diff --git a/daemons/controld/controld_fencing.c b/daemons/controld/controld_fencing.c index 3221272fc0..7b7a2f05b7 100644 --- a/daemons/controld/controld_fencing.c +++ b/daemons/controld/controld_fencing.c @@ -1,1055 +1,1051 @@ /* * Copyright 2004-2022 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU General Public License version 2 * or later (GPLv2+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include static void tengine_stonith_history_synced(stonith_t *st, stonith_event_t *st_event); /* * stonith failure counting * * We don't want to get stuck in a permanent fencing loop. Keep track of the * number of fencing failures for each target node, and the most we'll restart a * transition for. */ struct st_fail_rec { int count; }; static bool fence_reaction_panic = FALSE; static unsigned long int stonith_max_attempts = 10; static GHashTable *stonith_failures = NULL; void update_stonith_max_attempts(const char *value) { stonith_max_attempts = char2score(value); if (stonith_max_attempts < 1UL) { stonith_max_attempts = 10UL; } } void set_fence_reaction(const char *reaction_s) { if (pcmk__str_eq(reaction_s, "panic", pcmk__str_casei)) { fence_reaction_panic = TRUE; } else { if (!pcmk__str_eq(reaction_s, "stop", pcmk__str_casei)) { crm_warn("Invalid value '%s' for %s, using 'stop'", reaction_s, XML_CONFIG_ATTR_FENCE_REACTION); } fence_reaction_panic = FALSE; } } static gboolean too_many_st_failures(const char *target) { GHashTableIter iter; const char *key = NULL; struct st_fail_rec *value = NULL; if (stonith_failures == NULL) { return FALSE; } if (target == NULL) { g_hash_table_iter_init(&iter, stonith_failures); while (g_hash_table_iter_next(&iter, (gpointer *) &key, (gpointer *) &value)) { if (value->count >= stonith_max_attempts) { target = (const char*)key; goto too_many; } } } else { value = g_hash_table_lookup(stonith_failures, target); if ((value != NULL) && (value->count >= stonith_max_attempts)) { goto too_many; } } return FALSE; too_many: crm_warn("Too many failures (%d) to fence %s, giving up", value->count, target); return TRUE; } /*! * \internal * \brief Reset a stonith fail count * * \param[in] target Name of node to reset, or NULL for all */ void st_fail_count_reset(const char *target) { if (stonith_failures == NULL) { return; } if (target) { struct st_fail_rec *rec = NULL; rec = g_hash_table_lookup(stonith_failures, target); if (rec) { rec->count = 0; } } else { GHashTableIter iter; const char *key = NULL; struct st_fail_rec *rec = NULL; g_hash_table_iter_init(&iter, stonith_failures); while (g_hash_table_iter_next(&iter, (gpointer *) &key, (gpointer *) &rec)) { rec->count = 0; } } } static void st_fail_count_increment(const char *target) { struct st_fail_rec *rec = NULL; if (stonith_failures == NULL) { stonith_failures = pcmk__strkey_table(free, free); } rec = g_hash_table_lookup(stonith_failures, target); if (rec) { rec->count++; } else { rec = malloc(sizeof(struct st_fail_rec)); if(rec == NULL) { return; } rec->count = 1; g_hash_table_insert(stonith_failures, strdup(target), rec); } } /* end stonith fail count functions */ static void cib_fencing_updated(xmlNode *msg, int call_id, int rc, xmlNode *output, void *user_data) { if (rc < pcmk_ok) { crm_err("Fencing update %d for %s: failed - %s (%d)", call_id, (char *)user_data, pcmk_strerror(rc), rc); crm_log_xml_warn(msg, "Failed update"); abort_transition(INFINITY, tg_shutdown, "CIB update failed", NULL); } else { crm_info("Fencing update %d for %s: complete", call_id, (char *)user_data); } } static void send_stonith_update(crm_action_t *action, const char *target, const char *uuid) { int rc = pcmk_ok; crm_node_t *peer = NULL; /* We (usually) rely on the membership layer to do node_update_cluster, * and the peer status callback to do node_update_peer, because the node * might have already rejoined before we get the stonith result here. */ int flags = node_update_join | node_update_expected; /* zero out the node-status & remove all LRM status info */ xmlNode *node_state = NULL; CRM_CHECK(target != NULL, return); CRM_CHECK(uuid != NULL, return); /* Make sure the membership and join caches are accurate */ peer = crm_get_peer_full(0, target, CRM_GET_PEER_ANY); CRM_CHECK(peer != NULL, return); if (peer->state == NULL) { /* Usually, we rely on the membership layer to update the cluster state * in the CIB. However, if the node has never been seen, do it here, so * the node is not considered unclean. */ flags |= node_update_cluster; } if (peer->uuid == NULL) { crm_info("Recording uuid '%s' for node '%s'", uuid, target); peer->uuid = strdup(uuid); } crmd_peer_down(peer, TRUE); /* Generate a node state update for the CIB */ node_state = create_node_state_update(peer, flags, NULL, __func__); /* we have to mark whether or not remote nodes have already been fenced */ if (peer->flags & crm_remote_node) { char *now_s = pcmk__ttoa(time(NULL)); crm_xml_add(node_state, XML_NODE_IS_FENCED, now_s); free(now_s); } /* Force our known ID */ crm_xml_add(node_state, XML_ATTR_UUID, uuid); rc = fsa_cib_conn->cmds->update(fsa_cib_conn, XML_CIB_TAG_STATUS, node_state, cib_quorum_override | cib_scope_local | cib_can_create); /* Delay processing the trigger until the update completes */ crm_debug("Sending fencing update %d for %s", rc, target); fsa_register_cib_callback(rc, FALSE, strdup(target), cib_fencing_updated); /* Make sure it sticks */ /* fsa_cib_conn->cmds->bump_epoch(fsa_cib_conn, cib_quorum_override|cib_scope_local); */ controld_delete_node_state(peer->uname, controld_section_all, cib_scope_local); free_xml(node_state); return; } /*! * \internal * \brief Abort transition due to stonith failure * * \param[in] abort_action Whether to restart or stop transition * \param[in] target Don't restart if this (NULL for any) has too many failures * \param[in] reason Log this stonith action XML as abort reason (or NULL) */ static void abort_for_stonith_failure(enum transition_action abort_action, const char *target, xmlNode *reason) { /* If stonith repeatedly fails, we eventually give up on starting a new * transition for that reason. */ if ((abort_action != tg_stop) && too_many_st_failures(target)) { abort_action = tg_stop; } abort_transition(INFINITY, abort_action, "Stonith failed", reason); } /* * stonith cleanup list * * If the DC is shot, proper notifications might not go out. * The stonith cleanup list allows the cluster to (re-)send * notifications once a new DC is elected. */ static GList *stonith_cleanup_list = NULL; /*! * \internal * \brief Add a node to the stonith cleanup list * * \param[in] target Name of node to add */ void add_stonith_cleanup(const char *target) { stonith_cleanup_list = g_list_append(stonith_cleanup_list, strdup(target)); } /*! * \internal * \brief Remove a node from the stonith cleanup list * * \param[in] Name of node to remove */ void remove_stonith_cleanup(const char *target) { GList *iter = stonith_cleanup_list; while (iter != NULL) { GList *tmp = iter; char *iter_name = tmp->data; iter = iter->next; if (pcmk__str_eq(target, iter_name, pcmk__str_casei)) { crm_trace("Removing %s from the cleanup list", iter_name); stonith_cleanup_list = g_list_delete_link(stonith_cleanup_list, tmp); free(iter_name); } } } /*! * \internal * \brief Purge all entries from the stonith cleanup list */ void purge_stonith_cleanup() { if (stonith_cleanup_list) { GList *iter = NULL; for (iter = stonith_cleanup_list; iter != NULL; iter = iter->next) { char *target = iter->data; crm_info("Purging %s from stonith cleanup list", target); free(target); } g_list_free(stonith_cleanup_list); stonith_cleanup_list = NULL; } } /*! * \internal * \brief Send stonith updates for all entries in cleanup list, then purge it */ void execute_stonith_cleanup() { GList *iter; for (iter = stonith_cleanup_list; iter != NULL; iter = iter->next) { char *target = iter->data; crm_node_t *target_node = crm_get_peer(0, target); const char *uuid = crm_peer_uuid(target_node); crm_notice("Marking %s, target of a previous stonith action, as clean", target); send_stonith_update(NULL, target, uuid); free(target); } g_list_free(stonith_cleanup_list); stonith_cleanup_list = NULL; } /* end stonith cleanup list functions */ /* stonith API client * * Functions that need to interact directly with the fencer via its API */ static stonith_t *stonith_api = NULL; static crm_trigger_t *stonith_reconnect = NULL; static char *te_client_id = NULL; static gboolean fail_incompletable_stonith(crm_graph_t *graph) { GList *lpc = NULL; const char *task = NULL; xmlNode *last_action = NULL; if (graph == NULL) { return FALSE; } for (lpc = graph->synapses; lpc != NULL; lpc = lpc->next) { GList *lpc2 = NULL; synapse_t *synapse = (synapse_t *) lpc->data; if (pcmk_is_set(synapse->flags, pcmk__synapse_confirmed)) { continue; } for (lpc2 = synapse->actions; lpc2 != NULL; lpc2 = lpc2->next) { crm_action_t *action = (crm_action_t *) lpc2->data; if (action->type != action_type_crm || pcmk_is_set(action->flags, pcmk__graph_action_confirmed)) { continue; } task = crm_element_value(action->xml, XML_LRM_ATTR_TASK); if (task && pcmk__str_eq(task, CRM_OP_FENCE, pcmk__str_casei)) { crm__set_graph_action_flags(action, pcmk__graph_action_failed); last_action = action->xml; pcmk__update_graph(graph, action); crm_notice("Failing action %d (%s): fencer terminated", action->id, ID(action->xml)); } } } if (last_action != NULL) { crm_warn("Fencer failure resulted in unrunnable actions"); abort_for_stonith_failure(tg_restart, NULL, last_action); return TRUE; } return FALSE; } static void tengine_stonith_connection_destroy(stonith_t *st, stonith_event_t *e) { te_cleanup_stonith_history_sync(st, FALSE); if (pcmk_is_set(fsa_input_register, R_ST_REQUIRED)) { crm_crit("Fencing daemon connection failed"); mainloop_set_trigger(stonith_reconnect); } else { crm_info("Fencing daemon disconnected"); } if (stonith_api) { /* the client API won't properly reconnect notifications * if they are still in the table - so remove them */ if (stonith_api->state != stonith_disconnected) { stonith_api->cmds->disconnect(st); } stonith_api->cmds->remove_notification(stonith_api, NULL); } if (AM_I_DC) { fail_incompletable_stonith(transition_graph); trigger_graph(); } } /*! * \internal * \brief Handle an event notification from the fencing API * * \param[in] st Fencing API connection * \param[in] event Fencing API event notification */ static void handle_fence_notification(stonith_t *st, stonith_event_t *event) { bool succeeded = true; const char *executioner = "the cluster"; const char *client = "a client"; const char *reason = NULL; int exec_status; if (te_client_id == NULL) { te_client_id = crm_strdup_printf("%s.%lu", crm_system_name, (unsigned long) getpid()); } if (event == NULL) { crm_err("Notify data not found"); return; } if (event->executioner != NULL) { executioner = event->executioner; } if (event->client_origin != NULL) { client = event->client_origin; } exec_status = stonith__event_execution_status(event); if ((stonith__event_exit_status(event) != CRM_EX_OK) || (exec_status != PCMK_EXEC_DONE)) { succeeded = false; if (exec_status == PCMK_EXEC_DONE) { exec_status = PCMK_EXEC_ERROR; } } reason = stonith__event_exit_reason(event); crmd_alert_fencing_op(event); if (pcmk__str_eq("on", event->action, pcmk__str_none)) { // Unfencing doesn't need special handling, just a log message if (succeeded) { crm_notice("%s was unfenced by %s at the request of %s@%s", event->target, executioner, client, event->origin); - /* TODO: Hook up event->device */ } else { crm_err("Unfencing of %s by %s failed (%s%s%s) with exit status %d", event->target, executioner, pcmk_exec_status_str(exec_status), ((reason == NULL)? "" : ": "), ((reason == NULL)? "" : reason), stonith__event_exit_status(event)); } return; } if (succeeded && pcmk__str_eq(event->target, fsa_our_uname, pcmk__str_casei)) { /* We were notified of our own fencing. Most likely, either fencing was * misconfigured, or fabric fencing that doesn't cut cluster * communication is in use. * * Either way, shutting down the local host is a good idea, to require * administrator intervention. Also, other nodes would otherwise likely * set our status to lost because of the fencing callback and discard * our subsequent election votes as "not part of our cluster". */ crm_crit("We were allegedly just fenced by %s for %s!", executioner, event->origin); // Dumps blackbox if enabled if (fence_reaction_panic) { pcmk__panic(__func__); } else { crm_exit(CRM_EX_FATAL); } return; // Should never get here } /* Update the count of fencing failures for this target, in case we become * DC later. The current DC has already updated its fail count in * tengine_stonith_callback(). */ - if (!AM_I_DC - && pcmk__str_eq(event->operation, T_STONITH_NOTIFY_FENCE, - pcmk__str_none)) { - + if (!AM_I_DC) { if (succeeded) { st_fail_count_reset(event->target); } else { st_fail_count_increment(event->target); } } crm_notice("Peer %s was%s terminated (%s) by %s on behalf of %s@%s: " "%s%s%s%s " CRM_XS " event=%s", event->target, (succeeded? "" : " not"), event->action, executioner, client, event->origin, (succeeded? "OK" : pcmk_exec_status_str(exec_status)), ((reason == NULL)? "" : " ("), ((reason == NULL)? "" : reason), ((reason == NULL)? "" : ")"), event->id); if (succeeded) { crm_node_t *peer = pcmk__search_known_node_cache(0, event->target, CRM_GET_PEER_ANY); const char *uuid = NULL; if (peer == NULL) { return; } uuid = crm_peer_uuid(peer); if (AM_I_DC) { /* The DC always sends updates */ send_stonith_update(NULL, event->target, uuid); /* @TODO Ideally, at this point, we'd check whether the fenced node * hosted any guest nodes, and call remote_node_down() for them. * Unfortunately, the controller doesn't have a simple, reliable way * to map hosts to guests. It might be possible to track this in the * peer cache via crm_remote_peer_cache_refresh(). For now, we rely * on the scheduler creating fence pseudo-events for the guests. */ if (!pcmk__str_eq(client, te_client_id, pcmk__str_casei)) { /* Abort the current transition if it wasn't the cluster that * initiated fencing. */ crm_info("External fencing operation from %s fenced %s", client, event->target); abort_transition(INFINITY, tg_restart, "External Fencing Operation", NULL); } /* Assume it was our leader if we don't currently have one */ } else if (pcmk__str_eq(fsa_our_dc, event->target, pcmk__str_null_matches|pcmk__str_casei) && !pcmk_is_set(peer->flags, crm_remote_node)) { crm_notice("Fencing target %s %s our leader", event->target, (fsa_our_dc? "was" : "may have been")); /* Given the CIB resyncing that occurs around elections, * have one node update the CIB now and, if the new DC is different, * have them do so too after the election */ if (pcmk__str_eq(event->executioner, fsa_our_uname, pcmk__str_casei)) { send_stonith_update(NULL, event->target, uuid); } add_stonith_cleanup(event->target); } /* If the target is a remote node, and we host its connection, * immediately fail all monitors so it can be recovered quickly. * The connection won't necessarily drop when a remote node is fenced, * so the failure might not otherwise be detected until the next poke. */ if (pcmk_is_set(peer->flags, crm_remote_node)) { remote_ra_fail(event->target); } crmd_peer_down(peer, TRUE); } } /*! * \brief Connect to fencer * * \param[in] user_data If NULL, retry failures now, otherwise retry in main loop * * \return TRUE * \note If user_data is NULL, this will wait 2s between attempts, for up to * 30 attempts, meaning the controller could be blocked as long as 58s. */ static gboolean te_connect_stonith(gpointer user_data) { int rc = pcmk_ok; if (stonith_api == NULL) { stonith_api = stonith_api_new(); if (stonith_api == NULL) { crm_err("Could not connect to fencer: API memory allocation failed"); return TRUE; } } if (stonith_api->state != stonith_disconnected) { crm_trace("Already connected to fencer, no need to retry"); return TRUE; } if (user_data == NULL) { // Blocking (retry failures now until successful) rc = stonith_api_connect_retry(stonith_api, crm_system_name, 30); if (rc != pcmk_ok) { crm_err("Could not connect to fencer in 30 attempts: %s " CRM_XS " rc=%d", pcmk_strerror(rc), rc); } } else { // Non-blocking (retry failures later in main loop) rc = stonith_api->cmds->connect(stonith_api, crm_system_name, NULL); if (rc != pcmk_ok) { if (pcmk_is_set(fsa_input_register, R_ST_REQUIRED)) { crm_notice("Fencer connection failed (will retry): %s " CRM_XS " rc=%d", pcmk_strerror(rc), rc); mainloop_set_trigger(stonith_reconnect); } else { crm_info("Fencer connection failed (ignoring because no longer required): %s " CRM_XS " rc=%d", pcmk_strerror(rc), rc); } return TRUE; } } if (rc == pcmk_ok) { stonith_api->cmds->register_notification(stonith_api, T_STONITH_NOTIFY_DISCONNECT, tengine_stonith_connection_destroy); stonith_api->cmds->register_notification(stonith_api, T_STONITH_NOTIFY_FENCE, handle_fence_notification); stonith_api->cmds->register_notification(stonith_api, T_STONITH_NOTIFY_HISTORY_SYNCED, tengine_stonith_history_synced); te_trigger_stonith_history_sync(TRUE); crm_notice("Fencer successfully connected"); } return TRUE; } /*! \internal \brief Schedule fencer connection attempt in main loop */ void controld_trigger_fencer_connect() { if (stonith_reconnect == NULL) { stonith_reconnect = mainloop_add_trigger(G_PRIORITY_LOW, te_connect_stonith, GINT_TO_POINTER(TRUE)); } controld_set_fsa_input_flags(R_ST_REQUIRED); mainloop_set_trigger(stonith_reconnect); } void controld_disconnect_fencer(bool destroy) { if (stonith_api) { // Prevent fencer connection from coming up again controld_clear_fsa_input_flags(R_ST_REQUIRED); if (stonith_api->state != stonith_disconnected) { stonith_api->cmds->disconnect(stonith_api); } stonith_api->cmds->remove_notification(stonith_api, NULL); } if (destroy) { if (stonith_api) { stonith_api->cmds->free(stonith_api); stonith_api = NULL; } if (stonith_reconnect) { mainloop_destroy_trigger(stonith_reconnect); stonith_reconnect = NULL; } if (te_client_id) { free(te_client_id); te_client_id = NULL; } } } static gboolean do_stonith_history_sync(gpointer user_data) { if (stonith_api && (stonith_api->state != stonith_disconnected)) { stonith_history_t *history = NULL; te_cleanup_stonith_history_sync(stonith_api, FALSE); stonith_api->cmds->history(stonith_api, st_opt_sync_call | st_opt_broadcast, NULL, &history, 5); stonith_history_free(history); return TRUE; } else { crm_info("Skip triggering stonith history-sync as stonith is disconnected"); return FALSE; } } static void tengine_stonith_callback(stonith_t *stonith, stonith_callback_data_t *data) { char *uuid = NULL; int stonith_id = -1; int transition_id = -1; crm_action_t *action = NULL; const char *target = NULL; if ((data == NULL) || (data->userdata == NULL)) { crm_err("Ignoring fence operation %d result: " "No transition key given (bug?)", ((data == NULL)? -1 : data->call_id)); return; } if (!AM_I_DC) { const char *reason = stonith__exit_reason(data); if (reason == NULL) { reason = pcmk_exec_status_str(stonith__execution_status(data)); } crm_notice("Result of fence operation %d: %d (%s) " CRM_XS " key=%s", data->call_id, stonith__exit_status(data), reason, (const char *) data->userdata); return; } CRM_CHECK(decode_transition_key(data->userdata, &uuid, &transition_id, &stonith_id, NULL), goto bail); if (transition_graph->complete || (stonith_id < 0) || !pcmk__str_eq(uuid, te_uuid, pcmk__str_none) || (transition_graph->id != transition_id)) { crm_info("Ignoring fence operation %d result: " "Not from current transition " CRM_XS " complete=%s action=%d uuid=%s (vs %s) transition=%d (vs %d)", data->call_id, pcmk__btoa(transition_graph->complete), stonith_id, uuid, te_uuid, transition_id, transition_graph->id); goto bail; } action = controld_get_action(stonith_id); if (action == NULL) { crm_err("Ignoring fence operation %d result: " "Action %d not found in transition graph (bug?) " CRM_XS " uuid=%s transition=%d", data->call_id, stonith_id, uuid, transition_id); goto bail; } target = crm_element_value(action->xml, XML_LRM_ATTR_TARGET); if (target == NULL) { crm_err("Ignoring fence operation %d result: No target given (bug?)", data->call_id); goto bail; } stop_te_timer(action->timer); if (stonith__exit_status(data) == CRM_EX_OK) { const char *uuid = crm_element_value(action->xml, XML_LRM_ATTR_TARGET_UUID); const char *op = crm_meta_value(action->params, "stonith_action"); crm_info("Fence operation %d for %s succeeded", data->call_id, target); if (!(pcmk_is_set(action->flags, pcmk__graph_action_confirmed))) { te_action_confirmed(action, NULL); if (pcmk__str_eq("on", op, pcmk__str_casei)) { const char *value = NULL; char *now = pcmk__ttoa(time(NULL)); gboolean is_remote_node = FALSE; /* This check is not 100% reliable, since this node is not * guaranteed to have the remote node cached. However, it * doesn't have to be reliable, since the attribute manager can * learn a node's "remoteness" by other means sooner or later. * This allows it to learn more quickly if this node does have * the information. */ if (g_hash_table_lookup(crm_remote_peer_cache, uuid) != NULL) { is_remote_node = TRUE; } update_attrd(target, CRM_ATTR_UNFENCED, now, NULL, is_remote_node); free(now); value = crm_meta_value(action->params, XML_OP_ATTR_DIGESTS_ALL); update_attrd(target, CRM_ATTR_DIGESTS_ALL, value, NULL, is_remote_node); value = crm_meta_value(action->params, XML_OP_ATTR_DIGESTS_SECURE); update_attrd(target, CRM_ATTR_DIGESTS_SECURE, value, NULL, is_remote_node); } else if (!