diff --git a/cts/cli/regression.daemons.exp b/cts/cli/regression.daemons.exp index 494eef418a..5b1d67f22b 100644 --- a/cts/cli/regression.daemons.exp +++ b/cts/cli/regression.daemons.exp @@ -1,464 +1,438 @@ =#=#=#= Begin test: Get CIB manager metadata =#=#=#= 1.1 Cluster options used by Pacemaker's Cluster Information Base manager Cluster Information Base manager options Enable Access Control Lists (ACLs) for the CIB Enable Access Control Lists (ACLs) for the CIB Raise this if log has "Evicting client" messages for cluster daemon PIDs (a good value is the number of resources in the cluster multiplied by the number of nodes). Maximum IPC message backlog before disconnecting a cluster daemon =#=#=#= End test: Get CIB manager metadata - OK (0) =#=#=#= * Passed: pacemaker-based - Get CIB manager metadata =#=#=#= Begin test: Get controller metadata =#=#=#= 1.1 Cluster options used by Pacemaker's controller Pacemaker controller options Includes a hash which identifies the exact changeset the code was built from. Used for diagnostic purposes. Pacemaker version on cluster node elected Designated Controller (DC) Used for informational and diagnostic purposes. The messaging stack on which Pacemaker is currently running This optional value is mostly for users' convenience as desired in administration, but may also be used in Pacemaker configuration rules via the #cluster-name node attribute, and by higher-level tools and resource agents. An arbitrary name for the cluster The optimal value will depend on the speed and load of your network and the type of switches used. How long to wait for a response from other nodes during start-up Pacemaker is primarily event-driven, and looks ahead to know when to recheck cluster state for failure timeouts and most time-based rules. However, it will also recheck the cluster after this amount of inactivity, to evaluate rules with date specifications and serve as a fail-safe for certain types of scheduler bugs. Allowed values: Zero disables polling, while positive values are an interval in seconds (unless other units are specified, for example "5min"). Polling interval to recheck cluster state and evaluate rules with date specifications - - The cluster will slow down its recovery process when the amount of system resources used (currently CPU) approaches this limit - Maximum amount of system load that should be used by cluster nodes - - - - Maximum number of jobs that can be scheduled per node (defaults to 2x cores) - Maximum number of jobs that can be scheduled per node (defaults to 2x cores) - - A cluster node may receive notification of its own fencing if fencing is misconfigured, or if fabric fencing is in use that doesn't cut cluster communication. Use "stop" to attempt to immediately stop Pacemaker and stay stopped, or "panic" to attempt to immediately reboot the local node, falling back to stop on failure. Allowed values: stop, panic. How a cluster node should react if notified of its own fencing Declare an election failed if it is not decided within this much time. If you need to adjust this value, it probably indicates the presence of a bug. *** Advanced Use Only *** Exit immediately if shutdown does not complete within this much time. If you need to adjust this value, it probably indicates the presence of a bug. *** Advanced Use Only *** If you need to adjust this value, it probably indicates the presence of a bug. *** Advanced Use Only *** If you need to adjust this value, it probably indicates the presence of a bug. *** Advanced Use Only *** Delay cluster recovery for this much time to allow for additional events to occur. Useful if your configuration is sensitive to the order in which ping updates arrive. *** Advanced Use Only *** Enabling this option will slow down cluster recovery under all conditions If this is set to a positive value, lost nodes are assumed to self-fence using watchdog-based SBD within this much time. This does not require a fencing resource to be explicitly configured, though a fence_watchdog resource can be configured, to limit use to specific nodes. If this is set to 0 (the default), the cluster will never assume watchdog-based self-fencing. If this is set to a negative value, the cluster will use twice the local value of the `SBD_WATCHDOG_TIMEOUT` environment variable if that is positive, or otherwise treat this as 0. WARNING: When used, this timeout must be larger than `SBD_WATCHDOG_TIMEOUT` on all nodes that use watchdog-based SBD, and Pacemaker will refuse to start on any of those nodes where this is not true for the local value or SBD is not active. When this is set to a negative value, `SBD_WATCHDOG_TIMEOUT` must be set to the same value on all nodes that use SBD, otherwise data corruption or loss could occur. How long before nodes can be assumed to be safely down when watchdog-based self-fencing via SBD is in use How many times fencing can fail before it will no longer be immediately re-attempted on a target How many times fencing can fail before it will no longer be immediately re-attempted on a target - - What to do when the cluster does not have quorum. Allowed values: stop, freeze, ignore, demote, suicide. - What to do when the cluster does not have quorum - - - - - When true, resources active on a node when it is cleanly shut down are kept "locked" to that node (not allowed to run elsewhere) until they start again on that node after it rejoins (or for at most shutdown-lock-limit, if set). Stonith resources and Pacemaker Remote connections are never locked. Clone and bundle instances and the promoted role of promotable clones are currently never locked, though support could be added in a future release. - Whether to lock resources to a cleanly shut down node - - - - If shutdown-lock is true and this is set to a nonzero time duration, shutdown locks will expire after this much time has passed since the shutdown was initiated, even if the node has not rejoined. - Do not lock resources to a cleanly shut down node longer than this - + + The cluster will slow down its recovery process when the amount of system resources used (currently CPU) approaches this limit + Maximum amount of system load that should be used by cluster nodes + - - Fence nodes that do not join the controller process group within this much time after joining the cluster, to allow the cluster to continue managing resources. A value of 0 means never fence pending nodes. Setting the value to 2h means fence nodes after 2 hours. - How long to wait for a node that has joined the cluster to join the controller process group - + + Maximum number of jobs that can be scheduled per node (defaults to 2x cores) + Maximum number of jobs that can be scheduled per node (defaults to 2x cores) + =#=#=#= End test: Get controller metadata - OK (0) =#=#=#= * Passed: pacemaker-controld - Get controller metadata =#=#=#= Begin test: Get fencer metadata =#=#=#= 1.1 Instance attributes available for all "stonith"-class resources and used by Pacemaker's fence daemon, formerly known as stonithd Instance attributes available for all "stonith"-class resources Some devices do not support the standard 'port' parameter or may provide additional ones. Use this to specify an alternate, device-specific, parameter that should indicate the machine to be fenced. A value of "none" can be used to tell the cluster not to supply any additional parameters. *** Advanced Use Only *** An alternate parameter to supply instead of 'port' For example, "node1:1;node2:2,3" would tell the cluster to use port 1 for node1 and ports 2 and 3 for node2. A mapping of node names to port numbers for devices that do not support node names. For example, "node1,node2,node3". A list of nodes that can be targeted by this device (optional unless pcmk_host_list="static-list") Use "dynamic-list" to query the device via the 'list' command; "static-list" to check the pcmk_host_list attribute; "status" to query the device via the 'status' command; or "none" to assume every device can fence every node. Allowed values: dynamic-list, static-list, status, none. How to determine which nodes can be targeted by the device Enable a delay of no more than the time specified before executing fencing actions. Pacemaker derives the overall delay by taking the value of pcmk_delay_base and adding a random delay value such that the sum is kept below this maximum. Enable a base delay for fencing actions and specify base delay value. This enables a static delay for fencing actions, which can help avoid "death matches" where two nodes try to fence each other at the same time. If pcmk_delay_max is also used, a random delay will be added such that the total delay is kept below that value. This can be set to a single time value to apply to any node targeted by this device (useful if a separate device is configured for each target), or to a node map (for example, "node1:1s;node2:5") to set a different value for each target. Enable a base delay for fencing actions and specify base delay value. Cluster property concurrent-fencing="true" needs to be configured first. Then use this to specify the maximum number of actions can be performed in parallel on this device. A value of -1 means an unlimited number of actions can be performed in parallel. The maximum number of actions can be performed in parallel on this device Some devices do not support the standard commands or may provide additional ones. Use this to specify an alternate, device-specific, command that implements the 'reboot' action. *** Advanced Use Only *** An alternate command to run instead of 'reboot' Some devices need much more/less time to complete than normal. Use this to specify an alternate, device-specific, timeout for 'reboot' actions. *** Advanced Use Only *** Specify an alternate timeout to use for 'reboot' actions instead of stonith-timeout Some devices do not support multiple connections. Operations may "fail" if the device is busy with another task. In that case, Pacemaker will automatically retry the operation if there is time remaining. Use this option to alter the number of times Pacemaker tries a 'reboot' action before giving up. *** Advanced Use Only *** The maximum number of times to try the 'reboot' command within the timeout period Some devices do not support the standard commands or may provide additional ones. Use this to specify an alternate, device-specific, command that implements the 'off' action. *** Advanced Use Only *** An alternate command to run instead of 'off' Some devices need much more/less time to complete than normal. Use this to specify an alternate, device-specific, timeout for 'off' actions. *** Advanced Use Only *** Specify an alternate timeout to use for 'off' actions instead of stonith-timeout Some devices do not support multiple connections. Operations may "fail" if the device is busy with another task. In that case, Pacemaker will automatically retry the operation if there is time remaining. Use this option to alter the number of times Pacemaker tries a 'off' action before giving up. *** Advanced Use Only *** The maximum number of times to try the 'off' command within the timeout period Some devices do not support the standard commands or may provide additional ones. Use this to specify an alternate, device-specific, command that implements the 'on' action. *** Advanced Use Only *** An alternate command to run instead of 'on' Some devices need much more/less time to complete than normal. Use this to specify an alternate, device-specific, timeout for 'on' actions. *** Advanced Use Only *** Specify an alternate timeout to use for 'on' actions instead of stonith-timeout Some devices do not support multiple connections. Operations may "fail" if the device is busy with another task. In that case, Pacemaker will automatically retry the operation if there is time remaining. Use this option to alter the number of times Pacemaker tries a 'on' action before giving up. *** Advanced Use Only *** The maximum number of times to try the 'on' command within the timeout period Some devices do not support the standard commands or may provide additional ones. Use this to specify an alternate, device-specific, command that implements the 'list' action. *** Advanced Use Only *** An alternate command to run instead of 'list' Some devices need much more/less time to complete than normal. Use this to specify an alternate, device-specific, timeout for 'list' actions. *** Advanced Use Only *** Specify an alternate timeout to use for 'list' actions instead of stonith-timeout Some devices do not support multiple connections. Operations may "fail" if the device is busy with another task. In that case, Pacemaker will automatically retry the operation if there is time remaining. Use this option to alter the number of times Pacemaker tries a 'list' action before giving up. *** Advanced Use Only *** The maximum number of times to try the 'list' command within the timeout period Some devices do not support the standard commands or may provide additional ones. Use this to specify an alternate, device-specific, command that implements the 'monitor' action. *** Advanced Use Only *** An alternate command to run instead of 'monitor' Some devices need much more/less time to complete than normal. Use this to specify an alternate, device-specific, timeout for 'monitor' actions. *** Advanced Use Only *** Specify an alternate timeout to use for 'monitor' actions instead of stonith-timeout Some devices do not support multiple connections. Operations may "fail" if the device is busy with another task. In that case, Pacemaker will automatically retry the operation if there is time remaining. Use this option to alter the number of times Pacemaker tries a 'monitor' action before giving up. *** Advanced Use Only *** The maximum number of times to try the 'monitor' command within the timeout period Some devices do not support the standard commands or may provide additional ones. Use this to specify an alternate, device-specific, command that implements the 'status' action. *** Advanced Use Only *** An alternate command to run instead of 'status' Some devices need much more/less time to complete than normal. Use this to specify an alternate, device-specific, timeout for 'status' actions. *** Advanced Use Only *** Specify an alternate timeout to use for 'status' actions instead of stonith-timeout Some devices do not support multiple connections. Operations may "fail" if the device is busy with another task. In that case, Pacemaker will automatically retry the operation if there is time remaining. Use this option to alter the number of times Pacemaker tries a 'status' action before giving up. *** Advanced Use Only *** The maximum number of times to try the 'status' command within the timeout period =#=#=#= End test: Get fencer metadata - OK (0) =#=#=#= * Passed: pacemaker-fenced - Get fencer metadata =#=#=#= Begin test: Get scheduler metadata =#=#=#= 1.1 Cluster options used by Pacemaker's scheduler Pacemaker scheduler options What to do when the cluster does not have quorum. Allowed values: stop, freeze, ignore, demote, suicide. What to do when the cluster does not have quorum + + When true, resources active on a node when it is cleanly shut down are kept "locked" to that node (not allowed to run elsewhere) until they start again on that node after it rejoins (or for at most shutdown-lock-limit, if set). Stonith resources and Pacemaker Remote connections are never locked. Clone and bundle instances and the promoted role of promotable clones are currently never locked, though support could be added in a future release. + Whether to lock resources to a cleanly shut down node + + + + If shutdown-lock is true and this is set to a nonzero time duration, shutdown locks will expire after this much time has passed since the shutdown was initiated, even if the node has not rejoined. + Do not lock resources to a cleanly shut down node longer than this + + Whether resources can run on any node by default Whether resources can run on any node by default Whether the cluster should refrain from monitoring, starting, and stopping resources Whether the cluster should refrain from monitoring, starting, and stopping resources When true, the cluster will immediately ban a resource from a node if it fails to start there. When false, the cluster will instead check the resource's fail count against its migration-threshold. Whether a start failure should prevent a resource from being recovered on the same node Whether the cluster should check for active resources during start-up Whether the cluster should check for active resources during start-up - - When true, resources active on a node when it is cleanly shut down are kept "locked" to that node (not allowed to run elsewhere) until they start again on that node after it rejoins (or for at most shutdown-lock-limit, if set). Stonith resources and Pacemaker Remote connections are never locked. Clone and bundle instances and the promoted role of promotable clones are currently never locked, though support could be added in a future release. - Whether to lock resources to a cleanly shut down node - - - - If shutdown-lock is true and this is set to a nonzero time duration, shutdown locks will expire after this much time has passed since the shutdown was initiated, even if the node has not rejoined. - Do not lock resources to a cleanly shut down node longer than this - - If false, unresponsive nodes are immediately assumed to be harmless, and resources that were active on them may be recovered elsewhere. This can result in a "split-brain" situation, potentially leading to data loss and/or service unavailability. *** Advanced Use Only *** Whether nodes may be fenced as part of recovery Action to send to fence device when a node needs to be fenced ("poweroff" is a deprecated alias for "off"). Allowed values: reboot, off, poweroff. Action to send to fence device when a node needs to be fenced ("poweroff" is a deprecated alias for "off") How long to wait for on, off, and reboot fence actions to complete by default How long to wait for on, off, and reboot fence actions to complete by default This is set automatically by the cluster according to whether SBD is detected to be in use. User-configured values are ignored. The value `true` is meaningful if diskless SBD is used and `stonith-watchdog-timeout` is nonzero. In that case, if fencing is required, watchdog-based self-fencing will be performed via SBD without requiring a fencing resource explicitly configured. Whether watchdog integration is enabled Allow performing fencing operations in parallel Allow performing fencing operations in parallel Setting this to false may lead to a "split-brain" situation, potentially leading to data loss and/or service unavailability. *** Advanced Use Only *** Whether to fence unseen nodes at start-up Apply specified delay for the fencings that are targeting the lost nodes with the highest total resource priority in case we don't have the majority of the nodes in our cluster partition, so that the more significant nodes potentially win any fencing match, which is especially meaningful under split-brain of 2-node cluster. A promoted resource instance takes the base priority + 1 on calculation if the base priority is not 0. Any static/random delays that are introduced by `pcmk_delay_base/max` configured for the corresponding fencing resources will be added to this delay. This delay should be significantly greater than, safely twice, the maximum `pcmk_delay_base/max`. By default, priority fencing delay is disabled. Apply fencing delay targeting the lost nodes with the highest total resource priority Fence nodes that do not join the controller process group within this much time after joining the cluster, to allow the cluster to continue managing resources. A value of 0 means never fence pending nodes. Setting the value to 2h means fence nodes after 2 hours. How long to wait for a node that has joined the cluster to join the controller process group The node elected Designated Controller (DC) will consider an action failed if it does not get a response from the node executing the action within this time (after considering the action's own timeout). The "correct" value will depend on the speed and load of your network and cluster nodes. Maximum time for node-to-node communication The "correct" value will depend on the speed and load of your network and cluster nodes. If set to 0, the cluster will impose a dynamically calculated limit when any node has a high load. Maximum number of jobs that the cluster may execute in parallel across all nodes The number of live migration actions that the cluster is allowed to execute in parallel on a node (-1 means no limit) The number of live migration actions that the cluster is allowed to execute in parallel on a node (-1 means no limit) Whether the cluster should stop all active resources Whether the cluster should stop all active resources Whether to stop resources that were removed from the configuration Whether to stop resources that were removed from the configuration Whether to cancel recurring actions removed from the configuration Whether to cancel recurring actions removed from the configuration Values other than default are poorly tested and potentially dangerous. This option will be removed in a future release. *** Deprecated *** Whether to remove stopped resources from the executor Zero to disable, -1 to store unlimited. The number of scheduler inputs resulting in errors to save Zero to disable, -1 to store unlimited. The number of scheduler inputs resulting in warnings to save Zero to disable, -1 to store unlimited. The number of scheduler inputs without errors or warnings to save Requires external entities to create node attributes (named with the prefix "#health") with values "red", "yellow", or "green". Allowed values: none, migrate-on-red, only-green, progressive, custom. How cluster should react to node health attributes Only used when "node-health-strategy" is set to "progressive". Base health score assigned to a node Only used when "node-health-strategy" is set to "custom" or "progressive". The score to use for a node health attribute whose value is "green" Only used when "node-health-strategy" is set to "custom" or "progressive". The score to use for a node health attribute whose value is "yellow" Only used when "node-health-strategy" is set to "custom" or "progressive". The score to use for a node health attribute whose value is "red" How the cluster should allocate resources to nodes. Allowed values: default, utilization, minimal, balanced. How the cluster should allocate resources to nodes =#=#=#= End test: Get scheduler metadata - OK (0) =#=#=#= * Passed: pacemaker-schedulerd - Get scheduler metadata diff --git a/daemons/based/pacemaker-based.c b/daemons/based/pacemaker-based.c index d0d478d0d3..096ea488f8 100644 --- a/daemons/based/pacemaker-based.c +++ b/daemons/based/pacemaker-based.c @@ -1,432 +1,447 @@ /* * 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 #include #include #include #include #include #include #include #include #include #include #define SUMMARY "daemon for managing the configuration of a Pacemaker cluster" extern int init_remote_listener(int