUTES; } else if (pcmk__str_any_of(option_name, "-z", "--utilization", NULL)) { options.attr_set_type = PCMK_XE_UTILIZATION; } else if (pcmk__str_eq(option_name, "--element", pcmk__str_none)) { options.attr_set_type = ATTR_SET_ELEMENT; } return TRUE; } gboolean cmdline_config_cb(const gchar *option_name, const gchar *optarg, gpointer data, GError **error) { options.cmdline_config = true; if (pcmk__str_eq(option_name, "--class", pcmk__str_none)) { pcmk__str_update(&options.v_class, optarg); } else if (pcmk__str_eq(option_name, "--provider", pcmk__str_none)) { pcmk__str_update(&options.v_provider, optarg); } else { // --agent pcmk__str_update(&options.v_agent, optarg); } return TRUE; } gboolean option_cb(const gchar *option_name, const gchar *optarg, gpointer data, GError **error) { char *name = NULL; char *value = NULL; if (pcmk__scan_nvpair(optarg, &name, &value) != 2) { + free(name); + free(value); return FALSE; } if (options.cmdline_params == NULL) { options.cmdline_params = pcmk__strkey_table(free, free); } g_hash_table_replace(options.cmdline_params, name, value); return TRUE; } gboolean timeout_cb(const gchar *option_name, const gchar *optarg, gpointer data, GError **error) { long long timeout_ms = crm_get_msec(optarg); if (timeout_ms < 0) { // @COMPAT When we can break backward compatibilty, return FALSE crm_warn("Ignoring invalid timeout '%s'", optarg); options.timeout_ms = 0U; } else { options.timeout_ms = (guint) QB_MIN(timeout_ms, UINT_MAX); } return TRUE; } static int ban_or_move(pcmk__output_t *out, pcmk_resource_t *rsc, const char *move_lifetime) { int rc = pcmk_rc_ok; pcmk_node_t *current = NULL; unsigned int nactive = 0; CRM_CHECK(rsc != NULL, return EINVAL); current = pe__find_active_requires(rsc, &nactive); if (nactive == 1) { rc = cli_resource_ban(out, options.rsc_id, current->details->uname, move_lifetime, cib_conn, cib_sync_call, options.promoted_role_only, PCMK_ROLE_PROMOTED); } else if (pcmk_is_set(rsc->flags, pcmk_rsc_promotable)) { int count = 0; GList *iter = NULL; current = NULL; for(iter = rsc->children; iter; iter = iter->next) { pcmk_resource_t *child = (pcmk_resource_t *)iter->data; enum rsc_role_e child_role = child->fns->state(child, TRUE); if (child_role == pcmk_role_promoted) { count++; current = pcmk__current_node(child); } } if(count == 1 && current) { rc = cli_resource_ban(out, options.rsc_id, current->details->uname, move_lifetime, cib_conn, cib_sync_call, options.promoted_role_only, PCMK_ROLE_PROMOTED); } else { rc = EINVAL; g_set_error(&error, PCMK__EXITC_ERROR, CRM_EX_USAGE, _("Resource '%s' not moved: active in %d locations (promoted in %d).\n" "To prevent '%s' from running on a specific location, " "specify a node." "To prevent '%s' from being promoted at a specific " "location, specify a node and the --promoted option."), options.rsc_id, nactive, count, options.rsc_id, options.rsc_id); } } else { rc = EINVAL; g_set_error(&error, PCMK__EXITC_ERROR, CRM_EX_USAGE, _("Resource '%s' not moved: active in %d locations.\n" "To prevent '%s' from running on a specific location, " "specify a node."), options.rsc_id, nactive, options.rsc_id); } return rc; } static void cleanup(pcmk__output_t *out, pcmk_resource_t *rsc, pcmk_node_t *node) { int rc = pcmk_rc_ok; if (options.force == FALSE) { rsc = uber_parent(rsc); } crm_debug("Erasing failures of %s (%s requested) on %s", rsc->id, options.rsc_id, (options.host_uname? options.host_uname: "all nodes")); rc = cli_resource_delete(controld_api, options.host_uname, rsc, options.operation, options.interval_spec, TRUE, scheduler, options.force); if ((rc == pcmk_rc_ok) && !out->is_quiet(out)) { // Show any reasons why resource might stay stopped cli_resource_check(out, rsc, node); } if (rc == pcmk_rc_ok) { start_mainloop(controld_api); } } static int clear_constraints(pcmk__output_t *out, xmlNodePtr *cib_xml_copy) { GList *before = NULL; GList *after = NULL; GList *remaining = NULL; GList *ele = NULL; pcmk_node_t *dest = NULL; int rc = pcmk_rc_ok; if (!out->is_quiet(out)) { before = build_constraint_list(scheduler->input); } if (options.clear_expired) { rc = cli_resource_clear_all_expired(scheduler->input, cib_conn, cib_sync_call, options.rsc_id, options.host_uname, options.promoted_role_only); } else if (options.host_uname) { dest = pcmk_find_node(scheduler, options.host_uname); if (dest == NULL) { rc = pcmk_rc_node_unknown; if (!out->is_quiet(out)) { g_list_free(before); } return rc; } rc = cli_resource_clear(options.rsc_id, dest->details->uname, NULL, cib_conn, cib_sync_call, true, options.force); } else { rc = cli_resource_clear(options.rsc_id, NULL, scheduler->nodes, cib_conn, cib_sync_call, true, options.force); } if (!out->is_quiet(out)) { rc = cib_conn->cmds->query(cib_conn, NULL, cib_xml_copy, cib_scope_local | cib_sync_call); rc = pcmk_legacy2rc(rc); if (rc != pcmk_rc_ok) { g_set_error(&error, PCMK__RC_ERROR, rc, _("Could not get modified CIB: %s\n"), pcmk_rc_str(rc)); g_list_free(before); free_xml(*cib_xml_copy); *cib_xml_copy = NULL; return rc; } scheduler->input = *cib_xml_copy; cluster_status(scheduler); after = build_constraint_list(scheduler->input); remaining = pcmk__subtract_lists(before, after, (GCompareFunc) strcmp); for (ele = remaining; ele != NULL; ele = ele->next) { out->info(out, "Removing constraint: %s", (char *) ele->data); } g_list_free(before); g_list_free(after); g_list_free(remaining); } return rc; } static int initialize_scheduler_data(xmlNodePtr *cib_xml_copy) { int rc = pcmk_rc_ok; if (options.xml_file != NULL) { *cib_xml_copy = pcmk__xml_read(options.xml_file); if (*cib_xml_copy == NULL) { rc = pcmk_rc_cib_corrupt; } } else { rc = cib_conn->cmds->query(cib_conn, NULL, cib_xml_copy, cib_scope_local | cib_sync_call); rc = pcmk_legacy2rc(rc); } if (rc == pcmk_rc_ok) { scheduler = pe_new_working_set(); if (scheduler == NULL) { rc = ENOMEM; } else { pcmk__set_scheduler_flags(scheduler, pcmk_sched_no_counts |pcmk_sched_no_compat); scheduler->priv = out; rc = update_scheduler_input(scheduler, cib_xml_copy); } } if (rc != pcmk_rc_ok) { free_xml(*cib_xml_copy); *cib_xml_copy = NULL; return rc; } cluster_status(scheduler); return pcmk_rc_ok; } static void list_options(void) { switch (options.opt_list) { case pcmk__opt_fencing: exit_code = pcmk_rc2exitc(pcmk__list_fencing_params(out, options.all)); break; case pcmk__opt_primitive: exit_code = pcmk_rc2exitc(pcmk__list_primitive_meta(out, options.all)); break; default: exit_code = CRM_EX_SOFTWARE; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "BUG: Invalid option list type"); break; } } static int refresh(pcmk__output_t *out) { int rc = pcmk_rc_ok; const char *router_node = options.host_uname; int attr_options = pcmk__node_attr_none; if (options.host_uname) { pcmk_node_t *node = pcmk_find_node(scheduler, options.host_uname); if (pcmk__is_pacemaker_remote_node(node)) { node = pcmk__current_node(node->details->remote_rsc); if (node == NULL) { rc = ENXIO; g_set_error(&error, PCMK__RC_ERROR, rc, _("No cluster connection to Pacemaker Remote node %s detected"), options.host_uname); return rc; } router_node = node->details->uname; attr_options |= pcmk__node_attr_remote; } } if (controld_api == NULL) { out->info(out, "Dry run: skipping clean-up of %s due to CIB_file", options.host_uname? options.host_uname : "all nodes"); rc = pcmk_rc_ok; return rc; } crm_debug("Re-checking the state of all resources on %s", options.host_uname?options.host_uname:"all nodes"); rc = pcmk__attrd_api_clear_failures(NULL, options.host_uname, NULL, NULL, NULL, NULL, attr_options); if (pcmk_controld_api_reprobe(controld_api, options.host_uname, router_node) == pcmk_rc_ok) { start_mainloop(controld_api); } return rc; } static void refresh_resource(pcmk__output_t *out, pcmk_resource_t *rsc, pcmk_node_t *node) { int rc = pcmk_rc_ok; if (options.force == FALSE) { rsc = uber_parent(rsc); } crm_debug("Re-checking the state of %s (%s requested) on %s", rsc->id, options.rsc_id, (options.host_uname? options.host_uname: "all nodes")); rc = cli_resource_delete(controld_api, options.host_uname, rsc, NULL, 0, FALSE, scheduler, options.force); if ((rc == pcmk_rc_ok) && !out->is_quiet(out)) { // Show any reasons why resource might stay stopped cli_resource_check(out, rsc, node); } if (rc == pcmk_rc_ok) { start_mainloop(controld_api); } } static int set_property(void) { int rc = pcmk_rc_ok; xmlNode *msg_data = NULL; if (pcmk__str_empty(options.rsc_type)) { g_set_error(&error, PCMK__EXITC_ERROR, CRM_EX_USAGE, _("Must specify -t with resource type")); rc = ENXIO; return rc; } else if (pcmk__str_empty(options.prop_value)) { g_set_error(&error, PCMK__EXITC_ERROR, CRM_EX_USAGE, _("Must