(pcmk_is_set(action->flags, pcmk__graph_action_sent_update))) { send_stonith_update(action, target, uuid); crm__set_graph_action_flags(action, pcmk__graph_action_sent_update); } } st_fail_count_reset(target); } else { enum transition_action abort_action = tg_restart; int status = stonith__execution_status(data); const char *reason = stonith__exit_reason(data); if (reason == NULL) { if (status == PCMK_EXEC_DONE) { reason = "Agent returned error"; } else { reason = pcmk_exec_status_str(status); } } crm__set_graph_action_flags(action, pcmk__graph_action_failed); /* If no fence devices were available, there's no use in immediately * checking again, so don't start a new transition in that case. */ if (status == PCMK_EXEC_NO_FENCE_DEVICE) { crm_warn("Fence operation %d for %s failed: %s " "(aborting transition and giving up for now)", data->call_id, target, reason); abort_action = tg_stop; } else { crm_notice("Fence operation %d for %s failed: %s " "(aborting transition)", data->call_id, target, reason); } /* Increment the fail count now, so abort_for_stonith_failure() can * check it. Non-DC nodes will increment it in * handle_fence_notification(). */ st_fail_count_increment(target); abort_for_stonith_failure(abort_action, target, NULL); } pcmk__update_graph(transition_graph, action); trigger_graph(); bail: free(data->userdata); free(uuid); return; } static int fence_with_delay(const char *target, const char *type, const char *delay) { uint32_t options = st_opt_none; // Group of enum stonith_call_options int timeout_sec = (int) (transition_graph->stonith_timeout / 1000); int delay_i; if (crmd_join_phase_count(crm_join_confirmed) == 1) { stonith__set_call_options(options, target, st_opt_allow_suicide); } pcmk__scan_min_int(delay, &delay_i, 0); return stonith_api->cmds->fence_with_delay(stonith_api, options, target, type, timeout_sec, 0, delay_i); } gboolean te_fence_node(crm_graph_t *graph, crm_action_t *action) { int rc = 0; const char *id = NULL; const char *uuid = NULL; const char *target = NULL; const char *type = NULL; char *transition_key = NULL; const char *priority_delay = NULL; gboolean invalid_action = FALSE; id = ID(action->xml); target = crm_element_value(action->xml, XML_LRM_ATTR_TARGET); uuid = crm_element_value(action->xml, XML_LRM_ATTR_TARGET_UUID); type = crm_meta_value(action->params, "stonith_action"); CRM_CHECK(id != NULL, invalid_action = TRUE); CRM_CHECK(uuid != NULL, invalid_action = TRUE); CRM_CHECK(type != NULL, invalid_action = TRUE); CRM_CHECK(target != NULL, invalid_action = TRUE); if (invalid_action) { crm_log_xml_warn(action->xml, "BadAction"); return FALSE; } priority_delay = crm_meta_value(action->params, XML_CONFIG_ATTR_PRIORITY_FENCING_DELAY); crm_notice("Requesting fencing (%s) of node %s " CRM_XS " action=%s timeout=%u%s%s", type, target, id, transition_graph->stonith_timeout, priority_delay ? " priority_delay=" : "", priority_delay ? priority_delay : ""); /* Passing NULL means block until we can connect... */ te_connect_stonith(NULL); rc = fence_with_delay(target, type, priority_delay); transition_key = pcmk__transition_key(transition_graph->id, action->id, 0, te_uuid), stonith_api->cmds->register_callback(stonith_api, rc, (int) (transition_graph->stonith_timeout / 1000), st_opt_timeout_updates, transition_key, "tengine_stonith_callback", tengine_stonith_callback); return TRUE; } bool controld_verify_stonith_watchdog_timeout(const char *value) { gboolean rv = TRUE; if (stonith_api && (stonith_api->state != stonith_disconnected) && stonith__watchdog_fencing_enabled_for_node_api(stonith_api, fsa_our_uname)) { rv = pcmk__valid_sbd_timeout(value); } return rv; } /* end stonith API client functions */ /* * stonith history synchronization * * Each node's fencer keeps track of a cluster-wide fencing history. When a node * joins or leaves, we need to synchronize the history across all nodes. */ static crm_trigger_t *stonith_history_sync_trigger = NULL; static mainloop_timer_t *stonith_history_sync_timer_short = NULL; static mainloop_timer_t *stonith_history_sync_timer_long = NULL; void te_cleanup_stonith_history_sync(stonith_t *st, bool free_timers) { if (free_timers) { mainloop_timer_del(stonith_history_sync_timer_short); stonith_history_sync_timer_short = NULL; mainloop_timer_del(stonith_history_sync_timer_long); stonith_history_sync_timer_long = NULL; } else { mainloop_timer_stop(stonith_history_sync_timer_short); mainloop_timer_stop(stonith_history_sync_timer_long); } if (st) { st->cmds->remove_notification(st, T_STONITH_NOTIFY_HISTORY_SYNCED); } } static void tengine_stonith_history_synced(stonith_t *st, stonith_event_t *st_event) { te_cleanup_stonith_history_sync(st, FALSE); crm_debug("Fence-history synced - cancel all timers"); } static gboolean stonith_history_sync_set_trigger(gpointer user_data) { mainloop_set_trigger(stonith_history_sync_trigger); return FALSE; } void te_trigger_stonith_history_sync(bool long_timeout) { /* trigger a sync in 5s to give more nodes the * chance to show up so that we don't create * unnecessary stonith-history-sync traffic * * the long timeout of 30s is there as a fallback * so that after a successful connection to fenced * we will wait for 30s for the DC to trigger a * history-sync * if this doesn't happen we trigger a sync locally * (e.g. fenced segfaults and is restarted by pacemakerd) */ /* as we are finally checking the stonith-connection * in do_stonith_history_sync we should be fine * leaving stonith_history_sync_time & stonith_history_sync_trigger * around */ if (stonith_history_sync_trigger == NULL) { stonith_history_sync_trigger = mainloop_add_trigger(G_PRIORITY_LOW, do_stonith_history_sync, NULL); } if (long_timeout) { if(stonith_history_sync_timer_long == NULL) { stonith_history_sync_timer_long = mainloop_timer_add("history_sync_long", 30000, FALSE, stonith_history_sync_set_trigger, NULL); } crm_info("Fence history will be synchronized cluster-wide within 30 seconds"); mainloop_timer_start(stonith_history_sync_timer_long); } else { if(stonith_history_sync_timer_short == NULL) { stonith_history_sync_timer_short = mainloop_timer_add("history_sync_short", 5000, FALSE, stonith_history_sync_set_trigger, NULL); } crm_info("Fence history will be synchronized cluster-wide within 5 seconds"); mainloop_timer_start(stonith_history_sync_timer_short); } } /* end stonith history synchronization functions */ diff --git a/daemons/fenced/cts-fence-helper.c b/daemons/fenced/cts-fence-helper.c index e3113452ef..a906b9392b 100644 --- a/daemons/fenced/cts-fence-helper.c +++ b/daemons/fenced/cts-fence-helper.c @@ -1,686 +1,683 @@ /* * Copyright 2009-2022 the Pacemaker project contributors * * This source code is licensed under the GNU General Public License version 2 * or later (GPLv2+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include static GMainLoop *mainloop = NULL; static crm_trigger_t *trig = NULL; static int mainloop_iter = 0; static pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; typedef void (*mainloop_test_iteration_cb) (int check_event); #define MAINLOOP_DEFAULT_TIMEOUT 2 enum test_modes { test_standard = 0, // test using a specific developer environment test_passive, // watch notifications only test_api_sanity, // sanity-test stonith client API using fence_dummy test_api_mainloop, // sanity-test mainloop code with async responses }; static pcmk__cli_option_t long_options[] = { // long option, argument type, storage, short option, description, flags { "verbose", no_argument, NULL, 'V', NULL, pcmk__option_default }, { "version", no_argument, NULL, '$', NULL, pcmk__option_default }, { "help", no_argument, NULL, '?', NULL, pcmk__option_default }, { "passive", no_argument, NULL, 'p', NULL, pcmk__option_default }, { "api_test", no_argument, NULL, 't', NULL, pcmk__option_default }, { "mainloop_api_test", no_argument, NULL, 'm', NULL, pcmk__option_default }, { 0, 0, 0, 0 } }; static stonith_t *st = NULL; static struct pollfd pollfd; static const int st_opts = st_opt_sync_call; static int expected_notifications = 0; static int verbose = 0; static void mainloop_test_done(const char *origin, bool pass) { if (pass) { crm_info("SUCCESS - %s", origin); mainloop_iter++; mainloop_set_trigger(trig); result.execution_status = PCMK_EXEC_DONE; result.exit_status = CRM_EX_OK; } else { crm_err("FAILURE - %s (%d: %s)", origin, result.exit_status, pcmk_exec_status_str(result.execution_status)); crm_exit(CRM_EX_ERROR); } } static void dispatch_helper(int timeout) { int rc; crm_debug("Looking for notification"); pollfd.events = POLLIN; while (true) { rc = poll(&pollfd, 1, timeout); /* wait 10 minutes, -1 forever */ if (rc > 0) { if (!stonith_dispatch(st)) { break; } } else { break; } } } static void st_callback(stonith_t * st, stonith_event_t * e) { + char *desc = NULL; + if (st->state == stonith_disconnected) { crm_exit(CRM_EX_DISCONNECT); } - crm_notice("Operation '%s' targeting %s by %s for %s: %s (exit=%d, ref=%s)", - ((e->operation == NULL)? "unknown" : e->operation), - ((e->target == NULL)? "no node" : e->target), - ((e->executioner == NULL)? "any node" : e->executioner), - ((e->origin == NULL)? "unknown client" : e->origin), - pcmk_exec_status_str(stonith__event_execution_status(e)), - stonith__event_exit_status(e), - ((e->id == NULL)? "none" : e->id)); + desc = stonith__event_description(e); + crm_notice("%s", desc); + free(desc); if (expected_notifications) { expected_notifications--; } } static void st_global_callback(stonith_t * stonith, stonith_callback_data_t * data) { crm_notice("Call %d exited %d: %s (%s)", data->call_id, stonith__exit_status(data), stonith__execution_status(data), crm_str(stonith__exit_reason(data))); } static void passive_test(void) { int rc = 0; rc = st->cmds->connect(st, crm_system_name, &pollfd.fd); if (rc != pcmk_ok) { stonith_api_delete(st); crm_exit(CRM_EX_DISCONNECT); } st->cmds->register_notification(st, T_STONITH_NOTIFY_DISCONNECT, st_callback); st->cmds->register_notification(st, T_STONITH_NOTIFY_FENCE, st_callback); st->cmds->register_notification(st, STONITH_OP_DEVICE_ADD, st_callback); st->cmds->register_notification(st, STONITH_OP_DEVICE_DEL, st_callback); st->cmds->register_callback(st, 0, 120, st_opt_timeout_updates, NULL, "st_global_callback", st_global_callback); dispatch_helper(600 * 1000); } #define single_test(cmd, str, num_notifications, expected_rc) \ { \ int rc = 0; \ rc = cmd; \ expected_notifications = 0; \ if (num_notifications) { \ expected_notifications = num_notifications; \ dispatch_helper(500); \ } \ if (rc != expected_rc) { \ crm_err("FAILURE - expected rc %d != %d(%s) for cmd - %s", expected_rc, rc, pcmk_strerror(rc), str); \ crm_exit(CRM_EX_ERROR); \ } else if (expected_notifications) { \ crm_err("FAILURE - expected %d notifications, got only %d for cmd - %s", \ num_notifications, num_notifications - expected_notifications, str); \ crm_exit(CRM_EX_ERROR); \ } else { \ if (verbose) { \ crm_info("SUCCESS - %s: %d", str, rc); \ } else { \ crm_debug("SUCCESS - %s: %d", str, rc); \ } \ } \ }\ static void run_fence_failure_test(void) { stonith_key_value_t *params = NULL; params = stonith_key_value_add(params, PCMK_STONITH_HOST_MAP, "false_1_node1=1,2 false_1_node2=3,4"); params = stonith_key_value_add(params, "mode", "fail"); single_test(st-> cmds->register_device(st, st_opts, "test-id1", "stonith-ng", "fence_dummy", params), "Register device1 for failure test", 1, 0); single_test(st->cmds->fence(st, st_opts, "false_1_node2", "off", 3, 0), "Fence failure results off", 1, -ENODATA); single_test(st->cmds->fence(st, st_opts, "false_1_node2", "reboot", 3, 0), "Fence failure results reboot", 1, -ENODATA); single_test(st->cmds->remove_device(st, st_opts, "test-id1"), "Remove device1 for failure test", 1, 0); stonith_key_value_freeall(params, 1, 1); } static void run_fence_failure_rollover_test(void) { stonith_key_value_t *params = NULL; params = stonith_key_value_add(params, PCMK_STONITH_HOST_MAP, "false_1_node1=1,2 false_1_node2=3,4"); params = stonith_key_value_add(params, "mode", "fail"); single_test(st-> cmds->register_device(st, st_opts, "test-id1", "stonith-ng", "fence_dummy", params), "Register device1 for rollover test", 1, 0); stonith_key_value_freeall(params, 1, 1); params = NULL; params = stonith_key_value_add(params, PCMK_STONITH_HOST_MAP, "false_1_node1=1,2 false_1_node2=3,4"); params = stonith_key_value_add(params, "mode", "pass"); single_test(st-> cmds->register_device(st, st_opts, "test-id2", "stonith-ng", "fence_dummy", params), "Register device2 for rollover test", 1, 0); single_test(st->cmds->fence(st, st_opts, "false_1_node2", "off", 3, 0), "Fence rollover results off", 1, 0); /* Expect -ENODEV because fence_dummy requires 'on' to be executed on target */ single_test(st->cmds->fence(st, st_opts, "false_1_node2", "on", 3, 0), "Fence rollover results on", 1, -ENODEV); single_test(st->cmds->remove_device(st, st_opts, "test-id1"), "Remove device1 for rollover tests", 1, 0); single_test(st->cmds->remove_device(st, st_opts, "test-id2"), "Remove device2 for rollover tests", 1, 0); stonith_key_value_freeall(params, 1, 1); } static void run_standard_test(void) { stonith_key_value_t *params = NULL; params = stonith_key_value_add(params, PCMK_STONITH_HOST_MAP, "false_1_node1=1,2 false_1_node2=3,4"); params = stonith_key_value_add(params, "mode", "pass"); params = stonith_key_value_add(params, "mock_dynamic_hosts", "false_1_node1 false_1_node2"); single_test(st-> cmds->register_device(st, st_opts, "test-id", "stonith-ng", "fence_dummy", params), "Register", 1, 0); stonith_key_value_freeall(params, 1, 1); params = NULL; single_test(st->cmds->list(st, st_opts, "test-id", NULL, 1), "list", 1, 0); single_test(st->cmds->monitor(st, st_opts, "test-id", 1), "Monitor", 1, 0); single_test(st->cmds->status(st, st_opts, "test-id", "false_1_node2", 1), "Status false_1_node2", 1, 0); single_test(st->cmds->status(st, st_opts, "test-id", "false_1_node1", 1), "Status false_1_node1", 1, 0); single_test(st->cmds->fence(st, st_opts, "unknown-host", "off", 1, 0), "Fence unknown-host (expected failure)", 0, -ENODEV); single_test(st->cmds->fence(st, st_opts, "false_1_node1", "off", 1, 0), "Fence false_1_node1", 1, 0); /* Expect -ENODEV because fence_dummy requires 'on' to be executed on target */ single_test(st->cmds->fence(st, st_opts, "false_1_node1", "on", 1, 0), "Unfence false_1_node1", 1, -ENODEV); /* Confirm that an invalid level index is rejected */ single_test(st->cmds->register_level(st, st_opts, "node1", 999, params), "Attempt to register an invalid level index", 0, -EINVAL); single_test(st->cmds->remove_device(st, st_opts, "test-id"), "Remove test-id", 1, 0); stonith_key_value_freeall(params, 1, 1); } static void sanity_tests(void) { int rc = 0; rc = st->cmds->connect(st, crm_system_name, &pollfd.fd); if (rc != pcmk_ok) { stonith_api_delete(st); crm_exit(CRM_EX_DISCONNECT); } st->cmds->register_notification(st, T_STONITH_NOTIFY_DISCONNECT, st_callback); st->cmds->register_notification(st, T_STONITH_NOTIFY_FENCE, st_callback); st->cmds->register_notification(st, STONITH_OP_DEVICE_ADD, st_callback); st->cmds->register_notification(st, STONITH_OP_DEVICE_DEL, st_callback); st->cmds->register_callback(st, 0, 120, st_opt_timeout_updates, NULL, "st_global_callback", st_global_callback); crm_info("Starting API Sanity Tests"); run_standard_test(); run_fence_failure_test(); run_fence_failure_rollover_test(); crm_info("Sanity Tests Passed"); } static void standard_dev_test(void) { int rc = 0; char *tmp = NULL; stonith_key_value_t *params = NULL; rc = st->cmds->connect(st, crm_system_name, &pollfd.fd); if (rc != pcmk_ok) { stonith_api_delete(st); crm_exit(CRM_EX_DISCONNECT); } params = stonith_key_value_add(params, PCMK_STONITH_HOST_MAP, "some-host=pcmk-7 true_1_node1=3,4"); rc = st->cmds->register_device(st, st_opts, "test-id", "stonith-ng", "fence_xvm", params); crm_debug("Register: %d", rc); rc = st->cmds->list(st, st_opts, "test-id", &tmp, 10); crm_debug("List: %d output: %s", rc, tmp ? tmp : ""); rc = st->cmds->monitor(st, st_opts, "test-id", 10); crm_debug("Monitor: %d", rc); rc = st->cmds->status(st, st_opts, "test-id", "false_1_node2", 10); crm_debug("Status false_1_node2: %d", rc); rc = st->cmds->status(st, st_opts, "test-id", "false_1_node1", 10); crm_debug("Status false_1_node1: %d", rc); rc = st->cmds->fence(st, st_opts, "unknown-host", "off", 60, 0); crm_debug("Fence unknown-host: %d", rc); rc = st->cmds->status(st, st_opts, "test-id", "false_1_node1", 10); crm_debug("Status false_1_node1: %d", rc); rc = st->cmds->fence(st, st_opts, "false_1_node1", "off", 60, 0); crm_debug("Fence false_1_node1: %d", rc); rc = st->cmds->status(st, st_opts, "test-id", "false_1_node1", 10); crm_debug("Status false_1_node1: %d", rc); rc = st->cmds->fence(st, st_opts, "false_1_node1", "on", 10, 0); crm_debug("Unfence false_1_node1: %d", rc); rc = st->cmds->status(st, st_opts, "test-id", "false_1_node1", 10); crm_debug("Status false_1_node1: %d", rc); rc = st->cmds->fence(st, st_opts, "some-host", "off", 10, 0); crm_debug("Fence alias: %d", rc); rc = st->cmds->status(st, st_opts, "test-id", "some-host", 10); crm_debug("Status alias: %d", rc); rc = st->cmds->fence(st, st_opts, "false_1_node1", "on", 10, 0); crm_debug("Unfence false_1_node1: %d", rc); rc = st->cmds->remove_device(st, st_opts, "test-id"); crm_debug("Remove test-id: %d", rc); stonith_key_value_freeall(params, 1, 1); } static void iterate_mainloop_tests(gboolean event_ready); static void mainloop_callback(stonith_t * stonith, stonith_callback_data_t * data) { pcmk__set_result(&result, stonith__exit_status(data), stonith__execution_status(data), stonith__exit_reason(data)); iterate_mainloop_tests(TRUE); } static int register_callback_helper(int callid) { return st->cmds->register_callback(st, callid, MAINLOOP_DEFAULT_TIMEOUT, st_opt_timeout_updates, NULL, "callback", mainloop_callback); } static void test_async_fence_pass(int check_event) { int rc = 0; if (check_event) { mainloop_test_done(__func__, (result.exit_status == CRM_EX_OK)); return; } rc = st->cmds->fence(st, 0, "true_1_node1", "off", MAINLOOP_DEFAULT_TIMEOUT, 0); if (rc < 0) { crm_err("fence failed with rc %d", rc); mainloop_test_done(__func__, false); } register_callback_helper(rc); /* wait for event */ } #define CUSTOM_TIMEOUT_ADDITION 10 static void test_async_fence_custom_timeout(int check_event) { int rc = 0; static time_t begin = 0; if (check_event) { uint32_t diff = (time(NULL) - begin); if (result.execution_status != PCMK_EXEC_TIMEOUT) { mainloop_test_done(__func__, false); } else if (diff < CUSTOM_TIMEOUT_ADDITION + MAINLOOP_DEFAULT_TIMEOUT) { crm_err ("Custom timeout test failed, callback expiration should be updated to %d, actual timeout was %d", CUSTOM_TIMEOUT_ADDITION + MAINLOOP_DEFAULT_TIMEOUT, diff); mainloop_test_done(__func__, false); } else { mainloop_test_done(__func__, true); } return; } begin = time(NULL); rc = st->cmds->fence(st, 0, "custom_timeout_node1", "off", MAINLOOP_DEFAULT_TIMEOUT, 0); if (rc < 0) { crm_err("fence failed with rc %d", rc); mainloop_test_done(__func__, false); } register_callback_helper(rc); /* wait for event */ } static void test_async_fence_timeout(int check_event) { int rc = 0; if (check_event) { mainloop_test_done(__func__, (result.execution_status == PCMK_EXEC_NO_FENCE_DEVICE)); return; } rc = st->cmds->fence(st, 0, "false_1_node2", "off", MAINLOOP_DEFAULT_TIMEOUT, 0); if (rc < 0) { crm_err("fence failed with rc %d", rc); mainloop_test_done(__func__, false); } register_callback_helper(rc); /* wait for event */ } static void test_async_monitor(int check_event) { int rc = 0; if (check_event) { mainloop_test_done(__func__, (result.exit_status == CRM_EX_OK)); return; } rc = st->cmds->monitor(st, 0, "false_1", MAINLOOP_DEFAULT_TIMEOUT); if (rc < 0) { crm_err("monitor failed with rc %d", rc); mainloop_test_done(__func__, false); } register_callback_helper(rc); /* wait for event */ } static void test_register_async_devices(int check_event) { char buf[16] = { 0, }; stonith_key_value_t *params = NULL; params = stonith_key_value_add(params, PCMK_STONITH_HOST_MAP, "false_1_node1=1,2"); params = stonith_key_value_add(params, "mode", "fail"); st->cmds->register_device(st, st_opts, "false_1", "stonith-ng", "fence_dummy", params); stonith_key_value_freeall(params, 1, 1); params = NULL; params = stonith_key_value_add(params, PCMK_STONITH_HOST_MAP, "true_1_node1=1,2"); params = stonith_key_value_add(params, "mode", "pass"); st->cmds->register_device(st, st_opts, "true_1", "stonith-ng", "fence_dummy", params); stonith_key_value_freeall(params, 1, 1); params = NULL; params = stonith_key_value_add(params, PCMK_STONITH_HOST_MAP, "custom_timeout_node1=1,2"); params = stonith_key_value_add(params, "mode", "fail"); params = stonith_key_value_add(params, "delay", "1000"); snprintf(buf, sizeof(buf) - 1, "%d", MAINLOOP_DEFAULT_TIMEOUT + CUSTOM_TIMEOUT_ADDITION); params = stonith_key_value_add(params, "pcmk_off_timeout", buf); st->cmds->register_device(st, st_opts, "false_custom_timeout", "stonith-ng", "fence_dummy", params); stonith_key_value_freeall(params, 1, 1); mainloop_test_done(__func__, true); } static void try_mainloop_connect(int check_event) { int rc = stonith_api_connect_retry(st, crm_system_name, 10); if (rc == pcmk_ok) { mainloop_test_done(__func__, true); return; } crm_err("API CONNECTION FAILURE"); mainloop_test_done(__func__, false); } static void iterate_mainloop_tests(gboolean event_ready) { static mainloop_test_iteration_cb callbacks[] = { try_mainloop_connect, test_register_async_devices, test_async_monitor, test_async_fence_pass, test_async_fence_timeout, test_async_fence_custom_timeout, }; if (mainloop_iter == (sizeof(callbacks) / sizeof(mainloop_test_iteration_cb))) { /* all tests ran, everything passed */ crm_info("ALL MAINLOOP TESTS PASSED!"); crm_exit(CRM_EX_OK); } callbacks[mainloop_iter] (event_ready); } static gboolean trigger_iterate_mainloop_tests(gpointer user_data) { iterate_mainloop_tests(FALSE); return TRUE; } static void test_shutdown(int nsig) { int rc = 0; if (st) { rc = st->cmds->disconnect(st); crm_info("Disconnect: %d", rc); crm_debug("Destroy"); stonith_api_delete(st); } if (rc) { crm_exit(CRM_EX_ERROR); } } static void mainloop_tests(void) { trig = mainloop_add_trigger(G_PRIORITY_HIGH, trigger_iterate_mainloop_tests, NULL); mainloop_set_trigger(trig); mainloop_add_signal(SIGTERM, test_shutdown); crm_info("Starting"); mainloop = g_main_loop_new(NULL, FALSE); g_main_loop_run(mainloop); } int main(int argc, char **argv) { int argerr = 0; int flag; int option_index = 0; enum test_modes mode = test_standard; pcmk__cli_init_logging("cts-fence-helper", 0); pcmk__set_cli_options(NULL, " [options]", long_options, "inject commands into the Pacemaker fencer, " "and watch for events"); while (1) { flag = pcmk__next_cli_option(argc, argv, &option_index, NULL); if (flag == -1) { break; } switch (flag) { case 'V': verbose = 1; break; case '$': case '?': pcmk__cli_help(flag, CRM_EX_OK); break; case 'p': mode = test_passive; break; case 't': mode = test_api_sanity; break; case 'm': mode = test_api_mainloop; break; default: ++argerr; break; } } crm_log_init(NULL, LOG_INFO, TRUE, (verbose? TRUE : FALSE), argc, argv, FALSE); if (optind > argc) { ++argerr; } if (argerr) { pcmk__cli_help('?', CRM_EX_USAGE); } st = stonith_api_new(); if (st == NULL) { crm_err("Could not connect to fencer: API memory allocation failed"); crm_exit(CRM_EX_DISCONNECT); } switch (mode) { case test_standard: standard_dev_test(); break; case test_passive: passive_test(); break; case test_api_sanity: sanity_tests(); break; case test_api_mainloop: mainloop_tests(); break; } test_shutdown(0); return CRM_EX_OK; } diff --git a/daemons/fenced/fenced_commands.c b/daemons/fenced/fenced_commands.c index 331798dba4..a46cd0f668 100644 --- a/daemons/fenced/fenced_commands.c +++ b/daemons/fenced/fenced_commands.c @@ -1,3510 +1,3537 @@ /* * Copyright 2009-2022 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU General Public License version 2 * or later (GPLv2+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include GHashTable *device_list = NULL; GHashTable *topology = NULL; GList *cmd_list = NULL; static GHashTable *fenced_handlers = NULL; struct device_search_s { /* target of fence action */ char *host; /* requested fence action */ char *action; /* timeout to use if a device is queried dynamically for possible targets */ int per_device_timeout; /* number of registered fencing devices at time of request */ int replies_needed; /* number of device replies received so far */ int replies_received; /* whether the target is eligible to perform requested action (or off) */ bool allow_suicide; /* private data to pass to search callback function */ void *user_data; /* function to call when all replies have been received */ void (*callback) (GList * devices, void *user_data); /* devices capable of performing requested action (or off if remapping) */ GList *capable; }; static gboolean stonith_device_dispatch(gpointer user_data); static void st_child_done(int pid, const pcmk__action_result_t *result, void *user_data); static void stonith_send_reply(xmlNode * reply, int call_options, const char *remote_peer, pcmk__client_t *client); static void search_devices_record_result(struct device_search_s *search, const char *device, gboolean can_fence); static int get_agent_metadata(const char *agent, xmlNode **metadata); static void read_action_metadata(stonith_device_t *device); static enum fenced_target_by unpack_level_kind(xmlNode *level); typedef struct async_command_s { int id; int pid; int fd_stdout; int options; int default_timeout; /* seconds */ int timeout; /* seconds */ int start_delay; /* seconds */ int delay_id; char *op; char *origin; char *client; char *client_name; char *remote_op_id; char *victim; uint32_t victim_nodeid; char *action; char *device; GList *device_list; GList *device_next; void *internal_user_data; void (*done_cb) (int pid, const pcmk__action_result_t *result, void *user_data); guint timer_sigterm; guint timer_sigkill; /*! If the operation timed out, this is the last signal * we sent to the process to get it to terminate */ int last_timeout_signo; stonith_device_t *active_on; stonith_device_t *activating_on; } async_command_t; static xmlNode *construct_async_reply(async_command_t *cmd, const pcmk__action_result_t *result); static gboolean is_action_required(const char *action, stonith_device_t *device) { return device && device->automatic_unfencing && pcmk__str_eq(action, "on", pcmk__str_casei); } static int get_action_delay_max(stonith_device_t * device, const char * action) { const char *value = NULL; int delay_max = 0; if (!pcmk__is_fencing_action(action)) { return 0; } value = g_hash_table_lookup(device->params, PCMK_STONITH_DELAY_MAX); if (value) { delay_max = crm_parse_interval_spec(value) / 1000; } return delay_max; } static int get_action_delay_base(stonith_device_t *device, const char *action, const char *victim) { char *hash_value = NULL; int delay_base = 0; if (!pcmk__is_fencing_action(action)) { return 0; } hash_value = g_hash_table_lookup(device->params, PCMK_STONITH_DELAY_BASE); if (hash_value) { char *value = strdup(hash_value); char *valptr = value; CRM_ASSERT(value != NULL); if (victim) { for (char *val = strtok(value, "; \t"); val != NULL; val = strtok(NULL, "; \t")) { char *mapval = strchr(val, ':'); if (mapval == NULL || mapval[1] == 0) { crm_err("pcmk_delay_base: empty value in mapping", val); continue; } if (mapval != val && strncasecmp(victim, val, (size_t)(mapval - val)) == 0) { value = mapval + 1; crm_debug("pcmk_delay_base mapped to %s for %s", value, victim); break; } } } if (strchr(value, ':') == 0) { delay_base = crm_parse_interval_spec(value) / 1000; } free(valptr); } return delay_base; } /*! * \internal * \brief Override STONITH timeout with pcmk_*_timeout if available * * \param[in] device STONITH device to use * \param[in] action STONITH action name * \param[in] default_timeout Timeout to use if device does not have * a pcmk_*_timeout parameter for action * * \return Value of pcmk_(action)_timeout if available, otherwise default_timeout * \note For consistency, it would be nice if reboot/off/on timeouts could be * set the same way as start/stop/monitor timeouts, i.e. with an * entry in the fencing resource configuration. However that * is insufficient because fencing devices may be registered directly via * the fencer's register_device() API instead of going through the CIB * (e.g. stonith_admin uses it for its -R option, and the executor uses it * to ensure a device is registered when a command is issued). As device * properties, pcmk_*_timeout parameters can be grabbed by the fencer when * the device is registered, whether by CIB change or API call. */ static int get_action_timeout(stonith_device_t * device, const char *action, int default_timeout) { if (action && device && device->params) { char buffer[64] = { 0, }; const char *value = NULL; /* If "reboot" was requested but the device does not support it, * we will remap to "off", so check timeout for "off" instead */ if (pcmk__str_eq(action, "reboot", pcmk__str_casei) && !pcmk_is_set(device->flags, st_device_supports_reboot)) { crm_trace("%s doesn't support reboot, using timeout for off instead", device->id); action = "off"; } /* If the device config specified an action-specific timeout, use it */ snprintf(buffer, sizeof(buffer), "pcmk_%s_timeout", action); value = g_hash_table_lookup(device->params, buffer); if (value) { return atoi(value); } } return default_timeout; } static void free_async_command(async_command_t * cmd) { if (!cmd) { return; } if (cmd->delay_id) { g_source_remove(cmd->delay_id); } cmd_list = g_list_remove(cmd_list, cmd); g_list_free_full(cmd->device_list, free); free(cmd->device); free(cmd->action); free(cmd->victim); free(cmd->remote_op_id); free(cmd->client); free(cmd->client_name); free(cmd->origin); free(cmd->op); free(cmd); } static async_command_t * create_async_command(xmlNode * msg) { async_command_t *cmd = NULL; xmlNode *op = get_xpath_object("//@" F_STONITH_ACTION, msg, LOG_ERR); const char *action = crm_element_value(op, F_STONITH_ACTION); CRM_CHECK(action != NULL, crm_log_xml_warn(msg, "NoAction"); return NULL); crm_log_xml_trace(msg, "Command"); cmd = calloc(1, sizeof(async_command_t)); crm_element_value_int(msg, F_STONITH_CALLID, &(cmd->id)); crm_element_value_int(msg, F_STONITH_CALLOPTS, &(cmd->options)); crm_element_value_int(msg, F_STONITH_TIMEOUT, &(cmd->default_timeout)); cmd->timeout = cmd->default_timeout; // Value -1 means disable any static/random fencing delays crm_element_value_int(msg, F_STONITH_DELAY, &(cmd->start_delay)); cmd->origin = crm_element_value_copy(msg, F_ORIG); cmd->remote_op_id = crm_element_value_copy(msg, F_STONITH_REMOTE_OP_ID); cmd->client = crm_element_value_copy(msg, F_STONITH_CLIENTID); cmd->client_name = crm_element_value_copy(msg, F_STONITH_CLIENTNAME); cmd->op = crm_element_value_copy(msg, F_STONITH_OPERATION); cmd->action = strdup(action); cmd->victim = crm_element_value_copy(op, F_STONITH_TARGET); cmd->device = crm_element_value_copy(op, F_STONITH_DEVICE); CRM_CHECK(cmd->op != NULL, crm_log_xml_warn(msg, "NoOp"); free_async_command(cmd); return NULL); CRM_CHECK(cmd->client != NULL, crm_log_xml_warn(msg, "NoClient")); cmd->done_cb = st_child_done; cmd_list = g_list_append(cmd_list, cmd); return cmd; } static int get_action_limit(stonith_device_t * device) { const char *value = NULL; int action_limit = 1; value = g_hash_table_lookup(device->params, PCMK_STONITH_ACTION_LIMIT); if ((value == NULL) || (pcmk__scan_min_int(value, &action_limit, INT_MIN) != pcmk_rc_ok) || (action_limit == 0)) { action_limit = 1; } return action_limit; } static int get_active_cmds(stonith_device_t * device) { int counter = 0; GList *gIter = NULL; GList *gIterNext = NULL; CRM_CHECK(device != NULL, return 0); for (gIter = cmd_list; gIter != NULL; gIter = gIterNext) { async_command_t *cmd = gIter->data; gIterNext = gIter->next; if (cmd->active_on == device) { counter++; } } return counter; } static void fork_cb(int pid, void *user_data) { async_command_t *cmd = (async_command_t *) user_data; stonith_device_t * device = /* in case of a retry we've done the move from activating_on to active_on already */ cmd->activating_on?cmd->activating_on:cmd->active_on; CRM_ASSERT(device); crm_debug("Operation '%s' [%d]%s%s using %s now running with %ds timeout", cmd->action, pid, ((cmd->victim == NULL)? "" : " targeting "), ((cmd->victim == NULL)? "" : cmd->victim), device->id, cmd->timeout); cmd->active_on = device; cmd->activating_on = NULL; } static int get_agent_metadata_cb(gpointer data) { stonith_device_t *device = data; guint period_ms; switch (get_agent_metadata(device->agent, &device->agent_metadata)) { case pcmk_rc_ok: if (device->agent_metadata) { read_action_metadata(device); stonith__device_parameter_flags(&(device->flags), device->id, device->agent_metadata); } return G_SOURCE_REMOVE; case EAGAIN: period_ms = pcmk__mainloop_timer_get_period(device->timer); if (period_ms < 160 * 1000) { mainloop_timer_set_period(device->timer, 2 * period_ms); } return G_SOURCE_CONTINUE; default: return G_SOURCE_REMOVE; } } /*! * \internal * \brief Call a command's action callback for an internal (not library) result * * \param[in] cmd Command to report result for * \param[in] execution_status Execution status to use for result * \param[in] exit_status Exit status to use for result * \param[in] exit_reason Exit reason to use for result */ static void report_internal_result(async_command_t *cmd, int exit_status, int execution_status, const char *exit_reason) { pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; pcmk__set_result(&result, exit_status, execution_status, exit_reason); cmd->done_cb(0, &result, cmd); pcmk__reset_result(&result); } static gboolean stonith_device_execute(stonith_device_t * device) { int exec_rc = 0; const char *action_str = NULL; const char *host_arg = NULL; async_command_t *cmd = NULL; stonith_action_t *action = NULL; int active_cmds = 0; int action_limit = 0; GList *gIter = NULL; GList *gIterNext = NULL; CRM_CHECK(device != NULL, return FALSE); active_cmds = get_active_cmds(device); action_limit = get_action_limit(device); if (action_limit > -1 && active_cmds >= action_limit) { crm_trace("%s is over its action limit of %d (%u active action%s)", device->id, action_limit, active_cmds, pcmk__plural_s(active_cmds)); return TRUE; } for (gIter = device->pending_ops; gIter != NULL; gIter = gIterNext) { async_command_t *pending_op = gIter->data; gIterNext = gIter->next; if (pending_op && pending_op->delay_id) { crm_trace("Operation '%s'%s%s using %s was asked to run too early, " "waiting for start delay of %ds", pending_op->action, ((pending_op->victim == NULL)? "" : " targeting "), ((pending_op->victim == NULL)? "" : pending_op->victim), device->id, pending_op->start_delay); continue; } device->pending_ops = g_list_remove_link(device->pending_ops, gIter); g_list_free_1(gIter); cmd = pending_op; break; } if (cmd == NULL) { crm_trace("No actions using %s are needed", device->id); return TRUE; } if (pcmk__str_any_of(device->agent, STONITH_WATCHDOG_AGENT, STONITH_WATCHDOG_AGENT_INTERNAL, NULL)) { if (pcmk__is_fencing_action(cmd->action)) { if (node_does_watchdog_fencing(stonith_our_uname)) { pcmk__panic(__func__); goto done; } } else { crm_info("Faking success for %s watchdog operation", cmd->action); report_internal_result(cmd, CRM_EX_OK, PCMK_EXEC_DONE, NULL); goto done; } } #if SUPPORT_CIBSECRETS exec_rc = pcmk__substitute_secrets(device->id, device->params); if (exec_rc != pcmk_rc_ok) { if (pcmk__str_eq(cmd->action, "stop", pcmk__str_casei)) { crm_info("Proceeding with stop operation for %s " "despite being unable to load CIB secrets (%s)", device->id, pcmk_rc_str(exec_rc)); } else { crm_err("Considering %s unconfigured " "because unable to load CIB secrets: %s", device->id, pcmk_rc_str(exec_rc)); report_internal_result(cmd, CRM_EX_ERROR, PCMK_EXEC_NO_SECRETS, "Failed to get CIB secrets"); goto done; } } #endif action_str = cmd->action; if (pcmk__str_eq(cmd->action, "reboot", pcmk__str_casei) && !pcmk_is_set(device->flags, st_device_supports_reboot)) { crm_notice("Remapping 'reboot' action%s%s using %s to 'off' " "because agent '%s' does not support reboot", ((cmd->victim == NULL)? "" : " targeting "), ((cmd->victim == NULL)? "" : cmd->victim), device->id, device->agent); action_str = "off"; } if (pcmk_is_set(device->flags, st_device_supports_parameter_port)) { host_arg = "port"; } else if (pcmk_is_set(device->flags, st_device_supports_parameter_plug)) { host_arg = "plug"; } action = stonith_action_create(device->agent, action_str, cmd->victim, cmd->victim_nodeid, cmd->timeout, device->params, device->aliases, host_arg); /* for async exec, exec_rc is negative for early error exit otherwise handling of success/errors is done via callbacks */ cmd->activating_on = device; exec_rc = stonith_action_execute_async(action, (void *)cmd, cmd->done_cb, fork_cb); if (exec_rc < 0) { cmd->activating_on = NULL; cmd->done_cb(0, stonith__action_result(action), cmd); stonith__destroy_action(action); } done: /* Device might get triggered to work by multiple fencing commands * simultaneously. Trigger the device again to make sure any * remaining concurrent commands get executed. */ if (device->pending_ops) { mainloop_set_trigger(device->work); } return TRUE; } static gboolean stonith_device_dispatch(gpointer user_data) { return stonith_device_execute(user_data); } static gboolean start_delay_helper(gpointer data) { async_command_t *cmd = data; stonith_device_t *device = NULL; cmd->delay_id = 0; device = cmd->device ? g_hash_table_lookup(device_list, cmd->device) : NULL; if (device) { mainloop_set_trigger(device->work); } return FALSE; } static void schedule_stonith_command(async_command_t * cmd, stonith_device_t * device) { int delay_max = 0; int delay_base = 0; int requested_delay = cmd->start_delay; CRM_CHECK(cmd != NULL, return); CRM_CHECK(device != NULL, return); if (cmd->device) { free(cmd->device); } if (device->include_nodeid && cmd->victim) { crm_node_t *node = crm_get_peer(0, cmd->victim); cmd->victim_nodeid = node->id; } cmd->device = strdup(device->id); cmd->timeout = get_action_timeout(device, cmd->action, cmd->default_timeout); if (cmd->remote_op_id) { crm_debug("Scheduling '%s' action%s%s using %s for remote peer %s " "with op id %.8s and timeout %ds", cmd->action, cmd->victim ? " targeting " : "", cmd->victim ? cmd->victim : "", device->id, cmd->origin, cmd->remote_op_id, cmd->timeout); } else { crm_debug("Scheduling '%s' action%s%s using %s for %s with timeout %ds", cmd->action, cmd->victim ? " targeting " : "", cmd->victim ? cmd->victim : "", device->id, cmd->client, cmd->timeout); } device->pending_ops = g_list_append(device->pending_ops, cmd); mainloop_set_trigger(device->work); // Value -1 means disable any static/random fencing delays if (requested_delay < 0) { return; } delay_max = get_action_delay_max(device, cmd->action); delay_base = get_action_delay_base(device, cmd->action, cmd->victim); if (delay_max == 0) { delay_max = delay_base; } if (delay_max < delay_base) { crm_warn(PCMK_STONITH_DELAY_BASE " (%ds) is larger than " PCMK_STONITH_DELAY_MAX " (%ds) for %s using %s " "(limiting to maximum delay)", delay_base, delay_max, cmd->action, device->id); delay_base = delay_max; } if (delay_max > 0) { // coverity[dont_call] We're not using rand() for security cmd->start_delay += ((delay_max != delay_base)?(rand() % (delay_max - delay_base)):0) + delay_base; } if (cmd->start_delay > 0) { crm_notice("Delaying '%s' action%s%s using %s for %ds " CRM_XS " timeout=%ds requested_delay=%ds base=%ds max=%ds", cmd->action, cmd->victim ? " targeting " : "", cmd->victim ? cmd->victim : "", device->id, cmd->start_delay, cmd->timeout, requested_delay, delay_base, delay_max); cmd->delay_id = g_timeout_add_seconds(cmd->start_delay, start_delay_helper, cmd); } } static void free_device(gpointer data) { GList *gIter = NULL; stonith_device_t *device = data; g_hash_table_destroy(device->params); g_hash_table_destroy(device->aliases); for (gIter = device->pending_ops; gIter != NULL; gIter = gIter->next) { async_command_t *cmd = gIter->data; crm_warn("Removal of device '%s' purged operation '%s'", device->id, cmd->action); report_internal_result(cmd, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, "Device was removed before action could be executed"); } g_list_free(device->pending_ops); g_list_free_full(device->targets, free); if (device->timer) { mainloop_timer_stop(device->timer); mainloop_timer_del(device->timer); } mainloop_destroy_trigger(device->work); free_xml(device->agent_metadata); free(device->namespace); free(device->on_target_actions); free(device->agent); free(device->id); free(device); } void free_device_list(void) { if (device_list != NULL) { g_hash_table_destroy(device_list); device_list = NULL; } } void init_device_list(void) { if (device_list == NULL) { device_list = pcmk__strkey_table(NULL, free_device); } } static GHashTable * build_port_aliases(const char *hostmap, GList ** targets) { char *name = NULL; int last = 0, lpc = 0, max = 0, added = 0; GHashTable *aliases = pcmk__strikey_table(free, free); if (hostmap == NULL) { return aliases; } max = strlen(hostmap); for (; lpc <= max; lpc++) { switch (hostmap[lpc]) { /* Skip escaped chars */ case '\\': lpc++; break; /* Assignment chars */ case '=': case ':': if (lpc > last) { free(name); name = calloc(1, 1 + lpc - last); memcpy(name, hostmap + last, lpc - last); } last = lpc + 1; break; /* Delimeter chars */ /* case ',': Potentially used to specify multiple ports */ case 0: case ';': case ' ': case '\t': if (name) { char *value = NULL; int k = 0; value = calloc(1, 1 + lpc - last); memcpy(value, hostmap + last, lpc - last); for (int i = 0; value[i] != '\0'; i++) { if (value[i] != '\\') { value[k++] = value[i]; } } value[k] = '\0'; crm_debug("Adding alias '%s'='%s'", name, value); g_hash_table_replace(aliases, name, value); if (targets) { *targets = g_list_append(*targets, strdup(value)); } value = NULL; name = NULL; added++; } else if (lpc > last) { crm_debug("Parse error at offset %d near '%s'", lpc - last, hostmap + last); } last = lpc + 1; break; } if (hostmap[lpc] == 0) { break; } } if (added == 0) { crm_info("No host mappings detected in '%s'", hostmap); } free(name); return aliases; } GHashTable *metadata_cache = NULL; void free_metadata_cache(void) { if (metadata_cache != NULL) { g_hash_table_destroy(metadata_cache); metadata_cache = NULL; } } static void init_metadata_cache(void) { if (metadata_cache == NULL) { metadata_cache = pcmk__strkey_table(free, free); } } int get_agent_metadata(const char *agent, xmlNode ** metadata) { char *buffer = NULL; if (metadata == NULL) { return EINVAL; } *metadata = NULL; if (pcmk__str_eq(agent, STONITH_WATCHDOG_AGENT_INTERNAL, pcmk__str_none)) { return pcmk_rc_ok; } init_metadata_cache(); buffer = g_hash_table_lookup(metadata_cache, agent); if (buffer == NULL) { stonith_t *st = stonith_api_new(); int rc; if (st == NULL) { crm_warn("Could not get agent meta-data: " "API memory allocation failed"); return EAGAIN; } rc = st->cmds->metadata(st, st_opt_sync_call, agent, NULL, &buffer, 10); stonith_api_delete(st); if (rc || !buffer) { crm_err("Could not retrieve metadata for fencing agent %s", agent); return EAGAIN; } g_hash_table_replace(metadata_cache, strdup(agent), buffer); } *metadata = string2xml(buffer); return pcmk_rc_ok; } static gboolean is_nodeid_required(xmlNode * xml) { xmlXPathObjectPtr xpath = NULL; if (stand_alone) { return FALSE; } if (!xml) { return FALSE; } xpath = xpath_search(xml, "//parameter[@name='nodeid']"); if (numXpathResults(xpath) <= 0) { freeXpathObject(xpath); return FALSE; } freeXpathObject(xpath); return TRUE; } #define MAX_ACTION_LEN 256 static char * add_action(char *actions, const char *action) { int offset = 0; if (actions == NULL) { actions = calloc(1, MAX_ACTION_LEN); } else { offset = strlen(actions); } if (offset > 0) { offset += snprintf(actions+offset, MAX_ACTION_LEN - offset, " "); } offset += snprintf(actions+offset, MAX_ACTION_LEN - offset, "%s", action); return actions; } static void read_action_metadata(stonith_device_t *device) { xmlXPathObjectPtr xpath = NULL; int max = 0; int lpc = 0; if (device->agent_metadata == NULL) { return; } xpath = xpath_search(device->agent_metadata, "//action"); max = numXpathResults(xpath); if (max <= 0) { freeXpathObject(xpath); return; } for (lpc = 0; lpc < max; lpc++) { const char *action = NULL; xmlNode *match = getXpathResult(xpath, lpc); CRM_LOG_ASSERT(match != NULL); if(match == NULL) { continue; }; action = crm_element_value(match, "name"); if(pcmk__str_eq(action, "list", pcmk__str_casei)) { stonith__set_device_flags(device->flags, device->id, st_device_supports_list); } else if(pcmk__str_eq(action, "status", pcmk__str_casei)) { stonith__set_device_flags(device->flags, device->id, st_device_supports_status); } else if(pcmk__str_eq(action, "reboot", pcmk__str_casei)) { stonith__set_device_flags(device->flags, device->id, st_device_supports_reboot); } else if (pcmk__str_eq(action, "on", pcmk__str_casei)) { /* "automatic" means the cluster will unfence node when it joins */ /* "required" is a deprecated synonym for "automatic" */ if (pcmk__xe_attr_is_true(match, "automatic") || pcmk__xe_attr_is_true(match, "required")) { device->automatic_unfencing = TRUE; } } if (action && pcmk__xe_attr_is_true(match, "on_target")) { device->on_target_actions = add_action(device->on_target_actions, action); } } freeXpathObject(xpath); } /*! * \internal * \brief Set a pcmk_*_action parameter if not