diff --git a/daemons/controld/controld_cib.c b/daemons/controld/controld_cib.c index 6213adb5c1..2a75a91eec 100644 --- a/daemons/controld/controld_cib.c +++ b/daemons/controld/controld_cib.c @@ -1,1066 +1,1066 @@ /* * 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 #include /* sleep */ #include #include #include #include #include // Call ID of the most recent in-progress CIB resource update (or 0 if none) static int pending_rsc_update = 0; /*! * \internal * \brief Respond to a dropped CIB connection * * \param[in] user_data CIB connection that dropped */ static void handle_cib_disconnect(gpointer user_data) { CRM_LOG_ASSERT(user_data == controld_globals.cib_conn); controld_trigger_fsa(); controld_globals.cib_conn->state = cib_disconnected; if (pcmk_is_set(controld_globals.fsa_input_register, R_CIB_CONNECTED)) { // @TODO This should trigger a reconnect, not a shutdown crm_crit("Lost connection to the CIB manager, shutting down"); register_fsa_input(C_FSA_INTERNAL, I_ERROR, NULL); controld_clear_fsa_input_flags(R_CIB_CONNECTED); } else { // Expected crm_info("Disconnected from the CIB manager"); } } static void do_cib_updated(const char *event, xmlNode * msg) { const xmlNode *patchset = NULL; const char *client_name = NULL; crm_debug("Received CIB diff notification: DC=%s", pcmk__btoa(AM_I_DC)); if (cib__get_notify_patchset(msg, &patchset) != pcmk_rc_ok) { return; } if (cib__element_in_patchset(patchset, PCMK_XE_ALERTS) || cib__element_in_patchset(patchset, PCMK_XE_CRM_CONFIG)) { controld_trigger_config(); } if (!AM_I_DC) { // We're not in control of the join sequence return; } client_name = crm_element_value(msg, PCMK__XA_CIB_CLIENTNAME); if (!cib__client_triggers_refresh(client_name)) { // The CIB is still accurate return; } if (cib__element_in_patchset(patchset, PCMK_XE_NODES) || cib__element_in_patchset(patchset, PCMK_XE_STATUS)) { /* An unsafe client modified the PCMK_XE_NODES or PCMK_XE_STATUS * section. Ensure the node list is up-to-date, and start the join * process again so we get everyone's current resource history. */ if (client_name == NULL) { client_name = crm_element_value(msg, PCMK__XA_CIB_CLIENTID); } crm_notice("Populating nodes and starting an election after %s event " "triggered by %s", event, pcmk__s(client_name, "(unidentified client)")); populate_cib_nodes(node_update_quick|node_update_all, __func__); register_fsa_input(C_FSA_INTERNAL, I_ELECTION, NULL); } } void controld_disconnect_cib_manager(void) { cib_t *cib_conn = controld_globals.cib_conn; CRM_ASSERT(cib_conn != NULL); crm_debug("Disconnecting from the CIB manager"); controld_clear_fsa_input_flags(R_CIB_CONNECTED); cib_conn->cmds->del_notify_callback(cib_conn, PCMK__VALUE_CIB_DIFF_NOTIFY, do_cib_updated); cib_free_callbacks(cib_conn); if (cib_conn->state != cib_disconnected) { cib_conn->cmds->set_secondary(cib_conn, cib_scope_local|cib_discard_reply); cib_conn->cmds->signoff(cib_conn); } } /* A_CIB_STOP, A_CIB_START, O_CIB_RESTART */ void do_cib_control(long long action, enum crmd_fsa_cause cause, enum crmd_fsa_state cur_state, enum crmd_fsa_input current_input, fsa_data_t * msg_data) { static int cib_retries = 0; cib_t *cib_conn = controld_globals.cib_conn; void (*dnotify_fn) (gpointer user_data) = handle_cib_disconnect; void (*update_cb) (const char *event, xmlNodePtr msg) = do_cib_updated; int rc = pcmk_ok; CRM_ASSERT(cib_conn != NULL); if (pcmk_is_set(action, A_CIB_STOP)) { if ((cib_conn->state != cib_disconnected) && (pending_rsc_update != 0)) { crm_info("Waiting for resource update %d to complete", pending_rsc_update); crmd_fsa_stall(FALSE); return; } controld_disconnect_cib_manager(); } if (!pcmk_is_set(action, A_CIB_START)) { return; } if (cur_state == S_STOPPING) { crm_err("Ignoring request to connect to the CIB manager after " "shutdown"); return; } rc = cib_conn->cmds->signon(cib_conn, CRM_SYSTEM_CRMD, cib_command_nonblocking); if (rc != pcmk_ok) { // A short wait that usually avoids stalling the FSA sleep(1); rc = cib_conn->cmds->signon(cib_conn, CRM_SYSTEM_CRMD, cib_command_nonblocking); } if (rc != pcmk_ok) { crm_info("Could not connect to the CIB manager: %s", pcmk_strerror(rc)); } else if (cib_conn->cmds->set_connection_dnotify(cib_conn, dnotify_fn) != pcmk_ok) { crm_err("Could not set dnotify callback"); } else if (cib_conn->cmds->add_notify_callback(cib_conn, PCMK__VALUE_CIB_DIFF_NOTIFY, update_cb) != pcmk_ok) { crm_err("Could not set CIB notification callback (update)"); } else { controld_set_fsa_input_flags(R_CIB_CONNECTED); cib_retries = 0; } if (!pcmk_is_set(controld_globals.fsa_input_register, R_CIB_CONNECTED)) { cib_retries++; if (cib_retries < 30) { crm_warn("Couldn't complete CIB registration %d times... " "pause and retry", cib_retries); controld_start_wait_timer(); crmd_fsa_stall(FALSE); } else { crm_err("Could not complete CIB registration %d times... " "hard error", cib_retries); register_fsa_error(C_FSA_INTERNAL, I_ERROR, NULL); } } } #define MIN_CIB_OP_TIMEOUT (30) /*! * \internal * \brief Get the timeout (in seconds) that should be used with CIB operations * * \return The maximum of 30 seconds, the value of the PCMK_cib_timeout * environment variable, or 10 seconds times one more than the number of * nodes in the cluster. */ unsigned int cib_op_timeout(void) { // @COMPAT: Drop env_timeout at 3.0.0 static int env_timeout = -1; unsigned int calculated_timeout = 0; if (env_timeout == -1) { const char *env = pcmk__env_option(PCMK__ENV_CIB_TIMEOUT); pcmk__scan_min_int(env, &env_timeout, MIN_CIB_OP_TIMEOUT); crm_trace("Minimum CIB op timeout: %ds (environment: %s)", env_timeout, (env? env : "none")); } calculated_timeout = 10U * (1U + pcmk__cluster_num_active_nodes() + pcmk__cluster_num_remote_nodes()); calculated_timeout = QB_MAX(calculated_timeout, env_timeout); crm_trace("Calculated timeout: %us", calculated_timeout); if (controld_globals.cib_conn) { controld_globals.cib_conn->call_timeout = calculated_timeout; } return calculated_timeout; } /*! * \internal * \brief Get CIB call options to use local scope if primary is unavailable * * \return CIB call options */ int crmd_cib_smart_opt(void) { int call_opt = cib_none; if ((controld_globals.fsa_state == S_ELECTION) || (controld_globals.fsa_state == S_PENDING)) { crm_info("Sending update to local CIB in state: %s", fsa_state2string(controld_globals.fsa_state)); cib__set_call_options(call_opt, "update", cib_scope_local); } return call_opt; } static void cib_delete_callback(xmlNode *msg, int call_id, int rc, xmlNode *output, void *user_data) { char *desc = user_data; if (rc == 0) { crm_debug("Deletion of %s (via CIB call %d) succeeded", desc, call_id); } else { crm_warn("Deletion of %s (via CIB call %d) failed: %s " CRM_XS " rc=%d", desc, call_id, pcmk_strerror(rc), rc); } } // Searches for various portions of PCMK__XE_NODE_STATE to delete // Match a particular node's PCMK__XE_NODE_STATE (takes node name 1x) #define XPATH_NODE_STATE "//" PCMK__XE_NODE_STATE "[@" PCMK_XA_UNAME "='%s']" // Node's lrm section (name 1x) #define XPATH_NODE_LRM XPATH_NODE_STATE "/" PCMK__XE_LRM /* Node's PCMK__XE_LRM_RSC_OP entries and PCMK__XE_LRM_RESOURCE entries without * unexpired lock * (name 2x, (seconds_since_epoch - PCMK_OPT_SHUTDOWN_LOCK_LIMIT) 1x) */ #define XPATH_NODE_LRM_UNLOCKED XPATH_NODE_STATE "//" PCMK__XE_LRM_RSC_OP \ "|" XPATH_NODE_STATE \ "//" PCMK__XE_LRM_RESOURCE \ "[not(@" PCMK_OPT_SHUTDOWN_LOCK ") " \ "or " PCMK_OPT_SHUTDOWN_LOCK "<%lld]" // Node's PCMK__XE_TRANSIENT_ATTRIBUTES section (name 1x) #define XPATH_NODE_ATTRS XPATH_NODE_STATE "/" PCMK__XE_TRANSIENT_ATTRIBUTES // Everything under PCMK__XE_NODE_STATE (name 1x) #define XPATH_NODE_ALL XPATH_NODE_STATE "/*" /* Unlocked history + transient attributes * (name 2x, (seconds_since_epoch - PCMK_OPT_SHUTDOWN_LOCK_LIMIT) 1x, name 1x) */ #define XPATH_NODE_ALL_UNLOCKED XPATH_NODE_LRM_UNLOCKED "|" XPATH_NODE_ATTRS /*! * \internal * \brief Get the XPath and description of a node state section to be deleted * * \param[in] uname Desired node * \param[in] section Subsection of \c PCMK__XE_NODE_STATE to be deleted * \param[out] xpath Where to store XPath of \p section * \param[out] desc If not \c NULL, where to store description of \p section */ void controld_node_state_deletion_strings(const char *uname, enum controld_section_e section, char **xpath, char **desc) { const char *desc_pre = NULL; // Shutdown locks that started before this time are expired long long expire = (long long) time(NULL) - controld_globals.shutdown_lock_limit; switch (section) { case controld_section_lrm: *xpath = crm_strdup_printf(XPATH_NODE_LRM, uname); desc_pre = "resource history"; break; case controld_section_lrm_unlocked: *xpath = crm_strdup_printf(XPATH_NODE_LRM_UNLOCKED, uname, uname, expire); desc_pre = "resource history (other than shutdown locks)"; break; case controld_section_attrs: *xpath = crm_strdup_printf(XPATH_NODE_ATTRS, uname); desc_pre = "transient attributes"; break; case controld_section_all: *xpath = crm_strdup_printf(XPATH_NODE_ALL, uname); desc_pre = "all state"; break; case controld_section_all_unlocked: *xpath = crm_strdup_printf(XPATH_NODE_ALL_UNLOCKED, uname, uname, expire, uname); desc_pre = "all state (other than shutdown locks)"; break; default: // We called this function incorrectly CRM_ASSERT(false); break; } if (desc != NULL) { *desc = crm_strdup_printf("%s for node %s", desc_pre, uname); } } /*! * \internal * \brief Delete subsection of a node's CIB \c PCMK__XE_NODE_STATE * * \param[in] uname Desired node * \param[in] section Subsection of \c PCMK__XE_NODE_STATE to delete * \param[in] options CIB call options to use */ void controld_delete_node_state(const char *uname, enum controld_section_e section, int options) { cib_t *cib = controld_globals.cib_conn; char *xpath = NULL; char *desc = NULL; int cib_rc = pcmk_ok; CRM_ASSERT((uname != NULL) && (cib != NULL)); controld_node_state_deletion_strings(uname, section, &xpath, &desc); cib__set_call_options(options, "node state deletion", cib_xpath|cib_multiple); cib_rc = cib->cmds->remove(cib, xpath, NULL, options); fsa_register_cib_callback(cib_rc, desc, cib_delete_callback); crm_info("Deleting %s (via CIB call %d) " CRM_XS " xpath=%s", desc, cib_rc, xpath); // CIB library handles freeing desc free(xpath); } // Takes node name and resource ID #define XPATH_RESOURCE_HISTORY "//" PCMK__XE_NODE_STATE \ "[@" PCMK_XA_UNAME "='%s']/" \ PCMK__XE_LRM "/" PCMK__XE_LRM_RESOURCES \ "/" PCMK__XE_LRM_RESOURCE \ "[@" PCMK_XA_ID "='%s']" // @TODO could add "and @PCMK_OPT_SHUTDOWN_LOCK" to limit to locks /*! * \internal * \brief Clear resource history from CIB for a given resource and node * * \param[in] rsc_id ID of resource to be cleared * \param[in] node Node whose resource history should be cleared * \param[in] user_name ACL user name to use * \param[in] call_options CIB call options * * \return Standard Pacemaker return code */ int controld_delete_resource_history(const char *rsc_id, const char *node, const char *user_name, int call_options) { char *desc = NULL; char *xpath = NULL; int rc = pcmk_rc_ok; cib_t *cib = controld_globals.cib_conn; CRM_CHECK((rsc_id != NULL) && (node != NULL), return EINVAL); desc = crm_strdup_printf("resource history for %s on %s", rsc_id, node); if (cib == NULL) { crm_err("Unable to clear %s: no CIB connection", desc); free(desc); return ENOTCONN; } // Ask CIB to delete the entry xpath = crm_strdup_printf(XPATH_RESOURCE_HISTORY, node, rsc_id); cib->cmds->set_user(cib, user_name); rc = cib->cmds->remove(cib, xpath, NULL, call_options|cib_xpath); cib->cmds->set_user(cib, NULL); if (rc < 0) { rc = pcmk_legacy2rc(rc); crm_err("Could not delete resource status of %s on %s%s%s: %s " CRM_XS " rc=%d", rsc_id, node, (user_name? " for user " : ""), (user_name? user_name : ""), pcmk_rc_str(rc), rc); free(desc); free(xpath); return rc; } if (pcmk_is_set(call_options, cib_sync_call)) { if (pcmk_is_set(call_options, cib_dryrun)) { crm_debug("Deletion of %s would succeed", desc); } else { crm_debug("Deletion of %s succeeded", desc); } free(desc); } else { crm_info("Clearing %s (via CIB call %d) " CRM_XS " xpath=%s", desc, rc, xpath); fsa_register_cib_callback(rc, desc, cib_delete_callback); // CIB library handles freeing desc } free(xpath); return pcmk_rc_ok; } /*! * \internal * \brief Build XML and string of parameters meeting some criteria, for digest * * \param[in] op Executor event with parameter table to use * \param[in] metadata Parsed meta-data for executed resource agent * \param[in] param_type Flag used for selection criteria * \param[out] result Will be set to newly created XML with selected * parameters as attributes * * \return Newly allocated space-separated string of parameter names * \note Selection criteria varies by param_type: for the restart digest, we * want parameters that are *not* marked reloadable (OCF 1.1) or that * *are* marked unique (pre-1.1), for both string and XML results; for the * secure digest, we want parameters that *are* marked private for the * string, but parameters that are *not* marked private for the XML. * \note It is the caller's responsibility to free the string return value with * \p g_string_free() and the XML result with \p pcmk__xml_free(). */ static GString * build_parameter_list(const lrmd_event_data_t *op, const struct ra_metadata_s *metadata, enum ra_param_flags_e param_type, xmlNode **result) { GString *list = NULL; *result = pcmk__xe_create(NULL, PCMK_XE_PARAMETERS); /* Consider all parameters only except private ones to be consistent with * what scheduler does with calculate_secure_digest(). */ if (param_type == ra_param_private && compare_version(controld_globals.dc_version, "3.16.0") >= 0) { g_hash_table_foreach(op->params, hash2field, *result); pcmk__filter_op_for_digest(*result); } for (GList *iter = metadata->ra_params; iter != NULL; iter = iter->next) { struct ra_param_s *param = (struct ra_param_s *) iter->data; bool accept_for_list = false; bool accept_for_xml = false; switch (param_type) { case ra_param_reloadable: accept_for_list = !pcmk_is_set(param->rap_flags, param_type); accept_for_xml = accept_for_list; break; case ra_param_unique: accept_for_list = pcmk_is_set(param->rap_flags, param_type); accept_for_xml = accept_for_list; break; case ra_param_private: accept_for_list = pcmk_is_set(param->rap_flags, param_type); accept_for_xml = !accept_for_list; break; } if (accept_for_list) { crm_trace("Attr %s is %s", param->rap_name, ra_param_flag2text(param_type)); if (list == NULL) { // We will later search for " WORD ", so start list with a space pcmk__add_word(&list, 256, " "); } pcmk__add_word(&list, 0, param->rap_name); } else { crm_trace("Rejecting %s for %s", param->rap_name, ra_param_flag2text(param_type)); } if (accept_for_xml) { const char *v = g_hash_table_lookup(op->params, param->rap_name); if (v != NULL) { crm_trace("Adding attr %s=%s to the xml result", param->rap_name, v); crm_xml_add(*result, param->rap_name, v); } } else { crm_trace("Removing attr %s from the xml result", param->rap_name); pcmk__xe_remove_attr(*result, param->rap_name); } } if (list != NULL) { // We will later search for " WORD ", so end list with a space pcmk__add_word(&list, 0, " "); } return list; } static void