supply -v with new value")); rc = ENXIO; return rc; } CRM_LOG_ASSERT(options.prop_name != NULL); msg_data = pcmk__xe_create(NULL, options.rsc_type); crm_xml_add(msg_data, PCMK_XA_ID, options.rsc_id); crm_xml_add(msg_data, options.prop_name, options.prop_value); rc = cib_conn->cmds->modify(cib_conn, PCMK_XE_RESOURCES, msg_data, cib_sync_call); rc = pcmk_legacy2rc(rc); free_xml(msg_data); return rc; } static int show_metadata(pcmk__output_t *out, const char *agent_spec) { int rc = pcmk_rc_ok; char *standard = NULL; char *provider = NULL; char *type = NULL; char *metadata = NULL; lrmd_t *lrmd_conn = NULL; rc = lrmd__new(&lrmd_conn, NULL, NULL, 0); if (rc != pcmk_rc_ok) { g_set_error(&error, PCMK__RC_ERROR, rc, _("Could not create executor connection")); lrmd_api_delete(lrmd_conn); return rc; } rc = crm_parse_agent_spec(agent_spec, &standard, &provider, &type); rc = pcmk_legacy2rc(rc); if (rc == pcmk_rc_ok) { rc = lrmd_conn->cmds->get_metadata(lrmd_conn, standard, provider, type, &metadata, 0); rc = pcmk_legacy2rc(rc); if (metadata) { out->output_xml(out, PCMK_XE_METADATA, metadata); free(metadata); } else { /* We were given a validly formatted spec, but it doesn't necessarily * match up with anything that exists. Use ENXIO as the return code * here because that maps to an exit code of CRM_EX_NOSUCH, which * probably is the most common reason to get here. */ rc = ENXIO; g_set_error(&error, PCMK__RC_ERROR, rc, _("Metadata query for %s failed: %s"), agent_spec, pcmk_rc_str(rc)); } } else { rc = ENXIO; g_set_error(&error, PCMK__RC_ERROR, rc, _("'%s' is not a valid agent specification"), agent_spec); } lrmd_api_delete(lrmd_conn); return rc; } static void validate_cmdline_config(void) { // Cannot use both --resource and command-line resource configuration if (options.rsc_id != NULL) { g_set_error(&error, PCMK__EXITC_ERROR, CRM_EX_USAGE, _("--resource cannot be used with --class, --agent, and --provider")); // Not all commands support command-line resource configuration } else if (options.rsc_cmd != cmd_execute_agent) { g_set_error(&error, PCMK__EXITC_ERROR, CRM_EX_USAGE, _("--class, --agent, and --provider can only be used with " "--validate and --force-*")); // Not all of --class, --agent, and --provider need to be given. Not all // classes support the concept of a provider. Check that what we were given // is valid. } else if (pcmk__str_eq(options.v_class, "stonith", pcmk__str_none)) { if (options.v_provider != NULL) { g_set_error(&error, PCMK__EXITC_ERROR, CRM_EX_USAGE, _("stonith does not support providers")); } else if (stonith_agent_exists(options.v_agent, 0) == FALSE) { g_set_error(&error, PCMK__EXITC_ERROR, CRM_EX_USAGE, _("%s is not a known stonith agent"), options.v_agent ? options.v_agent : ""); } } else if (resources_agent_exists(options.v_class, options.v_provider, options.v_agent) == FALSE) { g_set_error(&error, PCMK__EXITC_ERROR, CRM_EX_USAGE, _("%s:%s:%s is not a known resource"), options.v_class ? options.v_class : "", options.v_provider ? options.v_provider : "", options.v_agent ? options.v_agent : ""); } if ((error == NULL) && (options.cmdline_params == NULL)) { options.cmdline_params = pcmk__strkey_table(free, free); } } /*! * \internal * \brief Get the enum pe_find flags for a given command * * \return enum pe_find flag group appropriate for \c options.rsc_cmd. */ static uint32_t get_find_flags(void) { switch (options.rsc_cmd) { case cmd_ban: case cmd_cleanup: case cmd_clear: case cmd_colocations: case cmd_digests: case cmd_execute_agent: case cmd_locate: case cmd_move: case cmd_refresh: case cmd_restart: case cmd_why: return pcmk_rsc_match_history|pcmk_rsc_match_anon_basename; // @COMPAT See note in is_scheduler_required() case cmd_delete: case cmd_delete_param: case cmd_get_param: case cmd_get_property: case cmd_query_xml_raw: case cmd_query_xml: case cmd_set_param: case cmd_set_property: return pcmk_rsc_match_history|pcmk_rsc_match_basename; default: return 0; } } /*! * \internal * \brief Check whether a node argument is required * * \return \c true if a \c --node argument is required, or \c false otherwise */ static bool is_node_required(void) { switch (options.rsc_cmd) { case cmd_digests: case cmd_fail: return true; default: return false; } } /*! * \internal * \brief Check whether a resource argument