already set * * \param[in,out] params Device parameters * \param[in] action Name of action * \param[in] value Value to use if action is not already set */ static void map_action(GHashTable *params, const char *action, const char *value) { char *key = crm_strdup_printf("pcmk_%s_action", action); if (g_hash_table_lookup(params, key)) { crm_warn("Ignoring %s='%s', see %s instead", STONITH_ATTR_ACTION_OP, value, key); free(key); } else { crm_warn("Mapping %s='%s' to %s='%s'", STONITH_ATTR_ACTION_OP, value, key, value); g_hash_table_insert(params, key, strdup(value)); } } /*! * \internal * \brief Create device parameter table from XML * * \param[in] name Device name (used for logging only) * \param[in,out] params Device parameters */ static GHashTable * xml2device_params(const char *name, xmlNode *dev) { GHashTable *params = xml2list(dev); const char *value; /* Action should never be specified in the device configuration, * but we support it for users who are familiar with other software * that worked that way. */ value = g_hash_table_lookup(params, STONITH_ATTR_ACTION_OP); if (value != NULL) { crm_warn("%s has '%s' parameter, which should never be specified in configuration", name, STONITH_ATTR_ACTION_OP); if (*value == '\0') { crm_warn("Ignoring empty '%s' parameter", STONITH_ATTR_ACTION_OP); } else if (strcmp(value, "reboot") == 0) { crm_warn("Ignoring %s='reboot' (see stonith-action cluster property instead)", STONITH_ATTR_ACTION_OP); } else if (strcmp(value, "off") == 0) { map_action(params, "reboot", value); } else { map_action(params, "off", value); map_action(params, "reboot", value); } g_hash_table_remove(params, STONITH_ATTR_ACTION_OP); } return params; } static const char * target_list_type(stonith_device_t * dev) { const char *check_type = NULL; check_type = g_hash_table_lookup(dev->params, PCMK_STONITH_HOST_CHECK); if (check_type == NULL) { if (g_hash_table_lookup(dev->params, PCMK_STONITH_HOST_LIST)) { check_type = "static-list"; } else if (g_hash_table_lookup(dev->params, PCMK_STONITH_HOST_MAP)) { check_type = "static-list"; } else if (pcmk_is_set(dev->flags, st_device_supports_list)) { check_type = "dynamic-list"; } else if (pcmk_is_set(dev->flags, st_device_supports_status)) { check_type = "status"; } else { check_type = PCMK__VALUE_NONE; } } return check_type; } static stonith_device_t * -build_device_from_xml(xmlNode * msg) +build_device_from_xml(xmlNode *dev) { const char *value; - xmlNode *dev = get_xpath_object("//" F_STONITH_DEVICE, msg, LOG_ERR); stonith_device_t *device = NULL; char *agent = crm_element_value_copy(dev, "agent"); CRM_CHECK(agent != NULL, return device); device = calloc(1, sizeof(stonith_device_t)); CRM_CHECK(device != NULL, {free(agent); return device;}); device->id = crm_element_value_copy(dev, XML_ATTR_ID); device->agent = agent; device->namespace = crm_element_value_copy(dev, "namespace"); device->params = xml2device_params(device->id, dev); value = g_hash_table_lookup(device->params, PCMK_STONITH_HOST_LIST); if (value) { device->targets = stonith__parse_targets(value); } value = g_hash_table_lookup(device->params, PCMK_STONITH_HOST_MAP); device->aliases = build_port_aliases(value, &(device->targets)); value = target_list_type(device); if (!pcmk__str_eq(value, "static-list", pcmk__str_casei) && device->targets) { /* Other than "static-list", dev-> targets is unnecessary. */ g_list_free_full(device->targets, free); device->targets = NULL; } switch (get_agent_metadata(device->agent, &device->agent_metadata)) { case pcmk_rc_ok: if (device->agent_metadata) { read_action_metadata(device); stonith__device_parameter_flags(&(device->flags), device->id, device->agent_metadata); } break; case EAGAIN: if (device->timer == NULL) { device->timer = mainloop_timer_add("get_agent_metadata", 10 * 1000, TRUE, get_agent_metadata_cb, device); } if (!mainloop_timer_running(device->timer)) { mainloop_timer_start(device->timer); } break; default: break; } value = g_hash_table_lookup(device->params, "nodeid"); if (!value) { device->include_nodeid = is_nodeid_required(device->agent_metadata); } value = crm_element_value(dev, "rsc_provides"); if (pcmk__str_eq(value, "unfencing", pcmk__str_casei)) { device->automatic_unfencing = TRUE; } if (is_action_required("on", device)) { crm_info("Fencing device '%s' requires unfencing", device->id); } if (device->on_target_actions) { crm_info("Fencing device '%s' requires actions (%s) to be executed " "on target", device->id, device->on_target_actions); } device->work = mainloop_add_trigger(G_PRIORITY_HIGH, stonith_device_dispatch, device); /* TODO: Hook up priority */ return device; } static void schedule_internal_command(const char *origin, stonith_device_t * device, const char *action, const char *victim, int timeout, void *internal_user_data, void (*done_cb) (int pid, const pcmk__action_result_t *result, void *user_data)) { async_command_t *cmd = NULL; cmd = calloc(1, sizeof(async_command_t)); cmd->id = -1; cmd->default_timeout = timeout ? timeout : 60; cmd->timeout = cmd->default_timeout; cmd->action = strdup(action); pcmk__str_update(&cmd->victim, victim); cmd->device = strdup(device->id); cmd->origin = strdup(origin); cmd->client = strdup(crm_system_name); cmd->client_name = strdup(crm_system_name); cmd->internal_user_data = internal_user_data; cmd->done_cb = done_cb; /* cmd, not internal_user_data, is passed to 'done_cb' as the userdata */ schedule_stonith_command(cmd, device); } // Fence agent status commands use custom exit status codes enum fence_status_code { fence_status_invalid = -1, fence_status_active = 0, fence_status_unknown = 1, fence_status_inactive = 2, }; static void status_search_cb(int pid, const pcmk__action_result_t *result, void *user_data) { async_command_t *cmd = user_data; struct device_search_s *search = cmd->internal_user_data; stonith_device_t *dev = cmd->device ? g_hash_table_lookup(device_list, cmd->device) : NULL; gboolean can = FALSE; free_async_command(cmd); if (!dev) { search_devices_record_result(search, NULL, FALSE); return; } mainloop_set_trigger(dev->work); if (result->execution_status != PCMK_EXEC_DONE) { crm_warn("Assuming %s cannot fence %s " "because status could not be executed: %s%s%s%s", dev->id, search->host, pcmk_exec_status_str(result->execution_status), ((result->exit_reason == NULL)? "" : " ("), ((result->exit_reason == NULL)? "" : result->exit_reason), ((result->exit_reason == NULL)? "" : ")")); search_devices_record_result(search, dev->id, FALSE); return; } switch (result->exit_status) { case fence_status_unknown: crm_trace("%s reported it cannot fence %s", dev->id, search->host); break; case fence_status_active: case fence_status_inactive: crm_trace("%s reported it can fence %s", dev->id, search->host); can = TRUE; break; default: crm_warn("Assuming %s cannot fence %s " "(status returned unknown code %d)", dev->id, search->host, result->exit_status); break; } search_devices_record_result(search, dev->id, can); } static void dynamic_list_search_cb(int pid, const pcmk__action_result_t *result, void *user_data) { async_command_t *cmd = user_data; struct device_search_s *search = cmd->internal_user_data; stonith_device_t *dev = cmd->device ? g_hash_table_lookup(device_list, cmd->device) : NULL; gboolean can_fence = FALSE; free_async_command(cmd); /* Host/alias must be in the list output to be eligible to be fenced * * Will cause problems if down'd nodes aren't listed or (for virtual nodes) * if the guest is still listed despite being moved to another machine */ if (!dev) { search_devices_record_result(search, NULL, FALSE); return; } mainloop_set_trigger(dev->work); if (pcmk__result_ok(result)) { crm_info("Refreshing target list for %s", dev->id); g_list_free_full(dev->targets, free); dev->targets = stonith__parse_targets(result->action_stdout); dev->targets_age = time(NULL); } else if (dev->targets != NULL) { if (result->execution_status == PCMK_EXEC_DONE) { crm_info("Reusing most recent target list for %s " "because list returned error code %d", dev->id, result->exit_status); } else { crm_info("Reusing most recent target list for %s " "because list could not be executed: %s%s%s%s", dev->id, pcmk_exec_status_str(result->execution_status), ((result->exit_reason == NULL)? "" : " ("), ((result->exit_reason == NULL)? "" : result->exit_reason), ((result->exit_reason == NULL)? "" : ")")); } } else { // We have never successfully executed list if (result->execution_status == PCMK_EXEC_DONE) { crm_warn("Assuming %s cannot fence %s " "because list returned error code %d", dev->id, search->host, result->exit_status); } else { crm_warn("Assuming %s cannot fence %s " "because list could not be executed: %s%s%s%s", dev->id, search->host, pcmk_exec_status_str(result->execution_status), ((result->exit_reason == NULL)? "" : " ("), ((result->exit_reason == NULL)? "" : result->exit_reason), ((result->exit_reason == NULL)? "" : ")")); } /* Fall back to pcmk_host_check="status" if the user didn't explicitly * specify "dynamic-list". */ if (g_hash_table_lookup(dev->params, PCMK_STONITH_HOST_CHECK) == NULL) { crm_notice("Switching to pcmk_host_check='status' for %s", dev->id); g_hash_table_replace(dev->params, strdup(PCMK_STONITH_HOST_CHECK), strdup("status")); } } if (dev->targets) { const char *alias = g_hash_table_lookup(dev->aliases, search->host); if (!alias) { alias = search->host; } if (pcmk__str_in_list(alias, dev->targets, pcmk__str_casei)) { can_fence = TRUE; } } search_devices_record_result(search, dev->id, can_fence); } /*! * \internal * \brief Returns true if any key in first is not in second or second has a different value for key */ static int device_params_diff(GHashTable *first, GHashTable *second) { char *key = NULL; char *value = NULL; GHashTableIter gIter; g_hash_table_iter_init(&gIter, first); while (g_hash_table_iter_next(&gIter, (void **)&key, (void **)&value)) { if(strstr(key, "CRM_meta") == key) { continue; } else if(strcmp(key, "crm_feature_set") == 0) { continue; } else { char *other_value = g_hash_table_lookup(second, key); if (!other_value || !pcmk__str_eq(other_value, value, pcmk__str_casei)) { crm_trace("Different value for %s: %s != %s", key, other_value, value); return 1; } } } return 0; } /*! * \internal * \brief Checks to see if an identical device already exists in the device_list */ static stonith_device_t * device_has_duplicate(stonith_device_t * device) { stonith_device_t *dup = g_hash_table_lookup(device_list, device->id); if (!dup) { crm_trace("No match for %s", device->id); return NULL; } else if (!pcmk__str_eq(dup->agent, device->agent, pcmk__str_casei)) { crm_trace("Different agent: %s != %s", dup->agent, device->agent); return NULL; } /* Use calculate_operation_digest() here? */ if (device_params_diff(device->params, dup->params) || device_params_diff(dup->params, device->params)) { return NULL; } crm_trace("Match"); return dup; } int -stonith_device_register(xmlNode * msg, const char **desc, gboolean from_cib) +stonith_device_register(xmlNode *dev, gboolean from_cib) { stonith_device_t *dup = NULL; - stonith_device_t *device = build_device_from_xml(msg); + stonith_device_t *device = build_device_from_xml(dev); guint ndevices = 0; int rv = pcmk_ok; CRM_CHECK(device != NULL, return -ENOMEM); /* do we have a watchdog-device? */ if (pcmk__str_eq(device->id, STONITH_WATCHDOG_ID, pcmk__str_none) || pcmk__str_any_of(device->agent, STONITH_WATCHDOG_AGENT, STONITH_WATCHDOG_AGENT_INTERNAL, NULL)) do { if (stonith_watchdog_timeout_ms <= 0) { crm_err("Ignoring watchdog fence device without " "stonith-watchdog-timeout set."); rv = -ENODEV; /* fall through to cleanup & return */ } else if (!pcmk__str_any_of(device->agent, STONITH_WATCHDOG_AGENT, STONITH_WATCHDOG_AGENT_INTERNAL, NULL)) { crm_err("Ignoring watchdog fence device with unknown " "agent '%s' unequal '" STONITH_WATCHDOG_AGENT "'.", device->agent?device->agent:""); rv = -ENODEV; /* fall through to cleanup & return */ } else if (!pcmk__str_eq(device->id, STONITH_WATCHDOG_ID, pcmk__str_none)) { crm_err("Ignoring watchdog fence device " "named %s !='"STONITH_WATCHDOG_ID"'.", device->id?device->id:""); rv = -ENODEV; /* fall through to cleanup & return */ } else { if (pcmk__str_eq(device->agent, STONITH_WATCHDOG_AGENT, pcmk__str_none)) { /* this either has an empty list or the targets configured for watchdog-fencing */ g_list_free_full(stonith_watchdog_targets, free); stonith_watchdog_targets = device->targets; device->targets = NULL; } if (node_does_watchdog_fencing(stonith_our_uname)) { g_list_free_full(device->targets, free); device->targets = stonith__parse_targets(stonith_our_uname); g_hash_table_replace(device->params, strdup(PCMK_STONITH_HOST_LIST), strdup(stonith_our_uname)); /* proceed as with any other stonith-device */ break; } crm_debug("Skip registration of watchdog fence device on node not in host-list."); /* cleanup and fall through to more cleanup and return */ device->targets = NULL; stonith_device_remove(device->id, from_cib); } free_device(device); return rv; } while (0); dup = device_has_duplicate(device); if (dup) { ndevices = g_hash_table_size(device_list); crm_debug("Device '%s' already in device list (%d active device%s)", device->id, ndevices, pcmk__plural_s(ndevices)); free_device(device); device = dup; dup = g_hash_table_lookup(device_list, device->id); dup->dirty = FALSE; } else { stonith_device_t *old = g_hash_table_lookup(device_list, device->id); if (from_cib && old && old->api_registered) { /* If the cib is writing over an entry that is shared with a stonith client, * copy any pending ops that currently exist on the old entry to the new one. * Otherwise the pending ops will be reported as failures */ crm_info("Overwriting existing entry for %s from CIB", device->id); device->pending_ops = old->pending_ops; device->api_registered = TRUE; old->pending_ops = NULL; if (device->pending_ops) { mainloop_set_trigger(device->work); } } g_hash_table_replace(device_list, device->id, device); ndevices = g_hash_table_size(device_list); crm_notice("Added '%s' to device list (%d active device%s)", device->id, ndevices, pcmk__plural_s(ndevices)); } - if (desc) { - *desc = device->id; - } if (from_cib) { device->cib_registered = TRUE; } else { device->api_registered = TRUE; } return pcmk_ok; } void stonith_device_remove(const char *id, bool from_cib) { stonith_device_t *device = g_hash_table_lookup(device_list, id); guint ndevices = 0; if (!device) { ndevices = g_hash_table_size(device_list); crm_info("Device '%s' not found (%d active device%s)", id, ndevices, pcmk__plural_s(ndevices)); return; } if (from_cib) { device->cib_registered = FALSE; } else { device->verified = FALSE; device->api_registered = FALSE; } if (!device->cib_registered && !device->api_registered) { g_hash_table_remove(device_list, id); ndevices = g_hash_table_size(device_list); crm_info("Removed '%s' from device list (%d active device%s)", id, ndevices, pcmk__plural_s(ndevices)); } else { crm_trace("Not removing '%s' from device list (%d active) because " "still registered via:%s%s", id, g_hash_table_size(device_list), (device->cib_registered? " cib" : ""), (device->api_registered? " api" : "")); } } /*! * \internal * \brief Return the number of stonith levels registered for a node * * \param[in] tp Node's topology table entry * * \return Number of non-NULL levels in topology entry * \note This function is used only for log messages. */ static int count_active_levels(stonith_topology_t * tp) { int lpc = 0; int count = 0; for (lpc = 0; lpc < ST_LEVEL_MAX; lpc++) { if (tp->levels[lpc] != NULL) { count++; } } return count; } static void free_topology_entry(gpointer data) { stonith_topology_t *tp = data; int lpc = 0; for (lpc = 0; lpc < ST_LEVEL_MAX; lpc++) { if (tp->levels[lpc] != NULL) { g_list_free_full(tp->levels[lpc], free); } } free(tp->target); free(tp->target_value); free(tp->target_pattern); free(tp->target_attribute); free(tp); } void free_topology_list(void) { if (topology != NULL) { g_hash_table_destroy(topology); topology = NULL; } } void init_topology_list(void) { if (topology == NULL) { topology = pcmk__strkey_table(NULL, free_topology_entry); } } char * stonith_level_key(xmlNode *level, enum fenced_target_by mode) { if (mode == fenced_target_by_unknown) { mode = unpack_level_kind(level); } switch (mode) { case fenced_target_by_name: return crm_element_value_copy(level, XML_ATTR_STONITH_TARGET); case fenced_target_by_pattern: return crm_element_value_copy(level, XML_ATTR_STONITH_TARGET_PATTERN); case fenced_target_by_attribute: return crm_strdup_printf("%s=%s", crm_element_value(level, XML_ATTR_STONITH_TARGET_ATTRIBUTE), crm_element_value(level, XML_ATTR_STONITH_TARGET_VALUE)); default: return crm_strdup_printf("unknown-%s", ID(level)); } } /*! * \internal * \brief Parse target identification from topology level XML * * \param[in] level Topology level XML to parse * * \return How to identify target of \p level */ static int unpack_level_kind(xmlNode *level) { if (crm_element_value(level, XML_ATTR_STONITH_TARGET) != NULL) { return fenced_target_by_name; } if (crm_element_value(level, XML_ATTR_STONITH_TARGET_PATTERN) != NULL) { return fenced_target_by_pattern; } if (!stand_alone /* if standalone, there's no attribute manager */ && (crm_element_value(level, XML_ATTR_STONITH_TARGET_ATTRIBUTE) != NULL) && (crm_element_value(level, XML_ATTR_STONITH_TARGET_VALUE) == NULL)) { return fenced_target_by_attribute; } return fenced_target_by_unknown; } static stonith_key_value_t * parse_device_list(const char *devices) { int lpc = 0; int max = 0; int last = 0; stonith_key_value_t *output = NULL; if (devices == NULL) { return output; } max = strlen(devices); for (lpc = 0; lpc <= max; lpc++) { if (devices[lpc] == ',' || devices[lpc] == 0) { char *line = strndup(devices + last, lpc - last); output = stonith_key_value_add(output, NULL, line); free(line); last = lpc + 1; } } return output; } +/*! + * \internal + * \brief Unpack essential information from topology request XML + * + * \param[in] xml Request XML to search + * \param[out] mode If not NULL, where to store level kind + * \param[out] target If not NULL, where to store representation of target + * \param[out] id If not NULL, where to store level number + * \param[out] desc If not NULL, where to store log-friendly level description + * + * \return Topology level XML from within \p xml, or NULL if not found + * \note The caller is responsible for freeing \p *target and \p *desc if set. + */ +static xmlNode * +unpack_level_request(xmlNode *xml, enum fenced_target_by *mode, char **target, + int *id, char **desc) +{ + enum fenced_target_by local_mode = fenced_target_by_unknown; + char *local_target = NULL; + int local_id = 0; + + /* The level element can be the top element or lower. If top level, don't + * search by xpath, because it might give multiple hits if the XML is the + * CIB. + */ + if ((xml != NULL) + && !pcmk__str_eq(TYPE(xml), XML_TAG_FENCING_LEVEL, pcmk__str_none)) { + xml = get_xpath_object("//" XML_TAG_FENCING_LEVEL, xml, LOG_WARNING); + } + + if (xml == NULL) { + if (desc != NULL) { + *desc = crm_strdup_printf("missing"); + } + } else { + local_mode = unpack_level_kind(xml); + local_target = stonith_level_key(xml, local_mode); + crm_element_value_int(xml, XML_ATTR_STONITH_INDEX, &local_id); + if (desc != NULL) { + *desc = crm_strdup_printf("%s[%d]", local_target, local_id); + } + } + + if (mode != NULL) { + *mode = local_mode; + } + if (id != NULL) { + *id = local_id; + } + + if (target != NULL) { + *target = local_target; + } else { + free(local_target); + } + + return xml; +} + /*! * \internal * \brief Register a fencing topology level for a target * * Given an XML request specifying the target name, level index, and device IDs * for the level, this will create an entry for the target in the global topology * table if one does not already exist, then append the specified device IDs to * the entry's device list for the specified level. * * \param[in] msg XML request for STONITH level registration * \param[out] desc If not NULL, set to string representation "TARGET[LEVEL]" * \param[out] result Where to set result of registration */ void fenced_register_level(xmlNode *msg, char **desc, pcmk__action_result_t *result) { int id = 0; xmlNode *level; enum fenced_target_by mode; char *target; stonith_topology_t *tp; stonith_key_value_t *dIter = NULL; stonith_key_value_t *devices = NULL; CRM_CHECK(result != NULL, return); - if (msg == NULL) { - fenced_set_protocol_error(result); - return; - } - - /* Allow the XML here to point to the level tag directly, or wrapped in - * another tag. If directly, don't search by xpath, because it might give - * multiple hits (e.g. if the XML is the CIB). - */ - if (pcmk__str_eq(TYPE(msg), XML_TAG_FENCING_LEVEL, pcmk__str_casei)) { - level = msg; - } else { - level = get_xpath_object("//" XML_TAG_FENCING_LEVEL, msg, LOG_WARNING); - } + level = unpack_level_request(msg, &mode, &target, &id, desc); if (level == NULL) { fenced_set_protocol_error(result); return; } - mode = unpack_level_kind(level); - target = stonith_level_key(level, mode); - crm_element_value_int(level, XML_ATTR_STONITH_INDEX, &id); - - if (desc) { - *desc = crm_strdup_printf("%s[%d]", target, id); - } - // Ensure a valid target was specified if (mode == fenced_target_by_unknown) { crm_warn("Ignoring registration for topology level '%s' " "without valid target", crm_str(ID(level))); free(target); crm_log_xml_trace(level, "Bad level"); pcmk__format_result(result, CRM_EX_INVALID_PARAM, PCMK_EXEC_INVALID, "Invalid target for topology level '%s'", crm_str(ID(level))); return; } // Ensure level ID is in allowed range if ((id <= 0) || (id >= ST_LEVEL_MAX)) { crm_warn("Ignoring topology registration for %s with invalid level %d", target, id); free(target); crm_log_xml_trace(level, "Bad level"); pcmk__format_result(result, CRM_EX_INVALID_PARAM, PCMK_EXEC_INVALID, "Invalid level number '%s' for topology level '%s'", crm_str(crm_element_value(level, XML_ATTR_STONITH_INDEX)), crm_str(ID(level))); return; } /* Find or create topology table entry */ tp = g_hash_table_lookup(topology, target); if (tp == NULL) { tp = calloc(1, sizeof(stonith_topology_t)); if (tp == NULL) { pcmk__set_result(result, CRM_EX_ERROR, PCMK_EXEC_ERROR, strerror(ENOMEM)); + free(target); return; } tp->kind = mode; tp->target = target; tp->target_value = crm_element_value_copy(level, XML_ATTR_STONITH_TARGET_VALUE); tp->target_pattern = crm_element_value_copy(level, XML_ATTR_STONITH_TARGET_PATTERN); tp->target_attribute = crm_element_value_copy(level, XML_ATTR_STONITH_TARGET_ATTRIBUTE); g_hash_table_replace(topology, tp->target, tp); crm_trace("Added %s (%d) to the topology (%d active entries)", target, (int) mode, g_hash_table_size(topology)); } else { free(target); } if (tp->levels[id] != NULL) { crm_info("Adding to the existing %s[%d] topology entry", tp->target, id); } devices = parse_device_list(crm_element_value(level, XML_ATTR_STONITH_DEVICES)); for (dIter = devices; dIter; dIter = dIter->next) { const char *device = dIter->value; crm_trace("Adding device '%s' for %s[%d]", device, tp->target, id); tp->levels[id] = g_list_append(tp->levels[id], strdup(device)); } stonith_key_value_freeall(devices, 1, 1); { int nlevels = count_active_levels(tp); crm_info("Target %s has %d active fencing level%s", tp->target, nlevels, pcmk__plural_s(nlevels)); } pcmk__set_result(result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); } /*! * \internal * \brief Unregister a fencing topology level for a target * * Given an XML request specifying the target name and level index (or 0 for all * levels), this will remove any corresponding entry for the target from the * global topology table. * * \param[in] msg XML request for STONITH level registration * \param[out] desc If not NULL, set to string representation "TARGET[LEVEL]" * \param[out] result Where to set result of unregistration */ void fenced_unregister_level(xmlNode *msg, char **desc, pcmk__action_result_t *result) { int id = -1; stonith_topology_t *tp; char *target; xmlNode *level = NULL; CRM_CHECK(result != NULL, return); - if (msg == NULL) { - fenced_set_protocol_error(result); - return; - } - - // Unlike additions, removal requests should always have one level tag - level = get_xpath_object("//" XML_TAG_FENCING_LEVEL, msg, LOG_WARNING); + level = unpack_level_request(msg, NULL, &target, &id, desc); if (level == NULL) { fenced_set_protocol_error(result); return; } - target = stonith_level_key(level, fenced_target_by_unknown); - crm_element_value_int(level, XML_ATTR_STONITH_INDEX, &id); - // Ensure level ID is in allowed range if ((id < 0) || (id >= ST_LEVEL_MAX)) { crm_warn("Ignoring topology unregistration for %s with invalid level %d", target, id); free(target); crm_log_xml_trace(level, "Bad level"); pcmk__format_result(result, CRM_EX_INVALID_PARAM, PCMK_EXEC_INVALID, "Invalid level number '%s' for topology level '%s'", crm_str(crm_element_value(level, XML_ATTR_STONITH_INDEX)), crm_str(ID(level))); return; } - if (desc) { - *desc = crm_strdup_printf("%s[%d]", target, id); - } - tp = g_hash_table_lookup(topology, target); if (tp == NULL) { guint nentries = g_hash_table_size(topology); crm_info("No fencing topology found for %s (%d active %s)", target, nentries, pcmk__plural_alt(nentries, "entry", "entries")); } else if (id == 0 && g_hash_table_remove(topology, target)) { guint nentries = g_hash_table_size(topology); crm_info("Removed all fencing topology entries related to %s " "(%d active %s remaining)", target, nentries, pcmk__plural_alt(nentries, "entry", "entries")); } else if (tp->levels[id] != NULL) { guint nlevels; g_list_free_full(tp->levels[id], free); tp->levels[id] = NULL; nlevels = count_active_levels(tp); crm_info("Removed level %d from fencing topology for %s " "(%d active level%s remaining)", id, target, nlevels, pcmk__plural_s(nlevels)); } free(target); pcmk__set_result(result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); } static char * list_to_string(GList *list, const char *delim, gboolean terminate_with_delim) { int max = g_list_length(list); size_t delim_len = delim?strlen(delim):0; size_t alloc_size = 1 + (max?((max-1+(terminate_with_delim?1:0))*delim_len):0); char *rv; GList *gIter; for (gIter = list; gIter != NULL; gIter = gIter->next) { const char *value = (const char *) gIter->data; alloc_size += strlen(value); } rv = calloc(alloc_size, sizeof(char)); if (rv) { char *pos = rv; const char *lead_delim = ""; for (gIter = list; gIter != NULL; gIter = gIter->next) { const char *value = (const char *) gIter->data; pos = &pos[sprintf(pos, "%s%s", lead_delim, value)]; lead_delim = delim; } if (max && terminate_with_delim) { sprintf(pos, "%s", delim); } } return rv; } /*! * \internal * \brief Execute a fence agent action directly (and asynchronously) * * Handle a STONITH_OP_EXEC API message by scheduling a requested agent action * directly on a specified device. Only list, monitor, and status actions are * expected to use this call, though it should work with any agent command. * * \param[in] msg Request XML specifying action * \param[out] result Where to store result of action * * \note If the action is monitor, the device must be registered via the API * (CIB registration is not sufficient), because monitor should not be * possible unless the device is "started" (API registered). */ static void execute_agent_action(xmlNode *msg, pcmk__action_result_t *result) { xmlNode *dev = get_xpath_object("//" F_STONITH_DEVICE, msg, LOG_ERR); xmlNode *op = get_xpath_object("//@" F_STONITH_ACTION, msg, LOG_ERR); const char *id = crm_element_value(dev, F_STONITH_DEVICE); const char *action = crm_element_value(op, F_STONITH_ACTION); async_command_t *cmd = NULL; stonith_device_t *device = NULL; if ((id == NULL) || (action == NULL)) { crm_info("Malformed API action request: device %s, action %s", (id? id : "not specified"), (action? action : "not specified")); fenced_set_protocol_error(result); return; } if (pcmk__str_eq(id, STONITH_WATCHDOG_ID, pcmk__str_none)) { // Watchdog agent actions are implemented internally if (stonith_watchdog_timeout_ms <= 0) { pcmk__set_result(result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, "Watchdog fence device not configured"); return; } else if (pcmk__str_eq(action, "list", pcmk__str_casei)) { pcmk__set_result(result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); pcmk__set_result_output(result, list_to_string(stonith_watchdog_targets, "\n", TRUE), NULL); return; } else if (pcmk__str_eq(action, "monitor", pcmk__str_casei)) { pcmk__set_result(result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); return; } } device = g_hash_table_lookup(device_list, id); if (device == NULL) { crm_info("Ignoring API '%s' action request because device %s not found", action, id); pcmk__format_result(result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, "'%s' not found", id); return; } else if (!device->api_registered && !strcmp(action, "monitor")) { // Monitors may run only on "started" (API-registered) devices crm_info("Ignoring API '%s' action request because device %s not active", action, id); pcmk__format_result(result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, "'%s' not active", id); return; } cmd = create_async_command(msg); if (cmd == NULL) { fenced_set_protocol_error(result); return; } schedule_stonith_command(cmd, device); pcmk__set_result(result, CRM_EX_OK, PCMK_EXEC_PENDING, NULL); } static void search_devices_record_result(struct device_search_s *search, const char *device, gboolean can_fence) { search->replies_received++; if (can_fence && device) { search->capable = g_list_append(search->capable, strdup(device)); } if (search->replies_needed == search->replies_received) { guint ndevices = g_list_length(search->capable); crm_debug("Search found %d device%s that can perform '%s' targeting %s", ndevices, pcmk__plural_s(ndevices), (search->action? search->action : "unknown action"), (search->host? search->host : "any node")); search->callback(search->capable, search->user_data); free(search->host); free(search->action); free(search); } } /*! * \internal * \brief Check whether the local host is allowed to execute a fencing action * * \param[in] device Fence device to check * \param[in] action Fence action to check * \param[in] target Hostname of fence target * \param[in] allow_suicide Whether self-fencing is allowed for this operation * * \return TRUE if local host is allowed to execute action, FALSE otherwise */ static gboolean localhost_is_eligible(const stonith_device_t *device, const char *action, const char *target, gboolean allow_suicide) { gboolean localhost_is_target = pcmk__str_eq(target, stonith_our_uname, pcmk__str_casei); if (device && action && device->on_target_actions && strstr(device->on_target_actions, action)) { if (!localhost_is_target) { crm_trace("Operation '%s' using %s can only be executed for " "local host, not %s", action, device->id, target); return FALSE; } } else if (localhost_is_target && !allow_suicide) { crm_trace("'%s' operation does not support self-fencing", action); return FALSE; } return TRUE; } static void can_fence_host_with_device(stonith_device_t * dev, struct device_search_s *search) { gboolean can = FALSE; const char *check_type = NULL; const char *host = search->host; const char *alias = NULL; CRM_LOG_ASSERT(dev != NULL); if (dev == NULL) { goto search_report_results; } else if (host == NULL) { can = TRUE; goto search_report_results; } /* Short-circuit query if this host is not allowed to perform the action */ if (pcmk__str_eq(search->action, "reboot", pcmk__str_casei)) { /* A "reboot" *might* get remapped to "off" then "on", so short-circuit * only if all three are disallowed. If only one or two are disallowed, * we'll report that with the results. We never allow suicide for * remapped "on" operations because the host is off at that point. */ if (!localhost_is_eligible(dev, "reboot", host, search->allow_suicide) && !localhost_is_eligible(dev, "off", host, search->allow_suicide) && !localhost_is_eligible(dev, "on", host, FALSE)) { goto search_report_results; } } else if (!localhost_is_eligible(dev, search->action, host, search->allow_suicide)) { goto search_report_results; } alias = g_hash_table_lookup(dev->aliases, host); if (alias == NULL) { alias = host; } check_type = target_list_type(dev); if (pcmk__str_eq(check_type, PCMK__VALUE_NONE, pcmk__str_casei)) { can = TRUE; } else if (pcmk__str_eq(check_type, "static-list", pcmk__str_casei)) { /* Presence in the hostmap is sufficient * Only use if all hosts on which the device can be active can always fence all listed hosts */ if (pcmk__str_in_list(host, dev->targets, pcmk__str_casei)) { can = TRUE; } else if (g_hash_table_lookup(dev->params, PCMK_STONITH_HOST_MAP) && g_hash_table_lookup(dev->aliases, host)) { can = TRUE; } } else if (pcmk__str_eq(check_type, "dynamic-list", pcmk__str_casei)) { time_t now = time(NULL); if (dev->targets == NULL || dev->targets_age + 60 < now) { int device_timeout = get_action_timeout(dev, "list", search->per_device_timeout); if (device_timeout > search->per_device_timeout) { crm_notice("Since the pcmk_list_timeout(%ds) parameter of %s is larger than stonith-timeout(%ds), timeout may occur", device_timeout, dev->id, search->per_device_timeout); } crm_trace("Running '%s' to check whether %s is eligible to fence %s (%s)", check_type, dev->id, search->host, search->action); schedule_internal_command(__func__, dev, "list", NULL, search->per_device_timeout, search, dynamic_list_search_cb); /* we'll respond to this search request async in the cb */ return; } if (pcmk__str_in_list(alias, dev->targets, pcmk__str_casei)) { can = TRUE; } } else if (pcmk__str_eq(check_type, "status", pcmk__str_casei)) { int device_timeout = get_action_timeout(dev, check_type, search->per_device_timeout); if (device_timeout > search->per_device_timeout) { crm_notice("Since the pcmk_status_timeout(%ds) parameter of %s is larger than stonith-timeout(%ds), timeout may occur", device_timeout, dev->id, search->per_device_timeout); } crm_trace("Running '%s' to check whether %s is eligible to fence %s (%s)", check_type, dev->id, search->host, search->action); schedule_internal_command(__func__, dev, "status", search->host, search->per_device_timeout, search, status_search_cb); /* we'll respond to this search request async in the cb */ return; } else { crm_err("Invalid value for " PCMK_STONITH_HOST_CHECK ": %s", check_type); check_type = "Invalid " PCMK_STONITH_HOST_CHECK; } if (pcmk__str_eq(host, alias, pcmk__str_casei)) { crm_info("%s is%s eligible to fence (%s) %s: %s", dev->id, (can? "" : " not"), search->action, host, check_type); } else { crm_info("%s is%s eligible to fence (%s) %s (aka. '%s'): %s", dev->id, (can? "" : " not"), search->action, host, alias, check_type); } search_report_results: search_devices_record_result(search, dev ? dev->id : NULL, can); } static void search_devices(gpointer key, gpointer value, gpointer user_data) { stonith_device_t *dev = value; struct device_search_s *search = user_data; can_fence_host_with_device(dev, search); } #define DEFAULT_QUERY_TIMEOUT 20 static void get_capable_devices(const char *host, const char *action, int timeout, bool suicide, void *user_data, void (*callback) (GList * devices, void *user_data)) { struct device_search_s *search; guint ndevices = g_hash_table_size(device_list); if (ndevices == 0) { callback(NULL, user_data); return; } search = calloc(1, sizeof(struct device_search_s)); if (!search) { crm_crit("Cannot search for capable fence devices: %s", strerror(ENOMEM)); callback(NULL, user_data); return; } pcmk__str_update(&search->host, host); pcmk__str_update(&search->action, action); search->per_device_timeout = timeout; search->allow_suicide = suicide; search->callback = callback; search->user_data = user_data; /* We are guaranteed this many replies, even if a device is * unregistered while the search is in progress. */ search->replies_needed = ndevices; crm_debug("Searching %d device%s to see which can execute '%s' targeting %s", ndevices, pcmk__plural_s(ndevices), (search->action? search->action : "unknown action"), (search->host? search->host : "any node")); g_hash_table_foreach(device_list, search_devices, search); } struct st_query_data { xmlNode *reply; char *remote_peer; char *client_id; char *target; char *action; int call_options; }; /*! * \internal * \brief Add action-specific attributes to query reply XML * * \param[in,out] xml XML to add attributes to * \param[in] action Fence action * \param[in] device Fence device * \param[in] target Fence target */ static void add_action_specific_attributes(xmlNode *xml, const char *action, stonith_device_t *device, const char *target) { int action_specific_timeout; int delay_max; int delay_base; CRM_CHECK(xml && action && device, return); if (is_action_required(action, device)) { crm_trace("Action '%s' is required using %s", action, device->id); crm_xml_add_int(xml, F_STONITH_DEVICE_REQUIRED, 1); } action_specific_timeout = get_action_timeout(device, action, 0); if (action_specific_timeout) { crm_trace("Action '%s' has timeout %dms using %s", action, action_specific_timeout, device->id); crm_xml_add_int(xml, F_STONITH_ACTION_TIMEOUT, action_specific_timeout); } delay_max = get_action_delay_max(device, action); if (delay_max > 0) { crm_trace("Action '%s' has maximum random delay %dms using %s", action, delay_max, device->id); crm_xml_add_int(xml, F_STONITH_DELAY_MAX, delay_max / 1000); } delay_base = get_action_delay_base(device, action, target); if (delay_base > 0) { crm_xml_add_int(xml, F_STONITH_DELAY_BASE, delay_base / 1000); } if ((delay_max > 0) && (delay_base == 0)) { crm_trace("Action '%s' has maximum random delay %dms using %s", action, delay_max, device->id); } else if ((delay_max == 0) && (delay_base > 0)) { crm_trace("Action '%s' has a static delay of %dms using %s", action, delay_base, device->id); } else if ((delay_max > 0) && (delay_base > 0)) { crm_trace("Action '%s' has a minimum delay of %dms and a randomly chosen " "maximum delay of %dms using %s", action, delay_base, delay_max, device->id); } } /*! * \internal * \brief Add "disallowed" attribute to query reply XML if appropriate * * \param[in,out] xml XML to add attribute to * \param[in] action Fence action * \param[in] device Fence device * \param[in] target Fence target * \param[in] allow_suicide Whether self-fencing is allowed */ static void add_disallowed(xmlNode *xml, const char *action, stonith_device_t *device, const char *target, gboolean allow_suicide) { if (!localhost_is_eligible(device, action, target, allow_suicide)) { crm_trace("Action '%s' using %s is disallowed for local host", action, device->id); pcmk__xe_set_bool_attr(xml, F_STONITH_ACTION_DISALLOWED, true); } } /*! * \internal * \brief Add child element with action-specific values to query reply XML * * \param[in,out] xml XML to add attribute to * \param[in] action Fence action * \param[in] device Fence device * \param[in] target Fence target * \param[in] allow_suicide Whether self-fencing is allowed */ static void add_action_reply(xmlNode *xml, const char *action, stonith_device_t *device, const char *target, gboolean allow_suicide) { xmlNode *child = create_xml_node(xml, F_STONITH_ACTION); crm_xml_add(child, XML_ATTR_ID, action); add_action_specific_attributes(child, action, device, target); add_disallowed(child, action, device, target, allow_suicide); } static void stonith_query_capable_device_cb(GList * devices, void *user_data) { struct st_query_data *query = user_data; int available_devices = 0; xmlNode *dev = NULL; xmlNode *list = NULL; GList *lpc = NULL; pcmk__client_t *client = NULL; if (query->client_id != NULL) { client = pcmk__find_client_by_id(query->client_id); if ((client == NULL) && (query->remote_peer == NULL)) { crm_trace("Skipping reply to %s: no longer a client", query->client_id); goto done; } } /* Pack the results into XML */ list = create_xml_node(NULL, __func__); crm_xml_add(list, F_STONITH_TARGET, query->target); for (lpc = devices; lpc != NULL; lpc = lpc->next) { stonith_device_t *device = g_hash_table_lookup(device_list, lpc->data); const char *action = query->action; if (!device) { /* It is possible the device got unregistered while * determining who can fence the target */ continue; } available_devices++; dev = create_xml_node(list, F_STONITH_DEVICE); crm_xml_add(dev, XML_ATTR_ID, device->id); crm_xml_add(dev, "namespace", device->namespace); crm_xml_add(dev, "agent", device->agent); crm_xml_add_int(dev, F_STONITH_DEVICE_VERIFIED, device->verified); /* If the originating fencer wants to reboot the node, and we have a * capable device that doesn't support "reboot", remap to "off" instead. */ if (!pcmk_is_set(device->flags, st_device_supports_reboot) && pcmk__str_eq(query->action, "reboot", pcmk__str_casei)) { crm_trace("%s doesn't support reboot, using values for off instead", device->id); action = "off"; } /* Add action-specific values if available */ add_action_specific_attributes(dev, action, device, query->target); if (pcmk__str_eq(query->action, "reboot", pcmk__str_casei)) { /* A "reboot" *might* get remapped to "off" then "on", so after * sending the "reboot"-specific values in the main element, we add * sub-elements for "off" and "on" values. * * We short-circuited earlier if "reboot", "off" and "on" are all * disallowed for the local host. However if only one or two are * disallowed, we send back the results and mark which ones are * disallowed. If "reboot" is disallowed, this might cause problems * with older fencer versions, which won't check for it. Older * versions will ignore "off" and "on", so they are not a problem. */ add_disallowed(dev, action, device, query->target, pcmk_is_set(query->call_options, st_opt_allow_suicide)); add_action_reply(dev, "off", device, query->target, pcmk_is_set(query->call_options, st_opt_allow_suicide)); add_action_reply(dev, "on", device, query->target, FALSE); } /* A query without a target wants device parameters */ if (query->target == NULL) { xmlNode *attrs = create_xml_node(dev, XML_TAG_ATTRS); g_hash_table_foreach(device->params, hash2field, attrs); } } crm_xml_add_int(list, F_STONITH_AVAILABLE_DEVICES, available_devices); if (query->target) { crm_debug("Found %d matching device%s for target '%s'", available_devices, pcmk__plural_s(available_devices), query->target); } else { crm_debug("%d device%s installed", available_devices, pcmk__plural_s(available_devices)); } if (list != NULL) { crm_log_xml_trace(list, "Add query results"); add_message_xml(query->reply, F_STONITH_CALLDATA, list); } stonith_send_reply(query->reply, query->call_options, query->remote_peer, client); done: free_xml(query->reply); free(query->remote_peer); free(query->client_id); free(query->target); free(query->action); free(query); free_xml(list); g_list_free_full(devices, free); } /*! * \internal * \brief Log the result of an asynchronous command * * \param[in] cmd Command the result is for * \param[in] result Result of command * \param[in] pid Process ID of command, if available * \param[in] next Alternate device that will be tried if command failed * \param[in] op_merged Whether this command was merged with an earlier one */ static void log_async_result(async_command_t *cmd, const pcmk__action_result_t *result, int pid, const char *next, bool op_merged) { int log_level = LOG_ERR; int output_log_level = LOG_NEVER; guint devices_remaining = g_list_length(cmd->device_next); GString *msg = g_string_sized_new(80); // Reasonable starting size // Choose log levels appropriately if we have a result if (pcmk__result_ok(result)) { log_level = (cmd->victim == NULL)? LOG_DEBUG : LOG_NOTICE; if ((result->action_stdout != NULL) && !pcmk__str_eq(cmd->action, "metadata", pcmk__str_casei)) { output_log_level = LOG_DEBUG; } next = NULL; } else { log_level = (cmd->victim == NULL)? LOG_NOTICE : LOG_ERR; if ((result->action_stdout != NULL) && !pcmk__str_eq(cmd->action, "metadata", pcmk__str_casei)) { output_log_level = LOG_WARNING; } } // Build the log message piece by piece g_string_printf(msg, "Operation '%s' ", cmd->action); if (pid != 0) { g_string_append_printf(msg, "[%d] ", pid); } if (cmd->victim != NULL) { g_string_append_printf(msg, "targeting %s ", cmd->victim); } if (cmd->device != NULL) { g_string_append_printf(msg, "using %s ", cmd->device); } // Add exit status or execution status as appropriate if (result->execution_status == PCMK_EXEC_DONE) { g_string_append_printf(msg, "returned %d", result->exit_status); } else { g_string_append_printf(msg, "could not be executed: %s", pcmk_exec_status_str(result->execution_status)); } // Add exit reason and next device if appropriate if (result->exit_reason != NULL) { g_string_append_printf(msg, " (%s)", result->exit_reason); } if (next != NULL) { g_string_append_printf(msg, ", retrying with %s", next); } if (devices_remaining > 0) { g_string_append_printf(msg, " (%u device%s remaining)", (unsigned int) devices_remaining, pcmk__plural_s(devices_remaining)); } g_string_append_printf(msg, " " CRM_XS " %scall %d from %s", (op_merged? "merged " : ""), cmd->id, cmd->client_name); // Log the result do_crm_log(log_level, "%s", msg->str); g_string_free(msg, TRUE); // Log the output (which may have multiple lines), if appropriate if (output_log_level != LOG_NEVER) { char *prefix = crm_strdup_printf("%s[%d]", cmd->device, pid); crm_log_output(output_log_level, prefix, result->action_stdout); free(prefix); } } /*! * \internal * \brief Reply to requester after asynchronous command completion * * \param[in] cmd Command that completed * \param[in] result Result of command * \param[in] pid Process ID of command, if available * \param[in] merged If true, command was merged with another, not executed */ static void send_async_reply(async_command_t *cmd, const pcmk__action_result_t *result, int pid, bool merged) { xmlNode *reply = NULL; pcmk__client_t *client = NULL; CRM_CHECK((cmd != NULL) && (result != NULL), return); log_async_result(cmd, result, pid, NULL, merged); if (cmd->client != NULL) { client = pcmk__find_client_by_id(cmd->client); if ((client == NULL) && (cmd->origin == NULL)) { crm_trace("Skipping reply to %s: no longer a client", cmd->client); return; } } reply = construct_async_reply(cmd, result); if (merged) { pcmk__xe_set_bool_attr(reply, F_STONITH_MERGED, true); } if (!stand_alone && pcmk__is_fencing_action(cmd->action) && pcmk__str_eq(cmd->origin, cmd->victim, pcmk__str_casei)) { /* The target was also the originator, so broadcast the result on its * behalf (since it will be unable to). */ crm_trace("Broadcast '%s' result for %s (target was also originator)", cmd->action, cmd->victim); crm_xml_add(reply, F_SUBTYPE, "broadcast"); crm_xml_add(reply, F_STONITH_OPERATION, T_STONITH_NOTIFY); send_cluster_message(NULL, crm_msg_stonith_ng, reply, FALSE); } else { // Reply only to the originator stonith_send_reply(reply, cmd->options, cmd->origin, client); } crm_log_xml_trace(reply, "Reply"); free_xml(reply); if (stand_alone) { /* Do notification with a clean data object */ xmlNode *notify_data = create_xml_node(NULL, T_STONITH_NOTIFY_FENCE); stonith__xe_set_result(notify_data, result); crm_xml_add(notify_data, F_STONITH_TARGET, cmd->victim); crm_xml_add(notify_data, F_STONITH_OPERATION, cmd->op); crm_xml_add(notify_data, F_STONITH_DELEGATE, "localhost"); crm_xml_add(notify_data, F_STONITH_DEVICE, cmd->device); crm_xml_add(notify_data, F_STONITH_REMOTE_OP_ID, cmd->remote_op_id); crm_xml_add(notify_data, F_STONITH_ORIGIN, cmd->client); fenced_send_notification(T_STONITH_NOTIFY_FENCE, result, notify_data); fenced_send_notification(T_STONITH_NOTIFY_HISTORY, NULL, NULL); } } static void cancel_stonith_command(async_command_t * cmd) { stonith_device_t *device; CRM_CHECK(cmd != NULL, return); if (!cmd->device) { return; } device = g_hash_table_lookup(device_list, cmd->device); if (device) { crm_trace("Cancel scheduled '%s' action using %s", cmd->action, device->id); device->pending_ops = g_list_remove(device->pending_ops, cmd); } } static void st_child_done(int pid, const pcmk__action_result_t *result, void *user_data) { stonith_device_t *device = NULL; stonith_device_t *next_device = NULL; async_command_t *cmd = user_data; GList *gIter = NULL; GList *gIterNext = NULL; CRM_CHECK(cmd != NULL, return); cmd->active_on = NULL; /* The device is ready to do something else now */ device = g_hash_table_lookup(device_list, cmd->device); if (device) { if (!device->verified && pcmk__result_ok(result) && (pcmk__strcase_any_of(cmd->action, "list", "monitor", "status", NULL))) { device->verified = TRUE; } mainloop_set_trigger(device->work); } if (pcmk__result_ok(result)) { GList *iter; /* see if there are any required devices left to execute for this op */ for (iter = cmd->device_next; iter != NULL; iter = iter->next) { next_device = g_hash_table_lookup(device_list, iter->data); if (next_device != NULL && is_action_required(cmd->action, next_device)) { cmd->device_next = iter->next; break; } next_device = NULL; } } else if ((cmd->device_next != NULL) && !is_action_required(cmd->action, device)) { /* if this device didn't work out, see if there are any others we can try. * if the failed device was 'required', we can't pick another device. */ next_device = g_hash_table_lookup(device_list, cmd->device_next->data); cmd->device_next = cmd->device_next->next; } /* this operation requires more fencing, hooray! */ if (next_device) { log_async_result(cmd, result, pid, next_device->id, false); schedule_stonith_command(cmd, next_device); /* Prevent cmd from being freed */ cmd = NULL; goto done; } send_async_reply(cmd, result, pid, false); if (!pcmk__result_ok(result)) { goto done; } /* Check to see if any operations are scheduled to do the exact * same thing that just completed. If so, rather than * performing the same fencing operation twice, return the result * of this operation for all pending commands it matches. */ for (gIter = cmd_list; gIter != NULL; gIter = gIterNext) { async_command_t *cmd_other = gIter->data; gIterNext = gIter->next; if (cmd == cmd_other) { continue; } /* A pending scheduled command matches the command that just finished if. * 1. The client connections are different. * 2. The node victim is the same. * 3. The fencing action is the same. * 4. The device scheduled to execute the action is the same. */ if (pcmk__str_eq(cmd->client, cmd_other->client, pcmk__str_casei) || !pcmk__str_eq(cmd->victim, cmd_other->victim, pcmk__str_casei) || !pcmk__str_eq(cmd->action, cmd_other->action, pcmk__str_casei) || !pcmk__str_eq(cmd->device, cmd_other->device, pcmk__str_casei)) { continue; } /* Duplicate merging will do the right thing for either type of remapped * reboot. If the executing fencer remapped an unsupported reboot to * off, then cmd->action will be reboot and will be merged with any * other reboot requests. If the originating fencer remapped a * topology reboot to off then on, we will get here once with * cmd->action "off" and once with "on", and they will be merged * separately with similar requests. */ crm_notice("Merging fencing action '%s' targeting %s originating from " "client %s with identical fencing request from client %s", cmd_other->action, cmd_other->victim, cmd_other->client_name, cmd->client_name); cmd_list = g_list_remove_link(cmd_list, gIter); send_async_reply(cmd_other, result, pid, true); cancel_stonith_command(cmd_other); free_async_command(cmd_other); g_list_free_1(gIter); } done: free_async_command(cmd); } static gint sort_device_priority(gconstpointer a, gconstpointer b) { const stonith_device_t *dev_a = a; const stonith_device_t *dev_b = b; if (dev_a->priority > dev_b->priority) { return -1; } else if (dev_a->priority < dev_b->priority) { return 1; } return 0; } static void stonith_fence_get_devices_cb(GList * devices, void *user_data) { async_command_t *cmd = user_data; stonith_device_t *device = NULL; guint ndevices = g_list_length(devices); crm_info("Found %d matching device%s for target '%s'", ndevices, pcmk__plural_s(ndevices), cmd->victim); if (devices != NULL) { /* Order based on priority */ devices = g_list_sort(devices, sort_device_priority); device = g_hash_table_lookup(device_list, devices->data); } if (device == NULL) { // No device found pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; pcmk__format_result(&result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, "No device configured for target '%s'", cmd->victim); send_async_reply(cmd, &result, 0, false); pcmk__reset_result(&result); free_async_command(cmd); g_list_free_full(devices, free); } else { // Device found, schedule it for fencing cmd->device_list = devices; cmd->device_next = devices->next; schedule_stonith_command(cmd, device); } } /*! * \internal * \brief Execute a fence action via the local node * * \param[in] msg Fencing request * \param[out] result Where to store result of fence action */ static void fence_locally(xmlNode *msg, pcmk__action_result_t *result) { const char *device_id = NULL; stonith_device_t *device = NULL; async_command_t *cmd = create_async_command(msg); xmlNode *dev = get_xpath_object("//@" F_STONITH_TARGET, msg, LOG_ERR); CRM_CHECK(result != NULL, return); if (cmd == NULL) { fenced_set_protocol_error(result); return; } device_id = crm_element_value(dev, F_STONITH_DEVICE); if (device_id != NULL) { device = g_hash_table_lookup(device_list, device_id); if (device == NULL) { crm_err("Requested device '%s' is not available", device_id); pcmk__format_result(result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, "Requested device '%s' not found", device_id); return; } schedule_stonith_command(cmd, device); } else { const char *host = crm_element_value(dev, F_STONITH_TARGET); if (pcmk_is_set(cmd->options, st_opt_cs_nodeid)) { int nodeid = 0; crm_node_t *node = NULL; pcmk__scan_min_int(host, &nodeid, 0); node = pcmk__search_known_node_cache(nodeid, NULL, CRM_GET_PEER_ANY); if (node != NULL) { host = node->uname; } } /* If we get to here, then self-fencing is implicitly allowed */ get_capable_devices(host, cmd->action, cmd->default_timeout, TRUE, cmd, stonith_fence_get_devices_cb); } pcmk__set_result(result, CRM_EX_OK, PCMK_EXEC_PENDING, NULL); } /*! * \internal * \brief Build an XML reply for a fencing operation * * \param[in] request Request that reply is for * \param[in] data If not NULL, add to reply as call data * \param[in] result Full result of fencing operation * * \return Newly created XML reply * \note The caller is responsible for freeing the result. * \note This has some overlap with construct_async_reply(), but that copies * values from an async_command_t, whereas this one copies them from the * request. */ xmlNode * fenced_construct_reply(xmlNode *request, xmlNode *data, pcmk__action_result_t *result) { xmlNode *reply = NULL; reply = create_xml_node(NULL, T_STONITH_REPLY); crm_xml_add(reply, "st_origin", __func__); crm_xml_add(reply, F_TYPE, T_STONITH_NG); stonith__xe_set_result(reply, result); if (request == NULL) { /* Most likely, this is the result of a stonith operation that was * initiated before we came up. Unfortunately that means we lack enough * information to provide clients with a full result. * * @TODO Maybe synchronize this information at start-up? */ crm_warn("Missing request information for client notifications for " "operation with result '%s' (initiated before we came up?)", pcmk_exec_status_str(result->execution_status)); } else { const char *name = NULL; const char *value = NULL; // Attributes to copy from request to reply const char *names[] = { F_STONITH_OPERATION, F_STONITH_CALLID, F_STONITH_CLIENTID, F_STONITH_CLIENTNAME, F_STONITH_REMOTE_OP_ID, F_STONITH_CALLOPTS }; for (int lpc = 0; lpc < PCMK__NELEM(names); lpc++) { name = names[lpc]; value = crm_element_value(request, name); crm_xml_add(reply, name, value); } if (data != NULL) { add_message_xml(reply, F_STONITH_CALLDATA, data); } } return reply; } /*! * \internal * \brief Build an XML reply to an asynchronous fencing command * * \param[in] cmd Fencing command that reply is for * \param[in] result Command result */ static xmlNode * construct_async_reply(async_command_t *cmd, const pcmk__action_result_t *result) { xmlNode *reply = create_xml_node(NULL, T_STONITH_REPLY); crm_xml_add(reply, "st_origin", __func__); crm_xml_add(reply, F_TYPE, T_STONITH_NG); crm_xml_add(reply, F_STONITH_OPERATION, cmd->op); crm_xml_add(reply, F_STONITH_DEVICE, cmd->device); crm_xml_add(reply, F_STONITH_REMOTE_OP_ID, cmd->remote_op_id); crm_xml_add(reply, F_STONITH_CLIENTID, cmd->client); crm_xml_add(reply, F_STONITH_CLIENTNAME, cmd->client_name); crm_xml_add(reply, F_STONITH_TARGET, cmd->victim); crm_xml_add(reply, F_STONITH_ACTION, cmd->op); crm_xml_add(reply, F_STONITH_ORIGIN, cmd->origin); crm_xml_add_int(reply, F_STONITH_CALLID, cmd->id); crm_xml_add_int(reply, F_STONITH_CALLOPTS, cmd->options); stonith__xe_set_result(reply, result); return reply; } bool fencing_peer_active(crm_node_t *peer) { if (peer == NULL) { return FALSE; } else if (peer->uname == NULL) { return FALSE; } else if (pcmk_is_set(peer->processes, crm_get_cluster_proc())) { return TRUE; } return FALSE; } void set_fencing_completed(remote_fencing_op_t *op) { struct timespec tv; qb_util_timespec_from_epoch_get(&tv); op->completed = tv.tv_sec; op->completed_nsec = tv.tv_nsec; } /*! * \internal * \brief Look for alternate node needed if local node shouldn't fence target * * \param[in] target Node that must be fenced * * \return Name of an alternate node that should fence \p target if any, * or NULL otherwise */ static const char * check_alternate_host(const char *target) { const char *alternate_host = NULL; crm_trace("Checking if we (%s) can fence %s", stonith_our_uname, target); if (find_topology_for_host(target) && pcmk__str_eq(target, stonith_our_uname, pcmk__str_casei)) { GHashTableIter gIter; crm_node_t *entry = NULL; g_hash_table_iter_init(&gIter, crm_peer_cache); while (g_hash_table_iter_next(&gIter, NULL, (void **)&entry)) { crm_trace("Checking for %s.%d != %s", entry->uname, entry->id, target); if (fencing_peer_active(entry) && !pcmk__str_eq(entry->uname, target, pcmk__str_casei)) { alternate_host = entry->uname; break; } } if (alternate_host == NULL) { crm_err("No alternate host available to handle request " "for self-fencing with topology"); g_hash_table_iter_init(&gIter, crm_peer_cache); while (g_hash_table_iter_next(&gIter, NULL, (void **)&entry)) { crm_notice("Peer[%d] %s", entry->id, entry->uname); } } } return alternate_host; } /*! * \internal * \brief Send a reply to a CPG peer or IPC client * * \param[in] reply XML reply to send * \param[in] call_options Send synchronously if st_opt_sync_call is set here * \param[in] remote_peer If not NULL, name of peer node to send CPG reply * \param[in] client If not NULL, client to send IPC reply */ static void stonith_send_reply(xmlNode *reply, int call_options, const char *remote_peer, pcmk__client_t *client) { CRM_CHECK((reply != NULL) && ((remote_peer != NULL) || (client != NULL)), return); if (remote_peer == NULL) { do_local_reply(reply, client, call_options); } else { send_cluster_message(crm_get_peer(0, remote_peer), crm_msg_stonith_ng, reply, FALSE); } } static void remove_relay_op(xmlNode * request) { xmlNode *dev = get_xpath_object("//@" F_STONITH_ACTION, request, LOG_TRACE); const char *relay_op_id = NULL; const char *op_id = NULL; const char *client_name = NULL; const char *target = NULL; remote_fencing_op_t *relay_op = NULL; if (dev) { target = crm_element_value(dev, F_STONITH_TARGET); } relay_op_id = crm_element_value(request, F_STONITH_REMOTE_OP_ID_RELAY); op_id = crm_element_value(request, F_STONITH_REMOTE_OP_ID); client_name = crm_element_value(request, F_STONITH_CLIENTNAME); /* Delete RELAY operation. */ if (relay_op_id && target && pcmk__str_eq(target, stonith_our_uname, pcmk__str_casei)) { relay_op = g_hash_table_lookup(stonith_remote_op_list, relay_op_id); if (relay_op) { GHashTableIter iter; remote_fencing_op_t *list_op = NULL; g_hash_table_iter_init(&iter, stonith_remote_op_list); /* If the operation to be deleted is registered as a duplicate, delete the registration. */ while (g_hash_table_iter_next(&iter, NULL, (void **)&list_op)) { GList *dup_iter = NULL; if (list_op != relay_op) { for (dup_iter = list_op->duplicates; dup_iter != NULL; dup_iter = dup_iter->next) { remote_fencing_op_t *other = dup_iter->data; if (other == relay_op) { other->duplicates = g_list_remove(other->duplicates, relay_op); break; } } } } crm_debug("Deleting relay op %s ('%s' targeting %s for %s), " "replaced by op %s ('%s' targeting %s for %s)", relay_op->id, relay_op->action, relay_op->target, relay_op->client_name, op_id, relay_op->action, target, client_name); g_hash_table_remove(stonith_remote_op_list, relay_op_id); } } } /*! * \internal * \brief Check whether an API request was sent by a privileged user * * API commands related to fencing configuration may be done only by privileged * IPC users (i.e. root or hacluster), because all other users should go through * the CIB to have ACLs applied. If no client was given, this is a peer request, * which is always allowed. * * \param[in] c IPC client that sent request (or NULL if sent by CPG peer) * \param[in] op Requested API operation (for logging only) * * \return true if sender is peer or privileged client, otherwise false */ static inline bool is_privileged(pcmk__client_t *c, const char *op) { if ((c == NULL) || pcmk_is_set(c->flags, pcmk__client_privileged)) { return true; } else { crm_warn("Rejecting IPC request '%s' from unprivileged client %s", crm_str(op), pcmk__client_name(c)); return false; } } // CRM_OP_REGISTER static xmlNode * handle_register_request(pcmk__request_t *request) { xmlNode *reply = create_xml_node(NULL, "reply"); CRM_ASSERT(request->ipc_client != NULL); crm_xml_add(reply, F_STONITH_OPERATION, CRM_OP_REGISTER); crm_xml_add(reply, F_STONITH_CLIENTID, request->ipc_client->id); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); pcmk__set_request_flags(request, pcmk__request_reuse_options); return reply; } // STONITH_OP_EXEC static xmlNode * handle_agent_request(pcmk__request_t *request) { execute_agent_action(request->xml, &request->result); if (request->result.execution_status == PCMK_EXEC_PENDING) { return NULL; } return fenced_construct_reply(request->xml, NULL, &request->result); } // STONITH_OP_TIMEOUT_UPDATE static xmlNode * handle_update_timeout_request(pcmk__request_t *request) { const char *call_id = crm_element_value(request->xml, F_STONITH_CALLID); const char *client_id = crm_element_value(request->xml, F_STONITH_CLIENTID); int op_timeout = 0; crm_element_value_int(request->xml, F_STONITH_TIMEOUT, &op_timeout); do_stonith_async_timeout_update(client_id, call_id, op_timeout); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); return NULL; } // STONITH_OP_QUERY static xmlNode * handle_query_request(pcmk__request_t *request) { int timeout = 0; xmlNode *dev = NULL; const char *action = NULL; const char *target = NULL; const char *client_id = crm_element_value(request->xml, F_STONITH_CLIENTID); struct st_query_data *query = NULL; if (request->peer != NULL) { // Record it for the future notification create_remote_stonith_op(client_id, request->xml, TRUE); } /* Delete the DC node RELAY operation. */ remove_relay_op(request->xml); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); dev = get_xpath_object("//@" F_STONITH_ACTION, request->xml, LOG_NEVER); if (dev != NULL) { const char *device = crm_element_value(dev, F_STONITH_DEVICE); if (pcmk__str_eq(device, "manual_ack", pcmk__str_casei)) { return NULL; // No query or reply necessary } target = crm_element_value(dev, F_STONITH_TARGET); action = crm_element_value(dev, F_STONITH_ACTION); } crm_log_xml_trace(request->xml, "Query"); query = calloc(1, sizeof(struct st_query_data)); CRM_ASSERT(query != NULL); query->reply = fenced_construct_reply(request->xml, NULL, &request->result); pcmk__str_update(&query->remote_peer, request->peer); pcmk__str_update(&query->client_id, client_id); pcmk__str_update(&query->target, target); pcmk__str_update(&query->action, action); query->call_options = request->call_options; crm_element_value_int(request->xml, F_STONITH_TIMEOUT, &timeout); get_capable_devices(target, action, timeout, pcmk_is_set(query->call_options, st_opt_allow_suicide), query, stonith_query_capable_device_cb); return NULL; } // T_STONITH_NOTIFY static xmlNode * handle_notify_request(pcmk__request_t *request) { const char *flag_name = NULL; CRM_ASSERT(request->ipc_client != NULL); flag_name = crm_element_value(request->xml, F_STONITH_NOTIFY_ACTIVATE); if (flag_name != NULL) { crm_debug("Enabling %s callbacks for client %s", flag_name, pcmk__request_origin(request)); pcmk__set_client_flags(request->ipc_client, get_stonith_flag(flag_name)); } flag_name = crm_element_value(request->xml, F_STONITH_NOTIFY_DEACTIVATE); if (flag_name != NULL) { crm_debug("Disabling %s callbacks for client %s", flag_name, pcmk__request_origin(request)); pcmk__clear_client_flags(request->ipc_client, get_stonith_flag(flag_name)); } pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); pcmk__set_request_flags(request, pcmk__request_reuse_options); return pcmk__ipc_create_ack(request->ipc_flags, "ack", CRM_EX_OK); } // STONITH_OP_RELAY static xmlNode * handle_relay_request(pcmk__request_t *request) { xmlNode *dev = get_xpath_object("//@" F_STONITH_TARGET, request->xml, LOG_TRACE); crm_notice("Received forwarded fencing request from " "%s %s to fence (%s) peer %s", pcmk__request_origin_type(request), pcmk__request_origin(request), crm_element_value(dev, F_STONITH_ACTION), crm_element_value(dev, F_STONITH_TARGET)); if (initiate_remote_stonith_op(NULL, request->xml, FALSE) == NULL) { fenced_set_protocol_error(&request->result); return fenced_construct_reply(request->xml, NULL, &request->result); } pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_PENDING, NULL); return NULL; } // STONITH_OP_FENCE static xmlNode * handle_fence_request(pcmk__request_t *request) { if ((request->peer != NULL) || stand_alone) { fence_locally(request->xml, &request->result); } else if (pcmk_is_set(request->call_options, st_opt_manual_ack)) { switch (fenced_handle_manual_confirmation(request->ipc_client, request->xml)) { case pcmk_rc_ok: pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); break; case EINPROGRESS: pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_PENDING, NULL); break; default: fenced_set_protocol_error(&request->result); break; } } else { const char *alternate_host = NULL; xmlNode *dev = get_xpath_object("//@" F_STONITH_TARGET, request->xml, LOG_TRACE); const char *target = crm_element_value(dev, F_STONITH_TARGET); const char *action = crm_element_value(dev, F_STONITH_ACTION); const char *device = crm_element_value(dev, F_STONITH_DEVICE); if (request->ipc_client != NULL) { int tolerance = 0; crm_notice("Client %s wants to fence (%s) %s using %s", pcmk__request_origin(request), action, target, (device? device : "any device")); crm_element_value_int(dev, F_STONITH_TOLERANCE, &tolerance); if (stonith_check_fence_tolerance(tolerance, target, action)) { pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); return