port, gboolean encrypted); gboolean cib_shutdown_flag = FALSE; int cib_status = pcmk_ok; crm_cluster_t *crm_cluster = NULL; GMainLoop *mainloop = NULL; gchar *cib_root = NULL; static gboolean preserve_status = FALSE; gboolean cib_writes_enabled = TRUE; gboolean stand_alone = FALSE; int remote_fd = 0; int remote_tls_fd = 0; GHashTable *config_hash = NULL; GHashTable *local_notify_queue = NULL; static void cib_init(void); void cib_shutdown(int nsig); static bool startCib(const char *filename); extern int write_cib_contents(gpointer p); static crm_exit_t exit_code = CRM_EX_OK; static void cib_enable_writes(int nsig) { crm_info("(Re)enabling disk writes"); cib_writes_enabled = TRUE; } /*! * \internal * \brief Set up options, users, and groups for stand-alone mode * * \param[out] error GLib error object * * \return Standard Pacemaker return code */ static int setup_stand_alone(GError **error) { int rc = 0; struct passwd *pwentry = NULL; preserve_status = TRUE; cib_writes_enabled = FALSE; errno = 0; pwentry = getpwnam(CRM_DAEMON_USER); if (pwentry == NULL) { exit_code = CRM_EX_FATAL; if (errno != 0) { g_set_error(error, PCMK__EXITC_ERROR, exit_code, "Error getting password DB entry for %s: %s", CRM_DAEMON_USER, strerror(errno)); return errno; } g_set_error(error, PCMK__EXITC_ERROR, exit_code, "Password DB entry for '%s' not found", CRM_DAEMON_USER); return ENXIO; } rc = setgid(pwentry->pw_gid); if (rc < 0) { exit_code = CRM_EX_FATAL; g_set_error(error, PCMK__EXITC_ERROR, exit_code, "Could not set group to %d: %s", pwentry->pw_gid, strerror(errno)); return errno; } rc = initgroups(CRM_DAEMON_USER, pwentry->pw_gid); if (rc < 0) { exit_code = CRM_EX_FATAL; g_set_error(error, PCMK__EXITC_ERROR, exit_code, "Could not setup groups for user %d: %s", pwentry->pw_uid, strerror(errno)); return errno; } rc = setuid(pwentry->pw_uid); if (rc < 0) { exit_code = CRM_EX_FATAL; g_set_error(error, PCMK__EXITC_ERROR, exit_code, "Could not set user to %d: %s", pwentry->pw_uid, strerror(errno)); return errno; } return pcmk_rc_ok; } +static void +based_metadata(void) +{ + const char *name = "pacemaker-based"; + const char *desc_short = "Cluster Information Base manager options"; + const char *desc_long = "Cluster options used by Pacemaker's Cluster " + "Information Base manager"; + + gchar *s = pcmk__cluster_option_metadata(name, desc_short, desc_long, + pcmk__opt_context_based); + + printf("%s", s); + g_free(s); +} + static GOptionEntry entries[] = { { "stand-alone", 's', G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &stand_alone, "(Advanced use only) Run in stand-alone mode", NULL }, { "disk-writes", 'w', G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &cib_writes_enabled, "(Advanced use only) Enable disk writes (enabled by default unless in " "stand-alone mode)", NULL }, { "cib-root", 'r', G_OPTION_FLAG_NONE, G_OPTION_ARG_FILENAME, &cib_root, "(Advanced use only) Directory where the CIB XML file should be located " "(default: " CRM_CONFIG_DIR ")", NULL }, { NULL } }; static pcmk__supported_format_t formats[] = { PCMK__SUPPORTED_FORMAT_NONE, PCMK__SUPPORTED_FORMAT_TEXT, PCMK__SUPPORTED_FORMAT_XML, { NULL, NULL, NULL } }; static GOptionContext * build_arg_context(pcmk__common_args_t *args, GOptionGroup **group) { GOptionContext *context = NULL; context = pcmk__build_arg_context(args, "text (default), xml", group, "[metadata]"); pcmk__add_main_args(context, entries); return context; } int main(int argc, char **argv) { int rc = pcmk_rc_ok; crm_ipc_t *old_instance = NULL; pcmk__output_t *out = NULL; GError *error = NULL; GOptionGroup *output_group = NULL; pcmk__common_args_t *args = pcmk__new_common_args(SUMMARY); gchar **processed_args = pcmk__cmdline_preproc(argv, "r"); GOptionContext *context = build_arg_context(args, &output_group); crm_log_preinit(NULL, argc, argv); pcmk__register_formats(output_group, formats); if (!g_option_context_parse_strv(context, &processed_args, &error)) { exit_code = CRM_EX_USAGE; goto done; } rc = pcmk__output_new(&out, args->output_ty, args->output_dest, argv); if (rc != pcmk_rc_ok) { exit_code = CRM_EX_ERROR; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Error creating output format %s: %s", args->output_ty, pcmk_rc_str(rc)); goto done; } if (args->version) { out->version(out, false); goto done; } mainloop_add_signal(SIGTERM, cib_shutdown); mainloop_add_signal(SIGPIPE, cib_enable_writes); cib_writer = mainloop_add_trigger(G_PRIORITY_LOW, write_cib_contents, NULL); if ((g_strv_length(processed_args) >= 2) && pcmk__str_eq(processed_args[1], "metadata", pcmk__str_none)) { - cib_metadata(); + based_metadata(); goto done; } pcmk__cli_init_logging("pacemaker-based", args->verbosity); crm_log_init(NULL, LOG_INFO, TRUE, FALSE, argc, argv, FALSE); crm_notice("Starting Pacemaker CIB manager"); old_instance = crm_ipc_new(PCMK__SERVER_BASED_RO, 0); if (old_instance == NULL) { /* crm_ipc_new() will have already logged an error message with * crm_err() */ exit_code = CRM_EX_FATAL; goto done; } if (pcmk__connect_generic_ipc(old_instance) == pcmk_rc_ok) { /* IPC end-point already up */ crm_ipc_close(old_instance); crm_ipc_destroy(old_instance); crm_err("pacemaker-based is already active, aborting startup"); goto done; } else { /* not up or not authentic, we'll proceed either way */ crm_ipc_destroy(old_instance); old_instance = NULL; } if (stand_alone) { rc = setup_stand_alone(&error); if (rc != pcmk_rc_ok) { goto done; } } if (cib_root == NULL) { cib_root = g_strdup(CRM_CONFIG_DIR); } else { crm_notice("Using custom config location: %s", cib_root); } if (!pcmk__daemon_can_write(cib_root, NULL)) { exit_code = CRM_EX_FATAL; crm_err("Terminating due to bad permissions on %s", cib_root); g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Bad permissions on %s (see logs for details)", cib_root); goto done; } crm_peer_init(); // Read initial CIB, connect to cluster, and start IPC servers cib_init(); // Run the main loop mainloop = g_main_loop_new(NULL, FALSE); crm_notice("Pacemaker CIB manager successfully started and accepting connections"); g_main_loop_run(mainloop); /* If main loop returned, clean up and exit. We disconnect in case * terminate_cib() was called with fast=-1. */ crm_cluster_disconnect(crm_cluster); pcmk__stop_based_ipc(ipcs_ro, ipcs_rw, ipcs_shm); done: g_strfreev(processed_args); pcmk__free_arg_context(context); crm_peer_destroy(); if (local_notify_queue != NULL) { g_hash_table_destroy(local_notify_queue); } if (config_hash != NULL) { g_hash_table_destroy(config_hash); } pcmk__client_cleanup(); pcmk_cluster_free(crm_cluster); g_free(cib_root); pcmk__output_and_clear_error(&error, out); if (out != NULL) { out->finish(out, exit_code, true, NULL); pcmk__output_free(out); } pcmk__unregister_formats(); crm_exit(exit_code); } #if SUPPORT_COROSYNC static void cib_cs_dispatch(cpg_handle_t handle, const struct cpg_name *groupName, uint32_t nodeid, uint32_t pid, void *msg, size_t msg_len) { uint32_t kind = 0; xmlNode *xml = NULL; const char *from = NULL; char *data = pcmk_message_common_cs(handle, nodeid, pid, msg, &kind, &from); if(data == NULL) { return; } if (kind == crm_class_cluster) { xml = string2xml(data); if (xml == NULL) { crm_err("Invalid XML: '%.120s'", data); free(data); return; } crm_xml_add(xml, PCMK__XA_SRC, from); cib_peer_callback(xml, NULL); } free_xml(xml); free(data); } static void cib_cs_destroy(gpointer user_data) { if (cib_shutdown_flag) { crm_info("Corosync disconnection complete"); } else { crm_crit("Lost connection to cluster layer, shutting down"); terminate_cib(__func__, CRM_EX_DISCONNECT); } } #endif static void cib_peer_update_callback(enum crm_status_type type, crm_node_t * node, const void *data) { switch (type) { case crm_status_processes: if (cib_legacy_mode() && !pcmk_is_set(node->processes, crm_get_cluster_proc())) { uint32_t old = data? *(const uint32_t *)data : 0; if ((node->processes ^ old) & crm_proc_cpg) { crm_info("Attempting to disable legacy mode after %s left the cluster", node->uname); legacy_mode = FALSE; } } break; case crm_status_uname: case crm_status_nstate: if (cib_shutdown_flag && (crm_active_peers() < 2) && (pcmk__ipc_client_count() == 0)) { crm_info("No more peers"); terminate_cib(__func__, -1); } break; } } static void cib_init(void) { crm_cluster = pcmk_cluster_new(); #if SUPPORT_COROSYNC if (is_corosync_cluster()) { crm_cluster->destroy = cib_cs_destroy; crm_cluster->cpg.cpg_deliver_fn = cib_cs_dispatch; crm_cluster->cpg.cpg_confchg_fn = pcmk_cpg_membership; } #endif // SUPPORT_COROSYNC config_hash = pcmk__strkey_table(free, free); if (startCib("cib.xml") == FALSE) { crm_crit("Cannot start CIB... terminating"); crm_exit(CRM_EX_NOINPUT); } if (!stand_alone) { crm_set_status_callback(&cib_peer_update_callback); if (!crm_cluster_connect(crm_cluster)) { crm_crit("Cannot sign in to the cluster... terminating"); crm_exit(CRM_EX_FATAL); } } pcmk__serve_based_ipc(&ipcs_ro, &ipcs_rw, &ipcs_shm, &ipc_ro_callbacks, &ipc_rw_callbacks); if (stand_alone) { based_is_primary = true; } } static bool startCib(const char *filename) { gboolean active = FALSE; xmlNode *cib = readCibXmlFile(cib_root, filename, !preserve_status); if (activateCibXml(cib, TRUE, "start") == 0) { int port = 0; active = TRUE; cib_read_config(config_hash, cib); pcmk__scan_port(crm_element_value(cib, "remote-tls-port"), &port); if (port >= 0) { remote_tls_fd = init_remote_listener(port, TRUE); } pcmk__scan_port(crm_element_value(cib, "remote-clear-port"), &port); if (port >= 0) { remote_fd = init_remote_listener(port, FALSE); } } return active; } diff --git a/daemons/controld/controld_control.c b/daemons/controld/controld_control.c index 754d118708..bcf1c7cbe5 100644 --- a/daemons/controld/controld_control.c +++ b/daemons/controld/controld_control.c @@ -1,871 +1,684 @@ /* * 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 #include #include #include static qb_ipcs_service_t *ipcs = NULL; static crm_trigger_t *config_read_trigger = NULL; #if SUPPORT_COROSYNC extern gboolean crm_connect_corosync(crm_cluster_t * cluster); #endif static void crm_shutdown(int nsig); static gboolean crm_read_options(gpointer user_data); /* A_HA_CONNECT */ void do_ha_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) { gboolean registered = FALSE; static crm_cluster_t *cluster = NULL; if (cluster == NULL) { cluster = pcmk_cluster_new(); } if (action & A_HA_DISCONNECT) { crm_cluster_disconnect(cluster); crm_info("Disconnected from the cluster"); controld_set_fsa_input_flags(R_HA_DISCONNECTED); } if (action & A_HA_CONNECT) { crm_set_status_callback(&peer_update_callback); crm_set_autoreap(FALSE); #if SUPPORT_COROSYNC if (is_corosync_cluster()) { registered = crm_connect_corosync(cluster); } #endif // SUPPORT_COROSYNC if (registered) { controld_election_init(cluster->uname); controld_globals.our_nodename = cluster->uname; controld_globals.our_uuid = cluster->uuid; if(cluster->uuid == NULL) { crm_err("Could not obtain local uuid"); registered = FALSE; } } if (!registered) { controld_set_fsa_input_flags(R_HA_DISCONNECTED); register_fsa_error(C_FSA_INTERNAL, I_ERROR, NULL); return; } populate_cib_nodes(node_update_none, __func__); controld_clear_fsa_input_flags(R_HA_DISCONNECTED); crm_info("Connected to the cluster"); } if (action & ~(A_HA_CONNECT | A_HA_DISCONNECT)) { crm_err("Unexpected action %s in %s", fsa_action2string(action), __func__); } } /* A_SHUTDOWN */ void do_shutdown(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) { /* just in case */ controld_set_fsa_input_flags(R_SHUTDOWN); controld_disconnect_fencer(FALSE); } /* A_SHUTDOWN_REQ */ void do_shutdown_req(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) { xmlNode *msg = NULL; controld_set_fsa_input_flags(R_SHUTDOWN); //controld_set_fsa_input_flags(R_STAYDOWN); crm_info("Sending shutdown request to all peers (DC is %s)", pcmk__s(controld_globals.dc_name, "not set")); msg = create_request(CRM_OP_SHUTDOWN_REQ, NULL, NULL, CRM_SYSTEM_CRMD, CRM_SYSTEM_CRMD, NULL); if (send_cluster_message(NULL, crm_msg_crmd, msg, TRUE) == FALSE) { register_fsa_error(C_FSA_INTERNAL, I_ERROR, NULL); } free_xml(msg); } void crmd_fast_exit(crm_exit_t exit_code) { if (pcmk_is_set(controld_globals.fsa_input_register, R_STAYDOWN)) { crm_warn("Inhibiting respawn "CRM_XS" remapping exit code %d to %d", exit_code, CRM_EX_FATAL); exit_code = CRM_EX_FATAL; } else if ((exit_code == CRM_EX_OK) && pcmk_is_set(controld_globals.fsa_input_register, R_IN_RECOVERY)) { crm_err("Could not recover from internal error"); exit_code = CRM_EX_ERROR; } if (controld_globals.logger_out != NULL) { controld_globals.logger_out->finish(controld_globals.logger_out, exit_code, true, NULL); pcmk__output_free(controld_globals.logger_out); controld_globals.logger_out = NULL; } crm_exit(exit_code); } crm_exit_t crmd_exit(crm_exit_t exit_code) { GMainLoop *mloop = controld_globals.mainloop; static bool in_progress = FALSE; if (in_progress && (exit_code == CRM_EX_OK)) { crm_debug("Exit is already in progress"); return exit_code; } else if(in_progress) { crm_notice("Error during shutdown process, exiting now with status %d (%s)", exit_code, crm_exit_str(exit_code)); crm_write_blackbox(SIGTRAP, NULL); crmd_fast_exit(exit_code); } in_progress = TRUE; crm_trace("Preparing to exit with status %d (%s)", exit_code, crm_exit_str(exit_code)); /* Suppress secondary errors resulting from us disconnecting everything */ controld_set_fsa_input_flags(R_HA_DISCONNECTED); /* Close all IPC servers and clients to ensure any and all shared memory files are cleaned up */ if(ipcs) { crm_trace("Closing IPC server"); mainloop_del_ipc_server(ipcs); ipcs = NULL; } controld_close_attrd_ipc(); controld_shutdown_schedulerd_ipc(); controld_disconnect_fencer(TRUE); if ((exit_code == CRM_EX_OK) && (controld_globals.mainloop == NULL)) { crm_debug("No mainloop detected"); exit_code = CRM_EX_ERROR; } /* On an error, just get out. * * Otherwise, make the effort to have mainloop exit gracefully so * that it (mostly) cleans up after itself and valgrind has less * to report on - allowing real errors stand out */ if (exit_code != CRM_EX_OK) { crm_notice("Forcing immediate exit with status %d (%s)", exit_code, crm_exit_str(exit_code)); crm_write_blackbox(SIGTRAP, NULL); crmd_fast_exit(exit_code); } /* Clean up as much memory as possible for valgrind */ for (GList *iter = controld_globals.fsa_message_queue; iter != NULL; iter = iter->next) { fsa_data_t *fsa_data = (fsa_data_t *) iter->data; crm_info("Dropping %s: [ state=%s cause=%s origin=%s ]", fsa_input2string(fsa_data->fsa_input), fsa_state2string(controld_globals.fsa_state), fsa_cause2string(fsa_data->fsa_cause), fsa_data->origin); delete_fsa_input(fsa_data); } controld_clear_fsa_input_flags(R_MEMBERSHIP); g_list_free(controld_globals.fsa_message_queue); controld_globals.fsa_message_queue = NULL; controld_free_node_pending_timers(); controld_election_fini(); /* Tear down the CIB manager connection, but don't free it yet -- it could * be used when we drain the mainloop later. */ controld_disconnect_cib_manager(); verify_stopped(controld_globals.fsa_state, LOG_WARNING); controld_clear_fsa_input_flags(R_LRM_CONNECTED); lrm_state_destroy_all(); mainloop_destroy_trigger(config_read_trigger); config_read_trigger = NULL; controld_destroy_fsa_trigger(); controld_destroy_transition_trigger(); pcmk__client_cleanup(); crm_peer_destroy(); controld_free_fsa_timers(); te_cleanup_stonith_history_sync(NULL, TRUE); controld_free_sched_timer(); free(controld_globals.our_nodename); controld_globals.our_nodename = NULL; free(controld_globals.our_uuid); controld_globals.our_uuid = NULL; free(controld_globals.dc_name); controld_globals.dc_name = NULL; free(controld_globals.dc_version); controld_globals.dc_version = NULL; free(controld_globals.cluster_name); controld_globals.cluster_name = NULL; free(controld_globals.te_uuid); controld_globals.te_uuid = NULL; free_max_generation(); controld_destroy_failed_sync_table(); controld_destroy_outside_events_table(); mainloop_destroy_signal(SIGPIPE); mainloop_destroy_signal(SIGUSR1); mainloop_destroy_signal(SIGTERM); mainloop_destroy_signal(SIGTRAP); /* leave SIGCHLD engaged as we might still want to drain some service-actions */ if (mloop) { GMainContext *ctx = g_main_loop_get_context(controld_globals.mainloop); /* Don't re-enter this block */ controld_globals.mainloop = NULL; /* no signals on final draining anymore */ mainloop_destroy_signal(SIGCHLD); crm_trace("Draining mainloop %d %d", g_main_loop_is_running(mloop), g_main_context_pending(ctx)); { int lpc = 0; while((g_main_context_pending(ctx) && lpc < 10)) { lpc++; crm_trace("Iteration %d", lpc); g_main_context_dispatch(ctx); } } crm_trace("Closing mainloop %d %d", g_main_loop_is_running(mloop), g_main_context_pending(ctx)); g_main_loop_quit(mloop); /* Won't do anything yet, since we're inside it now */ g_main_loop_unref(mloop); } else { mainloop_destroy_signal(SIGCHLD); } cib_delete(controld_globals.cib_conn); controld_globals.cib_conn = NULL; throttle_fini(); /* Graceful */ crm_trace("Done preparing for exit with status %d (%s)", exit_code, crm_exit_str(exit_code)); return exit_code; } /* A_EXIT_0, A_EXIT_1 */ void do_exit(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) { crm_exit_t exit_code = CRM_EX_OK; if (pcmk_is_set(action, A_EXIT_1)) { exit_code = CRM_EX_ERROR; crm_err("Exiting now due to errors"); } verify_stopped(cur_state, LOG_ERR); crmd_exit(exit_code); } static void sigpipe_ignore(int nsig) { return; } /* A_STARTUP */ void do_startup(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) { crm_debug("Registering Signal Handlers"); mainloop_add_signal(SIGTERM, crm_shutdown); mainloop_add_signal(SIGPIPE, sigpipe_ignore); config_read_trigger = mainloop_add_trigger(G_PRIORITY_HIGH, crm_read_options, NULL); controld_init_fsa_trigger(); controld_init_transition_trigger(); crm_debug("Creating CIB manager and executor objects"); controld_globals.cib_conn = cib_new(); lrm_state_init_local(); if (controld_init_fsa_timers() == FALSE) { register_fsa_error(C_FSA_INTERNAL, I_ERROR, NULL); } } // \return libqb error code (0 on success, -errno on error) static int32_t accept_controller_client(qb_ipcs_connection_t *c, uid_t uid, gid_t gid) { crm_trace("Accepting new IPC client connection"); if (pcmk__new_client(c, uid, gid) == NULL) { return -EIO; } return 0; } // \return libqb error code (0 on success, -errno on error) static int32_t dispatch_controller_ipc(qb_ipcs_connection_t * c, void *data, size_t size) { uint32_t id = 0; uint32_t flags = 0; pcmk__client_t *client = pcmk__find_client(c); xmlNode *msg = pcmk__client_data2xml(client, data, &id, &flags); if (msg == NULL) { pcmk__ipc_send_ack(client, id, flags, "ack", NULL, CRM_EX_PROTOCOL); return 0; } pcmk__ipc_send_ack(client, id, flags, "ack", NULL, CRM_EX_INDETERMINATE); CRM_ASSERT(client->user != NULL); pcmk__update_acl_user(msg, PCMK__XA_CRM_USER, client->user); crm_xml_add(msg, PCMK__XA_CRM_SYS_FROM, client->id); if (controld_authorize_ipc_message(msg, client, NULL)) { crm_trace("Processing IPC message from client %s", pcmk__client_name(client)); route_message(C_IPC_MESSAGE, msg); } controld_trigger_fsa(); free_xml(msg); return 0; } static int32_t ipc_client_disconnected(qb_ipcs_connection_t *c) { pcmk__client_t *client = pcmk__find_client(c); if (client) { crm_trace("Disconnecting %sregistered client %s (%p/%p)", (client->userdata? "" : "un"), pcmk__client_name(client), c, client); free(client->userdata); pcmk__free_client(client); controld_trigger_fsa(); } return 0; } static void ipc_connection_destroyed(qb_ipcs_connection_t *c) { crm_trace("Connection %p", c); ipc_client_disconnected(c); } /* A_STOP */ void do_stop(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) { crm_trace("Closing IPC server"); mainloop_del_ipc_server(ipcs); ipcs = NULL; register_fsa_input(C_FSA_INTERNAL, I_TERMINATE, NULL); } /* A_STARTED */ void do_started(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 struct qb_ipcs_service_handlers crmd_callbacks = { .connection_accept = accept_controller_client, .connection_created = NULL, .msg_process = dispatch_controller_ipc, .connection_closed = ipc_client_disconnected, .connection_destroyed = ipc_connection_destroyed }; if (cur_state != S_STARTING) { crm_err("Start cancelled... %s", fsa_state2string(cur_state)); return; } else if (!pcmk_is_set(controld_globals.fsa_input_register, R_MEMBERSHIP)) { crm_info("Delaying start, no membership data (%.16llx)", R_MEMBERSHIP); crmd_fsa_stall(TRUE); return; } else if (!pcmk_is_set(controld_globals.fsa_input_register, R_LRM_CONNECTED)) { crm_info("Delaying start, not connected to executor (%.16llx)", R_LRM_CONNECTED); crmd_fsa_stall(TRUE); return; } else if (!pcmk_is_set(controld_globals.fsa_input_register, R_CIB_CONNECTED)) { crm_info("Delaying start, CIB not connected (%.16llx)", R_CIB_CONNECTED); crmd_fsa_stall(TRUE); return; } else if (!pcmk_is_set(controld_globals.fsa_input_register, R_READ_CONFIG)) { crm_info("Delaying start, Config not read (%.16llx)", R_READ_CONFIG); crmd_fsa_stall(TRUE); return; } else if (!pcmk_is_set(controld_globals.fsa_input_register, R_PEER_DATA)) { crm_info("Delaying start, No peer data (%.16llx)", R_PEER_DATA); crmd_fsa_stall(TRUE); return; } crm_debug("Init server comms"); ipcs = pcmk__serve_controld_ipc(&crmd_callbacks); if (ipcs == NULL) { crm_err("Failed to create IPC server: shutting down and inhibiting respawn"); register_fsa_error(C_FSA_INTERNAL, I_ERROR, NULL); } else { crm_notice("Pacemaker controller successfully started and accepting connections"); } controld_set_fsa_input_flags(R_ST_REQUIRED); controld_timer_fencer_connect(GINT_TO_POINTER(TRUE)); controld_clear_fsa_input_flags(R_STARTING); register_fsa_input(msg_data->fsa_cause, I_PENDING, NULL); } /* A_RECOVER */ void do_recover(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) { controld_set_fsa_input_flags(R_IN_RECOVERY); crm_warn("Fast-tracking shutdown in response to errors"); register_fsa_input(C_FSA_INTERNAL, I_TERMINATE, NULL); } -static pcmk__cluster_option_t controller_options[] = { - /* name, old name, type, allowed values, - * default value, validator, - * short description, - * long description - */ - { - PCMK_OPT_DC_VERSION, NULL, "string", NULL, - PCMK__VALUE_NONE, NULL, - N_("Pacemaker version on cluster node elected Designated Controller (DC)"), - N_("Includes a hash which identifies the exact changeset the code was " - "built from. Used for diagnostic purposes.") - }, - { - PCMK_OPT_CLUSTER_INFRASTRUCTURE, NULL, "string", NULL, - "corosync", NULL, - N_("The messaging stack on which Pacemaker is currently running"), - N_("Used for informational and diagnostic purposes.") - }, - { - PCMK_OPT_CLUSTER_NAME, NULL, "string", NULL, - NULL, NULL, - N_("An arbitrary name for the cluster"), - N_("This optional value is mostly for users' convenience as desired " - "in administration, but may also be used in Pacemaker " - "configuration rules via the #cluster-name node attribute, and " - "by higher-level tools and resource agents.") - }, - { - PCMK_OPT_DC_DEADTIME, NULL, "time", NULL, - "20s", pcmk__valid_interval_spec, - N_("How long to wait for a response from other nodes during start-up"), - N_("The optimal value will depend on the speed and load of your network " - "and the type of