append_restart_list(lrmd_event_data_t *op, struct ra_metadata_s *metadata, xmlNode *update, const char *version) { GString *list = NULL; char *digest = NULL; xmlNode *restart = NULL; CRM_LOG_ASSERT(op->params != NULL); if (op->interval_ms > 0) { /* monitors are not reloadable */ return; } if (pcmk_is_set(metadata->ra_flags, ra_supports_reload_agent)) { /* Add parameters not marked reloadable to the PCMK__XA_OP_FORCE_RESTART * list */ list = build_parameter_list(op, metadata, ra_param_reloadable, &restart); } else if (pcmk_is_set(metadata->ra_flags, ra_supports_legacy_reload)) { /* @COMPAT pre-OCF-1.1 resource agents * * Before OCF 1.1, Pacemaker abused "unique=0" to indicate * reloadability. Add any parameters with unique="1" to the * PCMK__XA_OP_FORCE_RESTART list. */ list = build_parameter_list(op, metadata, ra_param_unique, &restart); } else { // Resource does not support agent reloads return; } - digest = calculate_operation_digest(restart, version); + digest = pcmk__digest_operation(restart); /* Add PCMK__XA_OP_FORCE_RESTART and PCMK__XA_OP_RESTART_DIGEST to indicate * the resource supports reload, no matter if it actually supports any * reloadable parameters */ crm_xml_add(update, PCMK__XA_OP_FORCE_RESTART, (list == NULL)? "" : (const char *) list->str); crm_xml_add(update, PCMK__XA_OP_RESTART_DIGEST, digest); if ((list != NULL) && (list->len > 0)) { crm_trace("%s: %s, %s", op->rsc_id, digest, (const char *) list->str); } else { crm_trace("%s: %s", op->rsc_id, digest); } if (list != NULL) { g_string_free(list, TRUE); } pcmk__xml_free(restart); free(digest); } static void append_secure_list(lrmd_event_data_t *op, struct ra_metadata_s *metadata, xmlNode *update, const char *version) { GString *list = NULL; char *digest = NULL; xmlNode *secure = NULL; CRM_LOG_ASSERT(op->params != NULL); /* To keep PCMK__XA_OP_SECURE_PARAMS short, we want it to contain the secure * parameters but PCMK__XA_OP_SECURE_DIGEST to be based on the insecure ones */ list = build_parameter_list(op, metadata, ra_param_private, &secure); if (list != NULL) { - digest = calculate_operation_digest(secure, version); + digest = pcmk__digest_operation(secure); crm_xml_add(update, PCMK__XA_OP_SECURE_PARAMS, (const char *) list->str); crm_xml_add(update, PCMK__XA_OP_SECURE_DIGEST, digest); crm_trace("%s: %s, %s", op->rsc_id, digest, (const char *) list->str); g_string_free(list, TRUE); } else { crm_trace("%s: no secure parameters", op->rsc_id); } pcmk__xml_free(secure); free(digest); } /*! * \internal * \brief Create XML for a resource history entry * * \param[in] func Function name of caller * \param[in,out] parent XML to add entry to * \param[in] rsc Affected resource * \param[in,out] op Action to add an entry for (or NULL to do nothing) * \param[in] node_name Node where action occurred */ void controld_add_resource_history_xml_as(const char *func, xmlNode *parent, const lrmd_rsc_info_t *rsc, lrmd_event_data_t *op, const char *node_name) { int target_rc = 0; xmlNode *xml_op = NULL; struct ra_metadata_s *metadata = NULL; const char *caller_version = NULL; lrm_state_t *lrm_state = NULL; if (op == NULL) { return; } target_rc = rsc_op_expected_rc(op); caller_version = g_hash_table_lookup(op->params, PCMK_XA_CRM_FEATURE_SET); CRM_CHECK(caller_version != NULL, caller_version = CRM_FEATURE_SET); xml_op = pcmk__create_history_xml(parent, op, caller_version, target_rc, controld_globals.our_nodename, func); if (xml_op == NULL) { return; } if ((rsc == NULL) || (op->params == NULL) || !crm_op_needs_metadata(rsc->standard, op->op_type)) { crm_trace("No digests needed for %s action on %s (params=%p rsc=%p)", op->op_type, op->rsc_id, op->params, rsc); return; } lrm_state = lrm_state_find(node_name); if (lrm_state == NULL) { crm_warn("Cannot calculate digests for operation " PCMK__OP_FMT " because we have no connection to executor for %s", op->rsc_id, op->op_type, op->interval_ms, node_name); return; } /* Ideally the metadata is cached, and the agent is just a fallback. * * @TODO Go through all callers and ensure they get metadata asynchronously * first. */ metadata = controld_get_rsc_metadata(lrm_state, rsc, controld_metadata_from_agent |controld_metadata_from_cache); if (metadata == NULL) { return; } crm_trace("Including additional digests for %s:%s:%s", rsc->standard, rsc->provider, rsc->type); append_restart_list(op, metadata, xml_op, caller_version); append_secure_list(op, metadata, xml_op, caller_version); return; } /*! * \internal * \brief Record an action as pending in the CIB, if appropriate * * \param[in] node_name Node where the action is pending * \param[in] rsc Resource that action is for * \param[in,out] op Pending action * * \return true if action was recorded in CIB, otherwise false */ bool controld_record_pending_op(const char *node_name, const lrmd_rsc_info_t *rsc, lrmd_event_data_t *op) { const char *record_pending = NULL; CRM_CHECK((node_name != NULL) && (rsc != NULL) && (op != NULL), return false); // Never record certain operation types as pending if ((op->op_type == NULL) || (op->params == NULL) || !controld_action_is_recordable(op->op_type)) { return false; } // Check action's PCMK_META_RECORD_PENDING meta-attribute (defaults to true) record_pending = crm_meta_value(op->params, PCMK_META_RECORD_PENDING); if ((record_pending != NULL) && !crm_is_true(record_pending)) { return false; } op->call_id = -1; op->t_run = time(NULL); op->t_rcchange = op->t_run; lrmd__set_result(op, PCMK_OCF_UNKNOWN, PCMK_EXEC_PENDING, NULL); crm_debug("Recording pending %s-interval %s for %s on %s in the CIB", pcmk__readable_interval(op->interval_ms), op->op_type, op->rsc_id, node_name); controld_update_resource_history(node_name, rsc, op, 0); return true; } static void cib_rsc_callback(xmlNode * msg, int call_id, int rc, xmlNode * output, void *user_data) { switch (rc) { case pcmk_ok: case -pcmk_err_diff_failed: case -pcmk_err_diff_resync: crm_trace("Resource history update completed (call=%d rc=%d)", call_id, rc); break; default: if (call_id > 0) { crm_warn("Resource history update %d failed: %s " CRM_XS " rc=%d", call_id, pcmk_strerror(rc), rc); } else { crm_warn("Resource history update failed: %s " CRM_XS " rc=%d", pcmk_strerror(rc), rc); } } if (call_id == pending_rsc_update) { pending_rsc_update = 0; controld_trigger_fsa(); } } /* Only successful stops, and probes that found the resource inactive, get locks * recorded in the history. This ensures the resource stays locked to the node * until it is active there again after the node comes back up. */ static bool should_preserve_lock(lrmd_event_data_t *op) { if (!pcmk_is_set(controld_globals.flags, controld_shutdown_lock_enabled)) { return false; } if (!strcmp(op->op_type, PCMK_ACTION_STOP) && (op->rc == PCMK_OCF_OK)) { return true; } if (!strcmp(op->op_type, PCMK_ACTION_MONITOR) && (op->rc == PCMK_OCF_NOT_RUNNING)) { return true; } return false; } /*! * \internal * \brief Request a CIB update * * \param[in] section Section of CIB to update * \param[in] data New XML of CIB section to update * \param[in] options CIB call options * \param[in] callback If not \c NULL, set this as the operation callback * * \return Standard Pacemaker return code * * \note If \p callback is \p cib_rsc_callback(), the CIB update's call ID is * stored in \p pending_rsc_update on success. */ int controld_update_cib(const char *section, xmlNode *data, int options, void (*callback)(xmlNode *, int, int, xmlNode *, void *)) { cib_t *cib = controld_globals.cib_conn; int cib_rc = -ENOTCONN; CRM_ASSERT(data != NULL); if (cib != NULL) { cib_rc = cib->cmds->modify(cib, section, data, options); if (cib_rc >= 0) { crm_debug("Submitted CIB update %d for %s section", cib_rc, section); } } if (callback == NULL) { if (cib_rc < 0) { crm_err("Failed to update CIB %s section: %s", section, pcmk_rc_str(pcmk_legacy2rc(cib_rc))); } } else { if ((cib_rc >= 0) && (callback == cib_rsc_callback)) { /* Checking for a particular callback is a little hacky, but it * didn't seem worth adding an output argument for cib_rc for just * one use case. */ pending_rsc_update = cib_rc; } fsa_register_cib_callback(cib_rc, NULL, callback); } return (cib_rc >= 0)? pcmk_rc_ok : pcmk_legacy2rc(cib_rc); } /*! * \internal * \brief Update resource history entry in CIB * * \param[in] node_name Node where action occurred * \param[in] rsc Resource that action is for * \param[in,out] op Action to record * \param[in] lock_time If nonzero, when resource was locked to node * * \note On success, the CIB update's call ID will be stored in * pending_rsc_update. */ void controld_update_resource_history(const char *node_name, const lrmd_rsc_info_t *rsc, lrmd_event_data_t *op, time_t lock_time) { xmlNode *update = NULL; xmlNode *xml = NULL; int call_opt = crmd_cib_smart_opt(); const char *node_id = NULL; const char *container = NULL; CRM_CHECK((node_name != NULL) && (op != NULL), return); if (rsc == NULL) { crm_warn("Resource %s no longer exists in the executor", op->rsc_id); controld_ack_event_directly(NULL, NULL, rsc, op, op->rsc_id); return; } // update = pcmk__xe_create(NULL, PCMK_XE_STATUS); // xml = pcmk__xe_create(update, PCMK__XE_NODE_STATE); if (pcmk__str_eq(node_name, controld_globals.our_nodename, pcmk__str_casei)) { node_id = controld_globals.our_uuid; } else { node_id = node_name; pcmk__xe_set_bool_attr(xml, PCMK_XA_REMOTE_NODE, true); } crm_xml_add(xml, PCMK_XA_ID, node_id); crm_xml_add(xml, PCMK_XA_UNAME, node_name); crm_xml_add(xml, PCMK_XA_CRM_DEBUG_ORIGIN, __func__); // xml = pcmk__xe_create(xml, PCMK__XE_LRM); crm_xml_add(xml, PCMK_XA_ID, node_id); // xml = pcmk__xe_create(xml, PCMK__XE_LRM_RESOURCES); // xml = pcmk__xe_create(xml, PCMK__XE_LRM_RESOURCE); crm_xml_add(xml, PCMK_XA_ID, op->rsc_id); crm_xml_add(xml, PCMK_XA_CLASS, rsc->standard); crm_xml_add(xml, PCMK_XA_PROVIDER, rsc->provider); crm_xml_add(xml, PCMK_XA_TYPE, rsc->type); if (lock_time != 0) { /* Actions on a locked resource should either preserve the lock by * recording it with the action result, or clear it. */ if (!should_preserve_lock(op)) { lock_time = 0; } crm_xml_add_ll(xml, PCMK_OPT_SHUTDOWN_LOCK, (long long) lock_time); } if (op->params != NULL) { container = g_hash_table_lookup(op->params, CRM_META "_" PCMK__META_CONTAINER); if (container != NULL) { crm_trace("Resource %s is a part of container resource %s", op->rsc_id, container); crm_xml_add(xml, PCMK__META_CONTAINER, container); } } // (possibly more than one) controld_add_resource_history_xml(xml, rsc, op, node_name); /* Update CIB asynchronously. Even if it fails, the resource state should be * discovered during the next election. Worst case, the node is wrongly * fenced for running a resource it isn't. */ crm_log_xml_trace(update, __func__); controld_update_cib(PCMK_XE_STATUS, update, call_opt, cib_rsc_callback); pcmk__xml_free(update); } /*! * \internal * \brief Erase an LRM history entry from the CIB, given the operation data * * \param[in] op Operation whose history should be deleted */ void controld_delete_action_history(const lrmd_event_data_t *op) { xmlNode *xml_top = NULL; CRM_CHECK(op != NULL, return); xml_top = pcmk__xe_create(NULL, PCMK__XE_LRM_RSC_OP); crm_xml_add_int(xml_top, PCMK__XA_CALL_ID, op->call_id); crm_xml_add(xml_top, PCMK__XA_TRANSITION_KEY, op->user_data); if (op->interval_ms > 0) { char *op_id = pcmk__op_key(op->rsc_id, op->op_type, op->interval_ms); /* Avoid deleting last_failure too (if it was a result of this recurring op failing) */ crm_xml_add(xml_top, PCMK_XA_ID, op_id); free(op_id); } crm_debug("Erasing resource operation history for " PCMK__OP_FMT " (call=%d)", op->rsc_id, op->op_type, op->interval_ms, op->call_id); controld_globals.cib_conn->cmds->remove(controld_globals.cib_conn, PCMK_XE_STATUS, xml_top, cib_none); crm_log_xml_trace(xml_top, "op:cancel"); pcmk__xml_free(xml_top); } /* Define xpath to find LRM resource history entry by node and resource */ #define XPATH_HISTORY \ "/" PCMK_XE_CIB "/" PCMK_XE_STATUS \ "/" PCMK__XE_NODE_STATE "[@" PCMK_XA_UNAME "='%s']" \ "/" PCMK__XE_LRM "/" PCMK__XE_LRM_RESOURCES \ "/" PCMK__XE_LRM_RESOURCE "[@" PCMK_XA_ID "='%s']" \ "/" PCMK__XE_LRM_RSC_OP /* ... and also by operation key */ #define XPATH_HISTORY_ID XPATH_HISTORY "[@" PCMK_XA_ID "='%s']" /* ... and also by operation key and operation call ID */ #define XPATH_HISTORY_CALL XPATH_HISTORY \ "[@" PCMK_XA_ID "='%s' and @" PCMK__XA_CALL_ID "='%d']" /* ... and also by operation key and original operation key */ #define XPATH_HISTORY_ORIG XPATH_HISTORY \ "[@" PCMK_XA_ID "='%s' and @" PCMK__XA_OPERATION_KEY "='%s']" /*! * \internal * \brief Delete a last_failure resource history entry from the CIB * * \param[in] rsc_id Name of resource to clear history for * \param[in] node Name of node to clear history for * \param[in] action If specified, delete only if this was failed action * \param[in] interval_ms If \p action is specified, it has this interval */ void controld_cib_delete_last_failure(const char *rsc_id, const char *node, const char *action, guint interval_ms) { char *xpath = NULL; char *last_failure_key = NULL; CRM_CHECK((rsc_id != NULL) && (node != NULL), return); // Generate XPath to match desired entry last_failure_key = pcmk__op_key(rsc_id, "last_failure", 0); if (action == NULL) { xpath = crm_strdup_printf(XPATH_HISTORY_ID, node, rsc_id, last_failure_key); } else { char *action_key = pcmk__op_key(rsc_id, action, interval_ms); xpath = crm_strdup_printf(XPATH_HISTORY_ORIG, node, rsc_id, last_failure_key, action_key); free(action_key); } free(last_failure_key); controld_globals.cib_conn->cmds->remove(controld_globals.cib_conn, xpath, NULL, cib_xpath); free(xpath); } /*! * \internal * \brief Delete resource history entry from the CIB, given operation key * * \param[in] rsc_id Name of resource to clear history for * \param[in] node Name of node to clear history for * \param[in] key Operation key of operation to clear history for * \param[in] call_id If specified, delete entry only if it has this call ID */ void controld_delete_action_history_by_key(const char *rsc_id, const char *node, const char *key, int call_id) { char *xpath = NULL; CRM_CHECK((rsc_id != NULL) && (node != NULL) && (key != NULL), return); if (call_id > 0) { xpath = crm_strdup_printf(XPATH_HISTORY_CALL, node, rsc_id, key, call_id); } else { xpath = crm_strdup_printf(XPATH_HISTORY_ID, node, rsc_id, key); } controld_globals.cib_conn->cmds->remove(controld_globals.cib_conn, xpath, NULL, cib_xpath); free(xpath); } diff --git a/daemons/fenced/fenced_commands.c b/daemons/fenced/fenced_commands.c index 01e2276bc8..8d884d27b6 100644 --- a/daemons/fenced/fenced_commands.c +++ b/daemons/fenced/fenced_commands.c @@ -1,3709 +1,3709 @@ /* * Copyright 2009-2024 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 #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_suicide; /* 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; int 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; 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_CALLOPT, &(cmd->options)); 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; 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; CRM_ASSERT(device); 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 += ((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); pcmk__xml_free(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? */ + // Use pcmk__digest_operation() 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_MAX; 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_MAX; 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 <= 0) || (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_MAX)) { 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_suicide 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_suicide) { 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_suicide) { 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_suicide)) { 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 suicide, 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_suicide = suicide; 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_suicide Whether self-fencing is allowed */ static void add_disallowed(xmlNode *xml, const char *action, const stonith_device_t *device, const char *target, gboolean allow_suicide) { if (!localhost_is_eligible(device, action, target, allow_suicide)) { 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_suicide 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_suicide) { 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_suicide); } /*! * \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_suicide)); add_action_reply(dev, PCMK_ACTION_OFF, device, query->target, pcmk_is_set(query->call_options, st_opt_allow_suicide)); 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: pcmk__xml_free(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"); pcmk__xml_free(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, 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"); CRM_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_suicide), 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; CRM_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); } pcmk__xml_free(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); } pcmk__xml_free(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) { int call_options = st_opt_none; bool is_reply = false; CRM_CHECK(message != NULL, return); if (get_xpath_object("//" PCMK__XE_ST_REPLY, message, LOG_NEVER) != NULL) { is_reply = true; } crm_element_value_int(message, PCMK__XA_ST_CALLOPT, &call_options); 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)) { CRM_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/include/crm/common/digest_internal.h b/include/crm/common/digest_internal.h index 2ed4238bd3..ad446fdef9 100644 --- a/include/crm/common/digest_internal.h +++ b/include/crm/common/digest_internal.h @@ -1,44 +1,45 @@ /* * Copyright 2015-2024 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. */ #ifndef PCMK__CRM_COMMON_DIGEST_INTERNAL__H #define PCMK__CRM_COMMON_DIGEST_INTERNAL__H /* * Internal-only functions to create digest strings from XML */ #include #include // xmlNode // Digest comparison results enum pcmk__digest_result { pcmk__digest_unknown, // No digest available for comparison pcmk__digest_match, // Digests match pcmk__digest_mismatch, // Any parameter changed (potentially reloadable) pcmk__digest_restart, // Parameters that require a restart changed }; // Information needed to compare operation digests typedef struct { enum pcmk__digest_result rc; // Result of digest comparison xmlNode *params_all; // All operation parameters xmlNode *params_secure; // Parameters marked private xmlNode *params_restart; // Parameters marked not reloadable char *digest_all_calc; // Digest of params_all char *digest_secure_calc; // Digest of params_secure char *digest_restart_calc; // Digest of params_restart } pcmk__op_digest_t; char *pcmk__digest_on_disk_cib(xmlNode *input); +char *pcmk__digest_operation(xmlNode *input); bool pcmk__verify_digest(xmlNode *input, const char *expected); #endif // PCMK__CRM_COMMON_DIGEST_INTERNAL__H diff --git a/lib/common/digest.c b/lib/common/digest.c index 31de289e88..59987c5bac 100644 --- a/lib/common/digest.c +++ b/lib/common/digest.c @@ -1,364 +1,378 @@ /* * Copyright 2015-2024 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 // GString, etc. #include // gnutls_hash_fast(), gnutls_hash_get_len() #include // gnutls_strerror() #include #include #include "crmcommon_private.h" #define BEST_EFFORT_STATUS 0 /*! * \internal * \brief Dump XML in a format used with v1 digests * * \param[in] xml Root of XML to dump * * \return Newly allocated buffer containing dumped XML */ static GString * dump_xml_for_digest(xmlNodePtr xml) { GString *buffer = g_string_sized_new(1024); /* for compatibility with the old result which is used for v1 digests */ g_string_append_c(buffer, ' '); pcmk__xml_string(xml, 0, buffer, 0); g_string_append_c(buffer, '\n'); return buffer; } /*! * \internal * \brief Calculate and return v1 digest of XML tree * * \param[in] input Root of XML to digest * \param[in] sort Whether to sort the XML before calculating digest * * \return Newly allocated string containing digest * * \note Example return value: "c048eae664dba840e1d2060f00299e9d" */ static char * calculate_xml_digest_v1(xmlNode *input, bool sort) { char *digest = NULL; GString *buffer = NULL; xmlNode *copy = NULL; if (sort) { crm_trace("Sorting xml..."); copy = sorted_xml(input, NULL, TRUE); crm_trace("Done"); input = copy; } buffer = dump_xml_for_digest(input); CRM_CHECK(buffer->len > 0, pcmk__xml_free(copy); g_string_free(buffer, TRUE); return NULL); digest = crm_md5sum((const char *) buffer->str); crm_log_xml_trace(input, "digest:source"); g_string_free(buffer, TRUE); pcmk__xml_free(copy); return digest; } /*! * \internal * \brief Calculate and return v2 digest of XML tree * * \param[in] source Root of XML to digest * \param[in] filter Whether to filter certain XML attributes * * \return Newly allocated string containing digest */ static char * calculate_xml_digest_v2(const xmlNode *source, bool filter) { char *digest = NULL; GString *buf = g_string_sized_new(1024); pcmk__xml_string(source, (filter? pcmk__xml_fmt_filtered : 0), buf, 0); digest = crm_md5sum(buf->str); pcmk__if_tracing( { char *trace_file = crm_strdup_printf("%s/digest-%s", pcmk__get_tmpdir(), digest); crm_trace("Saving %s.%s.%s to %s", crm_element_value(source, PCMK_XA_ADMIN_EPOCH), crm_element_value(source, PCMK_XA_EPOCH), crm_element_value(source, PCMK_XA_NUM_UPDATES), trace_file); save_xml_to_file(source, "digest input", trace_file); free(trace_file); }, {} ); g_string_free(buf, TRUE); return digest; } /*! * \internal * \brief Calculate and return the digest of a CIB, suitable for storing on disk * * \param[in] input Root of XML to digest * * \return Newly allocated string containing digest */ char * pcmk__digest_on_disk_cib(xmlNode *input) { /* Always use the v1 format for on-disk digests. * * Switching to v2 is a compatibility nightmare. * * We only use this once at startup. All other invocations are in a * separate child process. */ return calculate_xml_digest_v1(input, false); } +/*! + * \internal + * \brief Calculate and return digest of an operation XML element + * + * \param[in] input Root of XML to digest + * + * \return Newly allocated string containing digest + */ +char * +pcmk__digest_operation(xmlNode *input) +{ + return calculate_xml_digest_v1(input, true); +} + /*! * \brief Calculate and return digest of XML operation * * \param[in] input Root of XML to digest * \param[in] version Unused * * \return Newly allocated string containing digest */ char * calculate_operation_digest(xmlNode *input, const char *version) { /* We still need the sorting for operation digests */ return calculate_xml_digest_v1(input, true); } /*! * \brief Calculate and return digest of XML tree * * \param[in] input Root of XML to digest * \param[in] sort Whether to sort XML before calculating digest * \param[in] do_filter Whether to filter certain XML attributes * \param[in] version CRM feature set version (used to select v1/v2 digest) * * \return Newly allocated string containing digest */ char * calculate_xml_versioned_digest(xmlNode *input, gboolean sort, gboolean do_filter, const char *version) { /* * @COMPAT digests (on-disk or in diffs/patchsets) created <1.1.4; * removing this affects even full-restart upgrades from old versions * * The sorting associated with v1 digest creation accounted for 23% of * the CIB manager's CPU usage on the server. v2 drops this. * * The filtering accounts for an additional 2.5% and we may want to * remove it in future. * * v2 also uses the xmlBuffer contents directly to avoid additional copying */ if ((version == NULL) || (compare_version("3.0.5", version) > 0)) { crm_trace("Using v1 digest algorithm for %s", pcmk__s(version, "unknown feature set")); return calculate_xml_digest_v1(input, sort); } crm_trace("Using v2 digest algorithm for %s", version); return calculate_xml_digest_v2(input, do_filter); } /*! * \internal * \brief Check whether calculated digest of given XML matches expected digest * * \param[in] input Root of XML tree to digest * \param[in] expected Expected digest in on-disk format * * \return true if digests match, false on mismatch or error */ bool pcmk__verify_digest(xmlNode *input, const char *expected) { char *calculated = NULL; bool passed; if (input != NULL) { calculated = pcmk__digest_on_disk_cib(input); if (calculated == NULL) { crm_perror(LOG_ERR, "Could not calculate digest for comparison"); return false; } } passed = pcmk__str_eq(expected, calculated, pcmk__str_casei); if (passed) { crm_trace("Digest comparison passed: %s", calculated); } else { crm_err("Digest comparison failed: expected %s, calculated %s", expected, calculated); } free(calculated); return passed; } /*! * \internal * \brief Check whether an XML attribute should be excluded from CIB digests * * \param[in] name XML attribute name * * \return true if XML attribute should be excluded from CIB digest calculation */ bool pcmk__xa_filterable(const char *name) { static const char *filter[] = { PCMK_XA_CRM_DEBUG_ORIGIN, PCMK_XA_CIB_LAST_WRITTEN, PCMK_XA_UPDATE_ORIGIN, PCMK_XA_UPDATE_CLIENT, PCMK_XA_UPDATE_USER, }; for (int i = 0; i < PCMK__NELEM(filter); i++) { if (strcmp(name, filter[i]) == 0) { return true; } } return false; } char * crm_md5sum(const char *buffer) { unsigned int dlen = gnutls_hash_get_len(GNUTLS_DIG_MD5); unsigned char *raw_digest = NULL; char *digest = NULL; int rc = 0; if (dlen == 0) { return NULL; } if (buffer == NULL) { return NULL; } raw_digest = pcmk__assert_alloc(dlen, sizeof(unsigned char)); rc = gnutls_hash_fast(GNUTLS_DIG_MD5, buffer, strlen(buffer), raw_digest); if (rc < 0) { free(raw_digest); crm_err("Failed to calculate hash: %s", gnutls_strerror(rc)); return NULL; } digest = pcmk__assert_alloc(1 + (2 * dlen), sizeof(char)); for (int i = 0; i < dlen; i++) { sprintf(digest + (2 * i), "%02x", raw_digest[i]); } digest[(2 * dlen)] = 0; free(raw_digest); crm_trace("Digest %s.", digest); return digest; } // Return true if a is an attribute that should be filtered static bool should_filter_for_digest(xmlAttrPtr a, void *user_data) { if (strncmp((const char *) a->name, CRM_META "_", sizeof(CRM_META " ") - 1) == 0) { return true; } return pcmk__str_any_of((const char *) a->name, PCMK_XA_ID, PCMK_XA_CRM_FEATURE_SET, PCMK__XA_OP_DIGEST, PCMK__META_ON_NODE, PCMK__META_ON_NODE_UUID, "pcmk_external_ip", NULL); } /*! * \internal * \brief Remove XML attributes not needed for operation digest * * \param[in,out] param_set XML with operation parameters */ void pcmk__filter_op_for_digest(xmlNode *param_set) { char *key = NULL; char *timeout = NULL; guint interval_ms = 0; if (param_set == NULL) { return; } /* Timeout is useful for recurring operation digests, so grab it before * removing meta-attributes */ key = crm_meta_name(PCMK_META_INTERVAL); if (crm_element_value_ms(param_set, key, &interval_ms) != pcmk_ok) { interval_ms = 0; } free(key); key = NULL; if (interval_ms != 0) { key = crm_meta_name(PCMK_META_TIMEOUT); timeout = crm_element_value_copy(param_set, key); } // Remove all CRM_meta_* attributes and certain other attributes pcmk__xe_remove_matching_attrs(param_set, should_filter_for_digest, NULL); // Add timeout back for recurring operation digests if (timeout != NULL) { crm_xml_add(param_set, key, timeout); } free(timeout); free(key); } // Deprecated functions kept only for backward API compatibility // LCOV_EXCL_START #include char * calculate_on_disk_digest(xmlNode *input) { return calculate_xml_digest_v1(input, false); } // LCOV_EXCL_STOP // End deprecated API diff --git a/lib/pacemaker/pcmk_sched_actions.c b/lib/pacemaker/pcmk_sched_actions.c index 3fca1ea9ca..d28f636743 100644 --- a/lib/pacemaker/pcmk_sched_actions.c +++ b/lib/pacemaker/pcmk_sched_actions.c @@ -1,1941 +1,1941 @@ /* * 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 #include #include #include #include #include #include #include "libpacemaker_private.h" /*! * \internal * \brief Get the action flags relevant to ordering constraints * * \param[in,out] action Action to check * \param[in] node Node that *other* action in the ordering is on * (used only for clone resource actions) * * \return Action flags that should be used for orderings */ static uint32_t action_flags_for_ordering(pcmk_action_t *action, const pcmk_node_t *node) { bool runnable = false; uint32_t flags; // For non-resource actions, return the action flags if (action->rsc == NULL) { return action->flags; } /* For non-clone resources, or a clone action not assigned to a node, * return the flags as determined by the resource method without a node * specified. */ flags = action->rsc->private->cmds->action_flags(action, NULL); if ((node == NULL) || !pcmk__is_clone(action->rsc)) { return flags; } /* Otherwise (i.e., for clone resource actions on a specific node), first * remember whether the non-node-specific action is runnable. */ runnable = pcmk_is_set(flags, pcmk_action_runnable); // Then recheck the resource method with the node flags = action->rsc->private->cmds->action_flags(action, node); /* For clones in ordering constraints, the node-specific "runnable" doesn't * matter, just the non-node-specific setting (i.e., is the action runnable * anywhere). * * This applies only to runnable, and only for ordering constraints. This * function shouldn't be used for other types of constraints without * changes. Not very satisfying, but it's logical and appears to work well. */ if (runnable && !pcmk_is_set(flags, pcmk_action_runnable)) { pcmk__set_raw_action_flags(flags, action->rsc->id, pcmk_action_runnable); } return flags; } /*! * \internal * \brief Get action UUID that should be used with a resource ordering * * When an action is ordered relative to an action for a collective resource * (clone, group, or bundle), it actually needs to be ordered after all * instances of the collective have completed the relevant action (for example, * given "start CLONE then start RSC", RSC must wait until all instances of * CLONE have started). Given the UUID and resource of the first action in an * ordering, this returns the UUID of the action that should actually be used * for ordering (for example, "CLONE_started_0" instead of "CLONE_start_0"). * * \param[in] first_uuid UUID of first action in ordering * \param[in] first_rsc Resource of first action