fenced_construct_reply(request->xml, NULL, &request->result); } } else { crm_notice("Peer %s wants to fence (%s) '%s' with device '%s'", request->peer, action, target, (device == NULL)? "(any)" : device); } alternate_host = check_alternate_host(target); if ((alternate_host != NULL) && (request->ipc_client != NULL)) { const char *client_id = NULL; remote_fencing_op_t *op = NULL; crm_notice("Forwarding self-fencing request to peer %s " "due to topology", alternate_host); if (request->ipc_client->id == 0) { client_id = crm_element_value(request->xml, F_STONITH_CLIENTID); } else { client_id = request->ipc_client->id; } /* Create a duplicate fencing operation to relay with the client ID. * When a query response is received, this operation should be * deleted to avoid keeping the duplicate around. */ op = create_remote_stonith_op(client_id, request->xml, FALSE); crm_xml_add(request->xml, F_STONITH_OPERATION, STONITH_OP_RELAY); crm_xml_add(request->xml, F_STONITH_CLIENTID, request->ipc_client->id); crm_xml_add(request->xml, F_STONITH_REMOTE_OP_ID, op->id); send_cluster_message(crm_get_peer(0, alternate_host), crm_msg_stonith_ng, request->xml, FALSE); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_PENDING, NULL); } else if (initiate_remote_stonith_op(request->ipc_client, request->xml, FALSE) == NULL) { fenced_set_protocol_error(&request->result); } else { pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_PENDING, NULL); } } if (request->result.execution_status == PCMK_EXEC_PENDING) { return NULL; } return fenced_construct_reply(request->xml, NULL, &request->result); } // STONITH_OP_FENCE_HISTORY static xmlNode * handle_history_request(pcmk__request_t *request) { xmlNode *reply = NULL; xmlNode *data = NULL; stonith_fence_history(request->xml, &data, request->peer, request->call_options); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); if (!pcmk_is_set(request->call_options, st_opt_discard_reply)) { /* When the local node broadcasts its history, it sets * st_opt_discard_reply and doesn't need a reply. */ reply = fenced_construct_reply(request->xml, data, &request->result); } free_xml(data); return reply; } // STONITH_OP_DEVICE_ADD static xmlNode * handle_device_add_request(pcmk__request_t *request) { - const char *device_id = NULL; const char *op = crm_element_value(request->xml, F_STONITH_OPERATION); + xmlNode *dev = get_xpath_object("//" F_STONITH_DEVICE, request->xml, + LOG_ERR); if (is_privileged(request->ipc_client, op)) { - int rc = stonith_device_register(request->xml, &device_id, FALSE); + int rc = stonith_device_register(dev, FALSE); pcmk__set_result(&request->result, ((rc == pcmk_ok)? CRM_EX_OK : CRM_EX_ERROR), stonith__legacy2status(rc), ((rc == pcmk_ok)? NULL : pcmk_strerror(rc))); } else { pcmk__set_result(&request->result, CRM_EX_INSUFFICIENT_PRIV, PCMK_EXEC_INVALID, "Unprivileged users must register device via CIB"); } - fenced_send_device_notification(op, &request->result, device_id); + fenced_send_device_notification(op, &request->result, + (dev == NULL)? NULL : ID(dev)); return fenced_construct_reply(request->xml, NULL, &request->result); } // STONITH_OP_DEVICE_DEL static xmlNode * handle_device_delete_request(pcmk__request_t *request) { xmlNode *dev = get_xpath_object("//" F_STONITH_DEVICE, request->xml, LOG_ERR); const char *device_id = crm_element_value(dev, XML_ATTR_ID); const char *op = crm_element_value(request->xml, F_STONITH_OPERATION); if (is_privileged(request->ipc_client, op)) { stonith_device_remove(device_id, false); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); } else { pcmk__set_result(&request->result, CRM_EX_INSUFFICIENT_PRIV, PCMK_EXEC_INVALID, "Unprivileged users must delete device via CIB"); } fenced_send_device_notification(op, &request->result, device_id); return fenced_construct_reply(request->xml, NULL, &request->result); } // STONITH_OP_LEVEL_ADD static xmlNode * handle_level_add_request(pcmk__request_t *request) { - char *device_id = NULL; + char *desc = NULL; const char *op = crm_element_value(request->xml, F_STONITH_OPERATION); if (is_privileged(request->ipc_client, op)) { - fenced_register_level(request->xml, &device_id, &request->result); + fenced_register_level(request->xml, &desc, &request->result); } else { + unpack_level_request(request->xml, NULL, NULL, NULL, &desc); pcmk__set_result(&request->result, CRM_EX_INSUFFICIENT_PRIV, PCMK_EXEC_INVALID, "Unprivileged users must add level via CIB"); } - fenced_send_level_notification(op, &request->result, device_id); - free(device_id); + fenced_send_level_notification(op, &request->result, desc); + free(desc); return fenced_construct_reply(request->xml, NULL, &request->result); } // STONITH_OP_LEVEL_DEL static xmlNode * handle_level_delete_request(pcmk__request_t *request) { - char *device_id = NULL; + char *desc = NULL; const char *op = crm_element_value(request->xml, F_STONITH_OPERATION); if (is_privileged(request->ipc_client, op)) { - fenced_unregister_level(request->xml, &device_id, &request->result); + fenced_unregister_level(request->xml, &desc, &request->result); } else { + unpack_level_request(request->xml, NULL, NULL, NULL, &desc); pcmk__set_result(&request->result, CRM_EX_INSUFFICIENT_PRIV, PCMK_EXEC_INVALID, "Unprivileged users must delete level via CIB"); } - fenced_send_level_notification(op, &request->result, device_id); + fenced_send_level_notification(op, &request->result, desc); + free(desc); return fenced_construct_reply(request->xml, NULL, &request->result); } // CRM_OP_RM_NODE_CACHE static xmlNode * handle_cache_request(pcmk__request_t *request) { int node_id = 0; const char *name = NULL; crm_element_value_int(request->xml, XML_ATTR_ID, &node_id); name = crm_element_value(request->xml, XML_ATTR_UNAME); reap_crm_member(node_id, name); pcmk__set_result(&request->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); return NULL; } static xmlNode * handle_unknown_request(pcmk__request_t *request) { const char *op = crm_element_value(request->xml, F_STONITH_OPERATION); crm_err("Unknown IPC request %s from %s %s", op, pcmk__request_origin_type(request), pcmk__request_origin(request)); pcmk__format_result(&request->result, CRM_EX_PROTOCOL, PCMK_EXEC_INVALID, "Unknown IPC request type '%s' (bug?)", crm_str(op)); return fenced_construct_reply(request->xml, NULL, &request->result); } static void fenced_register_handlers(void) { pcmk__server_command_t handlers[] = { { CRM_OP_REGISTER, handle_register_request }, { STONITH_OP_EXEC, handle_agent_request }, { STONITH_OP_TIMEOUT_UPDATE, handle_update_timeout_request }, { STONITH_OP_QUERY, handle_query_request }, { T_STONITH_NOTIFY, handle_notify_request }, { STONITH_OP_RELAY, handle_relay_request }, { STONITH_OP_FENCE, handle_fence_request }, { STONITH_OP_FENCE_HISTORY, handle_history_request }, { STONITH_OP_DEVICE_ADD, handle_device_add_request }, { STONITH_OP_DEVICE_DEL, handle_device_delete_request }, { STONITH_OP_LEVEL_ADD, handle_level_add_request }, { STONITH_OP_LEVEL_DEL, handle_level_delete_request }, { CRM_OP_RM_NODE_CACHE, handle_cache_request }, { NULL, handle_unknown_request }, }; fenced_handlers = pcmk__register_handlers(handlers); } void fenced_unregister_handlers(void) { if (fenced_handlers != NULL) { g_hash_table_destroy(fenced_handlers); fenced_handlers = NULL; } } static void handle_request(pcmk__request_t *request) { xmlNode *reply = NULL; const char *reason = NULL; if (fenced_handlers == NULL) { fenced_register_handlers(); } reply = pcmk__process_request(request, fenced_handlers); if (reply != NULL) { if (pcmk_is_set(request->flags, pcmk__request_reuse_options) && (request->ipc_client != NULL)) { /* Certain IPC-only commands must reuse the call options from the * original request rather than the ones set by stonith_send_reply() * -> do_local_reply(). */ pcmk__ipc_send_xml(request->ipc_client, request->ipc_id, reply, request->ipc_flags); request->ipc_client->request_id = 0; } else { stonith_send_reply(reply, request->call_options, request->peer, request->ipc_client); } free_xml(reply); } reason = request->result.exit_reason; crm_debug("Processed %s request from %s %s: %s%s%s%s", request->op, pcmk__request_origin_type(request), pcmk__request_origin(request), pcmk_exec_status_str(request->result.execution_status), (reason == NULL)? "" : " (", (reason == NULL)? "" : reason, (reason == NULL)? "" : ")"); } static void handle_reply(pcmk__client_t *client, xmlNode *request, const char *remote_peer) { // Copy, because request might be freed before we want to log this char *op = crm_element_value_copy(request, F_STONITH_OPERATION); if (pcmk__str_eq(op, STONITH_OP_QUERY, pcmk__str_none)) { process_remote_stonith_query(request); } else if (pcmk__str_any_of(op, T_STONITH_NOTIFY, STONITH_OP_FENCE, NULL)) { fenced_process_fencing_reply(request); } else { crm_err("Ignoring unknown %s reply from %s %s", crm_str(op), ((client == NULL)? "peer" : "client"), ((client == NULL)? remote_peer : pcmk__client_name(client))); crm_log_xml_warn(request, "UnknownOp"); free(op); return; } crm_debug("Processed %s reply from %s %s", op, ((client == NULL)? "peer" : "client"), ((client == NULL)? remote_peer : pcmk__client_name(client))); free(op); } /*! * \internal * \brief Handle a message from an IPC client or CPG peer * * \param[in] client If not NULL, IPC client that sent message * \param[in] id If from IPC client, IPC message ID * \param[in] flags Message flags * \param[in] message Message XML * \param[in] remote_peer If not NULL, CPG peer that sent message */ void stonith_command(pcmk__client_t *client, uint32_t id, uint32_t flags, xmlNode *message, const char *remote_peer) { int call_options = st_opt_none; bool is_reply = false; CRM_CHECK(message != NULL, return); if (get_xpath_object("//" T_STONITH_REPLY, message, LOG_NEVER) != NULL) { is_reply = true; } crm_element_value_int(message, F_STONITH_CALLOPTS, &call_options); crm_debug("Processing %ssynchronous %s %s %u from %s %s", pcmk_is_set(call_options, st_opt_sync_call)? "" : "a", crm_element_value(message, F_STONITH_OPERATION), (is_reply? "reply" : "request"), id, ((client == NULL)? "peer" : "client"), ((client == NULL)? remote_peer : pcmk__client_name(client))); if (pcmk_is_set(call_options, st_opt_sync_call)) { CRM_ASSERT(client == NULL || client->request_id == id); } if (is_reply) { handle_reply(client, message, remote_peer); } else { pcmk__request_t request = { .ipc_client = client, .ipc_id = id, .ipc_flags = flags, .peer = remote_peer, .xml = message, .call_options = call_options, .result = PCMK__UNKNOWN_RESULT, }; request.op = crm_element_value(request.xml, F_STONITH_OPERATION); CRM_CHECK(request.op != NULL, return); if (pcmk_is_set(request.call_options, st_opt_sync_call)) { pcmk__set_request_flags(&request, pcmk__request_sync); } handle_request(&request); pcmk__reset_result(&request.result); } } diff --git a/daemons/fenced/fenced_remote.c b/daemons/fenced/fenced_remote.c index 366d453ec5..ac1510fd5a 100644 --- a/daemons/fenced/fenced_remote.c +++ b/daemons/fenced/fenced_remote.c @@ -1,2362 +1,2363 @@ /* * Copyright 2009-2022 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU General Public License version 2 * or later (GPLv2+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #define TIMEOUT_MULTIPLY_FACTOR 1.2 /* When one fencer queries its peers for devices able to handle a fencing * request, each peer will reply with a list of such devices available to it. * Each reply will be parsed into a peer_device_info_t, with each device's * information kept in a device_properties_t. */ typedef struct device_properties_s { /* Whether access to this device has been verified */ gboolean verified; /* The remaining members are indexed by the operation's "phase" */ /* Whether this device has been executed in each phase */ gboolean executed[st_phase_max]; /* Whether this device is disallowed from executing in each phase */ gboolean disallowed[st_phase_max]; /* Action-specific timeout for each phase */ int custom_action_timeout[st_phase_max]; /* Action-specific maximum random delay for each phase */ int delay_max[st_phase_max]; /* Action-specific base delay for each phase */ int delay_base[st_phase_max]; } device_properties_t; typedef struct { /* Name of peer that sent this result */ char *host; /* Only try peers for non-topology based operations once */ gboolean tried; /* Number of entries in the devices table */ int ndevices; /* Devices available to this host that are capable of fencing the target */ GHashTable *devices; } peer_device_info_t; GHashTable *stonith_remote_op_list = NULL; extern xmlNode *stonith_create_op(int call_id, const char *token, const char *op, xmlNode * data, int call_options); static void request_peer_fencing(remote_fencing_op_t *op, peer_device_info_t *peer); static void finalize_op(remote_fencing_op_t *op, xmlNode *data, bool dup); static void report_timeout_period(remote_fencing_op_t * op, int op_timeout); static int get_op_total_timeout(const remote_fencing_op_t *op, const peer_device_info_t *chosen_peer); static gint sort_strings(gconstpointer a, gconstpointer b) { return strcmp(a, b); } static void free_remote_query(gpointer data) { if (data != NULL) { peer_device_info_t *peer = data; g_hash_table_destroy(peer->devices); free(peer->host); free(peer); } } void free_stonith_remote_op_list() { if (stonith_remote_op_list != NULL) { g_hash_table_destroy(stonith_remote_op_list); stonith_remote_op_list = NULL; } } struct peer_count_data { const remote_fencing_op_t *op; gboolean verified_only; int count; }; /*! * \internal * \brief Increment a counter if a device has not been executed yet * * \param[in] key Device ID (ignored) * \param[in] value Device properties * \param[in] user_data Peer count data */ static void count_peer_device(gpointer key, gpointer value, gpointer user_data) { device_properties_t *props = (device_properties_t*)value; struct peer_count_data *data = user_data; if (!props->executed[data->op->phase] && (!data->verified_only || props->verified)) { ++(data->count); } } /*! * \internal * \brief Check the number of available devices in a peer's query results * * \param[in] op Operation that results are for * \param[in] peer Peer to count * \param[in] verified_only Whether to count only verified devices * * \return Number of devices available to peer that were not already executed */ static int count_peer_devices(const remote_fencing_op_t *op, const peer_device_info_t *peer, gboolean verified_only) { struct peer_count_data data; data.op = op; data.verified_only = verified_only; data.count = 0; if (peer) { g_hash_table_foreach(peer->devices, count_peer_device, &data); } return data.count; } /*! * \internal * \brief Search for a device in a query result * * \param[in] op Operation that result is for * \param[in] peer Query result for a peer * \param[in] device Device ID to search for * * \return Device properties if found, NULL otherwise */ static device_properties_t * find_peer_device(const remote_fencing_op_t *op, const peer_device_info_t *peer, const char *device) { device_properties_t *props = g_hash_table_lookup(peer->devices, device); return (props && !props->executed[op->phase] && !props->disallowed[op->phase])? props : NULL; } /*! * \internal * \brief Find a device in a peer's device list and mark it as executed * * \param[in] op Operation that peer result is for * \param[in,out] peer Peer with results to search * \param[in] device ID of device to mark as done * \param[in] verified_devices_only Only consider verified devices * * \return TRUE if device was found and marked, FALSE otherwise */ static gboolean grab_peer_device(const remote_fencing_op_t *op, peer_device_info_t *peer, const char *device, gboolean verified_devices_only) { device_properties_t *props = find_peer_device(op, peer, device); if ((props == NULL) || (verified_devices_only && !props->verified)) { return FALSE; } crm_trace("Removing %s from %s (%d remaining)", device, peer->host, count_peer_devices(op, peer, FALSE)); props->executed[op->phase] = TRUE; return TRUE; } static void clear_remote_op_timers(remote_fencing_op_t * op) { if (op->query_timer) { g_source_remove(op->query_timer); op->query_timer = 0; } if (op->op_timer_total) { g_source_remove(op->op_timer_total); op->op_timer_total = 0; } if (op->op_timer_one) { g_source_remove(op->op_timer_one); op->op_timer_one = 0; } } static void free_remote_op(gpointer data) { remote_fencing_op_t *op = data; crm_log_xml_debug(op->request, "Destroying"); clear_remote_op_timers(op); free(op->id); free(op->action); free(op->delegate); free(op->target); free(op->client_id); free(op->client_name); free(op->originator); if (op->query_results) { g_list_free_full(op->query_results, free_remote_query); } if (op->request) { free_xml(op->request); op->request = NULL; } if (op->devices_list) { g_list_free_full(op->devices_list, free); op->devices_list = NULL; } g_list_free_full(op->automatic_list, free); g_list_free(op->duplicates); pcmk__reset_result(&op->result); free(op); } void init_stonith_remote_op_hash_table(GHashTable **table) { if (*table == NULL) { *table = pcmk__strkey_table(NULL, free_remote_op); } } /*! * \internal * \brief Return an operation's originally requested action (before any remap) * * \param[in] op Operation to check * * \return Operation's original action */ static const char * op_requested_action(const remote_fencing_op_t *op) { return ((op->phase > st_phase_requested)? "reboot" : op->action); } /*! * \internal * \brief Remap a "reboot" operation to the "off" phase * * \param[in,out] op Operation to remap */ static void op_phase_off(remote_fencing_op_t *op) { crm_info("Remapping multiple-device reboot targeting %s to 'off' " CRM_XS " id=%.8s", op->target, op->id); op->phase = st_phase_off; /* Happily, "off" and "on" are shorter than "reboot", so we can reuse the * memory allocation at each phase. */ strcpy(op->action, "off"); } /*! * \internal * \brief Advance a remapped reboot operation to the "on" phase * * \param[in,out] op Operation to remap */ static void op_phase_on(remote_fencing_op_t *op) { GList *iter = NULL; crm_info("Remapped 'off' targeting %s complete, " "remapping to 'on' for %s " CRM_XS " id=%.8s", op->target, op->client_name, op->id); op->phase = st_phase_on; strcpy(op->action, "on"); /* Skip devices with automatic unfencing, because the cluster will handle it * when the node rejoins. */ for (iter = op->automatic_list; iter != NULL; iter = iter->next) { GList *match = g_list_find_custom(op->devices_list, iter->data, sort_strings); if (match) { op->devices_list = g_list_remove(op->devices_list, match->data); } } g_list_free_full(op->automatic_list, free); op->automatic_list = NULL; /* Rewind device list pointer */ op->devices = op->devices_list; } /*! * \internal * \brief Reset a remapped reboot operation * * \param[in,out] op Operation to reset */ static void undo_op_remap(remote_fencing_op_t *op) { if (op->phase > 0) { crm_info("Undoing remap of reboot targeting %s for %s " CRM_XS " id=%.8s", op->target, op->client_name, op->id); op->phase = st_phase_requested; strcpy(op->action, "reboot"); } } /*! * \internal * \brief Create notification data XML for a fencing operation result * * \param[in] op Fencer operation that completed * * \return Newly created XML to add as notification data * \note The caller is responsible for freeing the result. */ static xmlNode * fencing_result2xml(remote_fencing_op_t *op) { xmlNode *notify_data = create_xml_node(NULL, T_STONITH_NOTIFY_FENCE); crm_xml_add_int(notify_data, "state", op->state); crm_xml_add(notify_data, F_STONITH_TARGET, op->target); crm_xml_add(notify_data, F_STONITH_ACTION, op->action); crm_xml_add(notify_data, F_STONITH_DELEGATE, op->delegate); crm_xml_add(notify_data, F_STONITH_REMOTE_OP_ID, op->id); crm_xml_add(notify_data, F_STONITH_ORIGIN, op->originator); crm_xml_add(notify_data, F_STONITH_CLIENTID, op->client_id); crm_xml_add(notify_data, F_STONITH_CLIENTNAME, op->client_name); - stonith__xe_set_result(notify_data, &op->result); return notify_data; } /*! * \internal * \brief Broadcast a fence result notification to all CPG peers * * \param[in] op Fencer operation that completed * \param[in] op_merged Whether this operation is a duplicate of another */ void fenced_broadcast_op_result(remote_fencing_op_t *op, bool op_merged) { static int count = 0; xmlNode *bcast = create_xml_node(NULL, T_STONITH_REPLY); xmlNode *notify_data = fencing_result2xml(op); count++; crm_trace("Broadcasting result to peers"); crm_xml_add(bcast, F_TYPE, T_STONITH_NOTIFY); crm_xml_add(bcast, F_SUBTYPE, "broadcast"); crm_xml_add(bcast, F_STONITH_OPERATION, T_STONITH_NOTIFY); crm_xml_add_int(bcast, "count", count); if (op_merged) { pcmk__xe_set_bool_attr(bcast, F_STONITH_MERGED, true); } + stonith__xe_set_result(notify_data, &op->result); + add_message_xml(bcast, F_STONITH_CALLDATA, notify_data); send_cluster_message(NULL, crm_msg_stonith_ng, bcast, FALSE); free_xml(notify_data); free_xml(bcast); return; } /*! * \internal * \brief Reply to a local request originator and notify all subscribed clients * * \param[in] op Fencer operation that completed * \param[in] data Top-level XML to add notification to */ static void handle_local_reply_and_notify(remote_fencing_op_t *op, xmlNode *data) { xmlNode *notify_data = NULL; xmlNode *reply = NULL; pcmk__client_t *client = NULL; if (op->notify_sent == TRUE) { /* nothing to do */ return; } /* Do notification with a clean data object */ crm_xml_add_int(data, "state", op->state); crm_xml_add(data, F_STONITH_TARGET, op->target); crm_xml_add(data, F_STONITH_OPERATION, op->action); reply = fenced_construct_reply(op->request, data, &op->result); crm_xml_add(reply, F_STONITH_DELEGATE, op->delegate); /* Send fencing OP reply to local client that initiated fencing */ client = pcmk__find_client_by_id(op->client_id); if (client == NULL) { crm_trace("Skipping reply to %s: no longer a client", op->client_id); } else { do_local_reply(reply, client, op->call_options); } /* bcast to all local clients that the fencing operation happend */ notify_data = fencing_result2xml(op); fenced_send_notification(T_STONITH_NOTIFY_FENCE, &op->result, notify_data); free_xml(notify_data); fenced_send_notification(T_STONITH_NOTIFY_HISTORY, NULL, NULL); /* mark this op as having notify's already sent */ op->notify_sent = TRUE; free_xml(reply); } /*! * \internal * \brief Finalize all duplicates of a given fencer operation * * \param[in] op Fencer operation