switches used.") - }, - { - PCMK_OPT_CLUSTER_RECHECK_INTERVAL, NULL, "time", - N_("Zero disables polling, while positive values are an interval in " - "seconds (unless other units are specified, for example \"5min\")"), - "15min", pcmk__valid_interval_spec, - N_("Polling interval to recheck cluster state and evaluate rules " - "with date specifications"), - N_("Pacemaker is primarily event-driven, and looks ahead to know when to " - "recheck cluster state for failure timeouts and most time-based " - "rules. However, it will also recheck the cluster after this " - "amount of inactivity, to evaluate rules with date specifications " - "and serve as a fail-safe for certain types of scheduler bugs.") - }, - { - PCMK_OPT_LOAD_THRESHOLD, NULL, "percentage", NULL, - "80%", pcmk__valid_percentage, - N_("Maximum amount of system load that should be used by cluster nodes"), - N_("The cluster will slow down its recovery process when the amount of " - "system resources used (currently CPU) approaches this limit"), - }, - { - PCMK_OPT_NODE_ACTION_LIMIT, NULL, "integer", NULL, - "0", pcmk__valid_int, - N_("Maximum number of jobs that can be scheduled per node " - "(defaults to 2x cores)") - }, - { - PCMK_OPT_FENCE_REACTION, NULL, "select", "stop, panic", - "stop", NULL, - N_("How a cluster node should react if notified of its own fencing"), - N_("A cluster node may receive notification of its own fencing if fencing " - "is misconfigured, or if fabric fencing is in use that doesn't cut " - "cluster communication. Use \"stop\" to attempt to immediately " - "stop Pacemaker and stay stopped, or \"panic\" to attempt to " - "immediately reboot the local node, falling back to stop on " - "failure.") - }, - { - PCMK_OPT_ELECTION_TIMEOUT, NULL, "time", NULL, - "2min", pcmk__valid_interval_spec, - "*** Advanced Use Only ***", - N_("Declare an election failed if it is not decided within this much " - "time. If you need to adjust this value, it probably indicates " - "the presence of a bug.") - }, - { - PCMK_OPT_SHUTDOWN_ESCALATION, NULL, "time", NULL, - "20min", pcmk__valid_interval_spec, - "*** Advanced Use Only ***", - N_("Exit immediately if shutdown does not complete within this much " - "time. If you need to adjust this value, it probably indicates " - "the presence of a bug.") - }, - { - PCMK_OPT_JOIN_INTEGRATION_TIMEOUT, "crmd-integration-timeout", "time", - NULL, - "3min", pcmk__valid_interval_spec, - "*** Advanced Use Only ***", - N_("If you need to adjust this value, it probably indicates " - "the presence of a bug.") - }, - { - PCMK_OPT_JOIN_FINALIZATION_TIMEOUT, "crmd-finalization-timeout", - "time", NULL, - "30min", pcmk__valid_interval_spec, - "*** Advanced Use Only ***", - N_("If you need to adjust this value, it probably indicates " - "the presence of a bug.") - }, - { - PCMK_OPT_TRANSITION_DELAY, "crmd-transition-delay", "time", NULL, - "0s", pcmk__valid_interval_spec, - N_("*** Advanced Use Only *** Enabling this option will slow down " - "cluster recovery under all conditions"), - N_("Delay cluster recovery for this much time to allow for additional " - "events to occur. Useful if your configuration is sensitive to " - "the order in which ping updates arrive.") - }, - { - PCMK_OPT_STONITH_WATCHDOG_TIMEOUT, NULL, "time", NULL, - "0", controld_verify_stonith_watchdog_timeout, - N_("How long before nodes can be assumed to be safely down when " - "watchdog-based self-fencing via SBD is in use"), - N_("If this is set to a positive value, lost nodes are assumed to " - "self-fence using watchdog-based SBD within this much time. This " - "does not require a fencing resource to be explicitly configured, " - "though a fence_watchdog resource can be configured, to limit use " - "to specific nodes. If this is set to 0 (the default), the cluster " - "will never assume watchdog-based self-fencing. If this is set to a " - "negative value, the cluster will use twice the local value of the " - "`SBD_WATCHDOG_TIMEOUT` environment variable if that is positive, " - "or otherwise treat this as 0. WARNING: When used, this timeout " - "must be larger than `SBD_WATCHDOG_TIMEOUT` on all nodes that use " - "watchdog-based SBD, and Pacemaker will refuse to start on any of " - "those nodes where this is not true for the local value or SBD is " - "not active. When this is set to a negative value, " - "`SBD_WATCHDOG_TIMEOUT` must be set to the same value on all nodes " - "that use SBD, otherwise data corruption or loss could occur.") - }, - { - PCMK_OPT_STONITH_MAX_ATTEMPTS, NULL, "integer", NULL, - "10", pcmk__valid_positive_int, - N_("How many times fencing can fail before it will no longer be " - "immediately re-attempted on a target") - }, - - // Already documented in libpe_status (other values must be kept identical) - { - PCMK_OPT_NO_QUORUM_POLICY, NULL, "select", - "stop, freeze, ignore, demote, suicide", - "stop", pcmk__valid_no_quorum_policy, - N_("What to do when the cluster does not have quorum"), NULL - }, - { - PCMK_OPT_SHUTDOWN_LOCK, NULL, "boolean", NULL, - PCMK_VALUE_FALSE, pcmk__valid_boolean, - N_("Whether to lock resources to a cleanly shut down node"), - N_("When true, resources active on a node when it is cleanly shut down " - "are kept \"locked\" to that node (not allowed to run elsewhere) " - "until they start again on that node after it rejoins (or for at " - "most shutdown-lock-limit, if set). Stonith resources and " - "Pacemaker Remote connections are never locked. Clone and bundle " - "instances and the promoted role of promotable clones are " - "currently never locked, though support could be added in a future " - "release.") - }, - { - PCMK_OPT_SHUTDOWN_LOCK_LIMIT, NULL, "time", NULL, - "0", pcmk__valid_interval_spec, - N_("Do not lock resources to a cleanly shut down node longer than " - "this"), - N_("If shutdown-lock is true and this is set to a nonzero time " - "duration, shutdown locks will expire after this much time has " - "passed since the shutdown was initiated, even if the node has not " - "rejoined.") - }, - { - PCMK_OPT_NODE_PENDING_TIMEOUT, NULL, "time", NULL, - "0", pcmk__valid_interval_spec, - N_("How long to wait for a node that has joined the cluster to join " - "the controller process group"), - N_("Fence nodes that do not join the controller process group within " - "this much time after joining the cluster, to allow the cluster " - "to continue managing resources. A value of 0 means never fence " - "pending nodes. Setting the value to 2h means fence nodes after " - "2 hours.") - }, -}; - -void -crmd_metadata(void) -{ - const char *desc_short = "Pacemaker controller options"; - const char *desc_long = "Cluster options used by Pacemaker's controller"; - - gchar *s = pcmk__format_option_metadata("pacemaker-controld", desc_short, - desc_long, controller_options, - PCMK__NELEM(controller_options)); - printf("%s", s); - g_free(s); -} - static void config_query_callback(xmlNode * msg, int call_id, int rc, xmlNode * output, void *user_data) { const char *value = NULL; GHashTable *config_hash = NULL; crm_time_t *now = crm_time_new(NULL); xmlNode *crmconfig = NULL; xmlNode *alerts = NULL; if (rc != pcmk_ok) { fsa_data_t *msg_data = NULL; crm_err("Local CIB query resulted in an error: %s", pcmk_strerror(rc)); register_fsa_error(C_FSA_INTERNAL, I_ERROR, NULL); if (rc == -EACCES || rc == -pcmk_err_schema_validation) { crm_err("The cluster is mis-configured - shutting down and staying down"); controld_set_fsa_input_flags(R_STAYDOWN); } goto bail; } crmconfig = output; if ((crmconfig != NULL) && !pcmk__xe_is(crmconfig, PCMK_XE_CRM_CONFIG)) { crmconfig = first_named_child(crmconfig, PCMK_XE_CRM_CONFIG); } if (!crmconfig) { fsa_data_t *msg_data = NULL; crm_err("Local CIB query for " PCMK_XE_CRM_CONFIG " section failed"); register_fsa_error(C_FSA_INTERNAL, I_ERROR, NULL); goto bail; } crm_debug("Call %d : Parsing CIB options", call_id); config_hash = pcmk__strkey_table(free, free); pe_unpack_nvpairs(crmconfig, crmconfig, PCMK_XE_CLUSTER_PROPERTY_SET, NULL, config_hash, CIB_OPTIONS_FIRST, FALSE, now, NULL); // Validate all options, and use defaults if not already present in hash - pcmk__validate_cluster_options(config_hash, controller_options, - PCMK__NELEM(controller_options)); + pcmk__validate_cluster_options(config_hash); + + /* Validate the watchdog timeout in the context of the local node + * environment. If invalid, the controller will exit with a fatal error. + * + * We do this via a wrapper in the controller, so that we call + * pcmk__valid_stonith_watchdog_timeout() only if watchdog fencing is + * enabled for the local node. Otherwise, we may exit unnecessarily. + * + * A validator function in libcrmcommon can't act as such a wrapper, because + * it doesn't have a stonith API connection or the local node name. + */ + value = g_hash_table_lookup(config_hash, PCMK_OPT_STONITH_WATCHDOG_TIMEOUT); + controld_verify_stonith_watchdog_timeout(value); value = g_hash_table_lookup(config_hash, PCMK_OPT_NO_QUORUM_POLICY); if (pcmk__str_eq(value, "suicide", pcmk__str_casei) && pcmk__locate_sbd()) { controld_set_global_flags(controld_no_quorum_suicide); } value = g_hash_table_lookup(config_hash, PCMK_OPT_SHUTDOWN_LOCK); if (crm_is_true(value)) { controld_set_global_flags(controld_shutdown_lock_enabled); } else { controld_clear_global_flags(controld_shutdown_lock_enabled); } value = g_hash_table_lookup(config_hash, PCMK_OPT_SHUTDOWN_LOCK_LIMIT); pcmk_parse_interval_spec(value, &controld_globals.shutdown_lock_limit); controld_globals.shutdown_lock_limit /= 1000; value = g_hash_table_lookup(config_hash, PCMK_OPT_NODE_PENDING_TIMEOUT); pcmk_parse_interval_spec(value, &controld_globals.node_pending_timeout); controld_globals.node_pending_timeout /= 1000; value = g_hash_table_lookup(config_hash, PCMK_OPT_CLUSTER_NAME); pcmk__str_update(&(controld_globals.cluster_name), value); // Let subcomponents initialize their own static variables controld_configure_election(config_hash); controld_configure_fencing(config_hash); controld_configure_fsa_timers(config_hash); controld_configure_throttle(config_hash); alerts = first_named_child(output, PCMK_XE_ALERTS); crmd_unpack_alerts(alerts); controld_set_fsa_input_flags(R_READ_CONFIG); controld_trigger_fsa(); g_hash_table_destroy(config_hash); bail: crm_time_free(now); } /*! * \internal * \brief Trigger read and processing of the configuration * * \param[in] fn Calling function name * \param[in] line Line number where call occurred */ void controld_trigger_config_as(const char *fn, int line) { if (config_read_trigger != NULL) { crm_trace("%s:%d - Triggered config processing", fn, line); mainloop_set_trigger(config_read_trigger); } } gboolean crm_read_options(gpointer user_data) { cib_t *cib_conn = controld_globals.cib_conn; int call_id = cib_conn->cmds->query(cib_conn, "//" PCMK_XE_CRM_CONFIG " | //" PCMK_XE_ALERTS, NULL, cib_xpath|cib_scope_local); fsa_register_cib_callback(call_id, NULL, config_query_callback); crm_trace("Querying the CIB... call %d", call_id); return TRUE; } /* A_READCONFIG */ void do_read_config(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) { throttle_init(); controld_trigger_config(); } static void crm_shutdown(int nsig) { const char *value = NULL; guint default_period_ms = 0; if ((controld_globals.mainloop == NULL) || !g_main_loop_is_running(controld_globals.mainloop)) { crmd_exit(CRM_EX_OK); return; } if (pcmk_is_set(controld_globals.fsa_input_register, R_SHUTDOWN)) { crm_err("Escalating shutdown"); register_fsa_input_before(C_SHUTDOWN, I_ERROR, NULL); return; } controld_set_fsa_input_flags(R_SHUTDOWN); register_fsa_input(C_SHUTDOWN, I_SHUTDOWN, NULL); /* If shutdown timer doesn't have a period set, use the default * * @TODO: Evaluate whether this is still necessary. As long as * config_query_callback() has been run at least once, it doesn't look like * anything could have changed the timer period since then. */ - value = pcmk__cluster_option(NULL, controller_options, - PCMK__NELEM(controller_options), - PCMK_OPT_SHUTDOWN_ESCALATION); + value = pcmk__cluster_option(NULL, PCMK_OPT_SHUTDOWN_ESCALATION); pcmk_parse_interval_spec(value, &default_period_ms); controld_shutdown_start_countdown(default_period_ms); } diff --git a/daemons/controld/pacemaker-controld.c b/daemons/controld/pacemaker-controld.c index e4a72c2445..d80644d5ef 100644 --- a/daemons/controld/pacemaker-controld.c +++ b/daemons/controld/pacemaker-controld.c @@ -1,205 +1,219 @@ /* - * Copyright 2004-2023 the Pacemaker project contributors + * 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 #include #include #include #include #include #include #include #define SUMMARY "daemon for coordinating a Pacemaker cluster's response " \ "to events" _Noreturn void crmd_init(void); extern void init_dotfile(void); controld_globals_t controld_globals = { // Automatic initialization to 0, false, or NULL is fine for most members .fsa_state = S_STARTING, .fsa_actions = A_NOTHING, }; static pcmk__supported_format_t formats[] = { PCMK__SUPPORTED_FORMAT_NONE, PCMK__SUPPORTED_FORMAT_TEXT, PCMK__SUPPORTED_FORMAT_XML, { NULL, NULL, NULL } }; +static void +controld_metadata(void) +{ + const char *name = "pacemaker-controld"; + const char *desc_short = "Pacemaker controller options"; + const char *desc_long = "Cluster options used by Pacemaker's controller"; + + gchar *s = pcmk__cluster_option_metadata(name, desc_short, desc_long, + pcmk__opt_context_controld); + + printf("%s", s); + g_free(s); +} + static GOptionContext * build_arg_context(pcmk__common_args_t *args, GOptionGroup **group) { return pcmk__build_arg_context(args, "text (default), xml", group, "[metadata]"); } int main(int argc, char **argv) { int rc = pcmk_rc_ok; crm_exit_t exit_code = CRM_EX_OK; bool initialize = true; crm_ipc_t *old_instance = NULL; pcmk__output_t *out = NULL; GError *error = NULL; GOptionGroup *output_group = NULL; pcmk__common_args_t *args = pcmk__new_common_args(SUMMARY); gchar **processed_args = pcmk__cmdline_preproc(argv, NULL); GOptionContext *context = build_arg_context(args, &output_group); crm_log_preinit(NULL, argc, argv); pcmk__register_formats(output_group, formats); if (!g_option_context_parse_strv(context, &processed_args, &error)) { exit_code = CRM_EX_USAGE; goto done; } rc = pcmk__output_new(&out, args->output_ty, args->output_dest, argv); if (rc != pcmk_rc_ok) { exit_code = CRM_EX_ERROR; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Error creating output format %s: %s", args->output_ty, pcmk_rc_str(rc)); goto done; } if (args->version) { out->version(out, false); initialize = false; goto done; } if ((g_strv_length(processed_args) >= 2) && pcmk__str_eq(processed_args[1], "metadata", pcmk__str_none)) { - crmd_metadata(); + controld_metadata(); initialize = false; goto done; } pcmk__cli_init_logging("pacemaker-controld", args->verbosity); crm_log_init(NULL, LOG_INFO, TRUE, FALSE, argc, argv, FALSE); crm_notice("Starting Pacemaker controller"); old_instance = crm_ipc_new(CRM_SYSTEM_CRMD, 0); if (old_instance == NULL) { /* crm_ipc_new will have already printed an error message with crm_err. */ exit_code = CRM_EX_FATAL; goto done; } if (pcmk__connect_generic_ipc(old_instance) == pcmk_rc_ok) { /* IPC end-point already up */ crm_ipc_close(old_instance); crm_ipc_destroy(old_instance); crm_err("pacemaker-controld is already active, aborting startup"); initialize = false; goto done; } else { /* not up or not authentic, we'll proceed either way */ crm_ipc_destroy(old_instance); old_instance = NULL; } if (pcmk__daemon_can_write(PE_STATE_DIR, NULL) == FALSE) { exit_code = CRM_EX_FATAL; crm_err("Terminating due to bad permissions on " PE_STATE_DIR); g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Bad permissions on " PE_STATE_DIR " (see logs for details)"); goto done; } else if (pcmk__daemon_can_write(CRM_CONFIG_DIR, NULL) == FALSE) { exit_code = CRM_EX_FATAL; crm_err("Terminating due to bad permissions on " CRM_CONFIG_DIR); g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Bad permissions on " CRM_CONFIG_DIR " (see logs for details)"); goto done; } if (pcmk__log_output_new(&(controld_globals.logger_out)) != pcmk_rc_ok) { exit_code = CRM_EX_FATAL; goto done; } pcmk__output_set_log_level(controld_globals.logger_out, LOG_TRACE); done: g_strfreev(processed_args); pcmk__free_arg_context(context); pcmk__output_and_clear_error(&error, out); if (out != NULL) { out->finish(out, exit_code, true, NULL); pcmk__output_free(out); } pcmk__unregister_formats(); if ((exit_code == CRM_EX_OK) && initialize) { // Does not return crmd_init(); } crm_exit(exit_code); } void crmd_init(void) { crm_exit_t exit_code = CRM_EX_OK; enum crmd_fsa_state state; init_dotfile(); register_fsa_input(C_STARTUP, I_STARTUP, NULL); crm_peer_init(); state = s_crmd_fsa(C_STARTUP); if (state == S_PENDING || state == S_STARTING) { /* Create the mainloop and run it... */ crm_trace("Starting %s's mainloop", crm_system_name); controld_globals.mainloop = g_main_loop_new(NULL, FALSE); g_main_loop_run(controld_globals.mainloop); if (pcmk_is_set(controld_globals.fsa_input_register, R_STAYDOWN)) { crm_info("Inhibiting automated respawn"); exit_code = CRM_EX_FATAL; } } else { crm_err("Startup of %s failed. Current state: %s", crm_system_name, fsa_state2string(state)); exit_code = CRM_EX_ERROR; } crm_info("%s[%lu] exiting with status %d (%s)", crm_system_name, (unsigned long) getpid(), exit_code, crm_exit_str(exit_code)); crmd_fast_exit(exit_code); } diff --git a/daemons/controld/pacemaker-controld.h b/daemons/controld/pacemaker-controld.h index 2334cce321..ba8dc8f5b0 100644 --- a/daemons/controld/pacemaker-controld.h +++ b/daemons/controld/pacemaker-controld.h @@ -1,42 +1,41 @@ /* - * Copyright 2004-2023 the Pacemaker project contributors + * 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. */ #ifndef CRMD__H # define CRMD__H #include #include #include #include #include #include #include #include #include #include #include #include #include #include # define controld_trigger_config() \ controld_trigger_config_as(__func__, __LINE__) -void crmd_metadata(void); void controld_trigger_config_as(const char *fn, int line); void controld_election_init(const char *uname); void controld_configure_election(GHashTable *options); void controld_remove_voter(const char *uname); void controld_election_fini(void); void controld_stop_current_election_timeout(void); void set_join_state(const char *start_state, const char *node_name, const char *node_uuid, bool remote); #endif diff --git a/daemons/fenced/pacemaker-fenced.c b/daemons/fenced/pacemaker-fenced.c index 6cf51b67ba..ce9dca8554 100644 --- a/daemons/fenced/pacemaker-fenced.c +++ b/daemons/fenced/pacemaker-fenced.c @@ -1,973 +1,1002 @@ /* * 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 // PRIu32, PRIx32 #include #include #include #include #include #include #include #include #include #include #include #include #include #define SUMMARY "daemon for executing fencing devices in a Pacemaker cluster" char *stonith_our_uname = NULL; long stonith_watchdog_timeout_ms = 0; GList *stonith_watchdog_targets = NULL; static GMainLoop *mainloop = NULL; gboolean stand_alone = FALSE; gboolean stonith_shutdown_flag = FALSE; static qb_ipcs_service_t *ipcs = NULL; static pcmk__output_t *out = NULL; pcmk__supported_format_t formats[] = { PCMK__SUPPORTED_FORMAT_NONE, PCMK__SUPPORTED_FORMAT_TEXT, PCMK__SUPPORTED_FORMAT_XML, { NULL, NULL, NULL } }; static struct { bool no_cib_connect; gchar **log_files; } options; crm_exit_t exit_code = CRM_EX_OK; static void stonith_cleanup(void); static int32_t st_ipc_accept(qb_ipcs_connection_t * c, uid_t uid, gid_t gid) { if (stonith_shutdown_flag) { crm_info("Ignoring new client [%d] during shutdown", pcmk__client_pid(c)); return -EPERM; } if (pcmk__new_client(c, uid, gid) == NULL) { return -EIO; } return 0; } /* Exit code means? */ static int32_t st_ipc_dispatch(qb_ipcs_connection_t * qbc, void *data, size_t size) { uint32_t id = 0; uint32_t flags = 0; int call_options = 0; xmlNode *request = NULL; pcmk__client_t *c = pcmk__find_client(qbc); const char *op = NULL; if (c == NULL) { crm_info("Invalid client: %p", qbc); return 0; } request = pcmk__client_data2xml(c, data, &id, &flags); if (request == NULL) { pcmk__ipc_send_ack(c, id, flags, "nack", NULL, CRM_EX_PROTOCOL); return 0; } op = crm_element_value(request, PCMK__XA_CRM_TASK); if(pcmk__str_eq(op, CRM_OP_RM_NODE_CACHE, pcmk__str_casei)) { crm_xml_add(request, PCMK__XA_T, T_STONITH_NG); crm_xml_add(request, F_STONITH_OPERATION, op); crm_xml_add(request, F_STONITH_CLIENTID, c->id); crm_xml_add(request, F_STONITH_CLIENTNAME, pcmk__client_name(c)); crm_xml_add(request, F_STONITH_CLIENTNODE, stonith_our_uname); send_cluster_message(NULL, crm_msg_stonith_ng, request, FALSE); free_xml(request); return 0; } if (c->name == NULL) { const char *value = crm_element_value(request, F_STONITH_CLIENTNAME); if (value == NULL) { value = "unknown"; } c->name = crm_strdup_printf("%s.