in ordering * * \return Newly allocated copy of UUID to use with ordering * \note It is the caller's responsibility to free the return value. */ static char * action_uuid_for_ordering(const char *first_uuid, const pcmk_resource_t *first_rsc) { guint interval_ms = 0; char *uuid = NULL; char *rid = NULL; char *first_task_str = NULL; enum action_tasks first_task = pcmk_action_unspecified; enum action_tasks remapped_task = pcmk_action_unspecified; // Only non-notify actions for collective resources need remapping if ((strstr(first_uuid, PCMK_ACTION_NOTIFY) != NULL) || (first_rsc->private->variant < pcmk__rsc_variant_group)) { goto done; } // Only non-recurring actions need remapping CRM_ASSERT(parse_op_key(first_uuid, &rid, &first_task_str, &interval_ms)); if (interval_ms > 0) { goto done; } first_task = pcmk_parse_action(first_task_str); switch (first_task) { case pcmk_action_stop: case pcmk_action_start: case pcmk_action_notify: case pcmk_action_promote: case pcmk_action_demote: remapped_task = first_task + 1; break; case pcmk_action_stopped: case pcmk_action_started: case pcmk_action_notified: case pcmk_action_promoted: case pcmk_action_demoted: remapped_task = first_task; break; case pcmk_action_monitor: case pcmk_action_shutdown: case pcmk_action_fence: break; default: crm_err("Unknown action '%s' in ordering", first_task_str); break; } if (remapped_task != pcmk_action_unspecified) { /* If a clone or bundle has notifications enabled, the ordering will be * relative to when notifications have been sent for the remapped task. */ if (pcmk_is_set(first_rsc->flags, pcmk_rsc_notify) && (pcmk__is_clone(first_rsc) || pcmk__is_bundled(first_rsc))) { uuid = pcmk__notify_key(rid, "confirmed-post", pcmk_action_text(remapped_task)); } else { uuid = pcmk__op_key(rid, pcmk_action_text(remapped_task), 0); } pcmk__rsc_trace(first_rsc, "Remapped action UUID %s to %s for ordering purposes", first_uuid, uuid); } done: free(first_task_str); free(rid); return (uuid != NULL)? uuid : pcmk__str_copy(first_uuid); } /*! * \internal * \brief Get actual action that should be used with an ordering * * When an action is ordered relative to an action for a collective resource * (clone, group, or bundle), it actually needs to be ordered after all * instances of the collective have completed the relevant action (for example, * given "start CLONE then start RSC", RSC must wait until all instances of * CLONE have started). Given the first action in an ordering, this returns the * the action that should actually be used for ordering (for example, the * started action instead of the start action). * * \param[in] action First action in an ordering * * \return Actual action that should be used for the ordering */ static pcmk_action_t * action_for_ordering(pcmk_action_t *action) { pcmk_action_t *result = action; pcmk_resource_t *rsc = action->rsc; if (rsc == NULL) { return result; } if ((rsc->private->variant >= pcmk__rsc_variant_group) && (action->uuid != NULL)) { char *uuid = action_uuid_for_ordering(action->uuid, rsc); result = find_first_action(rsc->actions, uuid, NULL, NULL); if (result == NULL) { crm_warn("Not remapping %s to %s because %s does not have " "remapped action", action->uuid, uuid, rsc->id); result = action; } free(uuid); } return result; } /*! * \internal * \brief Wrapper for update_ordered_actions() method for readability * * \param[in,out] rsc Resource to call method for * \param[in,out] first 'First' action in an ordering * \param[in,out] then 'Then' action in an ordering * \param[in] node If not NULL, limit scope of ordering to this * node (only used when interleaving instances) * \param[in] flags Action flags for \p first for ordering purposes * \param[in] filter Action flags to limit scope of certain updates * (may include pcmk_action_optional to affect only * mandatory actions, and pe_action_runnable to * affect only runnable actions) * \param[in] type Group of enum pcmk__action_relation_flags to apply * \param[in,out] scheduler Scheduler data * * \return Group of enum pcmk__updated flags indicating what was updated */ static inline uint32_t update(pcmk_resource_t *rsc, pcmk_action_t *first, pcmk_action_t *then, const pcmk_node_t *node, uint32_t flags, uint32_t filter, uint32_t type, pcmk_scheduler_t *scheduler) { return rsc->private->cmds->update_ordered_actions(first, then, node, flags, filter, type, scheduler); } /*! * \internal * \brief Update flags for ordering's actions appropriately for ordering's flags * * \param[in,out] first First action in an ordering * \param[in,out] then Then action in an ordering * \param[in] first_flags Action flags for \p first for ordering purposes * \param[in] then_flags Action flags for \p then for ordering purposes * \param[in,out] order Action wrapper for \p first in ordering * \param[in,out] scheduler Scheduler data * * \return Group of enum pcmk__updated flags */ static uint32_t update_action_for_ordering_flags(pcmk_action_t *first, pcmk_action_t *then, uint32_t first_flags, uint32_t then_flags, pcmk__related_action_t *order, pcmk_scheduler_t *scheduler) { uint32_t changed = pcmk__updated_none; /* The node will only be used for clones. If interleaved, node will be NULL, * otherwise the ordering scope will be limited to the node. Normally, the * whole 'then' clone should restart if 'first' is restarted, so then->node * is needed. */ pcmk_node_t *node = then->node; if (pcmk_is_set(order->type, pcmk__ar_first_implies_same_node_then)) { /* For unfencing, only instances of 'then' on the same node as 'first' * (the unfencing operation) should restart, so reset node to * first->node, at which point this case is handled like a normal * pcmk__ar_first_implies_then. */ pcmk__clear_relation_flags(order->type, pcmk__ar_first_implies_same_node_then); pcmk__set_relation_flags(order->type, pcmk__ar_first_implies_then); node = first->node; pcmk__rsc_trace(then->rsc, "%s then %s: mapped " "pcmk__ar_first_implies_same_node_then to " "pcmk__ar_first_implies_then on %s", first->uuid, then->uuid, pcmk__node_name(node)); } if (pcmk_is_set(order->type, pcmk__ar_first_implies_then)) { if (then->rsc != NULL) { changed |= update(then->rsc, first, then, node, first_flags & pcmk_action_optional, pcmk_action_optional, pcmk__ar_first_implies_then, scheduler); } else if (!pcmk_is_set(first_flags, pcmk_action_optional) && pcmk_is_set(then->flags, pcmk_action_optional)) { pcmk__clear_action_flags(then, pcmk_action_optional); pcmk__set_updated_flags(changed, first, pcmk__updated_then); } pcmk__rsc_trace(then->rsc, "%s then %s: %s after pcmk__ar_first_implies_then", first->uuid, then->uuid, (changed? "changed" : "unchanged")); } if (pcmk_is_set(order->type, pcmk__ar_intermediate_stop) && (then->rsc != NULL)) { enum pe_action_flags restart = pcmk_action_optional |pcmk_action_runnable; changed |= update(then->rsc, first, then, node, first_flags, restart, pcmk__ar_intermediate_stop, scheduler); pcmk__rsc_trace(then->rsc, "%s then %s: %s after pcmk__ar_intermediate_stop", first->uuid, then->uuid, (changed? "changed" : "unchanged")); } if (pcmk_is_set(order->type, pcmk__ar_then_implies_first)) { if (first->rsc != NULL) { changed |= update(first->rsc, first, then, node, first_flags, pcmk_action_optional, pcmk__ar_then_implies_first, scheduler); } else if (!pcmk_is_set(first_flags, pcmk_action_optional) && pcmk_is_set(first->flags, pcmk_action_runnable)) { pcmk__clear_action_flags(first, pcmk_action_runnable); pcmk__set_updated_flags(changed, first, pcmk__updated_first); } pcmk__rsc_trace(then->rsc, "%s then %s: %s after pcmk__ar_then_implies_first", first->uuid, then->uuid, (changed? "changed" : "unchanged")); } if (pcmk_is_set(order->type, pcmk__ar_promoted_then_implies_first)) { if (then->rsc != NULL) { changed |= update(then->rsc, first, then, node, first_flags & pcmk_action_optional, pcmk_action_optional, pcmk__ar_promoted_then_implies_first, scheduler); } pcmk__rsc_trace(then->rsc, "%s then %s: %s after " "pcmk__ar_promoted_then_implies_first", first->uuid, then->uuid, (changed? "changed" : "unchanged")); } if (pcmk_is_set(order->type, pcmk__ar_min_runnable)) { if (then->rsc != NULL) { changed |= update(then->rsc, first, then, node, first_flags, pcmk_action_runnable, pcmk__ar_min_runnable, scheduler); } else if (pcmk_is_set(first_flags, pcmk_action_runnable)) { // We have another runnable instance of "first" then->runnable_before++; /* Mark "then" as runnable if it requires a certain number of * "before" instances to be runnable, and they now are. */ if ((then->runnable_before >= then->required_runnable_before) && !pcmk_is_set(then->flags, pcmk_action_runnable)) { pcmk__set_action_flags(then, pcmk_action_runnable); pcmk__set_updated_flags(changed, first, pcmk__updated_then); } } pcmk__rsc_trace(then->rsc, "%s then %s: %s after pcmk__ar_min_runnable", first->uuid, then->uuid, (changed? "changed" : "unchanged")); } if (pcmk_is_set(order->type, pcmk__ar_nested_remote_probe) && (then->rsc != NULL)) { if (!pcmk_is_set(first_flags, pcmk_action_runnable) && (first->rsc != NULL) && (first->rsc->running_on != NULL)) { pcmk__rsc_trace(then->rsc, "%s then %s: ignoring because first is stopping", first->uuid, then->uuid); order->type = (enum pe_ordering) pcmk__ar_none; } else { changed |= update(then->rsc, first, then, node, first_flags, pcmk_action_runnable, pcmk__ar_unrunnable_first_blocks, scheduler); } pcmk__rsc_trace(then->rsc, "%s then %s: %s after pcmk__ar_nested_remote_probe", first->uuid, then->uuid, (changed? "changed" : "unchanged")); } if (pcmk_is_set(order->type, pcmk__ar_unrunnable_first_blocks)) { if (then->rsc != NULL) { changed |= update(then->rsc, first, then, node, first_flags, pcmk_action_runnable, pcmk__ar_unrunnable_first_blocks, scheduler); } else if (!pcmk_is_set(first_flags, pcmk_action_runnable) && pcmk_is_set(then->flags, pcmk_action_runnable)) { pcmk__clear_action_flags(then, pcmk_action_runnable); pcmk__set_updated_flags(changed, first, pcmk__updated_then); } pcmk__rsc_trace(then->rsc, "%s then %s: %s after pcmk__ar_unrunnable_first_blocks", first->uuid, then->uuid, (changed? "changed" : "unchanged")); } if (pcmk_is_set(order->type, pcmk__ar_unmigratable_then_blocks)) { if (then->rsc != NULL) { changed |= update(then->rsc, first, then, node, first_flags, pcmk_action_optional, pcmk__ar_unmigratable_then_blocks, scheduler); } pcmk__rsc_trace(then->rsc, "%s then %s: %s after " "pcmk__ar_unmigratable_then_blocks", first->uuid, then->uuid, (changed? "changed" : "unchanged")); } if (pcmk_is_set(order->type, pcmk__ar_first_else_then)) { if (then->rsc != NULL) { changed |= update(then->rsc, first, then, node, first_flags, pcmk_action_optional, pcmk__ar_first_else_then, scheduler); } pcmk__rsc_trace(then->rsc, "%s then %s: %s after pcmk__ar_first_else_then", first->uuid, then->uuid, (changed? "changed" : "unchanged")); } if (pcmk_is_set(order->type, pcmk__ar_ordered)) { if (then->rsc != NULL) { changed |= update(then->rsc, first, then, node, first_flags, pcmk_action_runnable, pcmk__ar_ordered, scheduler); } pcmk__rsc_trace(then->rsc, "%s then %s: %s after pcmk__ar_ordered", first->uuid, then->uuid, (changed? "changed" : "unchanged")); } if (pcmk_is_set(order->type, pcmk__ar_asymmetric)) { if (then->rsc != NULL) { changed |= update(then->rsc, first, then, node, first_flags, pcmk_action_runnable, pcmk__ar_asymmetric, scheduler); } pcmk__rsc_trace(then->rsc, "%s then %s: %s after pcmk__ar_asymmetric", first->uuid, then->uuid, (changed? "changed" : "unchanged")); } if (pcmk_is_set(first->flags, pcmk_action_runnable) && pcmk_is_set(order->type, pcmk__ar_first_implies_then_graphed) && !pcmk_is_set(first_flags, pcmk_action_optional)) { pcmk__rsc_trace(then->rsc, "%s will be in graph because %s is required", then->uuid, first->uuid); pcmk__set_action_flags(then, pcmk_action_always_in_graph); // Don't bother marking 'then' as changed just for this } if (pcmk_is_set(order->type, pcmk__ar_then_implies_first_graphed) && !pcmk_is_set(then_flags, pcmk_action_optional)) { pcmk__rsc_trace(then->rsc, "%s will be in graph because %s is required", first->uuid, then->uuid); pcmk__set_action_flags(first, pcmk_action_always_in_graph); // Don't bother marking 'first' as changed just for this } if (pcmk_any_flags_set(order->type, pcmk__ar_first_implies_then |pcmk__ar_then_implies_first |pcmk__ar_intermediate_stop) && (first->rsc != NULL) && !pcmk_is_set(first->rsc->flags, pcmk_rsc_managed) && pcmk_is_set(first->rsc->flags, pcmk_rsc_blocked) && !pcmk_is_set(first->flags, pcmk_action_runnable) && pcmk__str_eq(first->task, PCMK_ACTION_STOP, pcmk__str_none)) { if (pcmk_is_set(then->flags, pcmk_action_runnable)) { pcmk__clear_action_flags(then, pcmk_action_runnable); pcmk__set_updated_flags(changed, first, pcmk__updated_then); } pcmk__rsc_trace(then->rsc, "%s then %s: %s after checking whether first " "is blocked, unmanaged, unrunnable stop", first->uuid, then->uuid, (changed? "changed" : "unchanged")); } return changed; } // Convenience macros for logging action properties #define action_type_str(flags) \ (pcmk_is_set((flags), pcmk_action_pseudo)? "pseudo-action" : "action") #define action_optional_str(flags) \ (pcmk_is_set((flags), pcmk_action_optional)? "optional" : "required") #define action_runnable_str(flags) \ (pcmk_is_set((flags), pcmk_action_runnable)? "runnable" : "unrunnable") #define action_node_str(a) \ (((a)->node == NULL)? "no node" : (a)->node->details->uname) /*! * \internal * \brief Update an action's flags for all orderings where it is "then" * * \param[in,out] then Action to update * \param[in,out] scheduler Scheduler data */ void pcmk__update_action_for_orderings(pcmk_action_t *then, pcmk_scheduler_t *scheduler) { GList *lpc = NULL; uint32_t changed = pcmk__updated_none; int last_flags = then->flags; pcmk__rsc_trace(then->rsc, "Updating %s %s (%s %s) on %s", action_type_str(then->flags), then->uuid, action_optional_str(then->flags), action_runnable_str(then->flags), action_node_str(then)); if (pcmk_is_set(then->flags, pcmk_action_min_runnable)) { /* Initialize current known "runnable before" actions. As * update_action_for_ordering_flags() is called for each of then's * before actions, this number will increment as runnable 'first' * actions are encountered. */ then->runnable_before = 0; if (then->required_runnable_before == 0) { /* @COMPAT This ordering constraint uses the deprecated * PCMK_XA_REQUIRE_ALL=PCMK_VALUE_FALSE attribute. Treat it like * PCMK_META_CLONE_MIN=1. */ then->required_runnable_before = 1; } /* The pcmk__ar_min_runnable clause of * update_action_for_ordering_flags() (called below) * will reset runnable if appropriate. */ pcmk__clear_action_flags(then, pcmk_action_runnable); } for (lpc = then->actions_before; lpc != NULL; lpc = lpc->next) { pcmk__related_action_t *other = lpc->data; pcmk_action_t *first = other->action; pcmk_node_t *then_node = then->node; pcmk_node_t *first_node = first->node; if ((first->rsc != NULL) && pcmk__is_group(first->rsc) && pcmk__str_eq(first->task, PCMK_ACTION_START, pcmk__str_none)) { first_node = first->rsc->private->fns->location(first->rsc, NULL, FALSE); if (first_node != NULL) { pcmk__rsc_trace(first->rsc, "Found %s for 'first' %s", pcmk__node_name(first_node), first->uuid); } } if (pcmk__is_group(then->rsc) && pcmk__str_eq(then->task, PCMK_ACTION_START, pcmk__str_none)) { then_node = then->rsc->private->fns->location(then->rsc, NULL, FALSE); if (then_node != NULL) { pcmk__rsc_trace(then->rsc, "Found %s for 'then' %s", pcmk__node_name(then_node), then->uuid); } } // Disable constraint if it only applies when on same node, but isn't if (pcmk_is_set(other->type, pcmk__ar_if_on_same_node) && (first_node != NULL) && (then_node != NULL) && !pcmk__same_node(first_node, then_node)) { pcmk__rsc_trace(then->rsc, "Disabled ordering %s on %s then %s on %s: " "not same node", other->action->uuid, pcmk__node_name(first_node), then->uuid, pcmk__node_name(then_node)); other->type = (enum pe_ordering) pcmk__ar_none; continue; } pcmk__clear_updated_flags(changed, then, pcmk__updated_first); if ((first->rsc != NULL) && pcmk_is_set(other->type, pcmk__ar_then_cancels_first) && !pcmk_is_set(then->flags, pcmk_action_optional)) { /* 'then' is required, so we must abandon 'first' * (e.g. a required stop cancels any agent reload). */ pcmk__set_action_flags(other->action, pcmk_action_optional); if (!strcmp(first->task, PCMK_ACTION_RELOAD_AGENT)) { pcmk__clear_rsc_flags(first->rsc, pcmk_rsc_reload); } } if ((first->rsc != NULL) && (then->rsc != NULL) && (first->rsc != then->rsc) && !is_parent(then->rsc, first->rsc)) { first = action_for_ordering(first); } if (first != other->action) { pcmk__rsc_trace(then->rsc, "Ordering %s after %s instead of %s", then->uuid, first->uuid, other->action->uuid); } pcmk__rsc_trace(then->rsc, "%s (%#.6x) then %s (%#.6x): type=%#.6x node=%s", first->uuid, first->flags, then->uuid, then->flags, other->type, action_node_str(first)); if (first == other->action) { /* 'first' was not remapped (e.g. from 'start' to 'running'), which * could mean it is a non-resource action, a primitive resource * action, or already expanded. */ uint32_t first_flags, then_flags; first_flags = action_flags_for_ordering(first, then_node); then_flags = action_flags_for_ordering(then, first_node); changed |= update_action_for_ordering_flags(first, then, first_flags, then_flags, other, scheduler); /* 'first' was for a complex resource (clone, group, etc), * create a new dependency if necessary */ } else if (order_actions(first, then, other->type)) { /* This was the first time 'first' and 'then' were associated, * start again to get the new actions_before list */ pcmk__set_updated_flags(changed, then, pcmk__updated_then); pcmk__rsc_trace(then->rsc, "Disabled ordering %s then %s in favor of %s " "then %s", other->action->uuid, then->uuid, first->uuid, then->uuid); other->type = (enum pe_ordering) pcmk__ar_none; } if (pcmk_is_set(changed, pcmk__updated_first)) { crm_trace("Re-processing %s and its 'after' actions " "because it changed", first->uuid); for (GList *lpc2 = first->actions_after; lpc2 != NULL; lpc2 = lpc2->next) { pcmk__related_action_t *other = lpc2->data; pcmk__update_action_for_orderings(other->action, scheduler); } pcmk__update_action_for_orderings(first, scheduler); } } if (pcmk_is_set(then->flags, pcmk_action_min_runnable)) { if (last_flags == then->flags) { pcmk__clear_updated_flags(changed, then, pcmk__updated_then); } else { pcmk__set_updated_flags(changed, then, pcmk__updated_then); } } if (pcmk_is_set(changed, pcmk__updated_then)) { crm_trace("Re-processing %s and its 'after' actions because it changed", then->uuid); if (pcmk_is_set(last_flags, pcmk_action_runnable) && !pcmk_is_set(then->flags, pcmk_action_runnable)) { pcmk__block_colocation_dependents(then); } pcmk__update_action_for_orderings(then, scheduler); for (lpc = then->actions_after; lpc != NULL; lpc = lpc->next) { pcmk__related_action_t *other = lpc->data; pcmk__update_action_for_orderings(other->action, scheduler); } } } static inline bool is_primitive_action(const pcmk_action_t *action) { return (action != NULL) && pcmk__is_primitive(action->rsc); } /*! * \internal * \brief Clear a single action flag and set reason text * * \param[in,out] action Action whose flag should be cleared * \param[in] flag Action flag that should be cleared * \param[in] reason Action that is the reason why flag is being cleared */ #define clear_action_flag_because(action, flag, reason) do { \ if (pcmk_is_set((action)->flags, (flag))) { \ pcmk__clear_action_flags(action, flag); \ if ((action)->rsc != (reason)->rsc) { \ char *reason_text = pe__action2reason((reason), (flag)); \ pe_action_set_reason((action), reason_text, false); \ free(reason_text); \ } \ } \ } while (0) /*! * \internal * \brief Update actions in an asymmetric ordering * * If the "first" action in an asymmetric ordering is unrunnable, make the * "second" action unrunnable as well, if appropriate. * * \param[in] first 'First' action in an asymmetric ordering * \param[in,out] then 'Then' action in an asymmetric ordering */ static void handle_asymmetric_ordering(const pcmk_action_t *first, pcmk_action_t *then) { /* Only resource actions after an unrunnable 'first' action need updates for * asymmetric ordering. */ if ((then->rsc == NULL) || pcmk_is_set(first->flags, pcmk_action_runnable)) { return; } // Certain optional 'then' actions are unaffected by unrunnable 'first' if (pcmk_is_set(then->flags, pcmk_action_optional)) { enum rsc_role_e then_rsc_role; then_rsc_role = then->rsc->private->fns->state(then->rsc, TRUE); if ((then_rsc_role == pcmk_role_stopped) && pcmk__str_eq(then->task, PCMK_ACTION_STOP, pcmk__str_none)) { /* If 'then' should stop after 'first' but is already stopped, the * ordering is irrelevant. */ return; } else if ((then_rsc_role >= pcmk_role_started) && pcmk__str_eq(then->task, PCMK_ACTION_START, pcmk__str_none) && pe__rsc_running_on_only(then->rsc, then->node)) { /* Similarly if 'then' should start after 'first' but is already * started on a single node. */ return; } } // 'First' can't run, so 'then' can't either clear_action_flag_because(then, pcmk_action_optional, first); clear_action_flag_because(then, pcmk_action_runnable, first); } /*! * \internal * \brief Set action bits appropriately when pcmk__ar_intermediate_stop is used * * \param[in,out] first 'First' action in ordering * \param[in,out] then 'Then' action in ordering * \param[in] filter What action flags to care about * * \note pcmk__ar_intermediate_stop is set for "stop resource before starting * it" and "stop later group member before stopping earlier group member" */ static void handle_restart_ordering(pcmk_action_t *first, pcmk_action_t *then, uint32_t filter) { const char *reason = NULL; CRM_ASSERT(is_primitive_action(first)); CRM_ASSERT(is_primitive_action(then)); // We need to update the action in two cases: // ... if 'then' is required if (pcmk_is_set(filter, pcmk_action_optional) && !pcmk_is_set(then->flags, pcmk_action_optional)) { reason = "restart"; } /* ... if 'then' is unrunnable action on same resource (if a resource * should restart but can't start, we still want to stop) */ if (pcmk_is_set(filter, pcmk_action_runnable) && !pcmk_is_set(then->flags, pcmk_action_runnable) && pcmk_is_set(then->rsc->flags, pcmk_rsc_managed) && (first->rsc == then->rsc)) { reason = "stop"; } if (reason == NULL) { return; } pcmk__rsc_trace(first->rsc, "Handling %s -> %s for %s", first->uuid, then->uuid, reason); // Make 'first' required if it is runnable if (pcmk_is_set(first->flags, pcmk_action_runnable)) { clear_action_flag_because(first, pcmk_action_optional, then); } // Make 'first' required if 'then' is required if (!pcmk_is_set(then->flags, pcmk_action_optional)) { clear_action_flag_because(first, pcmk_action_optional, then); } // Make 'first' unmigratable if 'then' is unmigratable if (!pcmk_is_set(then->flags, pcmk_action_migratable)) { clear_action_flag_because(first, pcmk_action_migratable, then); } // Make 'then' unrunnable if 'first' is required but unrunnable if (!pcmk_is_set(first->flags, pcmk_action_optional) && !pcmk_is_set(first->flags, pcmk_action_runnable)) { clear_action_flag_because(then, pcmk_action_runnable, first); } } /*! * \internal * \brief Update two actions according to an ordering between them * * Given information about an ordering of two actions, update the actions' flags * (and runnable_before members if appropriate) as appropriate for the ordering. * Effects may cascade to other orderings involving the actions as well. * * \param[in,out] first 'First' action in an ordering * \param[in,out] then 'Then' action in an ordering * \param[in] node If not NULL, limit scope of ordering to this node * (ignored) * \param[in] flags Action flags for \p first for ordering purposes * \param[in] filter Action flags to limit scope of certain updates (may * include pcmk_action_optional to affect only * mandatory actions, and pcmk_action_runnable to * affect only runnable actions) * \param[in] type Group of enum pcmk__action_relation_flags to apply * \param[in,out] scheduler Scheduler data * * \return Group of enum pcmk__updated flags indicating what was updated */ uint32_t pcmk__update_ordered_actions(pcmk_action_t *first, pcmk_action_t *then, const pcmk_node_t *node, uint32_t flags, uint32_t filter, uint32_t type, pcmk_scheduler_t *scheduler) { uint32_t changed = pcmk__updated_none; uint32_t then_flags = 0U; uint32_t first_flags = 0U; CRM_ASSERT((first != NULL) && (then != NULL) && (scheduler != NULL)); then_flags = then->flags; first_flags = first->flags; if (pcmk_is_set(type, pcmk__ar_asymmetric)) { handle_asymmetric_ordering(first, then); } if (pcmk_is_set(type, pcmk__ar_then_implies_first) && !pcmk_is_set(then_flags, pcmk_action_optional)) { // Then is required, and implies first should be, too if (pcmk_is_set(filter, pcmk_action_optional) && !pcmk_is_set(flags, pcmk_action_optional) && pcmk_is_set(first_flags, pcmk_action_optional)) { clear_action_flag_because(first, pcmk_action_optional, then); } if (pcmk_is_set(flags, pcmk_action_migratable) && !pcmk_is_set(then->flags, pcmk_action_migratable)) { clear_action_flag_because(first, pcmk_action_migratable, then); } } if (pcmk_is_set(type, pcmk__ar_promoted_then_implies_first) && (then->rsc != NULL) && (then->rsc->role == pcmk_role_promoted) && pcmk_is_set(filter, pcmk_action_optional) && !pcmk_is_set(then->flags, pcmk_action_optional)) { clear_action_flag_because(first, pcmk_action_optional, then); if (pcmk_is_set(first->flags, pcmk_action_migratable) && !pcmk_is_set(then->flags, pcmk_action_migratable)) { clear_action_flag_because(first, pcmk_action_migratable, then); } } if (pcmk_is_set(type, pcmk__ar_unmigratable_then_blocks) && pcmk_is_set(filter, pcmk_action_optional)) { if (!pcmk_all_flags_set(then->flags, pcmk_action_migratable |pcmk_action_runnable)) { clear_action_flag_because(first, pcmk_action_runnable, then); } if (!pcmk_is_set(then->flags, pcmk_action_optional)) { clear_action_flag_because(first, pcmk_action_optional, then); } } if (pcmk_is_set(type, pcmk__ar_first_else_then) && pcmk_is_set(filter, pcmk_action_optional) && !pcmk_is_set(first->flags, pcmk_action_runnable)) { clear_action_flag_because(then, pcmk_action_migratable, first); pcmk__clear_action_flags(then, pcmk_action_pseudo); } if (pcmk_is_set(type, pcmk__ar_unrunnable_first_blocks) && pcmk_is_set(filter, pcmk_action_runnable) && pcmk_is_set(then->flags, pcmk_action_runnable) && !pcmk_is_set(flags, pcmk_action_runnable)) { clear_action_flag_because(then, pcmk_action_runnable, first); clear_action_flag_because(then, pcmk_action_migratable, first); } if (pcmk_is_set(type, pcmk__ar_first_implies_then) && pcmk_is_set(filter, pcmk_action_optional) && pcmk_is_set(then->flags, pcmk_action_optional) && !pcmk_is_set(flags, pcmk_action_optional) && !pcmk_is_set(first->flags, pcmk_action_migratable)) { clear_action_flag_because(then, pcmk_action_optional, first); } if (pcmk_is_set(type, pcmk__ar_intermediate_stop)) { handle_restart_ordering(first, then, filter); } if (then_flags != then->flags) { pcmk__set_updated_flags(changed, first, pcmk__updated_then); pcmk__rsc_trace(then->rsc, "%s on %s: flags are now %#.6x (was %#.6x) " "because of 'first' %s (%#.6x)", then->uuid, pcmk__node_name(then->node), then->flags, then_flags, first->uuid, first->flags); if ((then->rsc != NULL) && (then->rsc->private->parent != NULL)) { // Required to handle "X_stop then X_start" for cloned groups pcmk__update_action_for_orderings(then, scheduler); } } if (first_flags != first->flags) { pcmk__set_updated_flags(changed, first, pcmk__updated_first); pcmk__rsc_trace(first->rsc, "%s on %s: flags are now %#.6x (was %#.6x) " "because of 'then' %s (%#.6x)", first->uuid, pcmk__node_name(first->node), first->flags, first_flags, then->uuid, then->flags); } return changed; } /*! * \internal * \brief Trace-log an action (optionally with its dependent actions) * * \param[in] pre_text If not NULL, prefix the log with this plus ": " * \param[in] action Action to log * \param[in] details If true, recursively log dependent actions */ void pcmk__log_action(const char *pre_text, const pcmk_action_t *action, bool details) { const char *node_uname = NULL; const char *node_uuid = NULL; const char *desc = NULL; CRM_CHECK(action != NULL, return); if (!pcmk_is_set(action->flags, pcmk_action_pseudo)) { if (action->node != NULL) { node_uname = action->node->details->uname; node_uuid = action->node->details->id; } else { node_uname = ""; } } switch (pcmk_parse_action(action->task)) { case pcmk_action_fence: case pcmk_action_shutdown: if (pcmk_is_set(action->flags, pcmk_action_pseudo)) { desc = "Pseudo "; } else if (pcmk_is_set(action->flags, pcmk_action_optional)) { desc = "Optional "; } else if (!pcmk_is_set(action->flags, pcmk_action_runnable)) { desc = "!!Non-Startable!! "; } else { desc = "(Provisional) "; } crm_trace("%s%s%sAction %d: %s%s%s%s%s%s", ((pre_text == NULL)? "" : pre_text), ((pre_text == NULL)? "" : ": "), desc, action->id, action->uuid, (node_uname? "\ton " : ""), (node_uname? node_uname : ""), (node_uuid? "\t\t(" : ""), (node_uuid? node_uuid : ""), (node_uuid? ")" : "")); break; default: if (pcmk_is_set(action->flags, pcmk_action_optional)) { desc = "Optional "; } else if (pcmk_is_set(action->flags, pcmk_action_pseudo)) { desc = "Pseudo "; } else if (!pcmk_is_set(action->flags, pcmk_action_runnable)) { desc = "!!Non-Startable!! "; } else { desc = "(Provisional) "; } crm_trace("%s%s%sAction %d: %s %s%s%s%s%s%s", ((pre_text == NULL)? "" : pre_text), ((pre_text == NULL)? "" : ": "), desc, action->id, action->uuid, (action->rsc? action->rsc->id : ""), (node_uname? "\ton " : ""), (node_uname? node_uname : ""), (node_uuid? "\t\t(" : ""), (node_uuid? node_uuid : ""), (node_uuid? ")" : "")); break; } if (details) { const GList *iter = NULL; const pcmk__related_action_t *other = NULL; crm_trace("\t\t====== Preceding Actions"); for (iter = action->actions_before; iter != NULL; iter = iter->next) { other = (const pcmk__related_action_t *) iter->data; pcmk__log_action("\t\t", other->action, false); } crm_trace("\t\t====== Subsequent Actions"); for (iter = action->actions_after; iter != NULL; iter = iter->next) { other = (const pcmk__related_action_t *) iter->data; pcmk__log_action("\t\t", other->action, false); } crm_trace("\t\t====== End"); } else { crm_trace("\t\t(before=%d, after=%d)", g_list_length(action->actions_before), g_list_length(action->actions_after)); } } /*! * \internal * \brief Create a new shutdown action for a node * * \param[in,out] node Node being shut down * * \return Newly created shutdown action for \p node */ pcmk_action_t * pcmk__new_shutdown_action(pcmk_node_t *node) { char *shutdown_id = NULL; pcmk_action_t *shutdown_op = NULL; CRM_ASSERT(node != NULL); shutdown_id = crm_strdup_printf("%s-%s", PCMK_ACTION_DO_SHUTDOWN, node->details->uname); shutdown_op = custom_action(NULL, shutdown_id, PCMK_ACTION_DO_SHUTDOWN, node, FALSE, node->details->data_set); pcmk__order_stops_before_shutdown(node, shutdown_op); pcmk__insert_meta(shutdown_op, PCMK__META_OP_NO_WAIT, PCMK_VALUE_TRUE); return shutdown_op; } /*! * \internal * \brief Calculate and add an operation digest to XML * * Calculate an operation digest, which enables us to later determine when a * restart is needed due to the resource's parameters being changed, and add it * to given XML. * * \param[in] op Operation result from executor * \param[in,out] update XML to add digest to */ static void add_op_digest_to_xml(const lrmd_event_data_t *op, xmlNode *update) { char *digest = NULL; xmlNode *args_xml = NULL; if (op->params == NULL) { return; } args_xml = pcmk__xe_create(NULL, PCMK_XE_PARAMETERS); g_hash_table_foreach(op->params, hash2field, args_xml); pcmk__filter_op_for_digest(args_xml); - digest = calculate_operation_digest(args_xml, NULL); + digest = pcmk__digest_operation(args_xml); crm_xml_add(update, PCMK__XA_OP_DIGEST, digest); pcmk__xml_free(args_xml); free(digest); } #define FAKE_TE_ID "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx" /*! * \internal * \brief Create XML for resource operation history update * * \param[in,out] parent Parent XML node to add to * \param[in,out] op Operation event data * \param[in] caller_version DC feature set * \param[in] target_rc Expected result of operation * \param[in] node Name of node on which operation was performed * \param[in] origin Arbitrary description of update source * * \return Newly created XML node for history update */ xmlNode * pcmk__create_history_xml(xmlNode *parent, lrmd_event_data_t *op, const char *caller_version, int target_rc, const char *node, const char *origin) { char *key = NULL; char *magic = NULL; char *op_id = NULL; char *op_id_additional = NULL; char *local_user_data = NULL; const char *exit_reason = NULL; xmlNode *xml_op = NULL; const char *task = NULL; CRM_CHECK(op != NULL, return NULL); crm_trace("Creating history XML for %s-interval %s action for %s on %s " "(DC version: %s, origin: %s)", pcmk__readable_interval(op->interval_ms), op->op_type, op->rsc_id, ((node == NULL)? "no node" : node), caller_version, origin); task = op->op_type; /* Record a successful agent reload as a start, and a failed one as a * monitor, to make life easier for the scheduler when determining the * current state. * * @COMPAT We should check "reload" here only if the operation was for a * pre-OCF-1.1 resource agent, but we don't know that here, and we should * only ever get results for actions scheduled by us, so we can reasonably * assume any "reload" is actually a pre-1.1 agent reload. */ if (pcmk__str_any_of(task, PCMK_ACTION_RELOAD, PCMK_ACTION_RELOAD_AGENT, NULL)) { if (op->op_status == PCMK_EXEC_DONE) { task = PCMK_ACTION_START; } else { task = PCMK_ACTION_MONITOR; } } key = pcmk__op_key(op->rsc_id, task, op->interval_ms); if (pcmk__str_eq(task, PCMK_ACTION_NOTIFY, pcmk__str_none)) { const char *n_type = crm_meta_value(op->params, "notify_type"); const char *n_task = crm_meta_value(op->params, "notify_operation"); CRM_LOG_ASSERT(n_type != NULL); CRM_LOG_ASSERT(n_task != NULL); op_id = pcmk__notify_key(op->rsc_id, n_type, n_task); if (op->op_status != PCMK_EXEC_PENDING) { /* Ignore notify errors. * * @TODO It might be better to keep the correct result here, and * ignore it in process_graph_event(). */ lrmd__set_result(op, PCMK_OCF_OK, PCMK_EXEC_DONE, NULL); } /* Migration history is preserved separately, which usually matters for * multiple nodes and is important for future cluster transitions. */ } else if (pcmk__str_any_of(op->op_type, PCMK_ACTION_MIGRATE_TO, PCMK_ACTION_MIGRATE_FROM, NULL)) { op_id = strdup(key); } else if (did_rsc_op_fail(op, target_rc)) { op_id = pcmk__op_key(op->rsc_id, "last_failure", 0); if (op->interval_ms == 0) { /* Ensure 'last' gets updated, in case PCMK_META_RECORD_PENDING is * true */ op_id_additional = pcmk__op_key(op->rsc_id, "last", 0); } exit_reason = op->exit_reason; } else if (op->interval_ms > 0) { op_id = strdup(key); } else { op_id = pcmk__op_key(op->rsc_id, "last", 0); } again: xml_op = pcmk__xe_first_child(parent, PCMK__XE_LRM_RSC_OP, PCMK_XA_ID, op_id); if (xml_op == NULL) { xml_op = pcmk__xe_create(parent, PCMK__XE_LRM_RSC_OP); } if (op->user_data == NULL) { crm_debug("Generating fake transition key for: " PCMK__OP_FMT " %d from %s", op->rsc_id, op->op_type, op->interval_ms, op->call_id, origin); local_user_data = pcmk__transition_key(-1, op->call_id, target_rc, FAKE_TE_ID); op->user_data = local_user_data; } if (magic == NULL) { magic = crm_strdup_printf("%d:%d;%s", op->op_status, op->rc, (const char *) op->user_data); } crm_xml_add(xml_op, PCMK_XA_ID, op_id); crm_xml_add(xml_op, PCMK__XA_OPERATION_KEY, key); crm_xml_add(xml_op, PCMK_XA_OPERATION, task); crm_xml_add(xml_op, PCMK_XA_CRM_DEBUG_ORIGIN, origin); crm_xml_add(xml_op, PCMK_XA_CRM_FEATURE_SET, caller_version); crm_xml_add(xml_op, PCMK__XA_TRANSITION_KEY, op->user_data); crm_xml_add(xml_op, PCMK__XA_TRANSITION_MAGIC, magic); crm_xml_add(xml_op, PCMK_XA_EXIT_REASON, pcmk__s(exit_reason, "")); crm_xml_add(xml_op, PCMK__META_ON_NODE, node); // For context during triage crm_xml_add_int(xml_op, PCMK__XA_CALL_ID, op->call_id); crm_xml_add_int(xml_op, PCMK__XA_RC_CODE, op->rc); crm_xml_add_int(xml_op, PCMK__XA_OP_STATUS, op->op_status); crm_xml_add_ms(xml_op, PCMK_META_INTERVAL, op->interval_ms); if (compare_version("2.1", caller_version) <= 0) { if (op->t_run || op->t_rcchange || op->exec_time || op->queue_time) { crm_trace("Timing data (" PCMK__OP_FMT "): last=%u change=%u exec=%u queue=%u", op->rsc_id, op->op_type, op->interval_ms, op->t_run, op->t_rcchange, op->exec_time, op->queue_time); if ((op->interval_ms != 0) && (op->t_rcchange != 0)) { // Recurring ops may have changed rc after initial run crm_xml_add_ll(xml_op, PCMK_XA_LAST_RC_CHANGE, (long long) op->t_rcchange); } else { crm_xml_add_ll(xml_op, PCMK_XA_LAST_RC_CHANGE, (long long) op->t_run); } crm_xml_add_int(xml_op, PCMK_XA_EXEC_TIME, op->exec_time); crm_xml_add_int(xml_op, PCMK_XA_QUEUE_TIME, op->queue_time); } } if (pcmk__str_any_of(op->op_type, PCMK_ACTION_MIGRATE_TO, PCMK_ACTION_MIGRATE_FROM, NULL)) { /* Record PCMK__META_MIGRATE_SOURCE and PCMK__META_MIGRATE_TARGET always * for migrate ops. */ const char *name = PCMK__META_MIGRATE_SOURCE; crm_xml_add(xml_op, name, crm_meta_value(op->params, name)); name = PCMK__META_MIGRATE_TARGET; crm_xml_add(xml_op, name, crm_meta_value(op->params, name)); } add_op_digest_to_xml(op, xml_op); if (op_id_additional) { free(op_id); op_id = op_id_additional; op_id_additional = NULL; goto again; } if (local_user_data) { free(local_user_data); op->user_data = NULL; } free(magic); free(op_id); free(key); return xml_op; } /*! * \internal * \brief Check whether an action shutdown-locks a resource to a node * * If the PCMK_OPT_SHUTDOWN_LOCK cluster property is set, resources will not be * recovered on a different node if cleanly stopped, and may start only on that * same node. This function checks whether that applies to a given action, so * that the transition graph can be marked appropriately. * * \param[in] action Action to check * * \return true if \p action locks its resource to the action's node, * otherwise false */ bool pcmk__action_locks_rsc_to_node(const pcmk_action_t *action) { // Only resource actions taking place on resource's lock node are locked if ((action == NULL) || (action->rsc == NULL) || !pcmk__same_node(action->node, action->rsc->lock_node)) { return false; } /* During shutdown, only stops are locked (otherwise, another action such as * a demote would cause the controller to clear the lock) */ if (action->node->details->shutdown && (action->task != NULL) && (strcmp(action->task, PCMK_ACTION_STOP) != 0)) { return false; } return true; } /* lowest to highest */ static gint sort_action_id(gconstpointer a, gconstpointer b) { const pcmk__related_action_t *action_wrapper2 = a; const pcmk__related_action_t *action_wrapper1 = b; if (a == NULL) { return 1; } if (b == NULL) { return -1; } if (action_wrapper1->action->id < action_wrapper2->action->id) { return 1; } if (action_wrapper1->action->id > action_wrapper2->action->id) { return -1; } return 0; } /*! * \internal * \brief Remove any duplicate action inputs, merging action flags * * \param[in,out] action Action whose inputs should be checked */ void pcmk__deduplicate_action_inputs(pcmk_action_t *action) { GList *item = NULL; GList *next = NULL; pcmk__related_action_t *last_input = NULL; action->actions_before = g_list_sort(action->actions_before, sort_action_id); for (item = action->actions_before; item != NULL; item = next) { pcmk__related_action_t *input = item->data; next = item->next; if ((last_input != NULL) && (input->action->id == last_input->action->id)) { crm_trace("Input %s (%d) duplicate skipped for action %s (%d)", input->action->uuid, input->action->id, action->uuid, action->id); /* For the purposes of scheduling, the ordering flags no longer * matter, but crm_simulate looks at certain ones when creating a * dot graph. Combining the flags is sufficient for that purpose. */ last_input->type |= input->type; if (input->state == pe_link_dumped) { last_input->state = pe_link_dumped; } free(item->data); action->actions_before = g_list_delete_link(action->actions_before, item); } else { last_input = input; input->state = pe_link_not_dumped; } } } /*! * \internal * \brief Output all scheduled actions * * \param[in,out] scheduler Scheduler data */ void pcmk__output_actions(pcmk_scheduler_t *scheduler) { pcmk__output_t *out = scheduler->priv; // Output node (non-resource) actions for (GList *iter = scheduler->actions; iter != NULL; iter = iter->next) { char *node_name = NULL; char *task = NULL; pcmk_action_t *action = (pcmk_action_t *) iter->data; if (action->rsc != NULL) { continue; // Resource actions will be output later } else if (pcmk_is_set(action->flags, pcmk_action_optional)) { continue; // This action was not scheduled } if (pcmk__str_eq(action->task, PCMK_ACTION_DO_SHUTDOWN, pcmk__str_none)) { task = strdup("Shutdown"); } 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); task = crm_strdup_printf("Fence (%s)", op); } else { continue; // Don't display other node action types } if (pcmk__is_guest_or_bundle_node(action->node)) { const pcmk_resource_t *remote = action->node->details->remote_rsc; node_name = crm_strdup_printf("%s (resource: %s)", pcmk__node_name(action->node), remote->container->id); } else if (action->node != NULL) { node_name = crm_strdup_printf("%s", pcmk__node_name(action->node)); } out->message(out, "node-action", task, node_name, action->reason); free(node_name); free(task); } // Output resource actions for (GList *iter = scheduler->resources; iter != NULL; iter = iter->next) { pcmk_resource_t *rsc = (pcmk_resource_t *) iter->data; rsc->private->cmds->output_actions(rsc); } } /*! * \internal * \brief Get action name needed to compare digest for configuration changes * * \param[in] task Action name from history * \param[in] interval_ms Action interval (in milliseconds) * * \return Action name whose digest should be compared */ static const char * task_for_digest(const char *task, guint interval_ms) { /* Certain actions need to be compared against the parameters used to start * the resource. */ if ((interval_ms == 0) && pcmk__str_any_of(task, PCMK_ACTION_MONITOR, PCMK_ACTION_MIGRATE_FROM, PCMK_ACTION_PROMOTE, NULL)) { task = PCMK_ACTION_START; } return task; } /*! * \internal * \brief Check whether only sanitized parameters to an action changed * * When collecting CIB files for troubleshooting, crm_report will mask * sensitive resource parameters. If simulations were run using that, affected * resources would appear to need a restart, which would complicate * troubleshooting. To avoid that, we save a "secure digest" of non-sensitive * parameters. This function used that digest to check whether only masked * parameters are different. * * \param[in] xml_op Resource history entry with secure digest * \param[in] digest_data Operation digest information being compared * \param[in] scheduler Scheduler data * * \return true if only sanitized parameters changed, otherwise false */ static bool only_sanitized_changed(const xmlNode *xml_op, const pcmk__op_digest_t *digest_data, const pcmk_scheduler_t *scheduler) { const char *digest_secure = NULL; if (!pcmk_is_set(scheduler->flags, pcmk_sched_sanitized)) { // The scheduler is not being run as a simulation return false; } digest_secure = crm_element_value(xml_op, PCMK__XA_OP_SECURE_DIGEST); return (digest_data->rc != pcmk__digest_match) && (digest_secure != NULL) && (digest_data->digest_secure_calc != NULL) && (strcmp(digest_data->digest_secure_calc, digest_secure) == 0); } /*! * \internal * \brief Force a restart due to a configuration change * * \param[in,out] rsc Resource that action is for * \param[in] task Name of action whose configuration changed * \param[in] interval_ms Action interval (in milliseconds) * \param[in,out] node Node where resource should be restarted */ static void force_restart(pcmk_resource_t *rsc, const char *task, guint interval_ms, pcmk_node_t *node) { char *key = pcmk__op_key(rsc->id, task, interval_ms); pcmk_action_t *required = custom_action(rsc, key, task, NULL, FALSE, rsc->private->scheduler); pe_action_set_reason(required, "resource definition change", true); trigger_unfencing(rsc, node, "Device parameters changed", NULL, rsc->private->scheduler); } /*! * \internal * \brief Schedule a reload of a resource on a node * * \param[in,out] data Resource to reload * \param[in] user_data Where resource should be reloaded */ static void schedule_reload(gpointer data, gpointer user_data) { pcmk_resource_t *rsc = data; const pcmk_node_t *node = user_data; pcmk_action_t *reload = NULL; // For collective resources, just call recursively for children if (rsc->private->variant > pcmk__rsc_variant_primitive) { g_list_foreach(rsc->children, schedule_reload, user_data); return; } // Skip the reload in certain situations if ((node == NULL) || !pcmk_is_set(rsc->flags, pcmk_rsc_managed) || pcmk_is_set(rsc->flags, pcmk_rsc_failed)) { pcmk__rsc_trace(rsc, "Skip reload of %s:%s%s %s", rsc->id, pcmk_is_set(rsc->flags, pcmk_rsc_managed)? "" : " unmanaged", pcmk_is_set(rsc->flags, pcmk_rsc_failed)? " failed" : "", (node == NULL)? "inactive" : node->details->uname); return; } /* If a resource's configuration changed while a start was pending, * force a full restart instead of a reload. */ if (pcmk_is_set(rsc->flags, pcmk_rsc_start_pending)) { pcmk__rsc_trace(rsc, "%s: preventing agent reload because start pending", rsc->id); custom_action(rsc, stop_key(rsc), PCMK_ACTION_STOP, node, FALSE, rsc->private->scheduler); return; } // Schedule