is required * * \return \c true if a \c --resource argument is required, or \c false * otherwise */ static bool is_resource_required(void) { if (options.cmdline_config) { return false; } switch (options.rsc_cmd) { case cmd_clear: return !options.clear_expired; case cmd_cleanup: case cmd_cts: case cmd_list_active_ops: case cmd_list_agents: case cmd_list_all_ops: case cmd_list_alternatives: case cmd_list_instances: case cmd_list_options: case cmd_list_providers: case cmd_list_resources: case cmd_list_standards: case cmd_metadata: case cmd_refresh: case cmd_wait: case cmd_why: return false; default: return true; } } /*! * \internal * \brief Check whether a CIB connection is required * * \return \c true if a CIB connection is required, or \c false otherwise */ static bool is_cib_required(void) { if (options.cmdline_config) { return false; } switch (options.rsc_cmd) { case cmd_list_agents: case cmd_list_alternatives: case cmd_list_options: case cmd_list_providers: case cmd_list_standards: case cmd_metadata: return false; default: return true; } } /*! * \internal * \brief Check whether a controller IPC connection is required * * \return \c true if a controller connection is required, or \c false otherwise */ static bool is_controller_required(void) { switch (options.rsc_cmd) { case cmd_cleanup: case cmd_refresh: return getenv("CIB_file") == NULL; case cmd_fail: return true; default: return false; } } /*! * \internal * \brief Check whether a scheduler IPC connection is required * * \return \c true if a scheduler connection is required, or \c false otherwise */ static bool is_scheduler_required(void) { if (options.cmdline_config) { return false; } /* @COMPAT cmd_delete does not actually need the scheduler and should not * set find_flags. However, crm_resource --delete currently throws a * "resource not found" error if the resource doesn't exist. This is * incorrect behavior (deleting a nonexistent resource should be considered * success); however, we shouldn't change it until 3.0.0. */ switch (options.rsc_cmd) { case cmd_list_agents: case cmd_list_alternatives: case cmd_list_options: case cmd_list_providers: case cmd_list_standards: case cmd_metadata: case cmd_wait: return false; default: return true; } } /*! * \internal * \brief Check whether the chosen command accepts clone instances * * \return \c true if \p options.rsc_cmd accepts or ignores clone instances, or * \c false otherwise */ static bool accept_clone_instance(void) { // @COMPAT At 3.0.0, add cmd_delete; for now, don't throw error switch (options.rsc_cmd) { case cmd_ban: case cmd_clear: case cmd_move: case cmd_restart: return false; default: return true; } } static GOptionContext * build_arg_context(pcmk__common_args_t *args, GOptionGroup **group) { GOptionContext *context = NULL; GOptionEntry extra_prog_entries[] = { { "quiet", 'Q', G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &(args->quiet), "Be less descriptive in output.", NULL }, { "resource", 'r', G_OPTION_FLAG_NONE, G_OPTION_ARG_STRING, &options.rsc_id, "Resource ID", "ID" }, { G_OPTION_REMAINING, 0, G_OPTION_FLAG_NONE, G_OPTION_ARG_STRING_ARRAY, &options.remainder, NULL, NULL }, { NULL } }; const char *description = "Examples:\n\n" "List the available OCF agents:\n\n" "\t# crm_resource --list-agents ocf\n\n" "List the available OCF agents from the linux-ha project:\n\n" "\t# crm_resource --list-agents ocf:heartbeat\n\n" "Move 'myResource' to a specific node:\n\n" "\t# crm_resource --resource myResource --move --node altNode\n\n" "Allow (but not force) 'myResource' to move back to its original " "location:\n\n" "\t# crm_resource --resource myResource --clear\n\n" "Stop 'myResource' (and anything that depends on it):\n\n" "\t# crm_resource --resource myResource --set-parameter " PCMK_META_TARGET_ROLE "--meta --parameter-value Stopped\n\n" "Tell the cluster not to manage 'myResource' (the cluster will not " "attempt to start or stop the\n" "resource under any circumstances; useful when performing maintenance " "tasks on a resource):\n\n" "\t# crm_resource --resource myResource --set-parameter " PCMK_META_IS_MANAGED "--meta --parameter-value false\n\n" "Erase the operation history of 'myResource' on 'aNode' (the cluster " "will 'forget' the existing\n" "resource state, including any errors, and attempt to recover