%u", value, c->pid); } crm_element_value_int(request, F_STONITH_CALLOPTS, &call_options); crm_trace("Flags %#08" PRIx32 "/%#08x for command %" PRIu32 " from client %s", flags, call_options, id, pcmk__client_name(c)); if (pcmk_is_set(call_options, st_opt_sync_call)) { CRM_ASSERT(flags & crm_ipc_client_response); CRM_LOG_ASSERT(c->request_id == 0); /* This means the client has two synchronous events in-flight */ c->request_id = id; /* Reply only to the last one */ } crm_xml_add(request, F_STONITH_CLIENTID, c->id); crm_xml_add(request, F_STONITH_CLIENTNAME, pcmk__client_name(c)); crm_xml_add(request, F_STONITH_CLIENTNODE, stonith_our_uname); crm_log_xml_trace(request, "ipc-received"); stonith_command(c, id, flags, request, NULL); free_xml(request); return 0; } /* Error code means? */ static int32_t st_ipc_closed(qb_ipcs_connection_t * c) { pcmk__client_t *client = pcmk__find_client(c); if (client == NULL) { return 0; } crm_trace("Connection %p closed", c); pcmk__free_client(client); /* 0 means: yes, go ahead and destroy the connection */ return 0; } static void st_ipc_destroy(qb_ipcs_connection_t * c) { crm_trace("Connection %p destroyed", c); st_ipc_closed(c); } static void stonith_peer_callback(xmlNode * msg, void *private_data) { const char *remote_peer = crm_element_value(msg, PCMK__XA_SRC); const char *op = crm_element_value(msg, F_STONITH_OPERATION); if (pcmk__str_eq(op, "poke", pcmk__str_none)) { return; } crm_log_xml_trace(msg, "Peer[inbound]"); stonith_command(NULL, 0, 0, msg, remote_peer); } #if SUPPORT_COROSYNC static void stonith_peer_ais_callback(cpg_handle_t handle, const struct cpg_name *groupName, uint32_t nodeid, uint32_t pid, void *msg, size_t msg_len) { uint32_t kind = 0; xmlNode *xml = NULL; const char *from = NULL; char *data = pcmk_message_common_cs(handle, nodeid, pid, msg, &kind, &from); if(data == NULL) { return; } if (kind == crm_class_cluster) { xml = string2xml(data); if (xml == NULL) { crm_err("Invalid XML: '%.120s'", data); free(data); return; } crm_xml_add(xml, PCMK__XA_SRC, from); stonith_peer_callback(xml, NULL); } free_xml(xml); free(data); return; } static void stonith_peer_cs_destroy(gpointer user_data) { crm_crit("Lost connection to cluster layer, shutting down"); stonith_shutdown(0); } #endif void do_local_reply(const xmlNode *notify_src, pcmk__client_t *client, int call_options) { /* send callback to originating child */ int local_rc = pcmk_rc_ok; int rid = 0; uint32_t ipc_flags = crm_ipc_server_event; if (pcmk_is_set(call_options, st_opt_sync_call)) { CRM_LOG_ASSERT(client->request_id); rid = client->request_id; client->request_id = 0; ipc_flags = crm_ipc_flags_none; } local_rc = pcmk__ipc_send_xml(client, rid, notify_src, ipc_flags); if (local_rc == pcmk_rc_ok) { crm_trace("Sent response %d to client %s", rid, pcmk__client_name(client)); } else { crm_warn("%synchronous reply to client %s failed: %s", (pcmk_is_set(call_options, st_opt_sync_call)? "S" : "As"), pcmk__client_name(client), pcmk_rc_str(local_rc)); } } uint64_t get_stonith_flag(const char *name) { if (pcmk__str_eq(name, T_STONITH_NOTIFY_FENCE, pcmk__str_casei)) { return st_callback_notify_fence; } else if (pcmk__str_eq(name, STONITH_OP_DEVICE_ADD, pcmk__str_casei)) { return st_callback_device_add; } else if (pcmk__str_eq(name, STONITH_OP_DEVICE_DEL, pcmk__str_casei)) { return st_callback_device_del; } else if (pcmk__str_eq(name, T_STONITH_NOTIFY_HISTORY, pcmk__str_casei)) { return st_callback_notify_history; } else if (pcmk__str_eq(name, T_STONITH_NOTIFY_HISTORY_SYNCED, pcmk__str_casei)) { return st_callback_notify_history_synced; } return st_callback_unknown; } static void stonith_notify_client(gpointer key, gpointer value, gpointer user_data) { const xmlNode *update_msg = user_data; pcmk__client_t *client = value; const char *type = NULL; CRM_CHECK(client != NULL, return); CRM_CHECK(update_msg != NULL, return); type = crm_element_value(update_msg, PCMK__XA_SUBT); CRM_CHECK(type != NULL, crm_log_xml_err(update_msg, "notify"); return); if (client->ipcs == NULL) { crm_trace("Skipping client with NULL channel"); return; } if (pcmk_is_set(client->flags, get_stonith_flag(type))) { int rc = pcmk__ipc_send_xml(client, 0, update_msg, crm_ipc_server_event); if (rc != pcmk_rc_ok) { crm_warn("%s notification of client %s failed: %s " CRM_XS " id=%.8s rc=%d", type, pcmk__client_name(client), pcmk_rc_str(rc), client->id, rc); } else { crm_trace("Sent %s notification to client %s", type, pcmk__client_name(client)); } } } void do_stonith_async_timeout_update(const char *client_id, const char *call_id, int timeout) { pcmk__client_t *client = NULL; xmlNode *notify_data = NULL; if (!timeout || !call_id || !client_id) { return; } client = pcmk__find_client_by_id(client_id); if (!client) { return; } notify_data = create_xml_node(NULL, T_STONITH_TIMEOUT_VALUE); crm_xml_add(notify_data, PCMK__XA_T, T_STONITH_TIMEOUT_VALUE); crm_xml_add(notify_data, F_STONITH_CALLID, call_id); crm_xml_add_int(notify_data, F_STONITH_TIMEOUT, timeout); crm_trace("timeout update is %d for client %s and call id %s", timeout, client_id, call_id); if (client) { pcmk__ipc_send_xml(client, 0, notify_data, crm_ipc_server_event); } free_xml(notify_data); } /*! * \internal * \brief Notify relevant IPC clients of a fencing operation result * * \param[in] type Notification type * \param[in] result Result of fencing operation (assume success if NULL) * \param[in] data If not NULL, add to notification as call data */ void fenced_send_notification(const char *type, const pcmk__action_result_t *result, xmlNode *data) { /* TODO: Standardize the contents of data */ xmlNode *update_msg = create_xml_node(NULL, "notify"); CRM_LOG_ASSERT(type != NULL); crm_xml_add(update_msg, PCMK__XA_T, T_STONITH_NOTIFY); crm_xml_add(update_msg, PCMK__XA_SUBT, type); crm_xml_add(update_msg, F_STONITH_OPERATION, type); stonith__xe_set_result(update_msg, result); if (data != NULL) { add_message_xml(update_msg, F_STONITH_CALLDATA, data); } crm_trace("Notifying clients"); pcmk__foreach_ipc_client(stonith_notify_client, update_msg); free_xml(update_msg); crm_trace("Notify complete"); } /*! * \internal * \brief Send notifications for a configuration change to subscribed clients * * \param[in] op Notification type (STONITH_OP_DEVICE_ADD, * STONITH_OP_DEVICE_DEL, STONITH_OP_LEVEL_ADD, or * STONITH_OP_LEVEL_DEL) * \param[in] result Operation result * \param[in] desc Description of what changed * \param[in] active Current number of devices or topologies in use */ static void send_config_notification(const char *op, const pcmk__action_result_t *result, const char *desc, int active) { xmlNode *notify_data = create_xml_node(NULL, op); CRM_CHECK(notify_data != NULL, return); crm_xml_add(notify_data, F_STONITH_DEVICE, desc); crm_xml_add_int(notify_data, F_STONITH_ACTIVE, active); fenced_send_notification(op, result, notify_data); free_xml(notify_data); } /*! * \internal * \brief Send notifications for a device change to subscribed clients * * \param[in] op Notification type (STONITH_OP_DEVICE_ADD or * STONITH_OP_DEVICE_DEL) * \param[in] result Operation result * \param[in] desc ID of device that changed */ void fenced_send_device_notification(const char *op, const pcmk__action_result_t *result, const char *desc) { send_config_notification(op, result, desc, g_hash_table_size(device_list)); } /*! * \internal * \brief Send notifications for a topology level change to subscribed clients * * \param[in] op Notification type (STONITH_OP_LEVEL_ADD or * STONITH_OP_LEVEL_DEL) * \param[in] result Operation result * \param[in] desc String representation of level ([]) */ void fenced_send_level_notification(const char *op, const pcmk__action_result_t *result, const char *desc) { send_config_notification(op, result, desc, g_hash_table_size(topology)); } /*! * \internal * \brief Check whether a node does watchdog-fencing * * \param[in] node Name of node to check * * \return TRUE if node found in stonith_watchdog_targets * or stonith_watchdog_targets is empty indicating * all nodes are doing watchdog-fencing */ gboolean node_does_watchdog_fencing(const char *node) { return ((stonith_watchdog_targets == NULL) || pcmk__str_in_list(node, stonith_watchdog_targets, pcmk__str_casei)); } void stonith_shutdown(int nsig) { crm_info("Terminating with %d clients", pcmk__ipc_client_count()); stonith_shutdown_flag = TRUE; if (mainloop != NULL && g_main_loop_is_running(mainloop)) { g_main_loop_quit(mainloop); } } static void stonith_cleanup(void) { fenced_cib_cleanup(); if (ipcs) { qb_ipcs_destroy(ipcs); } crm_peer_destroy(); pcmk__client_cleanup(); free_stonith_remote_op_list(); free_topology_list(); free_device_list(); free_metadata_cache(); fenced_unregister_handlers(); free(stonith_our_uname); stonith_our_uname = NULL; } static gboolean stand_alone_cpg_cb(const gchar *option_name, const gchar *optarg, gpointer data, GError **error) { stand_alone = FALSE; options.no_cib_connect = true; return TRUE; } struct qb_ipcs_service_handlers ipc_callbacks = { .connection_accept = st_ipc_accept, .connection_created = NULL, .msg_process = st_ipc_dispatch, .connection_closed = st_ipc_closed, .connection_destroyed = st_ipc_destroy }; /*! * \internal * \brief Callback for peer status changes * * \param[in] type What changed * \param[in] node What peer had the change * \param[in] data Previous value of what changed */ static void st_peer_update_callback(enum crm_status_type type, crm_node_t * node, const void *data) { if ((type != crm_status_processes) && !pcmk_is_set(node->flags, crm_remote_node)) { /* * This is a hack until we can send to a nodeid and/or we fix node name lookups * These messages are ignored in stonith_peer_callback() */ xmlNode *query = create_xml_node(NULL, "stonith_command"); crm_xml_add(query, PCMK__XA_T, T_STONITH_NG); crm_xml_add(query, F_STONITH_OPERATION, "poke"); crm_debug("Broadcasting our uname because of node %u", node->id); send_cluster_message(NULL, crm_msg_stonith_ng, query, FALSE); free_xml(query); } } static pcmk__cluster_option_t fencer_options[] = { /* name, old name, type, allowed values, * default value, validator, + * context, * short description, * long description */ { PCMK_STONITH_HOST_ARGUMENT, NULL, "string", NULL, "port", NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "An alternate parameter to supply instead of 'port'"), N_("Some devices do not support the standard 'port' parameter or may " "provide additional ones. Use this to specify an alternate, device-" "specific, parameter that should indicate the machine to be " "fenced. A value of \"none\" can be used to tell the cluster not " "to supply any additional parameters."), }, { PCMK_STONITH_HOST_MAP, NULL, "string", NULL, "", NULL, + pcmk__opt_context_none, N_("A mapping of node names to port numbers for devices that do not " "support node names."), N_("For example, \"node1:1;node2:2,3\" would tell the cluster to use " "port 1 for node1 and ports 2 and 3 for node2."), }, { PCMK_STONITH_HOST_LIST, NULL, "string", NULL, "", NULL, + pcmk__opt_context_none, N_("A list of nodes that can be targeted by this device (optional " "unless pcmk_host_list=\"static-list\")"), N_("For example, \"node1,node2,node3\"."), }, { PCMK_STONITH_HOST_CHECK, NULL, "select", "dynamic-list, static-list, status, none", "dynamic-list", NULL, + pcmk__opt_context_none, N_("How to determine which nodes can be targeted by the device"), N_("Use \"dynamic-list\" to query the device via the 'list' command; " "\"static-list\" to check the pcmk_host_list attribute; " "\"status\" to query the device via the 'status' command; or " "\"none\" to assume every device can fence every node."), }, { PCMK_STONITH_DELAY_MAX, NULL, "time", NULL, "0s", NULL, + pcmk__opt_context_none, N_("Enable a base delay for fencing actions and specify base delay " "value."), N_("Enable a delay of no more than the time specified before executing " "fencing actions. Pacemaker derives the overall delay by taking " "the value of pcmk_delay_base and adding a random delay value such " "that the sum is kept below this maximum."), }, { PCMK_STONITH_DELAY_BASE, NULL, "string", NULL, "0s", NULL, + pcmk__opt_context_none, N_("Enable a base delay for fencing actions and specify base delay " "value."), N_("This enables a static delay for fencing actions, which can help " "avoid \"death matches\" where two nodes try to fence each other " "at the same time. If pcmk_delay_max is also used, a random delay " "will be added such that the total delay is kept below that value. " "This can be set to a single time value to apply to any node " "targeted by this device (useful if a separate device is " "configured for each target), or to a node map (for example, " "\"node1:1s;node2:5\") to set a different value for each target."), }, { PCMK_STONITH_ACTION_LIMIT, NULL, "integer", NULL, "1", NULL, + pcmk__opt_context_none, N_("The maximum number of actions can be performed in parallel on this " "device"), N_("Cluster property concurrent-fencing=\"true\" needs to be " "configured first. Then use this to specify the maximum number of " "actions can be performed in parallel on this device. A value of " "-1 means an unlimited number of actions can be performed in " "parallel."), }, { "pcmk_reboot_action", NULL, "string", NULL, PCMK_ACTION_REBOOT, NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "An alternate command to run instead of 'reboot'"), N_("Some devices do not support the standard commands or may provide " "additional ones. Use this to specify an alternate, device-" "specific, command that implements the 'reboot' action."), }, { "pcmk_reboot_timeout", NULL, "time", NULL, "60s", NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "Specify an alternate timeout to use for 'reboot' actions instead " "of stonith-timeout"), N_("Some devices need much more/less time to complete than normal. " "Use this to specify an alternate, device-specific, timeout for " "'reboot' actions."), }, { "pcmk_reboot_retries", NULL, "integer", NULL, "2", NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "The maximum number of times to try the 'reboot' command within " "the timeout period"), N_("Some devices do not support multiple connections. Operations may " "\"fail\" if the device is busy with another task. In that case, " "Pacemaker will automatically retry the operation if there is time " "remaining. Use this option to alter the number of times Pacemaker " "tries a 'reboot' action before giving up."), }, { "pcmk_off_action", NULL, "string", NULL, PCMK_ACTION_OFF, NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "An alternate command to run instead of 'off'"), N_("Some devices do not support the standard commands or may provide " "additional ones. Use this to specify an alternate, device-" "specific, command that implements the 'off' action."), }, { "pcmk_off_timeout", NULL, "time", NULL, "60s", NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "Specify an alternate timeout to use for 'off' actions instead of " "stonith-timeout"), N_("Some devices need much more/less time to complete than normal. " "Use this to specify an alternate, device-specific, timeout for " "'off' actions."), }, { "pcmk_off_retries", NULL, "integer", NULL, "2", NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "The maximum number of times to try the 'off' command within the " "timeout period"), N_("Some devices do not support multiple connections. Operations may " "\"fail\" if the device is busy with another task. In that case, " "Pacemaker will automatically retry the operation if there is time " "remaining. Use this option to alter the number of times Pacemaker " "tries a 'off' action before giving up."), }, { "pcmk_on_action", NULL, "string", NULL, PCMK_ACTION_ON, NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "An alternate command to run instead of 'on'"), N_("Some devices do not support the standard commands or may provide " "additional ones. Use this to specify an alternate, device-" "specific, command that implements the 'on' action."), }, { "pcmk_on_timeout", NULL, "time", NULL, "60s", NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "Specify an alternate timeout to use for 'on' actions instead of " "stonith-timeout"), N_("Some devices need much more/less time to complete than normal. " "Use this to specify an alternate, device-specific, timeout for " "'on' actions."), }, { "pcmk_on_retries", NULL, "integer", NULL, "2", NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "The maximum number of times to try the 'on' command within the " "timeout period"), N_("Some devices do not support multiple connections. Operations may " "\"fail\" if the device is busy with another task. In that case, " "Pacemaker will automatically retry the operation if there is time " "remaining. Use this option to alter the number of times Pacemaker " "tries a 'on' action before giving up."), }, { "pcmk_list_action", NULL, "string", NULL, PCMK_ACTION_LIST, NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "An alternate command to run instead of 'list'"), N_("Some devices do not support the standard commands or may provide " "additional ones. Use this to specify an alternate, device-" "specific, command that implements the 'list' action."), }, { "pcmk_list_timeout", NULL, "time", NULL, "60s", NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "Specify an alternate timeout to use for 'list' actions instead of " "stonith-timeout"), N_("Some devices need much more/less time to complete than normal. " "Use this to specify an alternate, device-specific, timeout for " "'list' actions."), }, { "pcmk_list_retries", NULL, "integer", NULL, "2", NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "The maximum number of times to try the 'list' command within the " "timeout period"), N_("Some devices do not support multiple connections. Operations may " "\"fail\" if the device is busy with another task. In that case, " "Pacemaker will automatically retry the operation if there is time " "remaining. Use this option to alter the number of times Pacemaker " "tries a 'list' action before giving up."), }, { "pcmk_monitor_action", NULL, "string", NULL, PCMK_ACTION_MONITOR, NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "An alternate command to run instead of 'monitor'"), N_("Some devices do not support the standard commands or may provide " "additional ones. Use this to specify an alternate, device-" "specific, command that implements the 'monitor' action."), }, { "pcmk_monitor_timeout", NULL, "time", NULL, "60s", NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "Specify an alternate timeout to use for 'monitor' actions instead " "of stonith-timeout"), N_("Some devices need much more/less time to complete than normal. " "Use this to specify an alternate, device-specific, timeout for " "'monitor' actions."), }, { "pcmk_monitor_retries", NULL, "integer", NULL, "2", NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "The maximum number of times to try the 'monitor' command within " "the timeout period"), N_("Some devices do not support multiple connections. Operations may " "\"fail\" if the device is busy with another task. In that case, " "Pacemaker will automatically retry the operation if there is time " "remaining. Use this option to alter the number of times Pacemaker " "tries a 'monitor' action before giving up."), }, { "pcmk_status_action", NULL, "string", NULL, PCMK_ACTION_STATUS, NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "An alternate command to run instead of 'status'"), N_("Some devices do not support the standard commands or may provide " "additional ones. Use this to specify an alternate, device-" "specific, command that implements the 'status' action."), }, { "pcmk_status_timeout", NULL, "time", NULL, "60s", NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "Specify an alternate timeout to use for 'status' actions instead " "of stonith-timeout"), N_("Some devices need much more/less time to complete than normal. " "Use this to specify an alternate, device-specific, timeout for " "'status' actions."), }, { "pcmk_status_retries", NULL, "integer", NULL, "2", NULL, + pcmk__opt_context_none, N_("*** Advanced Use Only *** " "The maximum number of times to try the 'status' command within " "the timeout period"), N_("Some devices do not support multiple connections. Operations may " "\"fail\" if the device is busy with another task. In that case, " "Pacemaker will automatically retry the operation if there is time " "remaining. Use this option to alter the number of times Pacemaker " "tries a 'status' action before giving up."), }, }; void fencer_metadata(void) { + const char *name = "pacemaker-fenced"; const char *desc_short = N_("Instance attributes available for all " - "\"stonith\"-class resources"); - const char *desc_long = N_("Instance attributes available for all \"stonith\"-" - "class resources and used by Pacemaker's fence " - "daemon, formerly known as stonithd"); - - gchar *s = pcmk__format_option_metadata("pacemaker-fenced", desc_short, - desc_long, fencer_options, + "\"stonith\"-class resources"); + const char *desc_long = N_("Instance attributes available for all " + "\"stonith\"-class resources and used by " + "Pacemaker's fence daemon, formerly known as " + "stonithd"); + + gchar *s = pcmk__format_option_metadata(name, desc_short, desc_long, + pcmk__opt_context_none, + fencer_options, PCMK__NELEM(fencer_options)); printf("%s", s); g_free(s); } static GOptionEntry entries[] = { { "stand-alone", 's', G_OPTION_FLAG_NONE, G_OPTION_ARG_NONE, &stand_alone, N_("Deprecated (will be removed in a future release)"), NULL }, { "stand-alone-w-cpg", 'c', G_OPTION_FLAG_NO_ARG, G_OPTION_ARG_CALLBACK, stand_alone_cpg_cb, N_("Intended for use in regression testing only"), NULL }, { "logfile", 'l', G_OPTION_FLAG_NONE, G_OPTION_ARG_FILENAME_ARRAY, &options.log_files, N_("Send logs to the additional named logfile"), NULL }, { NULL } }; static GOptionContext * build_arg_context(pcmk__common_args_t *args, GOptionGroup **group) { GOptionContext *context = NULL; context = pcmk__build_arg_context(args, "text (default), xml", group, "[metadata]"); pcmk__add_main_args(context, entries); return context; } int main(int argc, char **argv) { int rc = pcmk_rc_ok; crm_cluster_t *cluster = NULL; crm_ipc_t *old_instance = NULL; GError *error = NULL; GOptionGroup *output_group = NULL; pcmk__common_args_t *args = pcmk__new_common_args(SUMMARY); gchar **processed_args = pcmk__cmdline_preproc(argv, "l"); GOptionContext *context = build_arg_context(args, &output_group); crm_log_preinit(NULL, argc, argv); pcmk__register_formats(output_group, formats); if (!g_option_context_parse_strv(context, &processed_args, &error)) { exit_code = CRM_EX_USAGE; goto done; } rc = pcmk__output_new(&out, args->output_ty, args->output_dest, argv); if (rc != pcmk_rc_ok) { exit_code = CRM_EX_ERROR; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Error creating output format %s: %s", args->output_ty, pcmk_rc_str(rc)); goto done; } if (args->version) { out->version(out, false); goto done; } if ((g_strv_length(processed_args) >= 2) && pcmk__str_eq(processed_args[1], "metadata", pcmk__str_none)) { fencer_metadata(); goto done; } // Open additional log files pcmk__add_logfiles(options.log_files, out); crm_log_init(NULL, LOG_INFO + args->verbosity, TRUE, (args->verbosity > 0), argc, argv, FALSE); crm_notice("Starting Pacemaker fencer"); old_instance = crm_ipc_new("stonith-ng", 0); if (old_instance == NULL) { /* crm_ipc_new() will have already logged an error message with * crm_err() */ exit_code = CRM_EX_FATAL; goto done; } if (pcmk__connect_generic_ipc(old_instance) == pcmk_rc_ok) { // IPC endpoint already up crm_ipc_close(old_instance); crm_ipc_destroy(old_instance); crm_err("pacemaker-fenced is already active, aborting startup"); goto done; } else { // Not up or not authentic, we'll proceed either way crm_ipc_destroy(old_instance); old_instance = NULL; } mainloop_add_signal(SIGTERM, stonith_shutdown); crm_peer_init(); rc = fenced_scheduler_init(); if (rc != pcmk_rc_ok) { exit_code = CRM_EX_FATAL; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Error initializing scheduler data: %s", pcmk_rc_str(rc)); goto done; } cluster = pcmk_cluster_new(); if (!stand_alone) { #if SUPPORT_COROSYNC if (is_corosync_cluster()) { cluster->destroy = stonith_peer_cs_destroy; cluster->cpg.cpg_deliver_fn = stonith_peer_ais_callback; cluster->cpg.cpg_confchg_fn = pcmk_cpg_membership; } #endif // SUPPORT_COROSYNC crm_set_status_callback(&st_peer_update_callback); if (crm_cluster_connect(cluster) == FALSE) { exit_code = CRM_EX_FATAL; crm_crit("Cannot sign in to the cluster... terminating"); goto done; } pcmk__str_update(&stonith_our_uname, cluster->uname); if (!options.no_cib_connect) { setup_cib(); } } else { pcmk__str_update(&stonith_our_uname, "localhost"); crm_warn("Stand-alone mode is deprecated and will be removed " "in a future release"); } init_device_list(); init_topology_list(); pcmk__serve_fenced_ipc(&ipcs, &ipc_callbacks); // Create the mainloop and run it... mainloop = g_main_loop_new(NULL, FALSE); crm_notice("Pacemaker fencer successfully started and accepting connections"); g_main_loop_run(mainloop); done: g_strfreev(processed_args); pcmk__free_arg_context(context); g_strfreev(options.log_files); stonith_cleanup(); pcmk_cluster_free(cluster); fenced_scheduler_cleanup(); pcmk__output_and_clear_error(&error, out); if (out != NULL) { out->finish(out, exit_code, true, NULL); pcmk__output_free(out); } pcmk__unregister_formats(); crm_exit(exit_code); } diff --git a/daemons/schedulerd/pacemaker-schedulerd.c