the reload pcmk__set_rsc_flags(rsc, pcmk_rsc_reload); reload = custom_action(rsc, reload_key(rsc), PCMK_ACTION_RELOAD_AGENT, node, FALSE, rsc->private->scheduler); pe_action_set_reason(reload, "resource definition change", FALSE); // Set orderings so that a required stop or demote cancels the reload pcmk__new_ordering(NULL, NULL, reload, rsc, stop_key(rsc), NULL, pcmk__ar_ordered|pcmk__ar_then_cancels_first, rsc->private->scheduler); pcmk__new_ordering(NULL, NULL, reload, rsc, demote_key(rsc), NULL, pcmk__ar_ordered|pcmk__ar_then_cancels_first, rsc->private->scheduler); } /*! * \internal * \brief Handle any configuration change for an action * * Given an action from resource history, if the resource's configuration * changed since the action was done, schedule any actions needed (restart, * reload, unfencing, rescheduling recurring actions, etc.). * * \param[in,out] rsc Resource that action is for * \param[in,out] node Node that action was on * \param[in] xml_op Action XML from resource history * * \return true if action configuration changed, otherwise false */ bool pcmk__check_action_config(pcmk_resource_t *rsc, pcmk_node_t *node, const xmlNode *xml_op) { guint interval_ms = 0; const char *task = NULL; const pcmk__op_digest_t *digest_data = NULL; CRM_CHECK((rsc != NULL) && (node != NULL) && (xml_op != NULL), return false); task = crm_element_value(xml_op, PCMK_XA_OPERATION); CRM_CHECK(task != NULL, return false); crm_element_value_ms(xml_op, PCMK_META_INTERVAL, &interval_ms); // If this is a recurring action, check whether it has been orphaned if (interval_ms > 0) { if (pcmk__find_action_config(rsc, task, interval_ms, false) != NULL) { pcmk__rsc_trace(rsc, "%s-interval %s for %s on %s is in configuration", pcmk__readable_interval(interval_ms), task, rsc->id, pcmk__node_name(node)); } else if (pcmk_is_set(rsc->private->scheduler->flags, pcmk_sched_cancel_removed_actions)) { pcmk__schedule_cancel(rsc, crm_element_value(xml_op, PCMK__XA_CALL_ID), task, interval_ms, node, "orphan"); return true; } else { pcmk__rsc_debug(rsc, "%s-interval %s for %s on %s is orphaned", pcmk__readable_interval(interval_ms), task, rsc->id, pcmk__node_name(node)); return true; } } crm_trace("Checking %s-interval %s for %s on %s for configuration changes", pcmk__readable_interval(interval_ms), task, rsc->id, pcmk__node_name(node)); task = task_for_digest(task, interval_ms); digest_data = rsc_action_digest_cmp(rsc, xml_op, node, rsc->private->scheduler); if (only_sanitized_changed(xml_op, digest_data, rsc->private->scheduler)) { if (!pcmk__is_daemon && (rsc->private->scheduler->priv != NULL)) { pcmk__output_t *out = rsc->private->scheduler->priv; out->info(out, "Only 'private' parameters to %s-interval %s for %s " "on %s changed: %s", pcmk__readable_interval(interval_ms), task, rsc->id, pcmk__node_name(node), crm_element_value(xml_op, PCMK__XA_TRANSITION_MAGIC)); } return false; } switch (digest_data->rc) { case pcmk__digest_restart: crm_log_xml_debug(digest_data->params_restart, "params:restart"); force_restart(rsc, task, interval_ms, node); return true; case pcmk__digest_unknown: case pcmk__digest_mismatch: // Changes that can potentially be handled by an agent reload if (interval_ms > 0) { /* Recurring actions aren't reloaded per se, they are just * re-scheduled so the next run uses the new parameters. * The old instance will be cancelled automatically. */ crm_log_xml_debug(digest_data->params_all, "params:reschedule"); pcmk__reschedule_recurring(rsc, task, interval_ms, node); } else if (crm_element_value(xml_op, PCMK__XA_OP_RESTART_DIGEST) != NULL) { // Agent supports reload, so use it trigger_unfencing(rsc, node, "Device parameters changed (reload)", NULL, rsc->private->scheduler); crm_log_xml_debug(digest_data->params_all, "params:reload"); schedule_reload((gpointer) rsc, (gpointer) node); } else { pcmk__rsc_trace(rsc, "Restarting %s " "because agent doesn't support reload", rsc->id); crm_log_xml_debug(digest_data->params_restart, "params:restart"); force_restart(rsc, task, interval_ms, node); } return true; default: break; } return false; } /*! * \internal * \brief Create a list of resource's action history entries, sorted by call ID * * \param[in] rsc_entry Resource's \c PCMK__XE_LRM_RSC_OP status XML * \param[out] start_index Where to store index of start-like action, if any * \param[out] stop_index Where to store index of stop action, if any */ static GList * rsc_history_as_list(const xmlNode *rsc_entry, int *start_index, int *stop_index) { GList *ops = NULL; for (xmlNode *rsc_op = pcmk__xe_first_child(rsc_entry, PCMK__XE_LRM_RSC_OP, NULL, NULL); rsc_op != NULL; rsc_op = pcmk__xe_next_same(rsc_op)) { ops = g_list_prepend(ops, rsc_op); } ops = g_list_sort(ops, sort_op_by_callid); calculate_active_ops(ops, start_index, stop_index); return ops; } /*! * \internal * \brief Process a resource's action history from the CIB status * * Given a resource's action history, if the resource's configuration * changed since the actions were done, schedule any actions needed (restart, * reload, unfencing, rescheduling recurring actions, clean-up, etc.). * (This also cancels recurring actions for maintenance mode, which is not * entirely related but convenient to do here.) * * \param[in] rsc_entry Resource's \c PCMK__XE_LRM_RSC_OP status XML * \param[in,out] rsc Resource whose history is being processed * \param[in,out] node Node whose history is being processed */ static void process_rsc_history(const xmlNode *rsc_entry, pcmk_resource_t *rsc, pcmk_node_t *node) { int offset = -1; int stop_index = 0; int start_index = 0; GList *sorted_op_list = NULL; if (pcmk_is_set(rsc->flags, pcmk_rsc_removed)) { if (pcmk__is_anonymous_clone(pe__const_top_resource(rsc, false))) { pcmk__rsc_trace(rsc, "Skipping configuration check " "for orphaned clone instance %s", rsc->id); } else { pcmk__rsc_trace(rsc, "Skipping configuration check and scheduling " "clean-up for orphaned resource %s", rsc->id); pcmk__schedule_cleanup(rsc, node, false); } return; } if (pe_find_node_id(rsc->running_on, node->details->id) == NULL) { if (pcmk__rsc_agent_changed(rsc, node, rsc_entry, false)) { pcmk__schedule_cleanup(rsc, node, false); } pcmk__rsc_trace(rsc, "Skipping configuration check for %s " "because no longer active on %s", rsc->id, pcmk__node_name(node)); return; } pcmk__rsc_trace(rsc, "Checking for configuration changes for %s on %s", rsc->id, pcmk__node_name(node)); if (pcmk__rsc_agent_changed(rsc, node, rsc_entry, true)) { pcmk__schedule_cleanup(rsc, node, false); } sorted_op_list = rsc_history_as_list(rsc_entry, &start_index, &stop_index); if (start_index < stop_index) { return; // Resource is stopped } for (GList *iter = sorted_op_list; iter != NULL; iter = iter->next) { xmlNode *rsc_op = (xmlNode *) iter->data; const char *task = NULL; guint interval_ms = 0; if (++offset < start_index) { // Skip actions that happened before a start continue; } task = crm_element_value(rsc_op, PCMK_XA_OPERATION); crm_element_value_ms(rsc_op, PCMK_META_INTERVAL, &interval_ms); if ((interval_ms > 0) && (pcmk_is_set(rsc->flags, pcmk_rsc_maintenance) || node->details->maintenance)) { // Maintenance mode cancels recurring operations pcmk__schedule_cancel(rsc, crm_element_value(rsc_op, PCMK__XA_CALL_ID), task, interval_ms, node, "maintenance mode"); } else if ((interval_ms > 0) || pcmk__strcase_any_of(task, PCMK_ACTION_MONITOR, PCMK_ACTION_START, PCMK_ACTION_PROMOTE, PCMK_ACTION_MIGRATE_FROM, NULL)) { /* If a resource operation failed, and the operation's definition * has changed, clear any fail count so they can be retried fresh. */ if (pe__bundle_needs_remote_name(rsc)) { /* We haven't assigned resources to nodes yet, so if the * REMOTE_CONTAINER_HACK is used, we may calculate the digest * based on the literal "#uname" value rather than the properly * substituted value. That would mistakenly make the action * definition appear to have been changed. Defer the check until * later in this case. */ pe__add_param_check(rsc_op, rsc, node, pcmk__check_active, rsc->private->scheduler); } else if (pcmk__check_action_config(rsc, node, rsc_op) && (pe_get_failcount(node, rsc, NULL, pcmk__fc_effective, NULL) != 0)) { pe__clear_failcount(rsc, node, "action definition changed", rsc->private->scheduler); } } } g_list_free(sorted_op_list); } /*! * \internal * \brief Process a node's action history from the CIB status * * Given a node's resource history, if the resource's configuration changed * since the actions were done, schedule any actions needed (restart, * reload, unfencing, rescheduling recurring actions, clean-up, etc.). * (This also cancels recurring actions for maintenance mode, which is not * entirely related but convenient to do here.) * * \param[in,out] node Node whose history is being processed * \param[in] lrm_rscs Node's \c PCMK__XE_LRM_RESOURCES from CIB status XML */ static void process_node_history(pcmk_node_t *node, const xmlNode *lrm_rscs) { crm_trace("Processing node history for %s", pcmk__node_name(node)); for (const xmlNode *rsc_entry = pcmk__xe_first_child(lrm_rscs, PCMK__XE_LRM_RESOURCE, NULL, NULL); rsc_entry != NULL; rsc_entry = pcmk__xe_next_same(rsc_entry)) { if (rsc_entry->children != NULL) { GList *result = pcmk__rscs_matching_id(pcmk__xe_id(rsc_entry), node->details->data_set); for (GList *iter = result; iter != NULL; iter = iter->next) { pcmk_resource_t *rsc = (pcmk_resource_t *) iter->data; if (pcmk__is_primitive(rsc)) { process_rsc_history(rsc_entry, rsc, node); } } g_list_free(result); } } } // XPath to find a node's resource history #define XPATH_NODE_HISTORY "/" PCMK_XE_CIB "/" PCMK_XE_STATUS \ "/" PCMK__XE_NODE_STATE \ "[@" PCMK_XA_UNAME "='%s']" \ "/" PCMK__XE_LRM "/" PCMK__XE_LRM_RESOURCES /*! * \internal * \brief Process any resource configuration changes in the CIB status * * Go through all nodes' resource history, and if a resource's configuration * changed since its actions were done, schedule any actions needed (restart, * reload, unfencing, rescheduling recurring actions, clean-up, etc.). * (This also cancels recurring actions for maintenance mode, which is not * entirely related but convenient to do here.) * * \param[in,out] scheduler Scheduler data */ void pcmk__handle_rsc_config_changes(pcmk_scheduler_t *scheduler) { crm_trace("Check resource and action configuration for changes"); /* Rather than iterate through the status section, iterate through the nodes * and search for the appropriate status subsection for each. This skips * orphaned nodes and lets us eliminate some cases before searching the XML. */ for (GList *iter = scheduler->nodes; iter != NULL; iter = iter->next) { pcmk_node_t *node = (pcmk_node_t *) iter->data; /* Don't bother checking actions for a node that can't run actions ... * unless it's in maintenance mode, in which case we still need to * cancel any existing recurring monitors. */ if (node->details->maintenance || pcmk__node_available(node, false, false)) { char *xpath = NULL; xmlNode *history = NULL; xpath = crm_strdup_printf(XPATH_NODE_HISTORY, node->details->uname); history = get_xpath_object(xpath, scheduler->input, LOG_NEVER); free(xpath); process_node_history(node, history); } } } diff --git a/lib/pengine/pe_digest.c b/lib/pengine/pe_digest.c index 6b1415efab..a5487a09ae 100644 --- a/lib/pengine/pe_digest.c +++ b/lib/pengine/pe_digest.c @@ -1,614 +1,611 @@ /* * 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 "pe_status_private.h" extern bool pcmk__is_daemon; /*! * \internal * \brief Free an operation digest cache entry * * \param[in,out] ptr Pointer to cache entry to free * * \note The argument is a gpointer so this can be used as a hash table * free function. */ void pe__free_digests(gpointer ptr) { pcmk__op_digest_t *data = ptr; if (data != NULL) { pcmk__xml_free(data->params_all); pcmk__xml_free(data->params_secure); pcmk__xml_free(data->params_restart); free(data->digest_all_calc); free(data->digest_restart_calc); free(data->digest_secure_calc); free(data); } } // Return true if XML attribute name is not substring of a given string static bool attr_not_in_string(xmlAttrPtr a, void *user_data) { bool filter = false; char *name = crm_strdup_printf(" %s ", (const char *) a->name); if (strstr((const char *) user_data, name) == NULL) { crm_trace("Filtering %s (not found in '%s')", (const char *) a->name, (const char *) user_data); filter = true; } free(name); return filter; } // Return true if XML attribute name is substring of a given string static bool attr_in_string(xmlAttrPtr a, void *user_data) { bool filter = false; char *name = crm_strdup_printf(" %s ", (const char *) a->name); if (strstr((const char *) user_data, name) != NULL) { crm_trace("Filtering %s (found in '%s')", (const char *) a->name, (const char *) user_data); filter = true; } free(name); return filter; } /*! * \internal * \brief Add digest of all parameters to a digest cache entry * * \param[out] data Digest cache entry to modify * \param[in,out] rsc Resource that action was for * \param[in] node Node action was performed on * \param[in] params Resource parameters evaluated for node * \param[in] task Name of action performed * \param[in,out] interval_ms Action's interval (will be reset if in overrides) * \param[in] xml_op Unused * \param[in] op_version CRM feature set to use for digest calculation * \param[in] overrides Key/value table to override resource parameters * \param[in,out] scheduler Scheduler data */ static void calculate_main_digest(pcmk__op_digest_t *data, pcmk_resource_t *rsc, const pcmk_node_t *node, GHashTable *params, const char *task, guint *interval_ms, const xmlNode *xml_op, const char *op_version, GHashTable *overrides, pcmk_scheduler_t *scheduler) { xmlNode *action_config = NULL; data->params_all = pcmk__xe_create(NULL, PCMK_XE_PARAMETERS); /* REMOTE_CONTAINER_HACK: Allow Pacemaker Remote nodes to run containers * that themselves are Pacemaker Remote nodes */ (void) pe__add_bundle_remote_name(rsc, data->params_all, PCMK_REMOTE_RA_ADDR); if (overrides != NULL) { // If interval was overridden, reset it const char *meta_name = CRM_META "_" PCMK_META_INTERVAL; const char *interval_s = g_hash_table_lookup(overrides, meta_name); if (interval_s != NULL) { long long value_ll; if ((pcmk__scan_ll(interval_s, &value_ll, 0LL) == pcmk_rc_ok) && (value_ll >= 0) && (value_ll <= G_MAXUINT)) { *interval_ms = (guint) value_ll; } } // Add overrides to list of all parameters g_hash_table_foreach(overrides, hash2field, data->params_all); } // Add provided instance parameters g_hash_table_foreach(params, hash2field, data->params_all); // Find action configuration XML in CIB action_config = pcmk__find_action_config(rsc, task, *interval_ms, true); /* Add action-specific resource instance attributes to the digest list. * * If this is a one-time action with action-specific instance attributes, * enforce a restart instead of reload-agent in case the main digest doesn't * match, even if the restart digest does. This ensures any changes of the * action-specific parameters get applied for this specific action, and * digests calculated for the resulting history will be correct. Default the * result to RSC_DIGEST_RESTART for the case where