the" "resource; useful when a resource\n" "had failed permanently and has been repaired by an administrator):\n\n" "\t# crm_resource --resource myResource --cleanup --node aNode\n\n"; context = pcmk__build_arg_context(args, "text (default), xml", group, NULL); g_option_context_set_description(context, description); /* Add the -Q option, which cannot be part of the globally supported options * because some tools use that flag for something else. */ pcmk__add_main_args(context, extra_prog_entries); pcmk__add_arg_group(context, "queries", "Queries:", "Show query help", query_entries); pcmk__add_arg_group(context, "commands", "Commands:", "Show command help", command_entries); pcmk__add_arg_group(context, "locations", "Locations:", "Show location help", location_entries); pcmk__add_arg_group(context, "advanced", "Advanced:", "Show advanced option help", advanced_entries); pcmk__add_arg_group(context, "additional", "Additional Options:", "Show additional options", addl_entries); return context; } int main(int argc, char **argv) { xmlNode *cib_xml_copy = NULL; pcmk_resource_t *rsc = NULL; pcmk_node_t *node = NULL; uint32_t find_flags = 0; int rc = pcmk_rc_ok; GOptionGroup *output_group = NULL; gchar **processed_args = NULL; GOptionContext *context = NULL; /* * Parse command line arguments */ args = pcmk__new_common_args(SUMMARY); processed_args = pcmk__cmdline_preproc(argv, "GHINSTdginpstuvx"); context = build_arg_context(args, &output_group); pcmk__register_formats(output_group, formats); if (!g_option_context_parse_strv(context, &processed_args, &error)) { exit_code = CRM_EX_USAGE; goto done; } pcmk__cli_init_logging("crm_resource", args->verbosity); rc = pcmk__output_new(&out, args->output_ty, args->output_dest, argv); if (rc != pcmk_rc_ok) { exit_code = CRM_EX_ERROR; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, _("Error creating output format %s: %s"), args->output_ty, pcmk_rc_str(rc)); goto done; } pe__register_messages(out); crm_resource_register_messages(out); lrmd__register_messages(out); pcmk__register_lib_messages(out); out->quiet = args->quiet; crm_log_args(argc, argv); /* * Validate option combinations */ // --expired without --clear/-U doesn't make sense if (options.clear_expired && (options.rsc_cmd != cmd_clear)) { exit_code = CRM_EX_USAGE; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, _("--expired requires --clear or -U")); goto done; } if ((options.remainder != NULL) && (options.override_params != NULL)) { // Commands that use positional arguments will create override_params for (gchar **s = options.remainder; *s; s++) { char *name = pcmk__assert_alloc(1, strlen(*s)); char *value = pcmk__assert_alloc(1, strlen(*s)); int rc = sscanf(*s, "%[^=]=%s", name, value); if (rc == 2) { g_hash_table_replace(options.override_params, name, value); } else { exit_code = CRM_EX_USAGE; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, _("Error parsing '%s' as a name=value pair"), argv[optind]); free(value); free(name); goto done; } } } else if (options.remainder != NULL) { gchar **strv = NULL; gchar *msg = NULL; int i = 1; int len = 0; for (gchar **s = options.remainder; *s; s++) { len++; } pcmk__assert(len > 0); /* Add 1 for the strv[0] string below, and add another 1 for the NULL * at the end of the array so g_strjoinv knows when to stop. */ strv = pcmk__assert_alloc(len+2, sizeof(char *)); strv[0] = strdup("non-option ARGV-elements:\n"); for (gchar **s = options.remainder; *s; s++) { strv[i] = crm_strdup_printf("[%d of %d] %s\n", i, len, *s); i++; } strv[i] = NULL; exit_code = CRM_EX_USAGE; msg = g_strjoinv("", strv); g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "%s", msg); g_free(msg); /* Don't try to free the last element, which is just NULL. */ for(i = 0; i < len+1; i++) { free(strv[i]); } free(strv); goto done; } if (pcmk__str_eq(args->output_ty, "xml", pcmk__str_none)) { switch (options.rsc_cmd) { /* These are the only commands that have historically used the * elements in their XML schema. For all others, use the simple list * argument. */ case cmd_get_param: case cmd_get_property: case cmd_list_instances: case cmd_list_standards: pcmk__output_enable_list_element(out); break; default: break; } } else if (pcmk__str_eq(args->output_ty, "text", pcmk__str_null_matches)) { switch (options.rsc_cmd) { case cmd_colocations: case cmd_list_resources: pcmk__output_text_set_fancy(out, true); break; default: break; } } if (args->version) { out->version(out, false); goto done; } if (options.cmdline_config) { /* A resource configuration was given on the command line. Sanity-check * the values and set error if they don't make sense. */ validate_cmdline_config(); if (error != NULL) { exit_code = CRM_EX_USAGE; goto done; } } else if (options.cmdline_params != NULL) { // @COMPAT @TODO error out here when we can break backward compatibility g_hash_table_destroy(options.cmdline_params); options.cmdline_params = NULL; } if (is_resource_required() && (options.rsc_id == NULL)) { exit_code = CRM_EX_USAGE; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, _("Must supply a resource id with -r")); goto done; } if (is_node_required() && (options.host_uname == NULL)) { exit_code = CRM_EX_USAGE; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, _("Must supply a node name with -N")); goto done; } /* * Set up necessary connections */ // Establish a connection to the CIB if needed if (is_cib_required()) { cib_conn = cib_new(); if ((cib_conn == NULL) || (cib_conn->cmds == NULL)) { exit_code = CRM_EX_DISCONNECT; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, _("Could not create CIB connection")); goto done; } rc = cib__signon_attempts(cib_conn, crm_system_name, cib_command, 5); rc = pcmk_legacy2rc(rc); if (rc != pcmk_rc_ok) { exit_code = pcmk_rc2exitc(rc); g_set_error(&error, PCMK__EXITC_ERROR, exit_code, _("Could not connect to the CIB: %s"), pcmk_rc_str(rc)); goto done; } } // Populate scheduler data from XML file if specified or CIB query otherwise if (is_scheduler_required()) { rc = initialize_scheduler_data(&cib_xml_copy); if (rc != pcmk_rc_ok) { exit_code = pcmk_rc2exitc(rc); goto done; } } find_flags = get_find_flags(); // If command requires that resource exist if specified, find it if ((find_flags != 0) && (options.rsc_id != NULL)) { rsc = pe_find_resource_with_flags(scheduler->resources, options.rsc_id, find_flags); if (rsc == NULL) { exit_code = CRM_EX_NOSUCH; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, _("Resource '%s' not found"), options.rsc_id); goto done; } /* The --ban, --clear, --move, and --restart commands do not work with * instances of clone resourcs. */ if (pcmk__is_clone(rsc->parent) && (strchr(options.rsc_id, ':') != NULL) && !accept_clone_instance()) { exit_code = CRM_EX_INVALID_PARAM; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, _("Cannot operate on clone resource instance '%s'"), options.rsc_id); goto done; } } // If user supplied a node name, check whether it exists if ((options.host_uname != NULL) && (scheduler != NULL)) { node = pcmk_find_node(scheduler, options.host_uname); if (node == NULL) { exit_code = CRM_EX_NOSUCH; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, _("Node '%s' not found"), options.host_uname); goto done; } } // Establish a connection to the controller if needed if (is_controller_required()) { rc = pcmk_new_ipc_api(&controld_api, pcmk_ipc_controld); if (rc != pcmk_rc_ok) { exit_code = pcmk_rc2exitc(rc); g_set_error(&error, PCMK__EXITC_ERROR, exit_code, _("Error connecting to the controller: %s"), pcmk_rc_str(rc)); goto done; } pcmk_register_ipc_callback(controld_api, controller_event_callback, NULL); rc = pcmk__connect_ipc(controld_api, pcmk_ipc_dispatch_main, 5); if (rc != pcmk_rc_ok) { exit_code = pcmk_rc2exitc(rc); g_set_error(&error, PCMK__EXITC_ERROR, exit_code, _("Error connecting to %s: %s"), pcmk_ipc_name(controld_api, true), pcmk_rc_str(rc)); goto done; } } /* * Handle requested command */ switch (options.rsc_cmd) { case cmd_list_resources: { GList *all = NULL; uint32_t show_opts = pcmk_show_inactive_rscs | pcmk_show_rsc_only | pcmk_show_pending; all = g_list_prepend(all, (gpointer) "*"); rc = out->message(out, "resource-list", scheduler, show_opts, true, all, all, false); g_list_free(all); if (rc == pcmk_rc_no_output) { rc = ENXIO; } break; } case cmd_list_instances: // coverity[var_deref_op] False positive rc = out->message(out, "resource-names-list", scheduler->resources); if (rc != pcmk_rc_ok) { rc = ENXIO; } break; case cmd_list_options: list_options(); break; case cmd_list_alternatives: rc = pcmk__list_alternatives(out, options.agent_spec); break; case cmd_list_agents: rc = pcmk__list_agents(out, options.agent_spec); break; case cmd_list_standards: rc = pcmk__list_standards(out); break; case cmd_list_providers: rc = pcmk__list_providers(out, options.agent_spec); break; case cmd_metadata: rc = show_metadata(out, options.agent_spec); break; case cmd_restart: /* We don't pass scheduler because rsc needs to stay valid for the * entire lifetime of cli_resource_restart(), but it will reset and * update the scheduler data multiple times, so it needs to use its * own copy. */ rc = cli_resource_restart(out, rsc, node, options.move_lifetime, options.timeout_ms, cib_conn, cib_sync_call, options.promoted_role_only, options.force); break; case cmd_wait: rc = wait_till_stable(out, options.timeout_ms, cib_conn); break; case cmd_execute_agent: if (options.cmdline_config) { exit_code = cli_resource_execute_from_params(out, NULL, options.v_class, options.v_provider, options.v_agent, options.operation, options.cmdline_params, options.override_params, options.timeout_ms, args->verbosity, options.force, options.check_level); } else { exit_code = cli_resource_execute(rsc, options.rsc_id, options.operation, options.override_params, options.timeout_ms, cib_conn, scheduler, args->verbosity, options.force, options.check_level); } goto done; case cmd_digests: node = pcmk_find_node(scheduler, options.host_uname); if (node == NULL) { rc = pcmk_rc_node_unknown; } else { rc = pcmk__resource_digests(out, rsc, node, options.override_params); } break; case cmd_colocations: rc = out->message(out, "locations-and-colocations", rsc, options.recursive, (bool) options.force); break; case cmd_cts: rc = pcmk_rc_ok; // coverity[var_deref_op] False positive g_list_foreach(scheduler->resources, (GFunc) cli_resource_print_cts, out); cli_resource_print_cts_constraints(scheduler); break; case cmd_fail: rc = cli_resource_fail(controld_api, options.host_uname, options.rsc_id, scheduler); if (rc == pcmk_rc_ok) { start_mainloop(controld_api); } break; case cmd_list_active_ops: rc = cli_resource_print_operations(options.rsc_id, options.host_uname, TRUE, scheduler); break; case cmd_list_all_ops: rc = cli_resource_print_operations(options.rsc_id, options.host_uname, FALSE, scheduler); break; case cmd_locate: { GList *nodes = cli_resource_search(rsc, options.rsc_id, scheduler); rc = out->message(out, "resource-search-list", nodes, options.rsc_id); g_list_free_full(nodes, free); break; } case cmd_query_xml: rc = cli_resource_print(rsc, scheduler, true); break; case cmd_query_xml_raw: rc = cli_resource_print(rsc, scheduler, false); break; case cmd_why: if ((options.host_uname != NULL) && (node == NULL)) { rc = pcmk_rc_node_unknown; } else { rc = out->message(out, "resource-reasons-list", scheduler->resources, rsc, node); } break; case cmd_clear: rc = clear_constraints(out, &cib_xml_copy); break; case cmd_move: if (options.host_uname == NULL) { rc = ban_or_move(out, rsc, options.move_lifetime); } else { rc = cli_resource_move(rsc, options.rsc_id, options.host_uname, options.move_lifetime, cib_conn, cib_sync_call, scheduler, options.promoted_role_only, options.force); } if (rc == EINVAL) { exit_code = CRM_EX_USAGE; goto done; } break; case cmd_ban: if (options.host_uname == NULL) { rc = ban_or_move(out, rsc, options.move_lifetime); } else if (node == NULL) { rc = pcmk_rc_node_unknown; } else { rc = cli_resource_ban(out, options.rsc_id, node->details->uname, options.move_lifetime, cib_conn, cib_sync_call, options.promoted_role_only, PCMK_ROLE_PROMOTED); } if (rc == EINVAL) { exit_code = CRM_EX_USAGE; goto done; } break; case cmd_get_property: rc = out->message(out, "property-list", rsc, options.prop_name); if (rc == pcmk_rc_no_output) { rc = ENXIO; } break; case cmd_set_property: rc = set_property(); break; case cmd_get_param: { unsigned int count = 0; GHashTable *params = NULL; // coverity[var_deref_op] False positive pcmk_node_t *current = rsc->fns->active_node(rsc, &count, NULL); bool free_params = true; const char* value = NULL; if (count > 1) { out->err(out, "%s is active on more than one node," " returning the default value for %s", rsc->id, pcmk__s(options.prop_name, "unspecified property")); current = NULL; } crm_debug("Looking up %s in %s", options.prop_name, rsc->id); if (pcmk__str_eq(options.attr_set_type, PCMK_XE_INSTANCE_ATTRIBUTES, pcmk__str_none)) { params = pe_rsc_params(rsc, current, scheduler); free_params = false; value = g_hash_table_lookup(params, options.prop_name); } else if (pcmk__str_eq(options.attr_set_type, PCMK_XE_META_ATTRIBUTES, pcmk__str_none)) { params = pcmk__strkey_table(free, free); get_meta_attributes(params, rsc, NULL, scheduler); value = g_hash_table_lookup(params, options.prop_name); } else if (pcmk__str_eq(options.attr_set_type, ATTR_SET_ELEMENT, pcmk__str_none)) { value = crm_element_value(rsc->xml, options.prop_name); free_params = false; } else { pe_rule_eval_data_t rule_data = { .now = scheduler->now, }; params = pcmk__strkey_table(free, free); pe__unpack_dataset_nvpairs(rsc->xml, PCMK_XE_UTILIZATION, &rule_data, params, NULL, FALSE, scheduler); value = g_hash_table_lookup(params, options.prop_name); } rc = out->message(out, "attribute-list", rsc, options.prop_name, value); if (free_params) { g_hash_table_destroy(params); } break; } case cmd_set_param: if (pcmk__str_empty(options.prop_value)) { exit_code = CRM_EX_USAGE; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, _("You need to supply a value with the -v option")); goto done; } /* coverity[var_deref_model] False positive */ rc = cli_resource_update_attribute(rsc, options.rsc_id, options.prop_set, options.attr_set_type, options.prop_id, options.prop_name, options.prop_value, options.recursive, cib_conn, options.force); break; case cmd_delete_param: /* coverity[var_deref_model] False positive */ rc = cli_resource_delete_attribute(rsc, options.rsc_id, options.prop_set, options.attr_set_type, options.prop_id, options.prop_name, cib_conn, cib_sync_call, options.force); break; case cmd_cleanup: if (rsc == NULL) { rc = cli_cleanup_all(controld_api, options.host_uname, options.operation, options.interval_spec, scheduler); if (rc == pcmk_rc_ok) { start_mainloop(controld_api); } } else { cleanup(out, rsc, node); } break; case cmd_refresh: if (rsc == NULL) { rc = refresh(out); } else { refresh_resource(out, rsc, node); } break; case cmd_delete: /* rsc_id was already checked for NULL much earlier when validating * command line arguments. */ if (options.rsc_type == NULL) { // @COMPAT @TODO change this to exit_code = CRM_EX_USAGE rc = ENXIO; g_set_error(&error, PCMK__RC_ERROR, rc, _("You need to specify a resource type with -t")); } else { rc = pcmk__resource_delete(cib_conn, cib_sync_call, options.rsc_id, options.rsc_type); if (rc != pcmk_rc_ok) { g_set_error(&error, PCMK__RC_ERROR, rc, _("Could not delete resource %s: %s"), options.rsc_id, pcmk_rc_str(rc)); } } break; default: exit_code = CRM_EX_USAGE; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, _("Unimplemented command: %d"), (int) options.rsc_cmd); goto done; } /* Convert rc into an exit code. */ if (rc != pcmk_rc_ok && rc != pcmk_rc_no_output) { exit_code = pcmk_rc2exitc(rc); } /* * Clean up and exit */ done: /* When we get here, exit_code has been set one of two ways - either at one of * the spots where there's a "goto done" (which itself could have happened either * directly or by calling pcmk_rc2exitc), or just up above after any of the break * statements. * * Thus, we can use just exit_code here to decide what to do. */ if (exit_code != CRM_EX_OK && exit_code != CRM_EX_USAGE) { if (error != NULL) { char *msg = crm_strdup_printf("%s\nError performing operation: %s", error->message, crm_exit_str(exit_code)); g_clear_error(&error); g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "%s", msg); free(msg); } else { g_set_error(&error, PCMK__EXITC_ERROR, exit_code, _("Error performing operation: %s"), crm_exit_str(exit_code)); } } g_free(options.host_uname); g_free(options.interval_spec); g_free(options.move_lifetime); g_free(options.operation); g_free(options.prop_id); free(options.prop_name); g_free(options.prop_set); g_free(options.prop_value); g_free(options.rsc_id); g_free(options.rsc_type); free(options.agent_spec); free(options.v_agent); free(options.v_class); free(options.v_provider); g_free(options.xml_file); g_strfreev(options.remainder); if (options.override_params != NULL) { g_hash_table_destroy(options.override_params); } /* options.cmdline_params does not need to be destroyed here. See the * comments in cli_resource_execute_from_params. */ g_strfreev(processed_args); g_option_context_free(context); return bye(exit_code); }