b/daemons/schedulerd/pacemaker-schedulerd.c index 3f2a3e861c..27c96da81e 100644 --- a/daemons/schedulerd/pacemaker-schedulerd.c +++ b/daemons/schedulerd/pacemaker-schedulerd.c @@ -1,181 +1,195 @@ /* - * Copyright 2004-2023 the Pacemaker project contributors + * 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 #include #include #include #include "pacemaker-schedulerd.h" #define SUMMARY "pacemaker-schedulerd - daemon for calculating a Pacemaker cluster's response to events" struct { gchar **remainder; } options; pcmk__output_t *logger_out = NULL; pcmk__output_t *out = NULL; static GMainLoop *mainloop = NULL; static qb_ipcs_service_t *ipcs = NULL; static crm_exit_t exit_code = CRM_EX_OK; pcmk__supported_format_t formats[] = { PCMK__SUPPORTED_FORMAT_NONE, PCMK__SUPPORTED_FORMAT_TEXT, PCMK__SUPPORTED_FORMAT_XML, { NULL, NULL, NULL } }; void pengine_shutdown(int nsig); +static void +scheduler_metadata(pcmk__output_t *out) +{ + const char *name = "pacemaker-schedulerd"; + const char *desc_short = "Pacemaker scheduler options"; + const char *desc_long = "Cluster options used by Pacemaker's scheduler"; + + gchar *s = pcmk__cluster_option_metadata(name, desc_short, desc_long, + pcmk__opt_context_schedulerd); + + out->output_xml(out, "metadata", s); + g_free(s); +} + static GOptionContext * build_arg_context(pcmk__common_args_t *args, GOptionGroup **group) { GOptionContext *context = NULL; GOptionEntry extra_prog_entries[] = { { G_OPTION_REMAINING, 0, G_OPTION_FLAG_NONE, G_OPTION_ARG_STRING_ARRAY, &options.remainder, NULL, NULL }, { NULL } }; context = pcmk__build_arg_context(args, "text (default), xml", group, "[metadata]"); pcmk__add_main_args(context, extra_prog_entries); return context; } int main(int argc, char **argv) { GError *error = NULL; int rc = pcmk_rc_ok; GOptionGroup *output_group = NULL; pcmk__common_args_t *args = pcmk__new_common_args(SUMMARY); gchar **processed_args = pcmk__cmdline_preproc(argv, NULL); GOptionContext *context = build_arg_context(args, &output_group); crm_log_preinit(NULL, argc, argv); mainloop_add_signal(SIGTERM, pengine_shutdown); pcmk__register_formats(output_group, formats); if (!g_option_context_parse_strv(context, &processed_args, &error)) { exit_code = CRM_EX_USAGE; goto done; } rc = pcmk__output_new(&out, args->output_ty, args->output_dest, argv); if ((rc != pcmk_rc_ok) || (out == NULL)) { exit_code = CRM_EX_FATAL; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Error creating output format %s: %s", args->output_ty, pcmk_rc_str(rc)); goto done; } pe__register_messages(out); pcmk__register_lib_messages(out); if (options.remainder) { if (g_strv_length(options.remainder) == 1 && pcmk__str_eq("metadata", options.remainder[0], pcmk__str_casei)) { - pe_metadata(out); + scheduler_metadata(out); goto done; } else { exit_code = CRM_EX_USAGE; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Unsupported extra command line parameters"); goto done; } } if (args->version) { out->version(out, false); goto done; } pcmk__cli_init_logging("pacemaker-schedulerd", args->verbosity); crm_log_init(NULL, LOG_INFO, TRUE, FALSE, argc, argv, FALSE); crm_notice("Starting Pacemaker scheduler"); if (pcmk__daemon_can_write(PE_STATE_DIR, NULL) == FALSE) { crm_err("Terminating due to bad permissions on " PE_STATE_DIR); exit_code = CRM_EX_FATAL; g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "ERROR: Bad permissions on %s (see logs for details)", PE_STATE_DIR); goto done; } ipcs = pcmk__serve_schedulerd_ipc(&ipc_callbacks); if (ipcs == NULL) { g_set_error(&error, PCMK__EXITC_ERROR, exit_code, "Failed to create pacemaker-schedulerd server: exiting and inhibiting respawn"); exit_code = CRM_EX_FATAL; goto done; } if (pcmk__log_output_new(&logger_out) != pcmk_rc_ok) { exit_code = CRM_EX_FATAL; goto done; } pe__register_messages(logger_out); pcmk__register_lib_messages(logger_out); pcmk__output_set_log_level(logger_out, LOG_TRACE); /* Create the mainloop and run it... */ mainloop = g_main_loop_new(NULL, FALSE); crm_notice("Pacemaker scheduler successfully started and accepting connections"); g_main_loop_run(mainloop); done: g_strfreev(options.remainder); g_strfreev(processed_args); pcmk__free_arg_context(context); pcmk__output_and_clear_error(&error, out); pengine_shutdown(0); } void pengine_shutdown(int nsig) { if (ipcs != NULL) { crm_trace("Closing IPC server"); mainloop_del_ipc_server(ipcs); ipcs = NULL; } if (logger_out != NULL) { logger_out->finish(logger_out, exit_code, true, NULL); pcmk__output_free(logger_out); logger_out = NULL; } if (out != NULL) { out->finish(out, exit_code, true, NULL); pcmk__output_free(out); out = NULL; } pcmk__unregister_formats(); crm_exit(exit_code); } diff --git a/daemons/schedulerd/schedulerd_messages.c b/daemons/schedulerd/schedulerd_messages.c index a2a9cd2b43..b598af2984 100644 --- a/daemons/schedulerd/schedulerd_messages.c +++ b/daemons/schedulerd/schedulerd_messages.c @@ -1,335 +1,336 @@ /* * 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 #include "pacemaker-schedulerd.h" static GHashTable *schedulerd_handlers = NULL; static pcmk_scheduler_t * init_working_set(void) { pcmk_scheduler_t *scheduler = pe_new_working_set(); CRM_ASSERT(scheduler != NULL); crm_config_error = FALSE; crm_config_warning = FALSE; was_processing_error = FALSE; was_processing_warning = FALSE; scheduler->priv = logger_out; return scheduler; } static xmlNode * handle_pecalc_request(pcmk__request_t *request) { static struct series_s { const char *name; const char *param; /* Maximum number of inputs of this kind to save to disk. * If -1, save all; if 0, save none. */ int wrap; } series[] = { { "pe-error", PCMK_OPT_PE_ERROR_SERIES_MAX, -1 }, { "pe-warn", PCMK_OPT_PE_WARN_SERIES_MAX, 5000 }, { "pe-input", PCMK_OPT_PE_INPUT_SERIES_MAX, 4000 }, }; xmlNode *msg = request->xml; xmlNode *xml_data = get_message_xml(msg, F_CRM_DATA); static char *last_digest = NULL; static char *filename = NULL; unsigned int seq; int series_id = 0; int series_wrap = 0; char *digest = NULL; const char *value = NULL; time_t execution_date = time(NULL); xmlNode *converted = NULL; xmlNode *reply = NULL; bool is_repoke = false; bool process = true; pcmk_scheduler_t *scheduler = init_working_set(); pcmk__ipc_send_ack(request->ipc_client, request->ipc_id, request->ipc_flags, "ack", NULL, CRM_EX_INDETERMINATE); digest = calculate_xml_versioned_digest(xml_data, FALSE, FALSE, CRM_FEATURE_SET); converted = copy_xml(xml_data); if (!cli_config_update(&converted, NULL, TRUE)) { scheduler->graph = create_xml_node(NULL, XML_TAG_GRAPH); crm_xml_add_int(scheduler->graph, "transition_id", 0); crm_xml_add_int(scheduler->graph, PCMK_OPT_CLUSTER_DELAY, 0); process = false; free(digest); } else if (pcmk__str_eq(digest, last_digest, pcmk__str_casei)) { is_repoke = true; free(digest); } else { free(last_digest); last_digest = digest; } if (process) { pcmk__schedule_actions(converted, pcmk_sched_no_counts |pcmk_sched_no_compat |pcmk_sched_show_utilization, scheduler); } // Get appropriate index into series[] array if (was_processing_error || crm_config_error) { series_id = 0; } else if (was_processing_warning || crm_config_warning) { series_id = 1; } else { series_id = 2; } - value = pe_pref(scheduler->config_hash, series[series_id].param); + value = pcmk__cluster_option(scheduler->config_hash, + series[series_id].param); if ((value == NULL) || (pcmk__scan_min_int(value, &series_wrap, -1) != pcmk_rc_ok)) { series_wrap = series[series_id].wrap; } if (pcmk__read_series_sequence(PE_STATE_DIR, series[series_id].name, &seq) != pcmk_rc_ok) { // @TODO maybe handle errors better ... seq = 0; } crm_trace("Series %s: wrap=%d, seq=%u, pref=%s", series[series_id].name, series_wrap, seq, value); scheduler->input = NULL; reply = create_reply(msg, scheduler->graph); if (reply == NULL) { pcmk__format_result(&request->result, CRM_EX_ERROR, PCMK_EXEC_ERROR, "Failed building ping reply for client %s", pcmk__client_name(request->ipc_client)); goto done; } if (series_wrap == 0) { // Don't save any inputs of this kind free(filename); filename = NULL; } else if (!is_repoke) { // Input changed, save to disk free(filename); filename = pcmk__series_filename(PE_STATE_DIR, series[series_id].name, seq, true); } crm_xml_add(reply, PCMK__XA_CRM_TGRAPH_IN, filename); crm_xml_add_int(reply, PCMK__XA_GRAPH_ERRORS, was_processing_error); crm_xml_add_int(reply, PCMK__XA_GRAPH_WARNINGS, was_processing_warning); crm_xml_add_int(reply, PCMK__XA_CONFIG_ERRORS, crm_config_error); crm_xml_add_int(reply, PCMK__XA_CONFIG_WARNINGS, crm_config_warning); pcmk__log_transition_summary(filename); if (series_wrap == 0) { crm_debug("Not saving input to disk (disabled by configuration)"); } else if (is_repoke) { crm_info("Input has not changed since last time, not saving to disk"); } else { unlink(filename); crm_xml_add_ll(xml_data, "execution-date", (long long) execution_date); write_xml_file(xml_data, filename, TRUE); pcmk__write_series_sequence(PE_STATE_DIR, series[series_id].name, ++seq, series_wrap); } pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); done: free_xml(converted); pe_free_working_set(scheduler); return reply; } static xmlNode * handle_unknown_request(pcmk__request_t *request) { pcmk__ipc_send_ack(request->ipc_client, request->ipc_id, request->ipc_flags, "ack", NULL, CRM_EX_INVALID_PARAM); pcmk__format_result(&request->result, CRM_EX_PROTOCOL, PCMK_EXEC_INVALID, "Unknown IPC request type '%s' (bug?)", pcmk__client_name(request->ipc_client)); return NULL; } static xmlNode * handle_hello_request(pcmk__request_t *request) { pcmk__ipc_send_ack(request->ipc_client, request->ipc_id, request->ipc_flags, "ack", NULL, CRM_EX_INDETERMINATE); crm_trace("Received IPC hello from %s", pcmk__client_name(request->ipc_client)); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); return NULL; } static void schedulerd_register_handlers(void) { pcmk__server_command_t handlers[] = { { CRM_OP_HELLO, handle_hello_request }, { CRM_OP_PECALC, handle_pecalc_request }, { NULL, handle_unknown_request }, }; schedulerd_handlers = pcmk__register_handlers(handlers); } static int32_t pe_ipc_accept(qb_ipcs_connection_t * c, uid_t uid, gid_t gid) { crm_trace("Connection %p", c); if (pcmk__new_client(c, uid, gid) == NULL) { return -EIO; } return 0; } static int32_t pe_ipc_dispatch(qb_ipcs_connection_t * qbc, void *data, size_t size) { uint32_t id = 0; uint32_t flags = 0; xmlNode *msg = NULL; pcmk__client_t *c = pcmk__find_client(qbc); const char *sys_to = NULL; CRM_CHECK(c != NULL, return 0); if (schedulerd_handlers == NULL) { schedulerd_register_handlers(); } msg = pcmk__client_data2xml(c, data, &id, &flags); if (msg == NULL) { pcmk__ipc_send_ack(c, id, flags, "ack", NULL, CRM_EX_PROTOCOL); return 0; } sys_to = crm_element_value(msg, PCMK__XA_CRM_SYS_TO); if (pcmk__str_eq(crm_element_value(msg, PCMK__XA_SUBT), PCMK__VALUE_RESPONSE, pcmk__str_none)) { pcmk__ipc_send_ack(c, id, flags, "ack", NULL, CRM_EX_INDETERMINATE); crm_info("Ignoring IPC reply from %s", pcmk__client_name(c)); } else if (!pcmk__str_eq(sys_to, CRM_SYSTEM_PENGINE, pcmk__str_none)) { pcmk__ipc_send_ack(c, id, flags, "ack", NULL, CRM_EX_INDETERMINATE); crm_info("Ignoring invalid IPC message: to '%s' not " CRM_SYSTEM_PENGINE, pcmk__s(sys_to, "")); } else { char *log_msg = NULL; const char *reason = NULL; xmlNode *reply = NULL; pcmk__request_t request = { .ipc_client = c, .ipc_id = id, .ipc_flags = flags, .peer = NULL, .xml = msg, .call_options = 0, .result = PCMK__UNKNOWN_RESULT, }; request.op = crm_element_value_copy(request.xml, PCMK__XA_CRM_TASK); CRM_CHECK(request.op != NULL, return 0); reply = pcmk__process_request(&request, schedulerd_handlers); if (reply != NULL) { pcmk__ipc_send_xml(c, id, reply, crm_ipc_server_event); free_xml(reply); } reason = request.result.exit_reason; log_msg = crm_strdup_printf("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)? "" : ")"); if (!pcmk__result_ok(&request.result)) { crm_warn("%s", log_msg); } else { crm_debug("%s", log_msg); } free(log_msg); pcmk__reset_request(&request); } free_xml(msg); return 0; } /* Error code means? */ static int32_t pe_ipc_closed(qb_ipcs_connection_t * c) { pcmk__client_t *client = pcmk__find_client(c); if (client == NULL) { return 0; } crm_trace("Connection %p", c); pcmk__free_client(client); return 0; } static void pe_ipc_destroy(qb_ipcs_connection_t * c) { crm_trace("Connection %p", c); pe_ipc_closed(c); } struct qb_ipcs_service_handlers ipc_callbacks = { .connection_accept = pe_ipc_accept, .connection_created = NULL, .msg_process = pe_ipc_dispatch, .connection_closed = pe_ipc_closed, .connection_destroyed = pe_ipc_destroy }; diff --git a/include/crm/cib/util.h b/include/crm/cib/util.h index 7497be54d6..a86be6149d 100644 --- a/include/crm/cib/util.h +++ b/include/crm/cib/util.h @@ -1,67 +1,64 @@ /* * 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. */ #ifndef PCMK__CRM_CIB_UTIL__H # define PCMK__CRM_CIB_UTIL__H #include // gboolean #include // xmlNode #include // cib_t #ifdef __cplusplus extern "C" { #endif /* Utility functions */ xmlNode *createEmptyCib(int cib_epoch); gboolean cib_version_details(xmlNode * cib, int *admin_epoch, int *epoch, int *updates); int update_attr_delegate(cib_t * the_cib, int call_options, const char *section, const char *node_uuid, const char *set_type, const char *set_name, const char *attr_id, const char *attr_name, const char *attr_value, gboolean to_console, const char *user_name, const char *node_type); int find_nvpair_attr_delegate(cib_t * the_cib, const char *attr, const char *section, const char *node_uuid, const char *set_type, const char *set_name, const char *attr_id, const char *attr_name, gboolean to_console, char **value, const char *user_name); int read_attr_delegate(cib_t * the_cib, const char *section, const char *node_uuid, const char *set_type, const char *set_name, const char *attr_id, const char *attr_name, char **attr_value, gboolean to_console, const char *user_name); int delete_attr_delegate(cib_t * the_cib, int options, const char *section, const char *node_uuid, const char *set_type, const char *set_name, const char *attr_id, const char *attr_name, const char *attr_value, gboolean to_console, const char *user_name); int query_node_uuid(cib_t * the_cib, const char *uname, char **uuid, int *is_remote_node); -void cib_metadata(void); -const char *cib_pref(GHashTable * options, const char *name); - int cib_apply_patch_event(xmlNode *event, xmlNode *input, xmlNode **output, int level); #if !defined(PCMK_ALLOW_DEPRECATED) || (PCMK_ALLOW_DEPRECATED == 1) #include #endif #ifdef __cplusplus } #endif #endif diff --git a/include/crm/cib/util_compat.h b/include/crm/cib/util_compat.h index e435483504..99b2551005 100644 --- a/include/crm/cib/util_compat.h +++ b/include/crm/cib/util_compat.h @@ -1,54 +1,60 @@ /* * Copyright 2021-2024 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_CIB_UTIL_COMPAT__H # define PCMK__CRM_CIB_UTIL_COMPAT__H #include // xmlNode #include // cib_t #include #ifdef __cplusplus extern "C" { #endif /** * \file * \brief Deprecated Pacemaker configuration utilities * \ingroup cib * \deprecated Do not include this header directly. The utilities in this * header, and the header itself, will be removed in a future * release. */ //! \deprecated Use pcmk_cib_xpath_for() instead const char *get_object_path(const char *object_type); //! \deprecated Use pcmk_cib_parent_name_for() instead const char *get_object_parent(const char *object_type); //! \deprecated Use pcmk_cib_xpath_for() instead xmlNode *get_object_root(const char *object_type, xmlNode *the_root); //! \deprecated Do not use int set_standby(cib_t *the_cib, const char *uuid, const char *scope, const char *standby_value); //! \deprecated Do not use int query_node_uname(cib_t * the_cib, const char *uuid, char **uname); //! \deprecated Do not use xmlNode *cib_get_generation(cib_t *cib); +//! \deprecated Do not use +const char *cib_pref(GHashTable * options, const char *name); + +//! \deprecated Do not use +void cib_metadata(void); + #ifdef __cplusplus } #endif #endif // PCMK__CRM_CIB_UTIL_COMPAT__H diff --git a/include/crm/common/options_internal.h b/include/crm/common/options_internal.h index c9066f86f4..a27547cd18 100644 --- a/include/crm/common/options_internal.h +++ b/include/crm/common/options_internal.h @@ -1,186 +1,204 @@ /* * Copyright 2006-2024 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__OPTIONS_INTERNAL__H # define PCMK__OPTIONS_INTERNAL__H # ifndef PCMK__CONFIG_H # define PCMK__CONFIG_H # include // _Noreturn # endif # include // GHashTable # include // bool #include // pcmk_parse_interval_spec() _Noreturn void pcmk__cli_help(char cmd); /* * Environment variable option handling */ const char *pcmk__env_option(const char *option); void pcmk__set_env_option(const char *option, const char *value, bool compat); bool pcmk__env_option_enabled(const char *daemon, const char *option); /* * Cluster option handling */ +/*! + * \internal + * \enum pcmk__opt_context + * \brief Context flags for options + */ +enum pcmk__opt_context { + // @COMPAT Used only for daemon metadata + pcmk__opt_context_none = 0, //!< No additional context + pcmk__opt_context_based = (1 << 1), //!< CIB manager metadata + pcmk__opt_context_controld = (1 << 2), //!< Controller metadata + pcmk__opt_context_schedulerd = (1 << 3), //!< Scheduler metadata +}; + typedef struct pcmk__cluster_option_s { const char *name; const char *alt_name; const char *type; const char *values; const char *default_value; bool (*is_valid)(const char *); + // @COMPAT context is used only for daemon meta-data + enum pcmk__opt_context context; + const char *description_short; const char *description_long; } pcmk__cluster_option_t; -const char *pcmk__cluster_option(GHashTable *options, - const pcmk__cluster_option_t *option_list, - int len, const char *name); +const char *pcmk__cluster_option(GHashTable *options, const char *name); gchar *pcmk__format_option_metadata(const char *name, const char *desc_short, const char *desc_long, + enum pcmk__opt_context filter, pcmk__cluster_option_t *option_list, int len); -void pcmk__validate_cluster_options(GHashTable *options, - pcmk__cluster_option_t *option_list, - int len); +gchar *pcmk__cluster_option_metadata(const char *name, const char *desc_short, + const char *desc_long, + enum pcmk__opt_context filter); + +void pcmk__validate_cluster_options(GHashTable *options); bool pcmk__valid_interval_spec(const char *value); bool pcmk__valid_boolean(const char *value); bool pcmk__valid_int(const char *value); bool pcmk__valid_positive_int(const char *value); bool pcmk__valid_no_quorum_policy(const char *value); bool pcmk__valid_percentage(const char *value); bool pcmk__valid_script(const char *value); +bool pcmk__valid_placement_strategy(const char *value); // from watchdog.c long pcmk__get_sbd_watchdog_timeout(void); bool pcmk__get_sbd_sync_resource_startup(void); long pcmk__auto_stonith_watchdog_timeout(void); bool pcmk__valid_stonith_watchdog_timeout(const char *value); // Constants for environment variable names #define PCMK__ENV_AUTHKEY_LOCATION "authkey_location" #define PCMK__ENV_BLACKBOX "blackbox" #define PCMK__ENV_CALLGRIND_ENABLED "callgrind_enabled" #define PCMK__ENV_CLUSTER_TYPE "cluster_type" #define PCMK__ENV_DEBUG "debug" #define PCMK__ENV_DH_MAX_BITS "dh_max_bits" #define PCMK__ENV_DH_MIN_BITS "dh_min_bits" #define PCMK__ENV_FAIL_FAST "fail_fast" #define PCMK__ENV_IPC_BUFFER "ipc_buffer" #define PCMK__ENV_IPC_TYPE "ipc_type" #define PCMK__ENV_LOGFACILITY "logfacility" #define PCMK__ENV_LOGFILE "logfile" #define PCMK__ENV_LOGFILE_MODE "logfile_mode" #define PCMK__ENV_LOGPRIORITY "logpriority" #define PCMK__ENV_NODE_ACTION_LIMIT "node_action_limit" #define PCMK__ENV_NODE_START_STATE "node_start_state" #define PCMK__ENV_PANIC_ACTION "panic_action" #define PCMK__ENV_REMOTE_ADDRESS "remote_address" #define PCMK__ENV_REMOTE_SCHEMA_DIR "remote_schema_directory" #define PCMK__ENV_REMOTE_PID1 "remote_pid1" #define PCMK__ENV_REMOTE_PORT "remote_port" #define PCMK__ENV_RESPAWNED "respawned" #define PCMK__ENV_SCHEMA_DIRECTORY "schema_directory" #define PCMK__ENV_SERVICE "service" #define PCMK__ENV_STDERR "stderr" #define PCMK__ENV_TLS_PRIORITIES "tls_priorities" #define PCMK__ENV_TRACE_BLACKBOX "trace_blackbox" #define PCMK__ENV_TRACE_FILES "trace_files" #define PCMK__ENV_TRACE_FORMATS "trace_formats" #define PCMK__ENV_TRACE_FUNCTIONS "trace_functions" #define PCMK__ENV_TRACE_TAGS "trace_tags" #define PCMK__ENV_VALGRIND_ENABLED "valgrind_enabled" // @COMPAT Drop at 3.0.0; default is plenty #define PCMK__ENV_CIB_TIMEOUT "cib_timeout" // @COMPAT Drop at 3.0.0; likely last used in 1.1.24 #define PCMK__ENV_MCP "mcp" // @COMPAT Drop at 3.0.0; added unused in 1.1.9 #define PCMK__ENV_QUORUM_TYPE "quorum_type" /* @COMPAT Drop at 3.0.0; added to debug shutdown issues when Pacemaker is * managed by systemd, but no longer useful. */ #define PCMK__ENV_SHUTDOWN_DELAY "shutdown_delay" // @COMPAT Deprecated since 2.1.0 #define PCMK__OPT_REMOVE_AFTER_STOP "remove-after-stop" // Constants for meta-attribute names #define PCMK__META_CLONE "clone" #define PCMK__META_CONTAINER "container" #define PCMK__META_DIGESTS_ALL "digests-all" #define PCMK__META_DIGESTS_SECURE "digests-secure" #define PCMK__META_INTERNAL_RSC "internal_rsc" #define PCMK__META_MIGRATE_SOURCE "migrate_source" #define PCMK__META_MIGRATE_TARGET "migrate_target" #define PCMK__META_ON_NODE "on_node" #define PCMK__META_ON_NODE_UUID "on_node_uuid" #define PCMK__META_OP_NO_WAIT "op_no_wait" #define PCMK__META_OP_TARGET_RC "op_target_rc" #define PCMK__META_PHYSICAL_HOST "physical-host" /* @TODO Plug these in. Currently, they're never set. These are op attrs for use * with https://projects.clusterlabs.org/T382. */ #define PCMK__META_CLEAR_FAILURE_OP "clear_failure_op" #define PCMK__META_CLEAR_FAILURE_INTERVAL "clear_failure_interval" // @COMPAT Deprecated meta-attribute since 2.1.0 #define PCMK__META_CAN_FAIL "can_fail" // @COMPAT Deprecated alias for PCMK__META_PROMOTED_MAX since 2.0.0 #define PCMK__META_PROMOTED_MAX_LEGACY "master-max" // @COMPAT Deprecated alias for PCMK__META_PROMOTED_NODE_MAX since 2.0.0 #define PCMK__META_PROMOTED_NODE_MAX_LEGACY "master-node-max" // @COMPAT Deprecated meta-attribute since 2.0.0 #define PCMK__META_RESTART_TYPE "restart-type" // @COMPAT Deprecated meta-attribute since 2.0.0 #define PCMK__META_ROLE_AFTER_FAILURE "role_after_failure" // Constants for enumerated values for various options #define PCMK__VALUE_CIB "cib" #define PCMK__VALUE_CLUSTER "cluster" #define PCMK__VALUE_CUSTOM "custom" #define PCMK__VALUE_EN "en" #define PCMK__VALUE_FENCING "fencing" #define PCMK__VALUE_GREEN "green" #define PCMK__VALUE_LOCAL "local" #define PCMK__VALUE_MIGRATE_ON_RED "migrate-on-red" #define PCMK__VALUE_NONE "none" #define PCMK__VALUE_NOTHING "nothing" #define PCMK__VALUE_ONLY_GREEN "only-green" #define PCMK__VALUE_PROGRESSIVE "progressive" #define PCMK__VALUE_QUORUM "quorum" #define PCMK__VALUE_RED "red" #define PCMK__VALUE_REQUEST "request" #define PCMK__VALUE_RESPONSE "response" #define PCMK__VALUE_UNFENCING "unfencing" #define PCMK__VALUE_YELLOW "yellow" #endif // PCMK__OPTIONS_INTERNAL__H diff --git a/include/crm/pengine/common.h b/include/crm/pengine/common.h index