the main digest doesn't * match. */ params = pcmk__unpack_action_rsc_params(action_config, node->details->attrs, scheduler); if ((*interval_ms == 0) && (g_hash_table_size(params) > 0)) { data->rc = pcmk__digest_restart; } g_hash_table_foreach(params, hash2field, data->params_all); g_hash_table_destroy(params); // Add action meta-attributes params = pcmk__unpack_action_meta(rsc, node, task, *interval_ms, action_config); g_hash_table_foreach(params, hash2metafield, data->params_all); g_hash_table_destroy(params); pcmk__filter_op_for_digest(data->params_all); - data->digest_all_calc = calculate_operation_digest(data->params_all, - op_version); + data->digest_all_calc = pcmk__digest_operation(data->params_all); } // Return true if XML attribute name is a Pacemaker-defined fencing parameter static bool is_fence_param(xmlAttrPtr attr, void *user_data) { return pcmk_stonith_param((const char *) attr->name); } /*! * \internal * \brief Add secure digest to a digest cache entry * * \param[out] data Digest cache entry to modify * \param[in] rsc Resource that action was for * \param[in] params Resource parameters evaluated for node * \param[in] xml_op XML of operation in CIB status (if available) * \param[in] op_version CRM feature set to use for digest calculation * \param[in] overrides Key/value hash table to override resource parameters */ static void calculate_secure_digest(pcmk__op_digest_t *data, const pcmk_resource_t *rsc, GHashTable *params, const xmlNode *xml_op, const char *op_version, GHashTable *overrides) { const char *class = crm_element_value(rsc->private->xml, PCMK_XA_CLASS); const char *secure_list = NULL; bool old_version = (compare_version(op_version, "3.16.0") < 0); if (xml_op == NULL) { secure_list = " passwd password user "; } else { secure_list = crm_element_value(xml_op, PCMK__XA_OP_SECURE_PARAMS); } if (old_version) { data->params_secure = pcmk__xe_create(NULL, PCMK_XE_PARAMETERS); if (overrides != NULL) { g_hash_table_foreach(overrides, hash2field, data->params_secure); } g_hash_table_foreach(params, hash2field, data->params_secure); } else { // Start with a copy of all parameters data->params_secure = pcmk__xml_copy(NULL, data->params_all); } if (secure_list != NULL) { pcmk__xe_remove_matching_attrs(data->params_secure, attr_in_string, (void *) secure_list); } if (old_version && pcmk_is_set(pcmk_get_ra_caps(class), pcmk_ra_cap_fence_params)) { /* For stonith resources, Pacemaker adds special parameters, * but these are not listed in fence agent meta-data, so with older * versions of DC, the controller will not hash them. That means we have * to filter them out before calculating our hash for comparison. */ pcmk__xe_remove_matching_attrs(data->params_secure, is_fence_param, NULL); } pcmk__filter_op_for_digest(data->params_secure); /* CRM_meta_timeout *should* be part of a digest for recurring operations. * However, with older versions of DC, the controller does not add timeout * to secure digests, because it only includes parameters declared by the * resource agent. * Remove any timeout that made it this far, to match. */ if (old_version) { pcmk__xe_remove_attr(data->params_secure, CRM_META "_" PCMK_META_TIMEOUT); } - data->digest_secure_calc = calculate_operation_digest(data->params_secure, - op_version); + data->digest_secure_calc = pcmk__digest_operation(data->params_secure); } /*! * \internal * \brief Add restart digest to a digest cache entry * * \param[out] data Digest cache entry to modify * \param[in] xml_op XML of operation in CIB status (if available) * \param[in] op_version CRM feature set to use for digest calculation * * \note This function doesn't need to handle overrides because it starts with * data->params_all, which already has overrides applied. */ static void calculate_restart_digest(pcmk__op_digest_t *data, const xmlNode *xml_op, const char *op_version) { const char *value = NULL; // We must have XML of resource operation history if (xml_op == NULL) { return; } // And the history must have a restart digest to compare against if (crm_element_value(xml_op, PCMK__XA_OP_RESTART_DIGEST) == NULL) { return; } // Start with a copy of all parameters data->params_restart = pcmk__xml_copy(NULL, data->params_all); // Then filter out reloadable parameters, if any value = crm_element_value(xml_op, PCMK__XA_OP_FORCE_RESTART); if (value != NULL) { pcmk__xe_remove_matching_attrs(data->params_restart, attr_not_in_string, (void *) value); } value = crm_element_value(xml_op, PCMK_XA_CRM_FEATURE_SET); - data->digest_restart_calc = calculate_operation_digest(data->params_restart, - value); + data->digest_restart_calc = pcmk__digest_operation(data->params_restart); } /*! * \internal * \brief Create a new digest cache entry with calculated digests * * \param[in,out] rsc Resource that action was for * \param[in] task Name of action performed * \param[in,out] interval_ms Action's interval (will be reset if in overrides) * \param[in] node Node action was performed on * \param[in] xml_op XML of operation in CIB status (if available) * \param[in] overrides Key/value table to override resource parameters * \param[in] calc_secure Whether to calculate secure digest * \param[in,out] scheduler Scheduler data * * \return Pointer to new digest cache entry (or NULL on memory error) * \note It is the caller's responsibility to free the result using * pe__free_digests(). */ pcmk__op_digest_t * pe__calculate_digests(pcmk_resource_t *rsc, const char *task, guint *interval_ms, const pcmk_node_t *node, const xmlNode *xml_op, GHashTable *overrides, bool calc_secure, pcmk_scheduler_t *scheduler) { pcmk__op_digest_t *data = NULL; const char *op_version = NULL; GHashTable *params = NULL; CRM_CHECK(scheduler != NULL, return NULL); data = calloc(1, sizeof(pcmk__op_digest_t)); if (data == NULL) { pcmk__sched_err("Could not allocate memory for operation digest"); return NULL; } data->rc = pcmk__digest_match; if (xml_op != NULL) { op_version = crm_element_value(xml_op, PCMK_XA_CRM_FEATURE_SET); } if ((op_version == NULL) && (scheduler->input != NULL)) { op_version = crm_element_value(scheduler->input, PCMK_XA_CRM_FEATURE_SET); } if (op_version == NULL) { op_version = CRM_FEATURE_SET; } params = pe_rsc_params(rsc, node, scheduler); calculate_main_digest(data, rsc, node, params, task, interval_ms, xml_op, op_version, overrides, scheduler); if (calc_secure) { calculate_secure_digest(data, rsc, params, xml_op, op_version, overrides); } calculate_restart_digest(data, xml_op, op_version); return data; } /*! * \internal * \brief Calculate action digests and store in node's digest cache * * \param[in,out] rsc Resource that action was for * \param[in] task Name of action performed * \param[in] interval_ms Action's interval * \param[in,out] node Node action was performed on * \param[in] xml_op XML of operation in CIB status (if available) * \param[in] calc_secure Whether to calculate secure digest * \param[in,out] scheduler Scheduler data * * \return Pointer to node's digest cache entry */ static pcmk__op_digest_t * rsc_action_digest(pcmk_resource_t *rsc, const char *task, guint interval_ms, pcmk_node_t *node, const xmlNode *xml_op, bool calc_secure, pcmk_scheduler_t *scheduler) { pcmk__op_digest_t *data = NULL; char *key = pcmk__op_key(rsc->id, task, interval_ms); data = g_hash_table_lookup(node->details->digest_cache, key); if (data == NULL) { data = pe__calculate_digests(rsc, task, &interval_ms, node, xml_op, NULL, calc_secure, scheduler); CRM_ASSERT(data != NULL); g_hash_table_insert(node->details->digest_cache, strdup(key), data); } free(key); return data; } /*! * \internal * \brief Calculate operation digests and compare against an XML history entry * * \param[in,out] rsc Resource to check * \param[in] xml_op Resource history XML * \param[in,out] node Node to use for digest calculation * \param[in,out] scheduler Scheduler data * * \return Pointer to node's digest cache entry, with comparison result set */ pcmk__op_digest_t * rsc_action_digest_cmp(pcmk_resource_t *rsc, const xmlNode *xml_op, pcmk_node_t *node, pcmk_scheduler_t *scheduler) { pcmk__op_digest_t *data = NULL; guint interval_ms = 0; const char *op_version; const char *task = crm_element_value(xml_op, PCMK_XA_OPERATION); const char *digest_all; const char *digest_restart; CRM_ASSERT(node != NULL); op_version = crm_element_value(xml_op, PCMK_XA_CRM_FEATURE_SET); digest_all = crm_element_value(xml_op, PCMK__XA_OP_DIGEST); digest_restart = crm_element_value(xml_op, PCMK__XA_OP_RESTART_DIGEST); crm_element_value_ms(xml_op, PCMK_META_INTERVAL, &interval_ms); data = rsc_action_digest(rsc, task, interval_ms, node, xml_op, pcmk_is_set(scheduler->flags, pcmk_sched_sanitized), scheduler); if (digest_restart && data->digest_restart_calc && strcmp(data->digest_restart_calc, digest_restart) != 0) { pcmk__rsc_info(rsc, "Parameters to %ums-interval %s action for %s on %s " "changed: hash was %s vs. now %s (restart:%s) %s", interval_ms, task, rsc->id, pcmk__node_name(node), pcmk__s(digest_restart, "missing"), data->digest_restart_calc, op_version, crm_element_value(xml_op, PCMK__XA_TRANSITION_MAGIC)); data->rc = pcmk__digest_restart; } else if (digest_all == NULL) { /* it is unknown what the previous op digest was */ data->rc = pcmk__digest_unknown; } else if (strcmp(digest_all, data->digest_all_calc) != 0) { /* Given a non-recurring operation with extra parameters configured, * in case that the main digest doesn't match, even if the restart * digest matches, enforce a restart rather than a reload-agent anyway. * So that it ensures any changes of the extra parameters get applied * for this specific operation, and the digests calculated for the * resulting PCMK__XE_LRM_RSC_OP will be correct. * Preserve the implied rc pcmk__digest_restart for the case that the * main digest doesn't match. */ if ((interval_ms == 0) && (data->rc == pcmk__digest_restart)) { pcmk__rsc_info(rsc, "Parameters containing extra ones to %ums-interval" " %s action for %s on %s " "changed: hash was %s vs. now %s (restart:%s) %s", interval_ms, task, rsc->id, pcmk__node_name(node), pcmk__s(digest_all, "missing"), data->digest_all_calc, op_version, crm_element_value(xml_op, PCMK__XA_TRANSITION_MAGIC)); } else { pcmk__rsc_info(rsc, "Parameters to %ums-interval %s action for %s on %s " "changed: hash was %s vs. now %s (%s:%s) %s", interval_ms, task, rsc->id, pcmk__node_name(node), pcmk__s(digest_all, "missing"), data->digest_all_calc, (interval_ms > 0)? "reschedule" : "reload", op_version, crm_element_value(xml_op, PCMK__XA_TRANSITION_MAGIC)); data->rc = pcmk__digest_mismatch; } } else { data->rc = pcmk__digest_match; } return data; } /*! * \internal * \brief Create an unfencing summary for use in special node attribute * * Create a string combining a fence device's resource ID, agent type, and * parameter digest (whether for all parameters or just non-private parameters). * This can be stored in a special node attribute, allowing us to detect changes * in either the agent type or parameters, to know whether unfencing must be * redone or can be safely skipped when the device's history is cleaned. * * \param[in] rsc_id Fence device resource ID * \param[in] agent_type Fence device agent * \param[in] param_digest Fence device parameter digest * * \return Newly allocated string with unfencing digest * \note The caller is responsible for freeing the result. */ static inline char * create_unfencing_summary(const char *rsc_id, const char *agent_type, const char *param_digest) { return crm_strdup_printf("%s:%s:%s", rsc_id, agent_type, param_digest); } /*! * \internal * \brief Check whether a node can skip unfencing * * Check whether a fence device's current definition matches a node's * stored summary of when it was last unfenced by the device. * * \param[in] rsc_id Fence device's resource ID * \param[in] agent Fence device's agent type * \param[in] digest_calc Fence device's current parameter digest * \param[in] node_summary Value of node's special unfencing node attribute * (a comma-separated list of unfencing summaries for * all devices that have unfenced this node) * * \return TRUE if digest matches, FALSE otherwise */ static bool unfencing_digest_matches(const char *rsc_id, const char *agent, const char *digest_calc, const char *node_summary) { bool matches = FALSE; if (rsc_id && agent && digest_calc && node_summary) { char *search_secure = create_unfencing_summary(rsc_id, agent, digest_calc); /* The digest was calculated including the device ID and agent, * so there is no risk of collision using strstr(). */ matches = (strstr(node_summary, search_secure) != NULL); crm_trace("Calculated unfencing digest '%s' %sfound in '%s'", search_secure, matches? "" : "not ", node_summary); free(search_secure); } return matches; } /* Magic string to use as action name for digest cache entries used for * unfencing checks. This is not a real action name (i.e. "on"), so * pcmk__check_action_config() won't confuse these entries with real actions. */ #define STONITH_DIGEST_TASK "stonith-on" /*! * \internal * \brief Calculate fence device digests and digest comparison result * * \param[in,out] rsc Fence device resource * \param[in] agent Fence device's agent type * \param[in,out] node Node with digest cache to use * \param[in,out] scheduler Scheduler data * * \return Node's digest cache entry */ pcmk__op_digest_t * pe__compare_fencing_digest(pcmk_resource_t *rsc, const char *agent, pcmk_node_t *node, pcmk_scheduler_t *scheduler) { const char *node_summary = NULL; // Calculate device's current parameter digests pcmk__op_digest_t *data = rsc_action_digest(rsc, STONITH_DIGEST_TASK, 0U, node, NULL, TRUE, scheduler); // Check whether node has special unfencing summary node attribute node_summary = pcmk__node_attr(node, CRM_ATTR_DIGESTS_ALL, NULL, pcmk__rsc_node_current); if (node_summary == NULL) { data->rc = pcmk__digest_unknown; return data; } // Check whether full parameter digest matches if (unfencing_digest_matches(rsc->id, agent, data->digest_all_calc, node_summary)) { data->rc = pcmk__digest_match; return data; } // Check whether secure parameter digest matches node_summary = pcmk__node_attr(node, CRM_ATTR_DIGESTS_SECURE, NULL, pcmk__rsc_node_current); if (unfencing_digest_matches(rsc->id, agent, data->digest_secure_calc, node_summary)) { data->rc = pcmk__digest_match; if (!pcmk__is_daemon && scheduler->priv != NULL) { pcmk__output_t *out = scheduler->priv; out->info(out, "Only 'private' parameters to %s " "for unfencing %s changed", rsc->id, pcmk__node_name(node)); } return data; } // Parameters don't match data->rc = pcmk__digest_mismatch; if (pcmk_is_set(scheduler->flags, pcmk_sched_sanitized) && (data->digest_secure_calc != NULL)) { if (scheduler->priv != NULL) { pcmk__output_t *out = scheduler->priv; char *digest = create_unfencing_summary(rsc->id, agent, data->digest_secure_calc); out->info(out, "Parameters to %s for unfencing " "%s changed, try '%s'", rsc->id, pcmk__node_name(node), digest); free(digest); } else if (!pcmk__is_daemon) { char *digest = create_unfencing_summary(rsc->id, agent, data->digest_secure_calc); printf("Parameters to %s for unfencing %s changed, try '%s'\n", rsc->id, pcmk__node_name(node), digest); free(digest); } } return data; }