a35be087f1..f5b0da8122 100644 --- a/include/crm/pengine/common.h +++ b/include/crm/pengine/common.h @@ -1,68 +1,66 @@ /* - * Copyright 2004-2023 the Pacemaker project contributors + * 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. */ #ifndef PCMK__CRM_PENGINE_COMMON__H # define PCMK__CRM_PENGINE_COMMON__H # include # include # include # include #ifdef __cplusplus extern "C" { #endif const char *task2text(enum action_tasks task); enum action_tasks text2task(const char *task); const char *fail2text(enum action_fail_response fail); -const char *pe_pref(GHashTable * options, const char *name); - typedef struct pe_re_match_data { char *string; int nregs; regmatch_t *pmatch; } pe_re_match_data_t; typedef struct pe_match_data { pe_re_match_data_t *re; GHashTable *params; GHashTable *meta; } pe_match_data_t; typedef struct pe_rsc_eval_data { const char *standard; const char *provider; const char *agent; } pe_rsc_eval_data_t; typedef struct pe_op_eval_data { const char *op_name; guint interval; } pe_op_eval_data_t; typedef struct pe_rule_eval_data { GHashTable *node_hash; // Only used with g_hash_table_lookup() enum rsc_role_e role; crm_time_t *now; // @COMPAT could be const pe_match_data_t *match_data; // @COMPAT could be const pe_rsc_eval_data_t *rsc_data; // @COMPAT could be const pe_op_eval_data_t *op_data; // @COMPAT could be const } pe_rule_eval_data_t; #if !defined(PCMK_ALLOW_DEPRECATED) || (PCMK_ALLOW_DEPRECATED == 1) #include #endif #ifdef __cplusplus } #endif #endif diff --git a/include/crm/pengine/common_compat.h b/include/crm/pengine/common_compat.h index 424a499fae..e6ce1e4260 100644 --- a/include/crm/pengine/common_compat.h +++ b/include/crm/pengine/common_compat.h @@ -1,75 +1,78 @@ /* - * Copyright 2004-2023 the Pacemaker project contributors + * 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. */ #ifndef PCMK__CRM_PENGINE_COMMON_COMPAT__H # define PCMK__CRM_PENGINE_COMMON_COMPAT__H #include #ifdef __cplusplus extern "C" { #endif /** * \file * \brief Deprecated Pacemaker scheduler utilities * \ingroup pengine * \deprecated Do not include this header directly. The utilities in this * header, and the header itself, will be removed in a future * release. */ //! \deprecated Use (pcmk_role_promoted + 1) instead #define RSC_ROLE_MAX (pcmk_role_promoted + 1) //! \deprecated Use pcmk_role_text(pcmk_role_unknown) instead #define RSC_ROLE_UNKNOWN_S pcmk_role_text(pcmk_role_unknown) //! \deprecated Use pcmk_role_text(pcmk_role_stopped) instead #define RSC_ROLE_STOPPED_S pcmk_role_text(pcmk_role_stopped) //! \deprecated Use pcmk_role_text(pcmk_role_started) instead #define RSC_ROLE_STARTED_S pcmk_role_text(pcmk_role_started) //! \deprecated Use pcmk_role_text(pcmk_role_unpromoted) instead #define RSC_ROLE_UNPROMOTED_S pcmk_role_text(pcmk_role_unpromoted) //! \deprecated Use pcmk_role_text(pcmk_role_promoted) instead #define RSC_ROLE_PROMOTED_S pcmk_role_text(pcmk_role_promoted) //! \deprecated Do not use #define RSC_ROLE_UNPROMOTED_LEGACY_S "Slave" //! \deprecated Do not use #define RSC_ROLE_SLAVE_S RSC_ROLE_UNPROMOTED_LEGACY_S //! \deprecated Do not use #define RSC_ROLE_PROMOTED_LEGACY_S "Master" //! \deprecated Do not use #define RSC_ROLE_MASTER_S RSC_ROLE_PROMOTED_LEGACY_S //! \deprecated Use pcmk_role_text() instead const char *role2text(enum rsc_role_e role); //! \deprecated Use pcmk_parse_role() instead enum rsc_role_e text2role(const char *role); //! \deprecated Use pcmk_multiply_active_text() instead static inline const char * recovery2text(enum rsc_recovery_type type) { return pcmk_multiply_active_text(type); } +//! \deprecated Do not use +const char *pe_pref(GHashTable * options, const char *name); + #ifdef __cplusplus } #endif #endif // PCMK__CRM_PENGINE_COMMON_COMPAT__H diff --git a/include/crm/pengine/internal.h b/include/crm/pengine/internal.h index 122af1ca6a..0a83062dab 100644 --- a/include/crm/pengine/internal.h +++ b/include/crm/pengine/internal.h @@ -1,454 +1,455 @@ /* * 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. */ #ifndef PE_INTERNAL__H # define PE_INTERNAL__H # include # include # include # include # include # include # include # include # include # include const char *pe__resource_description(const pcmk_resource_t *rsc, uint32_t show_opts); bool pe__clone_is_ordered(const pcmk_resource_t *clone); int pe__set_clone_flag(pcmk_resource_t *clone, enum pcmk__clone_flags flag); bool pe__clone_flag_is_set(const pcmk_resource_t *clone, uint32_t flags); bool pe__group_flag_is_set(const pcmk_resource_t *group, uint32_t flags); pcmk_resource_t *pe__last_group_member(const pcmk_resource_t *group); const pcmk_resource_t *pe__const_top_resource(const pcmk_resource_t *rsc, bool include_bundle); int pe__clone_max(const pcmk_resource_t *clone); int pe__clone_node_max(const pcmk_resource_t *clone); int pe__clone_promoted_max(const pcmk_resource_t *clone); int pe__clone_promoted_node_max(const pcmk_resource_t *clone); void pe__create_clone_notifications(pcmk_resource_t *clone); void pe__free_clone_notification_data(pcmk_resource_t *clone); void pe__create_clone_notif_pseudo_ops(pcmk_resource_t *clone, pcmk_action_t *start, pcmk_action_t *started, pcmk_action_t *stop, pcmk_action_t *stopped); pcmk_action_t *pe__new_rsc_pseudo_action(pcmk_resource_t *rsc, const char *task, bool optional, bool runnable); void pe__create_promotable_pseudo_ops(pcmk_resource_t *clone, bool any_promoting, bool any_demoting); bool pe_can_fence(const pcmk_scheduler_t *scheduler, const pcmk_node_t *node); void add_hash_param(GHashTable * hash, const char *name, const char *value); char *native_parameter(pcmk_resource_t *rsc, pcmk_node_t *node, gboolean create, const char *name, pcmk_scheduler_t *scheduler); pcmk_node_t *native_location(const pcmk_resource_t *rsc, GList **list, int current); -void pe_metadata(pcmk__output_t *out); -void verify_pe_options(GHashTable * options); - void native_add_running(pcmk_resource_t *rsc, pcmk_node_t *node, pcmk_scheduler_t *scheduler, gboolean failed); gboolean native_unpack(pcmk_resource_t *rsc, pcmk_scheduler_t *scheduler); gboolean group_unpack(pcmk_resource_t *rsc, pcmk_scheduler_t *scheduler); gboolean clone_unpack(pcmk_resource_t *rsc, pcmk_scheduler_t *scheduler); gboolean pe__unpack_bundle(pcmk_resource_t *rsc, pcmk_scheduler_t *scheduler); pcmk_resource_t *native_find_rsc(pcmk_resource_t *rsc, const char *id, const pcmk_node_t *node, int flags); gboolean native_active(pcmk_resource_t *rsc, gboolean all); gboolean group_active(pcmk_resource_t *rsc, gboolean all); gboolean clone_active(pcmk_resource_t *rsc, gboolean all); gboolean pe__bundle_active(pcmk_resource_t *rsc, gboolean all); //! \deprecated This function will be removed in a future release void native_print(pcmk_resource_t *rsc, const char *pre_text, long options, void *print_data); //! \deprecated This function will be removed in a future release void group_print(pcmk_resource_t *rsc, const char *pre_text, long options, void *print_data); //! \deprecated This function will be removed in a future release void clone_print(pcmk_resource_t *rsc, const char *pre_text, long options, void *print_data); //! \deprecated This function will be removed in a future release void pe__print_bundle(pcmk_resource_t *rsc, const char *pre_text, long options, void *print_data); gchar *pcmk__native_output_string(const pcmk_resource_t *rsc, const char *name, const pcmk_node_t *node, uint32_t show_opts, const char *target_role, bool show_nodes); int pe__name_and_nvpairs_xml(pcmk__output_t *out, bool is_list, const char *tag_name , size_t pairs_count, ...); char *pe__node_display_name(pcmk_node_t *node, bool print_detail); // Clone notifications (pe_notif.c) void pe__order_notifs_after_fencing(const pcmk_action_t *action, pcmk_resource_t *rsc, pcmk_action_t *stonith_op); // Resource output methods int pe__clone_xml(pcmk__output_t *out, va_list args); int pe__clone_default(pcmk__output_t *out, va_list args); int pe__group_xml(pcmk__output_t *out, va_list args); int pe__group_default(pcmk__output_t *out, va_list args); int pe__bundle_xml(pcmk__output_t *out, va_list args); int pe__bundle_html(pcmk__output_t *out, va_list args); int pe__bundle_text(pcmk__output_t *out, va_list args); int pe__node_html(pcmk__output_t *out, va_list args); int pe__node_text(pcmk__output_t *out, va_list args); int pe__node_xml(pcmk__output_t *out, va_list args); int pe__resource_xml(pcmk__output_t *out, va_list args); int pe__resource_html(pcmk__output_t *out, va_list args); int pe__resource_text(pcmk__output_t *out, va_list args); void native_free(pcmk_resource_t *rsc); void group_free(pcmk_resource_t *rsc); void clone_free(pcmk_resource_t *rsc); void pe__free_bundle(pcmk_resource_t *rsc); enum rsc_role_e native_resource_state(const pcmk_resource_t *rsc, gboolean current); enum rsc_role_e group_resource_state(const pcmk_resource_t *rsc, gboolean current); enum rsc_role_e clone_resource_state(const pcmk_resource_t *rsc, gboolean current); enum rsc_role_e pe__bundle_resource_state(const pcmk_resource_t *rsc, gboolean current); void pe__count_common(pcmk_resource_t *rsc); void pe__count_bundle(pcmk_resource_t *rsc); void common_free(pcmk_resource_t *rsc); pcmk_node_t *pe__copy_node(const pcmk_node_t *this_node); time_t get_effective_time(pcmk_scheduler_t *scheduler); /* Failure handling utilities (from failcounts.c) */ int pe_get_failcount(const pcmk_node_t *node, pcmk_resource_t *rsc, time_t *last_failure, uint32_t flags, const xmlNode *xml_op); pcmk_action_t *pe__clear_failcount(pcmk_resource_t *rsc, const pcmk_node_t *node, const char *reason, pcmk_scheduler_t *scheduler); /* Functions for finding/counting a resource's active nodes */ bool pe__count_active_node(const pcmk_resource_t *rsc, pcmk_node_t *node, pcmk_node_t **active, unsigned int *count_all, unsigned int *count_clean); pcmk_node_t *pe__find_active_requires(const pcmk_resource_t *rsc, unsigned int *count); /* Binary like operators for lists of nodes */ GHashTable *pe__node_list2table(const GList *list); pcmk_action_t *get_pseudo_op(const char *name, pcmk_scheduler_t *scheduler); gboolean order_actions(pcmk_action_t *lh_action, pcmk_action_t *rh_action, uint32_t flags); void pe__show_node_scores_as(const char *file, const char *function, int line, bool to_log, const pcmk_resource_t *rsc, const char *comment, GHashTable *nodes, pcmk_scheduler_t *scheduler); #define pe__show_node_scores(level, rsc, text, nodes, scheduler) \ pe__show_node_scores_as(__FILE__, __func__, __LINE__, \ (level), (rsc), (text), (nodes), (scheduler)) GHashTable *pcmk__unpack_action_meta(pcmk_resource_t *rsc, const pcmk_node_t *node, const char *action_name, guint interval_ms, const xmlNode *action_config); GHashTable *pcmk__unpack_action_rsc_params(const xmlNode *action_xml, GHashTable *node_attrs, pcmk_scheduler_t *data_set); xmlNode *pcmk__find_action_config(const pcmk_resource_t *rsc, const char *action_name, guint interval_ms, bool include_disabled); enum rsc_start_requirement pcmk__action_requires(const pcmk_resource_t *rsc, const char *action_name); enum action_fail_response pcmk__parse_on_fail(const pcmk_resource_t *rsc, const char *action_name, guint interval_ms, const char *value); enum rsc_role_e pcmk__role_after_failure(const pcmk_resource_t *rsc, const char *action_name, enum action_fail_response on_fail, GHashTable *meta); pcmk_action_t *custom_action(pcmk_resource_t *rsc, char *key, const char *task, const pcmk_node_t *on_node, gboolean optional, pcmk_scheduler_t *scheduler); # define delete_key(rsc) pcmk__op_key(rsc->id, PCMK_ACTION_DELETE, 0) # define delete_action(rsc, node, optional) custom_action( \ rsc, delete_key(rsc), PCMK_ACTION_DELETE, node, \ optional, rsc->cluster); # define stop_key(rsc) pcmk__op_key(rsc->id, PCMK_ACTION_STOP, 0) # define stop_action(rsc, node, optional) custom_action( \ rsc, stop_key(rsc), PCMK_ACTION_STOP, node, \ optional, rsc->cluster); # define reload_key(rsc) pcmk__op_key(rsc->id, PCMK_ACTION_RELOAD_AGENT, 0) # define start_key(rsc) pcmk__op_key(rsc->id, PCMK_ACTION_START, 0) # define start_action(rsc, node, optional) custom_action( \ rsc, start_key(rsc), PCMK_ACTION_START, node, \ optional, rsc->cluster) # define promote_key(rsc) pcmk__op_key(rsc->id, PCMK_ACTION_PROMOTE, 0) # define promote_action(rsc, node, optional) custom_action( \ rsc, promote_key(rsc), PCMK_ACTION_PROMOTE, node, \ optional, rsc->cluster) # define demote_key(rsc) pcmk__op_key(rsc->id, PCMK_ACTION_DEMOTE, 0) # define demote_action(rsc, node, optional) custom_action( \ rsc, demote_key(rsc), PCMK_ACTION_DEMOTE, node, \ optional, rsc->cluster) extern int pe_get_configured_timeout(pcmk_resource_t *rsc, const char *action, pcmk_scheduler_t *scheduler); pcmk_action_t *find_first_action(const GList *input, const char *uuid, const char *task, const pcmk_node_t *on_node); enum action_tasks get_complex_task(const pcmk_resource_t *rsc, const char *name); GList *find_actions(GList *input, const char *key, const pcmk_node_t *on_node); GList *find_actions_exact(GList *input, const char *key, const pcmk_node_t *on_node); GList *pe__resource_actions(const pcmk_resource_t *rsc, const pcmk_node_t *node, const char *task, bool require_node); extern void pe_free_action(pcmk_action_t *action); void resource_location(pcmk_resource_t *rsc, const pcmk_node_t *node, int score, const char *tag, pcmk_scheduler_t *scheduler); extern int pe__is_newer_op(const xmlNode *xml_a, const xmlNode *xml_b, bool same_node_default); extern gint sort_op_by_callid(gconstpointer a, gconstpointer b); gboolean get_target_role(const pcmk_resource_t *rsc, enum rsc_role_e *role); void pe__set_next_role(pcmk_resource_t *rsc, enum rsc_role_e role, const char *why); pcmk_resource_t *find_clone_instance(const pcmk_resource_t *rsc, const char *sub_id); extern void destroy_ticket(gpointer data); pcmk_ticket_t *ticket_new(const char *ticket_id, pcmk_scheduler_t *scheduler); // Resources for manipulating resource names const char *pe_base_name_end(const char *id); char *clone_strip(const char *last_rsc_id); char *clone_zero(const char *last_rsc_id); static inline bool pe_base_name_eq(const pcmk_resource_t *rsc, const char *id) { if (id && rsc && rsc->id) { // Number of characters in rsc->id before any clone suffix size_t base_len = pe_base_name_end(rsc->id) - rsc->id + 1; return (strlen(id) == base_len) && !strncmp(id, rsc->id, base_len); } return false; } int pe__target_rc_from_xml(const xmlNode *xml_op); gint pe__cmp_node_name(gconstpointer a, gconstpointer b); bool is_set_recursive(const pcmk_resource_t *rsc, long long flag, bool any); 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); void pe__free_digests(gpointer ptr); 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_action_t *pe_fence_op(pcmk_node_t *node, const char *op, bool optional, const char *reason, bool priority_delay, pcmk_scheduler_t *scheduler); void trigger_unfencing(pcmk_resource_t *rsc, pcmk_node_t *node, const char *reason, pcmk_action_t *dependency, pcmk_scheduler_t *scheduler); char *pe__action2reason(const pcmk_action_t *action, enum pe_action_flags flag); void pe_action_set_reason(pcmk_action_t *action, const char *reason, bool overwrite); void pe__add_action_expected_result(pcmk_action_t *action, int expected_result); void pe__set_resource_flags_recursive(pcmk_resource_t *rsc, uint64_t flags); void pe__clear_resource_flags_recursive(pcmk_resource_t *rsc, uint64_t flags); void pe__clear_resource_flags_on_all(pcmk_scheduler_t *scheduler, uint64_t flag); gboolean add_tag_ref(GHashTable * tags, const char * tag_name, const char * obj_ref); //! \deprecated This function will be removed in a future release void print_rscs_brief(GList *rsc_list, const char * pre_text, long options, void * print_data, gboolean print_all); int pe__rscs_brief_output(pcmk__output_t *out, GList *rsc_list, unsigned int options); void pe_fence_node(pcmk_scheduler_t *scheduler, pcmk_node_t *node, const char *reason, bool priority_delay); pcmk_node_t *pe_create_node(const char *id, const char *uname, const char *type, const char *score, pcmk_scheduler_t *scheduler); //! \deprecated This function will be removed in a future release void common_print(pcmk_resource_t *rsc, const char *pre_text, const char *name, const pcmk_node_t *node, long options, void *print_data); int pe__common_output_text(pcmk__output_t *out, const pcmk_resource_t *rsc, const char *name, const pcmk_node_t *node, unsigned int options); int pe__common_output_html(pcmk__output_t *out, const pcmk_resource_t *rsc, const char *name, const pcmk_node_t *node, unsigned int options); GList *pe__bundle_containers(const pcmk_resource_t *bundle); int pe__bundle_max(const pcmk_resource_t *rsc); bool pe__node_is_bundle_instance(const pcmk_resource_t *bundle, const pcmk_node_t *node); pcmk_resource_t *pe__bundled_resource(const pcmk_resource_t *rsc); const pcmk_resource_t *pe__get_rsc_in_container(const pcmk_resource_t *instance); pcmk_resource_t *pe__first_container(const pcmk_resource_t *bundle); void pe__foreach_bundle_replica(pcmk_resource_t *bundle, bool (*fn)(pcmk__bundle_replica_t *, void *), void *user_data); void pe__foreach_const_bundle_replica(const pcmk_resource_t *bundle, bool (*fn)(const pcmk__bundle_replica_t *, void *), void *user_data); pcmk_resource_t *pe__find_bundle_replica(const pcmk_resource_t *bundle, const pcmk_node_t *node); bool pe__bundle_needs_remote_name(pcmk_resource_t *rsc); const char *pe__add_bundle_remote_name(pcmk_resource_t *rsc, pcmk_scheduler_t *scheduler, xmlNode *xml, const char *field); const char *pe__node_attribute_calculated(const pcmk_node_t *node, const char *name, const pcmk_resource_t *rsc, enum pcmk__rsc_node node_type, bool force_host); const char *pe_node_attribute_raw(const pcmk_node_t *node, const char *name); bool pe__is_universal_clone(const pcmk_resource_t *rsc, const pcmk_scheduler_t *scheduler); void pe__add_param_check(const xmlNode *rsc_op, pcmk_resource_t *rsc, pcmk_node_t *node, enum pcmk__check_parameters, pcmk_scheduler_t *scheduler); void pe__foreach_param_check(pcmk_scheduler_t *scheduler, void (*cb)(pcmk_resource_t*, pcmk_node_t*, const xmlNode*, enum pcmk__check_parameters)); void pe__free_param_checks(pcmk_scheduler_t *scheduler); bool pe__shutdown_requested(const pcmk_node_t *node); void pe__update_recheck_time(time_t recheck, pcmk_scheduler_t *scheduler, const char *reason); /*! * \internal * \brief Register xml formatting message functions. * * \param[in,out] out Output object to register messages with */ void pe__register_messages(pcmk__output_t *out); void pe__unpack_dataset_nvpairs(const xmlNode *xml_obj, const char *set_name, const pe_rule_eval_data_t *rule_data, GHashTable *hash, const char *always_first, gboolean overwrite, pcmk_scheduler_t *scheduler); bool pe__resource_is_disabled(const pcmk_resource_t *rsc); void pe__clear_resource_history(pcmk_resource_t *rsc, const pcmk_node_t *node); GList *pe__rscs_with_tag(pcmk_scheduler_t *scheduler, const char *tag_name); GList *pe__unames_with_tag(pcmk_scheduler_t *scheduler, const char *tag_name); bool pe__rsc_has_tag(pcmk_scheduler_t *scheduler, const char *rsc, const char *tag); bool pe__uname_has_tag(pcmk_scheduler_t *scheduler, const char *node, const char *tag); bool pe__rsc_running_on_only(const pcmk_resource_t *rsc, const pcmk_node_t *node); bool pe__rsc_running_on_any(pcmk_resource_t *rsc, GList *node_list); GList *pe__filter_rsc_list(GList *rscs, GList *filter); GList * pe__build_node_name_list(pcmk_scheduler_t *scheduler, const char *s); GList * pe__build_rsc_list(pcmk_scheduler_t *scheduler, const char *s); bool pcmk__rsc_filtered_by_node(pcmk_resource_t *rsc, GList *only_node); gboolean pe__bundle_is_filtered(const pcmk_resource_t *rsc, GList *only_rsc, gboolean check_parent); gboolean pe__clone_is_filtered(const pcmk_resource_t *rsc, GList *only_rsc, gboolean check_parent); gboolean pe__group_is_filtered(const pcmk_resource_t *rsc, GList *only_rsc, gboolean check_parent); gboolean pe__native_is_filtered(const pcmk_resource_t *rsc, GList *only_rsc, gboolean check_parent); xmlNode *pe__failed_probe_for_rsc(const pcmk_resource_t *rsc, const char *name); const char *pe__clone_child_id(const pcmk_resource_t *rsc); int pe__sum_node_health_scores(const pcmk_node_t *node, int base_health); int pe__node_health(pcmk_node_t *node); static inline enum pcmk__health_strategy pe__health_strategy(pcmk_scheduler_t *scheduler) { - return pcmk__parse_health_strategy(pe_pref(scheduler->config_hash, - PCMK_OPT_NODE_HEALTH_STRATEGY)); + const char *strategy = pcmk__cluster_option(scheduler->config_hash, + PCMK_OPT_NODE_HEALTH_STRATEGY); + + return pcmk__parse_health_strategy(strategy); } static inline int pe__health_score(const char *option, pcmk_scheduler_t *scheduler) { - return char2score(pe_pref(scheduler->config_hash, option)); + const char *value = pcmk__cluster_option(scheduler->config_hash, option); + + return char2score(value); } #endif diff --git a/lib/cib/cib_utils.c b/lib/cib/cib_utils.c index ec45b5c56f..cf92cd0219 100644 --- a/lib/cib/cib_utils.c +++ b/lib/cib/cib_utils.c @@ -1,1117 +1,1089 @@ /* * Original copyright 2004 International Business Machines * Later changes copyright 2008-2024 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include gboolean cib_version_details(xmlNode * cib, int *admin_epoch, int *epoch, int *updates) { *epoch = -1; *updates = -1; *admin_epoch = -1; if (cib == NULL) { return FALSE; } else { crm_element_value_int(cib, PCMK_XA_EPOCH, epoch); crm_element_value_int(cib, PCMK_XA_NUM_UPDATES, updates); crm_element_value_int(cib, PCMK_XA_ADMIN_EPOCH, admin_epoch); } return TRUE; } gboolean cib_diff_version_details(xmlNode * diff, int *admin_epoch, int *epoch, int *updates, int *_admin_epoch, int *_epoch, int *_updates) { int add[] = { 0, 0, 0 }; int del[] = { 0, 0, 0 }; xml_patch_versions(diff, add, del); *admin_epoch = add[0]; *epoch = add[1]; *updates = add[2]; *_admin_epoch = del[0]; *_epoch = del[1]; *_updates = del[2]; return TRUE; } /*! * \internal * \brief Get the XML patchset from a CIB diff notification * * \param[in] msg CIB diff notification * \param[out] patchset Where to store XML patchset * * \return Standard Pacemaker return code */ int cib__get_notify_patchset(const xmlNode *msg, const xmlNode **patchset) { int rc = pcmk_err_generic; CRM_ASSERT(patchset != NULL); *patchset = NULL; if (msg == NULL) { crm_err("CIB diff notification received with no XML"); return ENOMSG; } if ((crm_element_value_int(msg, F_CIB_RC, &rc) != 0) || (rc != pcmk_ok)) { crm_warn("Ignore failed CIB update: %s " CRM_XS " rc=%d", pcmk_strerror(rc), rc); crm_log_xml_debug(msg, "failed"); return pcmk_legacy2rc(rc); } *patchset = get_message_xml(msg, F_CIB_UPDATE_RESULT); if (*patchset == NULL) { crm_err("CIB diff notification received with no patchset"); return ENOMSG; } return pcmk_rc_ok; } #define XPATH_DIFF_V1 "//" F_CIB_UPDATE_RESULT "//" PCMK__XE_DIFF_ADDED /*! * \internal * \brief Check whether a given CIB element was modified in a CIB patchset (v1) * * \param[in] patchset CIB XML patchset * \param[in] element XML tag of CIB element to check (\c NULL is equivalent * to \c PCMK_XE_CIB) * * \return \c true if \p element was modified, or \c false otherwise */ static bool element_in_patchset_v1(const xmlNode *patchset, const char *element) { char *xpath = crm_strdup_printf(XPATH_DIFF_V1 "//%s", pcmk__s(element, PCMK_XE_CIB)); xmlXPathObject *xpath_obj = xpath_search(patchset, xpath); free(xpath); if (xpath_obj == NULL) { return false; } freeXpathObject(xpath_obj); return true; } /*! * \internal * \brief Check whether a given CIB element was modified in a CIB patchset (v2) * * \param[in] patchset CIB XML patchset * \param[in] element XML tag of CIB element to check (\c NULL is equivalent * to \c PCMK_XE_CIB). Supported values include any CIB * element supported by \c pcmk__cib_abs_xpath_for(). * * \return \c true if \p element was modified, or \c false otherwise */ static bool element_in_patchset_v2(const xmlNode *patchset, const char *element) { const char *element_xpath = pcmk__cib_abs_xpath_for(element); const char *parent_xpath = pcmk_cib_parent_name_for(element); char *element_regex = NULL; bool rc = false; CRM_CHECK(element_xpath != NULL, return false); // Unsupported element // Matches if and only if element_xpath is part of a changed path element_regex = crm_strdup_printf("^%s(/|$)", element_xpath); for (const xmlNode *change = first_named_child(patchset, XML_DIFF_CHANGE); change != NULL; change = crm_next_same_xml(change)) { const char *op = crm_element_value(change, F_CIB_OPERATION); const char *diff_xpath = crm_element_value(change, PCMK_XA_PATH); if (pcmk__str_eq(diff_xpath, element_regex, pcmk__str_regex)) { // Change to an existing element rc = true; break; } if (pcmk__str_eq(op, "create", pcmk__str_none) && pcmk__str_eq(diff_xpath, parent_xpath, pcmk__str_none) && pcmk__xe_is(pcmk__xml_first_child(change), element)) { // Newly added element rc = true; break; } } free(element_regex); return rc; } /*! * \internal * \brief Check whether a given CIB element was modified in a CIB patchset * * \param[in] patchset CIB XML patchset * \param[in] element XML tag of CIB element to check (\c NULL is equivalent * to \c PCMK_XE_CIB). Supported values include any CIB * element supported by \c pcmk__cib_abs_xpath_for(). * * \return \c true if \p element was modified, or \c false otherwise */ bool cib__element_in_patchset(const xmlNode *patchset, const char *element) { int format = 1; CRM_ASSERT(patchset != NULL); crm_element_value_int(patchset, PCMK_XA_FORMAT, &format); switch (format) { case 1: return element_in_patchset_v1(patchset, element); case 2: return element_in_patchset_v2(patchset, element); default: crm_warn("Unknown patch format: %d", format); return false; } } /*! * \brief Create XML for a new (empty) CIB * * \param[in] cib_epoch What to use as \c PCMK_XA_EPOCH CIB attribute * * \return Newly created XML for empty CIB * \note It is the caller's responsibility to free the result with free_xml(). */ xmlNode * createEmptyCib(int cib_epoch) { xmlNode *cib_root = NULL, *config = NULL; cib_root = create_xml_node(NULL, PCMK_XE_CIB); crm_xml_add(cib_root, PCMK_XA_CRM_FEATURE_SET, CRM_FEATURE_SET); crm_xml_add(cib_root, PCMK_XA_VALIDATE_WITH, xml_latest_schema()); crm_xml_add_int(cib_root, PCMK_XA_EPOCH, cib_epoch); crm_xml_add_int(cib_root, PCMK_XA_NUM_UPDATES, 0); crm_xml_add_int(cib_root, PCMK_XA_ADMIN_EPOCH, 0); config = create_xml_node(cib_root, PCMK_XE_CONFIGURATION); create_xml_node(cib_root, PCMK_XE_STATUS); create_xml_node(config, PCMK_XE_CRM_CONFIG); create_xml_node(config, PCMK_XE_NODES); create_xml_node(config, PCMK_XE_RESOURCES); create_xml_node(config, PCMK_XE_CONSTRAINTS); #if PCMK__RESOURCE_STICKINESS_DEFAULT != 0 { xmlNode *rsc_defaults = create_xml_node(config, PCMK_XE_RSC_DEFAULTS); xmlNode *meta = create_xml_node(rsc_defaults, PCMK_XE_META_ATTRIBUTES); xmlNode *nvpair = create_xml_node(meta, PCMK_XE_NVPAIR); crm_xml_add(meta, PCMK_XA_ID, "build-resource-defaults"); crm_xml_add(nvpair, PCMK_XA_ID, "build-" PCMK_META_RESOURCE_STICKINESS); crm_xml_add(nvpair, PCMK_XA_NAME, PCMK_META_RESOURCE_STICKINESS); crm_xml_add_int(nvpair, PCMK_XA_VALUE, PCMK__RESOURCE_STICKINESS_DEFAULT); } #endif return cib_root; } static bool cib_acl_enabled(xmlNode *xml, const char *user) { bool rc = FALSE; if(pcmk_acl_required(user)) { const char *value = NULL; GHashTable *options = pcmk__strkey_table(free, free); cib_read_config(options, xml); - value = cib_pref(options, PCMK_OPT_ENABLE_ACL); + value = pcmk__cluster_option(options, PCMK_OPT_ENABLE_ACL); rc = crm_is_true(value); g_hash_table_destroy(options); } crm_trace("CIB ACL is %s", rc ? "enabled" : "disabled"); return rc; } /*! * \internal * \brief Determine whether to perform operations on a scratch copy of the CIB * * \param[in] op CIB operation * \param[in] section CIB section * \param[in] call_options CIB call options * * \return \p true if we should make a copy of the CIB, or \p false otherwise */ static bool should_copy_cib(const char *op, const char *section, int call_options) { if (pcmk_is_set(call_options, cib_dryrun)) { // cib_dryrun implies a scratch copy by definition; no side effects return true; } if (pcmk__str_eq(op, PCMK__CIB_REQUEST_COMMIT_TRANSACT, pcmk__str_none)) { /* Commit-transaction must make a copy for atomicity. We must revert to * the original CIB if the entire transaction cannot be applied * successfully. */ return true; } if (pcmk_is_set(call_options, cib_transaction)) { /* If cib_transaction is set, then we're in the process of committing a * transaction. The commit-transaction request already made a scratch * copy, and we're accumulating changes in that copy. */ return false; } if (pcmk__str_eq(section, PCMK_XE_STATUS, pcmk__str_none)) { /* Copying large CIBs accounts for a huge percentage of our CIB usage, * and this avoids some of it. * * @TODO: Is this safe? See discussion at * https://github.com/ClusterLabs/pacemaker/pull/3094#discussion_r1211400690. */ return false; } // Default behavior is to operate on a scratch copy return true; } int cib_perform_op(cib_t *cib, const char *op, int call_options, cib__op_fn_t fn, bool is_query, const char *section, xmlNode *req, xmlNode *input, bool manage_counters, bool *config_changed, xmlNode **current_cib, xmlNode **result_cib, xmlNode **diff, xmlNode **output) { int rc = pcmk_ok; bool check_schema = true; bool make_copy = true; xmlNode *top = NULL; xmlNode *scratch = NULL; xmlNode *patchset_cib = NULL; xmlNode *local_diff = NULL; const char *user = crm_element_value(req, F_CIB_USER); bool with_digest = false; crm_trace("Begin %s%s%s op", (pcmk_is_set(call_options, cib_dryrun)? "dry run of " : ""), (is_query? "read-only " : ""), op); CRM_CHECK(output != NULL, return -ENOMSG); CRM_CHECK(current_cib != NULL, return -ENOMSG); CRM_CHECK(result_cib != NULL, return -ENOMSG); CRM_CHECK(config_changed != NULL, return -ENOMSG); if(output) { *output = NULL; } *result_cib = NULL; *config_changed = false; if (fn == NULL) { return -EINVAL; } if (is_query) { xmlNode *cib_ro = *current_cib; xmlNode *cib_filtered = NULL; if (cib_acl_enabled(cib_ro, user) && xml_acl_filtered_copy(user, *current_cib, *current_cib, &cib_filtered)) { if (cib_filtered == NULL) { crm_debug("Pre-filtered the entire cib"); return -EACCES; } cib_ro = cib_filtered; crm_log_xml_trace(cib_ro, "filtered"); } rc = (*fn) (op, call_options, section, req, input, cib_ro, result_cib, output); if(output == NULL || *output == NULL) { /* nothing */ } else if(cib_filtered == *output) { cib_filtered = NULL; /* Let them have this copy */ } else if (*output == *current_cib) { /* They already know not to free it */ } else if(cib_filtered && (*output)->doc == cib_filtered->doc) { /* We're about to free the document of which *output is a part */ *output = copy_xml(*output); } else if ((*output)->doc == (*current_cib)->doc) { /* Give them a copy they can free */ *output = copy_xml(*output); } free_xml(cib_filtered); return rc; } make_copy = should_copy_cib(op, section, call_options); if (!make_copy) { /* Conditional on v2 patch style */ scratch = *current_cib; // Make a copy of the top-level element to store version details top = create_xml_node(NULL, (const char *) scratch->name); copy_in_properties(top, scratch); patchset_cib = top; xml_track_changes(scratch, user, NULL, cib_acl_enabled(scratch, user)); rc = (*fn) (op, call_options, section, req, input, scratch, &scratch, output); /* If scratch points to a new object now (for example, after an erase * operation), then *current_cib should point to the same object. */ *current_cib = scratch; } else { scratch = copy_xml(*current_cib); patchset_cib = *current_cib; xml_track_changes(scratch, user, NULL, cib_acl_enabled(scratch, user)); rc = (*fn) (op, call_options, section, req, input, *current_cib, &scratch, output); if ((scratch != NULL) && !xml_tracking_changes(scratch)) { crm_trace("Inferring changes after %s op", op); xml_track_changes(scratch, user, *current_cib, cib_acl_enabled(*current_cib, user)); xml_calculate_changes(*current_cib, scratch); } CRM_CHECK(*current_cib != scratch, return -EINVAL); } xml_acl_disable(scratch); /* Allow the system to make any additional changes */ if (rc == pcmk_ok && scratch == NULL) { rc = -EINVAL; goto done; } else if(rc == pcmk_ok && xml_acl_denied(scratch)) { crm_trace("ACL rejected part or all of the proposed changes"); rc = -EACCES; goto done; } else if (rc != pcmk_ok) { goto done; } /* If the CIB is from a file, we don't need to check that the feature set is * supported. All we care about in that case is the schema version, which * is checked elsewhere. */ if (scratch && (cib == NULL || cib->variant != cib_file)) { const char *new_version = crm_element_value(scratch, PCMK_XA_CRM_FEATURE_SET); rc = pcmk__check_feature_set(new_version); if (rc != pcmk_rc_ok) { pcmk__config_err("Discarding update with feature set '%s' greater than our own '%s'", new_version, CRM_FEATURE_SET); rc = pcmk_rc2legacy(rc); goto done; } } if (patchset_cib != NULL) { int old = 0; int new = 0; crm_element_value_int(scratch, PCMK_XA_ADMIN_EPOCH, &new); crm_element_value_int(patchset_cib, PCMK_XA_ADMIN_EPOCH, &old); if (old > new) { crm_err("%s went backwards: %d -> %d (Opts: %#x)", PCMK_XA_ADMIN_EPOCH, old, new, call_options); crm_log_xml_warn(req, "Bad Op"); crm_log_xml_warn(input, "Bad Data"); rc = -pcmk_err_old_data; } else if (old == new) { crm_element_value_int(scratch, PCMK_XA_EPOCH, &new); crm_element_value_int(patchset_cib, PCMK_XA_EPOCH, &old); if (old > new) { crm_err("%s went backwards: %d -> %d (Opts: %#x)", PCMK_XA_EPOCH, old, new, call_options); crm_log_xml_warn(req, "Bad Op"); crm_log_xml_warn(input, "Bad Data"); rc = -pcmk_err_old_data; } } } crm_trace("Massaging CIB contents"); pcmk__strip_xml_text(scratch); fix_plus_plus_recursive(scratch); if (!make_copy) { /* At this point, patchset_cib is just the PCMK_XE_CIB tag and its * properties. * * The v1 format would barf on this, but we know the v2 patch * format only needs it for the top-level version fields */ local_diff = xml_create_patchset(2, patchset_cib, scratch, config_changed, manage_counters); } else { static time_t expires = 0; time_t tm_now = time(NULL); if (expires < tm_now) { expires = tm_now + 60; /* Validate clients are correctly applying v2-style diffs at most once a minute */ with_digest = true; } local_diff = xml_create_patchset(0, patchset_cib, scratch, config_changed, manage_counters); } pcmk__log_xml_changes(LOG_TRACE, scratch); xml_accept_changes(scratch); if(local_diff) { patchset_process_digest(local_diff, patchset_cib, scratch, with_digest); pcmk__log_xml_patchset(LOG_INFO, local_diff); crm_log_xml_trace(local_diff, "raw patch"); } if (make_copy && (local_diff != NULL)) { // Original to compare against doesn't exist pcmk__if_tracing( { // Validate the calculated patch set int test_rc = pcmk_ok; int format = 1; xmlNode *cib_copy = copy_xml(patchset_cib); crm_element_value_int(local_diff, PCMK_XA_FORMAT, &format); test_rc = xml_apply_patchset(cib_copy, local_diff, manage_counters); if (test_rc != pcmk_ok) { save_xml_to_file(cib_copy, "PatchApply:calculated", NULL); save_xml_to_file(patchset_cib, "PatchApply:input", NULL); save_xml_to_file(scratch, "PatchApply:actual", NULL); save_xml_to_file(local_diff, "PatchApply:diff", NULL); crm_err("v%d patchset error, patch failed to apply: %s " "(%d)", format, pcmk_rc_str(pcmk_legacy2rc(test_rc)), test_rc); } free_xml(cib_copy); }, {} ); } if (pcmk__str_eq(section, PCMK_XE_STATUS, pcmk__str_casei)) { /* Throttle the amount of costly validation we perform due to status updates * a) we don't really care whats in the status section * b) we don't validate any of its contents at the moment anyway */ check_schema = false; } /* === scratch must not be modified after this point === * Exceptions, anything in: static filter_t filter[] = { { 0, PCMK_XA_CRM_DEBUG_ORIGIN }, { 0, PCMK_XA_CIB_LAST_WRITTEN }, { 0, PCMK_XA_UPDATE_ORIGIN }, { 0, PCMK_XA_UPDATE_CLIENT }, { 0, PCMK_XA_UPDATE_USER }, }; */ if (*config_changed && !pcmk_is_set(call_options, cib_no_mtime)) { const char *schema = crm_element_value(scratch, PCMK_XA_VALIDATE_WITH); pcmk__xe_add_last_written(scratch); if (schema) { static int minimum_schema = 0; int current_schema = get_schema_version(schema); if (minimum_schema == 0) { minimum_schema = get_schema_version("pacemaker-1.2"); } /* Does the CIB support the "update-*" attributes... */ if (current_schema >= minimum_schema) { /* Ensure values of origin, client, and user in scratch match * the values in req */ const char *origin = crm_element_value(req, PCMK__XA_SRC); const char *client = crm_element_value(req, F_CIB_CLIENTNAME); if (origin != NULL) { crm_xml_add(scratch, PCMK_XA_UPDATE_ORIGIN, origin); } else { xml_remove_prop(scratch, PCMK_XA_UPDATE_ORIGIN); } if (client != NULL) { crm_xml_add(scratch, PCMK_XA_UPDATE_CLIENT, user); } else { xml_remove_prop(scratch, PCMK_XA_UPDATE_CLIENT); } if (user != NULL) { crm_xml_add(scratch, PCMK_XA_UPDATE_USER, user); } else { xml_remove_prop(scratch, PCMK_XA_UPDATE_USER); } } } } crm_trace("Perform validation: %s", pcmk__btoa(check_schema)); if ((rc == pcmk_ok) && check_schema && !validate_xml(scratch, NULL, true)) { const char *current_schema = crm_element_value(scratch, PCMK_XA_VALIDATE_WITH); crm_warn("Updated CIB does not validate against %s schema", pcmk__s(current_schema, "unspecified")); rc = -pcmk_err_schema_validation; } done: *result_cib = scratch; /* @TODO: This may not work correctly with !make_copy, since we don't * keep the original CIB. */ if ((rc != pcmk_ok) && cib_acl_enabled(patchset_cib, user) && xml_acl_filtered_copy(user, patchset_cib, scratch, result_cib)) { if (*result_cib == NULL) { crm_debug("Pre-filtered the entire cib result"); } free_xml(scratch); } if(diff) { *diff = local_diff; } else { free_xml(local_diff); } free_xml(top); crm_trace("Done"); return rc; } int cib__create_op(cib_t *cib, const char *op, const char *host, const char *section, xmlNode *data, int call_options, const char *user_name, const char *client_name, xmlNode **op_msg) { CRM_CHECK((cib != NULL) && (op_msg != NULL), return -EPROTO); *op_msg = create_xml_node(NULL, T_CIB_COMMAND); if (*op_msg == NULL) { return -EPROTO; } cib->call_id++; if (cib->call_id < 1) { cib->call_id = 1; } crm_xml_add(*op_msg, PCMK__XA_T, T_CIB); crm_xml_add(*op_msg, F_CIB_OPERATION, op); crm_xml_add(*op_msg, F_CIB_HOST, host); crm_xml_add(*op_msg, F_CIB_SECTION, section); crm_xml_add(*op_msg, F_CIB_USER, user_name); crm_xml_add(*op_msg, F_CIB_CLIENTNAME, client_name); crm_xml_add_int(*op_msg, F_CIB_CALLID, cib->call_id); crm_trace("Sending call options: %.8lx, %d", (long)call_options, call_options); crm_xml_add_int(*op_msg, F_CIB_CALLOPTS, call_options); if (data != NULL) { add_message_xml(*op_msg, F_CIB_CALLDATA, data); } if (pcmk_is_set(call_options, cib_inhibit_bcast)) { CRM_CHECK(pcmk_is_set(call_options, cib_scope_local), free_xml(*op_msg); return -EPROTO); } return pcmk_ok; } /*! * \internal * \brief Check whether a CIB request is supported in a transaction * * \param[in] request CIB request * * \return Standard Pacemaker return code */ static int validate_transaction_request(const xmlNode *request) { const char *op = crm_element_value(request, F_CIB_OPERATION); const char *host = crm_element_value(request, F_CIB_HOST); const cib__operation_t *operation = NULL; int rc = cib__get_operation(op, &operation); if (rc != pcmk_rc_ok) { // cib__get_operation() logs error return rc; } if (!pcmk_is_set(operation->flags, cib__op_attr_transaction)) { crm_err("Operation %s is not supported in CIB transactions", op); return EOPNOTSUPP; } if (host != NULL) { crm_err("Operation targeting a specific node (%s) is not supported in " "a CIB transaction", host); return EOPNOTSUPP; } return pcmk_rc_ok; } /*! * \internal * \brief Append a CIB request to a CIB transaction * * \param[in,out] cib CIB client whose transaction to extend * \param[in,out] request Request to add to transaction * * \return Legacy Pacemaker return code */ int cib__extend_transaction(cib_t *cib, xmlNode *request) { int rc = pcmk_rc_ok; CRM_ASSERT((cib != NULL) && (request != NULL)); rc = validate_transaction_request(request); if ((rc == pcmk_rc_ok) && (cib->transaction == NULL)) { rc = pcmk_rc_no_transaction; } if (rc == pcmk_rc_ok) { add_node_copy(cib->transaction, request); } else { const char *op = crm_element_value(request, F_CIB_OPERATION); const char *client_id = NULL; cib->cmds->client_id(cib, NULL, &client_id); crm_err("Failed to add '%s' operation to transaction for client %s: %s", op, pcmk__s(client_id, "(unidentified)"), pcmk_rc_str(rc)); crm_log_xml_info(request, "failed"); } return pcmk_rc2legacy(rc); } void cib_native_callback(cib_t * cib, xmlNode * msg, int call_id, int rc) { xmlNode *output = NULL; cib_callback_client_t *blob = NULL; if (msg != NULL) { crm_element_value_int(msg, F_CIB_RC, &rc); crm_element_value_int(msg, F_CIB_CALLID, &call_id); output = get_message_xml(msg, F_CIB_CALLDATA); } blob = cib__lookup_id(call_id); if (blob == NULL) { crm_trace("No callback found for call %d", call_id); } if (cib == NULL) { crm_debug("No cib object supplied"); } if (rc == -pcmk_err_diff_resync) { /* This is an internal value that clients do not and should not care about */ rc = pcmk_ok; } if (blob && blob->callback && (rc == pcmk_ok || blob->only_success == FALSE)) { crm_trace("Invoking callback %s for call %d", pcmk__s(blob->id, "without ID"), call_id); blob->callback(msg, call_id, rc, output, blob->user_data); } else if (cib && cib->op_callback == NULL && rc != pcmk_ok) { crm_warn("CIB command failed: %s", pcmk_strerror(rc)); crm_log_xml_debug(msg, "Failed CIB Update"); } /* This may free user_data, so do it after the callback */ if (blob) { remove_cib_op_callback(call_id, FALSE); } if (cib && cib->op_callback != NULL) { crm_trace("Invoking global callback for call %d", call_id); cib->op_callback(msg, call_id, rc, output); } crm_trace("OP callback activated for %d", call_id); } void cib_native_notify(gpointer data, gpointer user_data) { xmlNode *msg = user_data; cib_notify_client_t *entry = data; const char *event = NULL; if (msg == NULL) { crm_warn("Skipping callback - NULL message"); return; } event = crm_element_value(msg, PCMK__XA_SUBT); if (entry == NULL) { crm_warn("Skipping callback - NULL callback client"); return; } else if (entry->callback == NULL) { crm_warn("Skipping callback - NULL callback"); return; } else if (!pcmk__str_eq(entry->event, event, pcmk__str_casei)) { crm_trace("Skipping callback - event mismatch %p/%s vs. %s", entry, entry->event, event); return; } crm_trace("Invoking callback for %p/%s event...", entry, event); entry->callback(event, msg); crm_trace("Callback invoked..."); } -static pcmk__cluster_option_t cib_opts[] = { - /* name, legacy name, type, allowed values, - * default value, validator, - * short description, - * long description - */ - { - PCMK_OPT_ENABLE_ACL, NULL, "boolean", NULL, - PCMK_VALUE_FALSE, pcmk__valid_boolean, - N_("Enable Access Control Lists (ACLs) for the CIB"), - NULL - }, - { - PCMK_OPT_CLUSTER_IPC_LIMIT, NULL, "integer", NULL, - "500", pcmk__valid_positive_int, - N_("Maximum IPC message backlog before disconnecting a cluster daemon"), - N_("Raise this if log has \"Evicting client\" messages for cluster daemon" - " PIDs (a good value is the number of resources in the cluster" - " multiplied by the number of nodes).") - }, -}; - -void -cib_metadata(void) -{ - const char *desc_short = "Cluster Information Base manager options"; - const char *desc_long = "Cluster options used by Pacemaker's Cluster " - "Information Base manager"; - - gchar *s = pcmk__format_option_metadata("pacemaker-based", desc_short, - desc_long, cib_opts, - PCMK__NELEM(cib_opts)); - printf("%s", s); - g_free(s); -} - -static void -verify_cib_options(GHashTable *options) -{ - pcmk__validate_cluster_options(options, cib_opts, PCMK__NELEM(cib_opts)); -} - -const char * -cib_pref(GHashTable * options, const char *name) -{ - return pcmk__cluster_option(options, cib_opts, PCMK__NELEM(cib_opts), - name); -} - gboolean cib_read_config(GHashTable * options, xmlNode * current_cib) { xmlNode *config = NULL; crm_time_t *now = NULL; if (options == NULL || current_cib == NULL) { return FALSE; } now = crm_time_new(NULL); g_hash_table_remove_all(options); config = pcmk_find_cib_element(current_cib, PCMK_XE_CRM_CONFIG); if (config) { pe_unpack_nvpairs(current_cib, config, PCMK_XE_CLUSTER_PROPERTY_SET, NULL, options, CIB_OPTIONS_FIRST, TRUE, now, NULL); } - verify_cib_options(options); + pcmk__validate_cluster_options(options); crm_time_free(now); return TRUE; } int cib_internal_op(cib_t * cib, const char *op, const char *host, const char *section, xmlNode * data, xmlNode ** output_data, int call_options, const char *user_name) { int (*delegate) (cib_t * cib, const char *op, const char *host, const char *section, xmlNode * data, xmlNode ** output_data, int call_options, const char *user_name) = cib->delegate_fn; if(user_name == NULL) { user_name = getenv("CIB_user"); } return delegate(cib, op, host, section, data, output_data, call_options, user_name); } /*! * \brief Apply a CIB update patch to a given CIB * * \param[in] event CIB update patch * \param[in] input CIB to patch * \param[out] output Resulting CIB after patch * \param[in] level Log the patch at this log level (unless LOG_CRIT) * * \return Legacy Pacemaker return code * \note sbd calls this function */ int cib_apply_patch_event(xmlNode *event, xmlNode *input, xmlNode **output, int level) { int rc = pcmk_err_generic; xmlNode *diff = NULL; CRM_ASSERT(event); CRM_ASSERT(input); CRM_ASSERT(output); crm_element_value_int(event, F_CIB_RC, &rc); diff = get_message_xml(event, F_CIB_UPDATE_RESULT); if (rc < pcmk_ok || diff == NULL) { return rc; } if (level > LOG_CRIT) { pcmk__log_xml_patchset(level, diff); } if (input != NULL) { rc = cib_process_diff(NULL, cib_none, NULL, event, diff, input, output, NULL); if (rc != pcmk_ok) { crm_debug("Update didn't apply: %s (%d) %p", pcmk_strerror(rc), rc, *output); if (rc == -pcmk_err_old_data) { crm_trace("Masking error, we already have the supplied update"); return pcmk_ok; } free_xml(*output); *output = NULL; return rc; } } return rc; } #define log_signon_query_err(out, fmt, args...) do { \ if (out != NULL) { \ out->err(out, fmt, ##args); \ } else { \ crm_err(fmt, ##args); \ } \ } while (0) int cib__signon_query(pcmk__output_t *out, cib_t **cib, xmlNode **cib_object) { int rc = pcmk_rc_ok; cib_t *cib_conn = NULL; CRM_ASSERT(cib_object != NULL); if (cib == NULL) { cib_conn = cib_new(); } else { if (*cib == NULL) { *cib = cib_new(); } cib_conn = *cib; } if (cib_conn == NULL) { return ENOMEM; } if (cib_conn->state == cib_disconnected) { rc = cib_conn->cmds->signon(cib_conn, crm_system_name, cib_command); rc = pcmk_legacy2rc(rc); } if (rc != pcmk_rc_ok) { log_signon_query_err(out, "Could not connect to the CIB: %s", pcmk_rc_str(rc)); goto done; } if (out != NULL) { out->transient(out, "Querying CIB..."); } rc = cib_conn->cmds->query(cib_conn, NULL, cib_object, cib_scope_local|cib_sync_call); rc = pcmk_legacy2rc(rc); if (rc != pcmk_rc_ok) { log_signon_query_err(out, "CIB query failed: %s", pcmk_rc_str(rc)); } done: if (cib == NULL) { cib__clean_up_connection(&cib_conn); } if ((rc == pcmk_rc_ok) && (*cib_object == NULL)) { return pcmk_rc_no_input; } return rc; } int cib__clean_up_connection(cib_t **cib) { int rc; if (*cib == NULL) { return pcmk_rc_ok; } rc = (*cib)->cmds->signoff(*cib); cib_delete(*cib); *cib = NULL; return pcmk_legacy2rc(rc); } // Deprecated functions kept only for backward API compatibility // LCOV_EXCL_START #include xmlNode * cib_get_generation(cib_t * cib) { xmlNode *the_cib = NULL; xmlNode *generation = create_xml_node(NULL, XML_CIB_TAG_GENERATION_TUPPLE); cib->cmds->query(cib, NULL, &the_cib, cib_scope_local | cib_sync_call); if (the_cib != NULL) { copy_in_properties(generation, the_cib); free_xml(the_cib); } return generation; } const char * get_object_path(const char *object_type) { return pcmk_cib_xpath_for(object_type); } const char * get_object_parent(const char *object_type) { return pcmk_cib_parent_name_for(object_type); } xmlNode * get_object_root(const char *object_type, xmlNode *the_root) { return pcmk_find_cib_element(the_root, object_type); } +const char * +cib_pref(GHashTable * options, const char *name) +{ + return pcmk__cluster_option(options, name); +} + +void +cib_metadata(void) +{ + const char *name = "pacemaker-based"; + const char *desc_short = "Cluster Information Base manager options"; + const char *desc_long = "Cluster options used by Pacemaker's Cluster " + "Information Base manager"; + + gchar *s = pcmk__cluster_option_metadata(name, desc_short, desc_long, + pcmk__opt_context_based); + + printf("%s", s); + g_free(s); +} + // LCOV_EXCL_STOP // End deprecated API diff --git a/lib/common/options.c b/lib/common/options.c index 20afde6eee..274e21a41f 100644 --- a/lib/common/options.c +++ b/lib/common/options.c @@ -1,596 +1,1152 @@ /* * 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. */ #ifndef _GNU_SOURCE # define _GNU_SOURCE #endif #include #include #include #include #include #include #include #include void pcmk__cli_help(char cmd) { if (cmd == 'v' || cmd == '$') { printf("Pacemaker %s\n", PACEMAKER_VERSION); printf("Written by Andrew Beekhof and " "the Pacemaker project contributors\n"); } else if (cmd == '!') { printf("Pacemaker %s (Build: %s): %s\n", PACEMAKER_VERSION, BUILD_VERSION, CRM_FEATURES); } crm_exit(CRM_EX_OK); while(1); // above does not return } +/* + * Option metadata + */ + +static pcmk__cluster_option_t cluster_options[] = { + /* name, old name, type, allowed values, + * default value, validator, + * context, + * short description, + * long description + */ + { + PCMK_OPT_DC_VERSION, NULL, "string", NULL, + PCMK__VALUE_NONE, NULL, + pcmk__opt_context_controld, + N_("Pacemaker version on cluster node elected Designated Controller " + "(DC)"), + N_("Includes a hash which identifies the exact changeset the code was " + "built from. Used for diagnostic purposes."), + }, + { + PCMK_OPT_CLUSTER_INFRASTRUCTURE, NULL, "string", NULL, + "corosync", NULL, + pcmk__opt_context_controld, + N_("The messaging stack on which Pacemaker is currently running"), + N_("Used for informational and diagnostic purposes."), + }, + { + PCMK_OPT_CLUSTER_NAME, NULL, "string", NULL, + NULL, NULL, + pcmk__opt_context_controld, + N_("An arbitrary name for the cluster"), + N_("This optional value is mostly for users' convenience as desired " + "in administration, but may also be used in Pacemaker " + "configuration rules via the #cluster-name node attribute, and " + "by higher-level tools and resource agents."), + }, + { + PCMK_OPT_DC_DEADTIME, NULL, "time", NULL, + "20s", pcmk__valid_interval_spec, + pcmk__opt_context_controld, + N_("How long to wait for a response from other nodes during start-up"), + N_("The optimal value will depend on the speed and load of your " + "network and the type of switches used."), + }, + { + PCMK_OPT_CLUSTER_RECHECK_INTERVAL, NULL, "time", + N_("Zero disables polling, while positive values are an interval in " + "seconds (unless other units are specified, for example \"5min\")"), + "15min", pcmk__valid_interval_spec, + pcmk__opt_context_controld, + N_("Polling interval to recheck cluster state and evaluate rules " + "with date specifications"), + N_("Pacemaker is primarily event-driven, and looks ahead to know when " + "to recheck cluster state for failure timeouts and most time-based " + "rules. However, it will also recheck the cluster after this " + "amount of inactivity, to evaluate rules with date specifications " + "and serve as a fail-safe for certain types of scheduler bugs."), + }, + { + PCMK_OPT_FENCE_REACTION, NULL, "select", "stop, panic", + "stop", NULL, + pcmk__opt_context_controld, + N_("How a cluster node should react if notified of its own fencing"), + N_("A cluster node may receive notification of its own fencing if " + "fencing is misconfigured, or if fabric fencing is in use that " + "doesn't cut cluster communication. Use \"stop\" to attempt to " + "immediately stop Pacemaker and stay stopped, or \"panic\" to " + "attempt to immediately reboot the local node, falling back to " + "stop on failure."), + }, + { + PCMK_OPT_ELECTION_TIMEOUT, NULL, "time", NULL, + "2min", pcmk__valid_interval_spec, + pcmk__opt_context_controld, + N_("*** Advanced Use Only ***"), + N_("Declare an election failed if it is not decided within this much " + "time. If you need to adjust this value, it probably indicates " + "the presence of a bug."), + }, + { + PCMK_OPT_SHUTDOWN_ESCALATION, NULL, "time", NULL, + "20min", pcmk__valid_interval_spec, + pcmk__opt_context_controld, + N_("*** Advanced Use Only ***"), + N_("Exit immediately if shutdown does not complete within this much " + "time. If you need to adjust this value, it probably indicates " + "the presence of a bug."), + }, + { + PCMK_OPT_JOIN_INTEGRATION_TIMEOUT, "crmd-integration-timeout", "time", + NULL, + "3min", pcmk__valid_interval_spec, + pcmk__opt_context_controld, + N_("*** Advanced Use Only ***"), + N_("If you need to adjust this value, it probably indicates " + "the presence of a bug."), + }, + { + PCMK_OPT_JOIN_FINALIZATION_TIMEOUT, "crmd-finalization-timeout", + "time", NULL, + "30min", pcmk__valid_interval_spec, + pcmk__opt_context_controld, + N_("*** Advanced Use Only ***"), + N_("If you need to adjust this value, it probably indicates " + "the presence of a bug."), + }, + { + PCMK_OPT_TRANSITION_DELAY, "crmd-transition-delay", "time", NULL, + "0s", pcmk__valid_interval_spec, + pcmk__opt_context_controld, + N_("*** Advanced Use Only *** " + "Enabling this option will slow down cluster recovery under all " + "conditions"), + N_("Delay cluster recovery for this much time to allow for additional " + "events to occur. Useful if your configuration is sensitive to " + "the order in which ping updates arrive."), + }, + { + PCMK_OPT_NO_QUORUM_POLICY, NULL, "select", + "stop, freeze, ignore, demote, suicide", + "stop", pcmk__valid_no_quorum_policy, + pcmk__opt_context_schedulerd, + N_("What to do when the cluster does not have quorum"), + NULL, + }, + { + PCMK_OPT_SHUTDOWN_LOCK, NULL, "boolean", NULL, + PCMK_VALUE_FALSE, pcmk__valid_boolean, + pcmk__opt_context_schedulerd, + N_("Whether to lock resources to a cleanly shut down node"), + N_("When true, resources active on a node when it is cleanly shut down " + "are kept \"locked\" to that node (not allowed to run elsewhere) " + "until they start again on that node after it rejoins (or for at " + "most shutdown-lock-limit, if set). Stonith resources and " + "Pacemaker Remote connections are never locked. Clone and bundle " + "instances and the promoted role of promotable clones are " + "currently never locked, though support could be added in a future " + "release."), + }, + { + PCMK_OPT_SHUTDOWN_LOCK_LIMIT, NULL, "time", NULL, + "0", pcmk__valid_interval_spec, + pcmk__opt_context_schedulerd, + N_("Do not lock resources to a cleanly shut down node longer than " + "this"), + N_("If shutdown-lock is true and this is set to a nonzero time " + "duration, shutdown locks will expire after this much time has " + "passed since the shutdown was initiated, even if the node has not " + "rejoined."), + }, + { + PCMK_OPT_ENABLE_ACL, NULL, "boolean", NULL, + PCMK_VALUE_FALSE, pcmk__valid_boolean, + pcmk__opt_context_based, + N_("Enable Access Control Lists (ACLs) for the CIB"), + NULL, + }, + { + PCMK_OPT_SYMMETRIC_CLUSTER, NULL, "boolean", NULL, + PCMK_VALUE_TRUE, pcmk__valid_boolean, + pcmk__opt_context_schedulerd, + N_("Whether resources can run on any node by default"), + NULL, + }, + { + PCMK_OPT_MAINTENANCE_MODE, NULL, "boolean", NULL, + PCMK_VALUE_FALSE, pcmk__valid_boolean, + pcmk__opt_context_schedulerd, + N_("Whether the cluster should refrain from monitoring, starting, and " + "stopping resources"), + NULL, + }, + { + PCMK_OPT_START_FAILURE_IS_FATAL, NULL, "boolean", NULL, + PCMK_VALUE_TRUE, pcmk__valid_boolean, + pcmk__opt_context_schedulerd, + N_("Whether a start failure should prevent a resource from being " + "recovered on the same node"), + N_("When true, the cluster will immediately ban a resource from a node " + "if it fails to start there. When false, the cluster will instead " + "check the resource's fail count against its migration-threshold.") + }, + { + PCMK_OPT_ENABLE_STARTUP_PROBES, NULL, "boolean", NULL, + PCMK_VALUE_TRUE, pcmk__valid_boolean, + pcmk__opt_context_schedulerd, + N_("Whether the cluster should check for active resources during " + "start-up"), + NULL, + }, + + // Fencing-related options + { + PCMK_OPT_STONITH_ENABLED, NULL, "boolean", NULL, + PCMK_VALUE_TRUE, pcmk__valid_boolean, + pcmk__opt_context_schedulerd, + N_("*** Advanced Use Only *** " + "Whether nodes may be fenced as part of recovery"), + N_("If false, unresponsive nodes are immediately assumed to be " + "harmless, and resources that were active on them may be recovered " + "elsewhere. This can result in a \"split-brain\" situation, " + "potentially leading to data loss and/or service unavailability."), + }, + { + PCMK_OPT_STONITH_ACTION, NULL, "select", "reboot, off, poweroff", + PCMK_ACTION_REBOOT, pcmk__is_fencing_action, + pcmk__opt_context_schedulerd, + N_("Action to send to fence device when a node needs to be fenced " + "(\"poweroff\" is a deprecated alias for \"off\")"), + NULL, + }, + { + PCMK_OPT_STONITH_TIMEOUT, NULL, "time", NULL, + "60s", pcmk__valid_interval_spec, + pcmk__opt_context_schedulerd, + N_("How long to wait for on, off, and reboot fence actions to complete " + "by default"), + NULL, + }, + { + PCMK_OPT_HAVE_WATCHDOG, NULL, "boolean", NULL, + PCMK_VALUE_FALSE, pcmk__valid_boolean, + pcmk__opt_context_schedulerd, + N_("Whether watchdog integration is enabled"), + N_("This is set automatically by the cluster according to whether SBD " + "is detected to be in use. User-configured values are ignored. " + "The value `true` is meaningful if diskless SBD is used and " + "`stonith-watchdog-timeout` is nonzero. In that case, if fencing " + "is required, watchdog-based self-fencing will be performed via " + "SBD without requiring a fencing resource explicitly configured."), + }, + { + /* @COMPAT Currently, unparsable values default to -1 (auto-calculate), + * while missing values default to 0 (disable). All values are accepted + * (unless the controller finds that the value conflicts with the + * SBD_WATCHDOG_TIMEOUT). + * + * At a compatibility break: properly validate as a timeout, let + * either negative values or a particular string like "auto" mean auto- + * calculate, and use 0 as the single default for when the option either + * is unset or fails to validate. + */ + PCMK_OPT_STONITH_WATCHDOG_TIMEOUT, NULL, "time", NULL, + "0", NULL, + pcmk__opt_context_controld, + N_("How long before nodes can be assumed to be safely down when " + "watchdog-based self-fencing via SBD is in use"), + N_("If this is set to a positive value, lost nodes are assumed to " + "self-fence using watchdog-based SBD within this much time. This " + "does not require a fencing resource to be explicitly configured, " + "though a fence_watchdog resource can be configured, to limit use " + "to specific nodes. If this is set to 0 (the default), the cluster " + "will never assume watchdog-based self-fencing. If this is set to a " + "negative value, the cluster will use twice the local value of the " + "`SBD_WATCHDOG_TIMEOUT` environment variable if that is positive, " + "or otherwise treat this as 0. WARNING: When used, this timeout " + "must be larger than `SBD_WATCHDOG_TIMEOUT` on all nodes that use " + "watchdog-based SBD, and Pacemaker will refuse to start on any of " + "those nodes where this is not true for the local value or SBD is " + "not active. When this is set to a negative value, " + "`SBD_WATCHDOG_TIMEOUT` must be set to the same value on all nodes " + "that use SBD, otherwise data corruption or loss could occur."), + }, + { + PCMK_OPT_STONITH_MAX_ATTEMPTS, NULL, "integer", NULL, + "10", pcmk__valid_positive_int, + pcmk__opt_context_controld, + N_("How many times fencing can fail before it will no longer be " + "immediately re-attempted on a target"), + NULL, + }, + { + PCMK_OPT_CONCURRENT_FENCING, NULL, "boolean", NULL, + PCMK__CONCURRENT_FENCING_DEFAULT, pcmk__valid_boolean, + pcmk__opt_context_schedulerd, + N_("Allow performing fencing operations in parallel"), + NULL, + }, + { + PCMK_OPT_STARTUP_FENCING, NULL, "boolean", NULL, + PCMK_VALUE_TRUE, pcmk__valid_boolean, + pcmk__opt_context_schedulerd, + N_("*** Advanced Use Only *** " + "Whether to fence unseen nodes at start-up"), + N_("Setting this to false may lead to a \"split-brain\" situation, " + "potentially leading to data loss and/or service unavailability."), + }, + { + PCMK_OPT_PRIORITY_FENCING_DELAY, NULL, "time", NULL, + "0", pcmk__valid_interval_spec, + pcmk__opt_context_schedulerd, + N_("Apply fencing delay targeting the lost nodes with the highest " + "total resource priority"), + N_("Apply specified delay for the fencings that are targeting the lost " + "nodes with the highest total resource priority in case we don't " + "have the majority of the nodes in our cluster partition, so that " + "the more significant nodes potentially win any fencing match, " + "which is especially meaningful under split-brain of 2-node " + "cluster. A promoted resource instance takes the base priority + 1 " + "on calculation if the base priority is not 0. Any static/random " + "delays that are introduced by `pcmk_delay_base/max` configured " + "for the corresponding fencing resources will be added to this " + "delay. This delay should be significantly greater than, safely " + "twice, the maximum `pcmk_delay_base/max`. By default, priority " + "fencing delay is disabled."), + }, + { + PCMK_OPT_NODE_PENDING_TIMEOUT, NULL, "time", NULL, + "0", pcmk__valid_interval_spec, + pcmk__opt_context_schedulerd, + N_("How long to wait for a node that has joined the cluster to join " + "the controller process group"), + N_("Fence nodes that do not join the controller process group within " + "this much time after joining the cluster, to allow the cluster " + "to continue managing resources. A value of 0 means never fence " + "pending nodes. Setting the value to 2h means fence nodes after " + "2 hours."), + }, + { + PCMK_OPT_CLUSTER_DELAY, NULL, "time", NULL, + "60s", pcmk__valid_interval_spec, + pcmk__opt_context_schedulerd, + N_("Maximum time for node-to-node communication"), + N_("The node elected Designated Controller (DC) will consider an action " + "failed if it does not get a response from the node executing the " + "action within this time (after considering the action's own " + "timeout). The \"correct\" value will depend on the speed and " + "load of your network and cluster nodes.") + }, + + // Limits + { + PCMK_OPT_LOAD_THRESHOLD, NULL, "percentage", NULL, + "80%", pcmk__valid_percentage, + pcmk__opt_context_controld, + N_("Maximum amount of system load that should be used by cluster " + "nodes"), + N_("The cluster will slow down its recovery process when the amount of " + "system resources used (currently CPU) approaches this limit"), + }, + { + PCMK_OPT_NODE_ACTION_LIMIT, NULL, "integer", NULL, + "0", pcmk__valid_int, + pcmk__opt_context_controld, + N_("Maximum number of jobs that can be scheduled per node (defaults to " + "2x cores)"), + NULL, + }, + { + PCMK_OPT_BATCH_LIMIT, NULL, "integer", NULL, + "0", pcmk__valid_int, + pcmk__opt_context_schedulerd, + N_("Maximum number of jobs that the cluster may execute in parallel " + "across all nodes"), + N_("The \"correct\" value will depend on the speed and load of your " + "network and cluster nodes. If set to 0, the cluster will " + "impose a dynamically calculated limit when any node has a " + "high load."), + }, + { + PCMK_OPT_MIGRATION_LIMIT, NULL, "integer", NULL, + "-1", pcmk__valid_int, + pcmk__opt_context_schedulerd, + N_("The number of live migration actions that the cluster is allowed " + "to execute in parallel on a node (-1 means no limit)"), + NULL, + }, + { + PCMK_OPT_CLUSTER_IPC_LIMIT, NULL, "integer", NULL, + "500", pcmk__valid_positive_int, + pcmk__opt_context_based, + N_("Maximum IPC message backlog before disconnecting a cluster daemon"), + N_("Raise this if log has \"Evicting client\" messages for cluster " + "daemon PIDs (a good value is the number of resources in the " + "cluster multiplied by the number of nodes)."), + }, + + // Orphans and stopping + { + PCMK_OPT_STOP_ALL_RESOURCES, NULL, "boolean", NULL, + PCMK_VALUE_FALSE, pcmk__valid_boolean, + pcmk__opt_context_schedulerd, + N_("Whether the cluster should stop all active resources"), + NULL, + }, + { + PCMK_OPT_STOP_ORPHAN_RESOURCES, NULL, "boolean", NULL, + PCMK_VALUE_TRUE, pcmk__valid_boolean, + pcmk__opt_context_schedulerd, + N_("Whether to stop resources that were removed from the " + "configuration"), + NULL, + }, + { + PCMK_OPT_STOP_ORPHAN_ACTIONS, NULL, "boolean", NULL, + PCMK_VALUE_TRUE, pcmk__valid_boolean, + pcmk__opt_context_schedulerd, + N_("Whether to cancel recurring actions removed from the " + "configuration"), + NULL, + }, + { + PCMK__OPT_REMOVE_AFTER_STOP, NULL, "boolean", NULL, + PCMK_VALUE_FALSE, pcmk__valid_boolean, + pcmk__opt_context_schedulerd, + N_("*** Deprecated *** " + "Whether to remove stopped resources from the executor"), + N_("Values other than default are poorly tested and potentially " + "dangerous. This option will be removed in a future release."), + }, + + // Storing inputs + { + PCMK_OPT_PE_ERROR_SERIES_MAX, NULL, "integer", NULL, + "-1", pcmk__valid_int, + pcmk__opt_context_schedulerd, + N_("The number of scheduler inputs resulting in errors to save"), + N_("Zero to disable, -1 to store unlimited."), + }, + { + PCMK_OPT_PE_WARN_SERIES_MAX, NULL, "integer", NULL, + "5000", pcmk__valid_int, + pcmk__opt_context_schedulerd, + N_("The number of scheduler inputs resulting in warnings to save"), + N_("Zero to disable, -1 to store unlimited."), + }, + { + PCMK_OPT_PE_INPUT_SERIES_MAX, NULL, "integer", NULL, + "4000", pcmk__valid_int, + pcmk__opt_context_schedulerd, + N_("The number of scheduler inputs without errors or warnings to save"), + N_("Zero to disable, -1 to store unlimited."), + }, + + // Node health + { + PCMK_OPT_NODE_HEALTH_STRATEGY, NULL, "select", + PCMK__VALUE_NONE ", " PCMK__VALUE_MIGRATE_ON_RED ", " + PCMK__VALUE_ONLY_GREEN ", " PCMK__VALUE_PROGRESSIVE ", " + PCMK__VALUE_CUSTOM, + PCMK__VALUE_NONE, pcmk__validate_health_strategy, + pcmk__opt_context_schedulerd, + N_("How cluster should react to node health attributes"), + N_("Requires external entities to create node attributes (named with " + "the prefix \"#health\") with values \"red\", \"yellow\", or " + "\"green\".") + }, + { + PCMK_OPT_NODE_HEALTH_BASE, NULL, "integer", NULL, + "0", pcmk__valid_int, + pcmk__opt_context_schedulerd, + N_("Base health score assigned to a node"), + N_("Only used when \"node-health-strategy\" is set to " + "\"progressive\"."), + }, + { + PCMK_OPT_NODE_HEALTH_GREEN, NULL, "integer", NULL, + "0", pcmk__valid_int, + pcmk__opt_context_schedulerd, + N_("The score to use for a node health attribute whose value is " + "\"green\""), + N_("Only used when \"node-health-strategy\" is set to \"custom\" or " + "\"progressive\"."), + }, + { + PCMK_OPT_NODE_HEALTH_YELLOW, NULL, "integer", NULL, + "0", pcmk__valid_int, + pcmk__opt_context_schedulerd, + N_("The score to use for a node health attribute whose value is " + "\"yellow\""), + N_("Only used when \"node-health-strategy\" is set to \"custom\" or " + "\"progressive\"."), + }, + { + PCMK_OPT_NODE_HEALTH_RED, NULL, "integer", NULL, + "-INFINITY", pcmk__valid_int, + pcmk__opt_context_schedulerd, + N_("The score to use for a node health attribute whose value is " + "\"red\""), + N_("Only used when \"node-health-strategy\" is set to \"custom\" or " + "\"progressive\".") + }, + + // Placement strategy + { + PCMK_OPT_PLACEMENT_STRATEGY, NULL, "select", + "default, utilization, minimal, balanced", + "default", pcmk__valid_placement_strategy, + pcmk__opt_context_schedulerd, + N_("How the cluster should allocate resources to nodes"), + NULL, + }, +}; + + /* * Environment variable option handling */ /*! * \internal * \brief Get the value of a Pacemaker environment variable option * * If an environment variable option is set, with either a PCMK_ or (for * backward compatibility) HA_ prefix, log and return the value. * * \param[in] option Environment variable name (without prefix) * * \return Value of environment variable option, or NULL in case of * option name too long or value not found */ const char * pcmk__env_option(const char *option) { const char *const prefixes[] = {"PCMK_", "HA_"}; char env_name[NAME_MAX]; const char *value = NULL; CRM_CHECK(!pcmk__str_empty(option), return NULL); for (int i = 0; i < PCMK__NELEM(prefixes); i++) { int rv = snprintf(env_name, NAME_MAX, "%s%s", prefixes[i], option); if (rv < 0) { crm_err("Failed to write %s%s to buffer: %s", prefixes[i], option, strerror(errno)); return NULL; } if (rv >= sizeof(env_name)) { crm_trace("\"%s%s\" is too long", prefixes[i], option); continue; } value = getenv(env_name); if (value != NULL) { crm_trace("Found %s = %s", env_name, value); return value; } } crm_trace("Nothing found for %s", option); return NULL; } /*! * \brief Set or unset a Pacemaker environment variable option * * Set an environment variable option with a \c "PCMK_" prefix and optionally * an \c "HA_" prefix for backward compatibility. * * \param[in] option Environment variable name (without prefix) * \param[in] value New value (or NULL to unset) * \param[in] compat If false and \p value is not \c NULL, set only * \c "PCMK_