that completed * \param[in] data Top-level XML to add notification to */ static void finalize_op_duplicates(remote_fencing_op_t *op, xmlNode *data) { for (GList *iter = op->duplicates; iter != NULL; iter = iter->next) { remote_fencing_op_t *other = iter->data; if (other->state == st_duplicate) { other->state = op->state; crm_debug("Performing duplicate notification for %s@%s: %s " CRM_XS " id=%.8s", other->client_name, other->originator, pcmk_exec_status_str(op->result.execution_status), other->id); pcmk__copy_result(&op->result, &other->result); finalize_op(other, data, true); } else { // Possible if (for example) it timed out already crm_err("Skipping duplicate notification for %s@%s " CRM_XS " state=%s id=%.8s", other->client_name, other->originator, stonith_op_state_str(other->state), other->id); } } } static char * delegate_from_xml(xmlNode *xml) { xmlNode *match = get_xpath_object("//@" F_STONITH_DELEGATE, xml, LOG_NEVER); if (match == NULL) { return crm_element_value_copy(xml, F_ORIG); } else { return crm_element_value_copy(match, F_STONITH_DELEGATE); } } /*! * \internal * \brief Finalize a peer fencing operation * * Clean up after a fencing operation completes. This function has two code * paths: the executioner uses it to broadcast the result to CPG peers, and then * each peer (including the executioner) uses it to process that broadcast and * notify its IPC clients of the result. * * \param[in] op Fencer operation that completed * \param[in] data If not NULL, XML reply of last delegated fencing operation * \param[in] dup Whether this operation is a duplicate of another * (in which case, do not broadcast the result) * * \note The operation result should be set before calling this function. */ static void finalize_op(remote_fencing_op_t *op, xmlNode *data, bool dup) { int level = LOG_ERR; const char *subt = NULL; xmlNode *local_data = NULL; gboolean op_merged = FALSE; CRM_CHECK((op != NULL), return); if (op->notify_sent) { // Most likely, this is a timed-out action that eventually completed crm_notice("Operation '%s'%s%s by %s for %s@%s%s: " "Result arrived too late " CRM_XS " id=%.8s", op->action, (op->target? " targeting " : ""), (op->target? op->target : ""), (op->delegate? op->delegate : "unknown node"), op->client_name, op->originator, (op_merged? " (merged)" : ""), op->id); return; } set_fencing_completed(op); clear_remote_op_timers(op); undo_op_remap(op); if (data == NULL) { data = create_xml_node(NULL, "remote-op"); local_data = data; } else if (op->delegate == NULL) { switch (op->result.execution_status) { case PCMK_EXEC_NO_FENCE_DEVICE: break; case PCMK_EXEC_INVALID: if (op->result.exit_status == CRM_EX_EXPIRED) { break; } // else fall through default: op->delegate = delegate_from_xml(data); break; } } if (dup || (crm_element_value(data, F_STONITH_MERGED) != NULL)) { op_merged = true; } /* Tell everyone the operation is done, we will continue * with doing the local notifications once we receive * the broadcast back. */ subt = crm_element_value(data, F_SUBTYPE); if (!dup && !pcmk__str_eq(subt, "broadcast", pcmk__str_casei)) { /* Defer notification until the bcast message arrives */ fenced_broadcast_op_result(op, op_merged); free_xml(local_data); return; } if (pcmk__result_ok(&op->result) || dup || !pcmk__str_eq(op->originator, stonith_our_uname, pcmk__str_casei)) { level = LOG_NOTICE; } do_crm_log(level, "Operation '%s'%s%s by %s for %s@%s%s: %s (%s%s%s) " CRM_XS " id=%.8s", op->action, (op->target? " targeting " : ""), (op->target? op->target : ""), (op->delegate? op->delegate : "unknown node"), op->client_name, op->originator, (op_merged? " (merged)" : ""), crm_exit_str(op->result.exit_status), pcmk_exec_status_str(op->result.execution_status), ((op->result.exit_reason == NULL)? "" : ": "), ((op->result.exit_reason == NULL)? "" : op->result.exit_reason), op->id); handle_local_reply_and_notify(op, data); if (!dup) { finalize_op_duplicates(op, data); } /* Free non-essential parts of the record * Keep the record around so we can query the history */ if (op->query_results) { g_list_free_full(op->query_results, free_remote_query); op->query_results = NULL; } if (op->request) { free_xml(op->request); op->request = NULL; } free_xml(local_data); } /*! * \internal * \brief Finalize a watchdog fencer op after the waiting time expires * * \param[in] userdata Fencer operation that completed * * \return G_SOURCE_REMOVE (which tells glib not to restart timer) */ static gboolean remote_op_watchdog_done(gpointer userdata) { remote_fencing_op_t *op = userdata; op->op_timer_one = 0; crm_notice("Self-fencing (%s) by %s for %s assumed complete " CRM_XS " id=%.8s", op->action, op->target, op->client_name, op->id); op->state = st_done; pcmk__set_result(&op->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); finalize_op(op, NULL, false); return G_SOURCE_REMOVE; } static gboolean remote_op_timeout_one(gpointer userdata) { remote_fencing_op_t *op = userdata; op->op_timer_one = 0; crm_notice("Peer's '%s' action targeting %s for client %s timed out " CRM_XS " id=%.8s", op->action, op->target, op->client_name, op->id); pcmk__set_result(&op->result, CRM_EX_ERROR, PCMK_EXEC_TIMEOUT, "Peer did not return fence result within timeout"); // Try another device, if appropriate request_peer_fencing(op, NULL); return FALSE; } /*! * \internal * \brief Finalize a remote fencer operation that timed out * * \param[in] op Fencer operation that timed out * \param[in] reason Readable description of what step timed out */ static void finalize_timed_out_op(remote_fencing_op_t *op, const char *reason) { op->op_timer_total = 0; crm_debug("Action '%s' targeting %s for client %s timed out " CRM_XS " id=%.8s", op->action, op->target, op->client_name, op->id); if (op->phase == st_phase_on) { /* A remapped reboot operation timed out in the "on" phase, but the * "off" phase completed successfully, so quit trying any further * devices, and return success. */ op->state = st_done; pcmk__set_result(&op->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); } else { op->state = st_failed; pcmk__set_result(&op->result, CRM_EX_ERROR, PCMK_EXEC_TIMEOUT, reason); } finalize_op(op, NULL, false); } /*! * \internal * \brief Finalize a remote fencer operation that timed out * * \param[in] userdata Fencer operation that timed out * * \return G_SOURCE_REMOVE (which tells glib not to restart timer) */ static gboolean remote_op_timeout(gpointer userdata) { remote_fencing_op_t *op = userdata; if (op->state == st_done) { crm_debug("Action '%s' targeting %s for client %s already completed " CRM_XS " id=%.8s", op->action, op->target, op->client_name, op->id); } else { finalize_timed_out_op(userdata, "Fencing did not complete within a " "total timeout based on the " "configured timeout and retries for " "any devices attempted"); } return G_SOURCE_REMOVE; } static gboolean remote_op_query_timeout(gpointer data) { remote_fencing_op_t *op = data; op->query_timer = 0; if (op->state == st_done) { crm_debug("Operation %.8s targeting %s already completed", op->id, op->target); } else if (op->state == st_exec) { crm_debug("Operation %.8s targeting %s already in progress", op->id, op->target); } else if (op->query_results) { // Query succeeded, so attempt the actual fencing crm_debug("Query %.8s targeting %s complete (state=%s)", op->id, op->target, stonith_op_state_str(op->state)); request_peer_fencing(op, NULL); } else { crm_debug("Query %.8s targeting %s timed out (state=%s)", op->id, op->target, stonith_op_state_str(op->state)); if (op->op_timer_total) { g_source_remove(op->op_timer_total); op->op_timer_total = 0; } finalize_timed_out_op(op, "No capable peers replied to device query " "within timeout"); } return FALSE; } static gboolean topology_is_empty(stonith_topology_t *tp) { int i; if (tp == NULL) { return TRUE; } for (i = 0; i < ST_LEVEL_MAX; i++) { if (tp->levels[i] != NULL) { return FALSE; } } return TRUE; } /*! * \internal * \brief Add a device to an operation's automatic unfencing list * * \param[in,out] op Operation to modify * \param[in] device Device ID to add */ static void add_required_device(remote_fencing_op_t *op, const char *device) { GList *match = g_list_find_custom(op->automatic_list, device, sort_strings); if (!match) { op->automatic_list = g_list_prepend(op->automatic_list, strdup(device)); } } /*! * \internal * \brief Remove a device from the automatic unfencing list * * \param[in,out] op Operation to modify * \param[in] device Device ID to remove */ static void remove_required_device(remote_fencing_op_t *op, const char *device) { GList *match = g_list_find_custom(op->automatic_list, device, sort_strings); if (match) { op->automatic_list = g_list_remove(op->automatic_list, match->data); } } /* deep copy the device list */ static void set_op_device_list(remote_fencing_op_t * op, GList *devices) { GList *lpc = NULL; if (op->devices_list) { g_list_free_full(op->devices_list, free); op->devices_list = NULL; } for (lpc = devices; lpc != NULL; lpc = lpc->next) { op->devices_list = g_list_append(op->devices_list, strdup(lpc->data)); } op->devices = op->devices_list; } /*! * \internal * \brief Check whether a node matches a topology target * * \param[in] tp Topology table entry to check * \param[in] node Name of node to check * * \return TRUE if node matches topology target */ static gboolean topology_matches(const stonith_topology_t *tp, const char *node) { regex_t r_patt; CRM_CHECK(node && tp && tp->target, return FALSE); switch (tp->kind) { case fenced_target_by_attribute: /* This level targets by attribute, so tp->target is a NAME=VALUE pair * of a permanent attribute applied to targeted nodes. The test below * relies on the locally cached copy of the CIB, so if fencing needs to * be done before the initial CIB is received or after a malformed CIB * is received, then the topology will be unable to be used. */ if (node_has_attr(node, tp->target_attribute, tp->target_value)) { crm_notice("Matched %s with %s by attribute", node, tp->target); return TRUE; } break; case fenced_target_by_pattern: /* This level targets node names matching a pattern, so tp->target * (and tp->target_pattern) is a regular expression. */ if (regcomp(&r_patt, tp->target_pattern, REG_EXTENDED|REG_NOSUB)) { crm_info("Bad regex '%s' for fencing level", tp->target); } else { int status = regexec(&r_patt, node, 0, NULL, 0); regfree(&r_patt); if (status == 0) { crm_notice("Matched %s with %s by name", node, tp->target); return TRUE; } } break; case fenced_target_by_name: crm_trace("Testing %s against %s", node, tp->target); return pcmk__str_eq(tp->target, node, pcmk__str_casei); default: break; } crm_trace("No match for %s with %s", node, tp->target); return FALSE; } stonith_topology_t * find_topology_for_host(const char *host) { GHashTableIter tIter; stonith_topology_t *tp = g_hash_table_lookup(topology, host); if(tp != NULL) { crm_trace("Found %s for %s in %d entries", tp->target, host, g_hash_table_size(topology)); return tp; } g_hash_table_iter_init(&tIter, topology); while (g_hash_table_iter_next(&tIter, NULL, (gpointer *) & tp)) { if (topology_matches(tp, host)) { crm_trace("Found %s for %s in %d entries", tp->target, host, g_hash_table_size(topology)); return tp; } } crm_trace("No matches for %s in %d topology entries", host, g_hash_table_size(topology)); return NULL; } /*! * \internal * \brief Set fencing operation's device list to target's next topology level * * \param[in,out] op Remote fencing operation to modify * \param[in] empty_ok If true, an operation without a target (i.e. * queries) or a target without a topology will get a * pcmk_rc_ok return value instead of ENODEV * * \return Standard Pacemaker return value */ static int advance_topology_level(remote_fencing_op_t *op, bool empty_ok) { stonith_topology_t *tp = NULL; if (op->target) { tp = find_topology_for_host(op->target); } if (topology_is_empty(tp)) { return empty_ok? pcmk_rc_ok : ENODEV; } CRM_ASSERT(tp->levels != NULL); stonith__set_call_options(op->call_options, op->id, st_opt_topology); /* This is a new level, so undo any remapping left over from previous */ undo_op_remap(op); do { op->level++; } while (op->level < ST_LEVEL_MAX && tp->levels[op->level] == NULL); if (op->level < ST_LEVEL_MAX) { crm_trace("Attempting fencing level %d targeting %s (%d devices) " "for client %s@%s (id=%.8s)", op->level, op->target, g_list_length(tp->levels[op->level]), op->client_name, op->originator, op->id); set_op_device_list(op, tp->levels[op->level]); // The requested delay has been applied for the first fencing level if (op->level > 1 && op->delay > 0) { op->delay = 0; } if (g_list_next(op->devices_list) && pcmk__str_eq(op->action, "reboot", pcmk__str_casei)) { /* A reboot has been requested for a topology level with multiple * devices. Instead of rebooting the devices sequentially, we will * turn them all off, then turn them all on again. (Think about * switched power outlets for redundant power supplies.) */ op_phase_off(op); } return pcmk_rc_ok; } crm_info("All fencing options targeting %s for client %s@%s failed " CRM_XS " id=%.8s", op->target, op->client_name, op->originator, op->id); return ENODEV; } /*! * \brief Check to see if this operation is a duplicate of another in flight * operation. If so merge this operation into the inflight operation, and mark * it as a duplicate. */ static void merge_duplicates(remote_fencing_op_t * op) { GHashTableIter iter; remote_fencing_op_t *other = NULL; time_t now = time(NULL); g_hash_table_iter_init(&iter, stonith_remote_op_list); while (g_hash_table_iter_next(&iter, NULL, (void **)&other)) { const char *other_action = op_requested_action(other); if (!strcmp(op->id, other->id)) { continue; // Don't compare against self } if (other->state > st_exec) { crm_trace("%.8s not duplicate of %.8s: not in progress", op->id, other->id); continue; } if (!pcmk__str_eq(op->target, other->target, pcmk__str_casei)) { crm_trace("%.8s not duplicate of %.8s: node %s vs. %s", op->id, other->id, op->target, other->target); continue; } if (!pcmk__str_eq(op->action, other_action, pcmk__str_casei)) { crm_trace("%.8s not duplicate of %.8s: action %s vs. %s", op->id, other->id, op->action, other_action); continue; } if (pcmk__str_eq(op->client_name, other->client_name, pcmk__str_casei)) { crm_trace("%.8s not duplicate of %.8s: same client %s", op->id, other->id, op->client_name); continue; } if (pcmk__str_eq(other->target, other->originator, pcmk__str_casei)) { crm_trace("%.8s not duplicate of %.8s: suicide for %s", op->id, other->id, other->target); continue; } if (!fencing_peer_active(crm_get_peer(0, other->originator))) { crm_notice("Failing action '%s' targeting %s originating from " "client %s@%s: Originator is dead " CRM_XS " id=%.8s", other->action, other->target, other->client_name, other->originator, other->id); crm_trace("%.8s not duplicate of %.8s: originator dead", op->id, other->id); other->state = st_failed; continue; } if ((other->total_timeout > 0) && (now > (other->total_timeout + other->created))) { crm_trace("%.8s not duplicate of %.8s: old (%ld vs. %ld + %d)", op->id, other->id, now, other->created, other->total_timeout); continue; } /* There is another in-flight request to fence the same host * Piggyback on that instead. If it fails, so do we. */ other->duplicates = g_list_append(other->duplicates, op); if (other->total_timeout == 0) { other->total_timeout = op->total_timeout = TIMEOUT_MULTIPLY_FACTOR * get_op_total_timeout(op, NULL); crm_trace("Best guess as to timeout used for %.8s: %d", other->id, other->total_timeout); } crm_notice("Merging fencing action '%s' targeting %s originating from " "client %s with identical request from %s@%s " CRM_XS " original=%.8s duplicate=%.8s total_timeout=%ds", op->action, op->target, op->client_name, other->client_name, other->originator, op->id, other->id, other->total_timeout); report_timeout_period(op, other->total_timeout); op->state = st_duplicate; } } static uint32_t fencing_active_peers(void) { uint32_t count = 0; crm_node_t *entry; GHashTableIter gIter; g_hash_table_iter_init(&gIter, crm_peer_cache); while (g_hash_table_iter_next(&gIter, NULL, (void **)&entry)) { if(fencing_peer_active(entry)) { count++; } } return count; } /*! * \internal * \brief Process a manual confirmation of a pending fence action * * \param[in] client IPC client that sent confirmation * \param[in] msg Request XML with manual confirmation * * \return Standard Pacemaker return code */ int fenced_handle_manual_confirmation(pcmk__client_t *client, xmlNode *msg) { remote_fencing_op_t *op = NULL; xmlNode *dev = get_xpath_object("//@" F_STONITH_TARGET, msg, LOG_ERR); CRM_CHECK(dev != NULL, return EPROTO); crm_notice("Received manual confirmation that %s has been fenced", crm_str(crm_element_value(dev, F_STONITH_TARGET))); op = initiate_remote_stonith_op(client, msg, TRUE); if (op == NULL) { return EPROTO; } op->state = st_done; set_fencing_completed(op); op->delegate = strdup("a human"); // For the fencer's purposes, the fencing operation is done pcmk__set_result(&op->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); finalize_op(op, msg, false); /* For the requester's purposes, the operation is still pending. The * actual result will be sent asynchronously via the operation's done_cb(). */ return EINPROGRESS; } /*! * \internal * \brief Create a new remote stonith operation * * \param[in] client ID of local stonith client that initiated the operation * \param[in] request The request from the client that started the operation * \param[in] peer TRUE if this operation is owned by another stonith peer * (an operation owned by one peer is stored on all peers, * but only the owner executes it; all nodes get the results * once the owner finishes execution) */ void * create_remote_stonith_op(const char *client, xmlNode * request, gboolean peer) { remote_fencing_op_t *op = NULL; xmlNode *dev = get_xpath_object("//@" F_STONITH_TARGET, request, LOG_NEVER); int call_options = 0; const char *operation = NULL; init_stonith_remote_op_hash_table(&stonith_remote_op_list); /* If this operation is owned by another node, check to make * sure we haven't already created this operation. */ if (peer && dev) { const char *op_id = crm_element_value(dev, F_STONITH_REMOTE_OP_ID); CRM_CHECK(op_id != NULL, return NULL); op = g_hash_table_lookup(stonith_remote_op_list, op_id); if (op) { crm_debug("Reusing existing remote fencing op %.8s for %s", op_id, ((client == NULL)? "unknown client" : client)); return op; } } op = calloc(1, sizeof(remote_fencing_op_t)); CRM_ASSERT(op != NULL); crm_element_value_int(request, F_STONITH_TIMEOUT, &(op->base_timeout)); // Value -1 means disable any static/random fencing delays crm_element_value_int(request, F_STONITH_DELAY, &(op->delay)); if (peer && dev) { op->id = crm_element_value_copy(dev, F_STONITH_REMOTE_OP_ID); } else { op->id = crm_generate_uuid(); } g_hash_table_replace(stonith_remote_op_list, op->id, op); op->state = st_query; op->replies_expected = fencing_active_peers(); op->action = crm_element_value_copy(dev, F_STONITH_ACTION); op->originator = crm_element_value_copy(dev, F_STONITH_ORIGIN); op->delegate = crm_element_value_copy(dev, F_STONITH_DELEGATE); /* May not be set */ op->created = time(NULL); if (op->originator == NULL) { /* Local or relayed request */ op->originator = strdup(stonith_our_uname); } CRM_LOG_ASSERT(client != NULL); if (client) { op->client_id = strdup(client); } /* For a RELAY operation, set fenced on the client. */ operation = crm_element_value(request, F_STONITH_OPERATION); if (pcmk__str_eq(operation, STONITH_OP_RELAY, pcmk__str_none)) { op->client_name = crm_strdup_printf("%s.%lu", crm_system_name, (unsigned long) getpid()); } else { op->client_name = crm_element_value_copy(request, F_STONITH_CLIENTNAME); } op->target = crm_element_value_copy(dev, F_STONITH_TARGET); op->request = copy_xml(request); /* TODO: Figure out how to avoid this */ crm_element_value_int(request, F_STONITH_CALLOPTS, &call_options); op->call_options = call_options; crm_element_value_int(request, F_STONITH_CALLID, &(op->client_callid)); crm_trace("%s new fencing op %s ('%s' targeting %s for client %s, " "base timeout %d, %u %s expected)", (peer && dev)? "Recorded" : "Generated", op->id, op->action, op->target, op->client_name, op->base_timeout, op->replies_expected, pcmk__plural_alt(op->replies_expected, "reply", "replies")); if (op->call_options & st_opt_cs_nodeid) { int nodeid; crm_node_t *node; pcmk__scan_min_int(op->target, &nodeid, 0); node = pcmk__search_known_node_cache(nodeid, NULL, CRM_GET_PEER_ANY); /* Ensure the conversion only happens once */ stonith__clear_call_options(op->call_options, op->id, st_opt_cs_nodeid); if (node && node->uname) { free(op->target); op->target = strdup(node->uname); } else { crm_warn("Could not expand nodeid '%s' into a host name", op->target); } } /* check to see if this is a duplicate operation of another in-flight operation */ merge_duplicates(op); if (op->state != st_duplicate) { /* kick history readers */ fenced_send_notification(T_STONITH_NOTIFY_HISTORY, NULL, NULL); } /* safe to trim as long as that doesn't touch pending ops */ stonith_fence_history_trim(); return op; } /*! * \internal * \brief Create a peer fencing operation from a request, and initiate it * * \param[in] client IPC client that made request (NULL to get from request) * \param[in] request Request XML * \param[in] manual_ack Whether this is a manual action confirmation * * \return Newly created operation on success, otherwise NULL */ remote_fencing_op_t * initiate_remote_stonith_op(pcmk__client_t *client, xmlNode *request, gboolean manual_ack) { int query_timeout = 0; xmlNode *query = NULL; const char *client_id = NULL; remote_fencing_op_t *op = NULL; const char *relay_op_id = NULL; const char *operation = NULL; if (client) { client_id = client->id; } else { client_id = crm_element_value(request, F_STONITH_CLIENTID); } CRM_LOG_ASSERT(client_id != NULL); op = create_remote_stonith_op(client_id, request, FALSE); op->owner = TRUE; if (manual_ack) { return op; } CRM_CHECK(op->action, return NULL); if (advance_topology_level(op, true) != pcmk_rc_ok) { op->state = st_failed; } switch (op->state) { case st_failed: // advance_topology_level() exhausted levels pcmk__set_result(&op->result, CRM_EX_ERROR, PCMK_EXEC_ERROR, "All topology levels failed"); crm_warn("Could not request peer fencing (%s) targeting %s " CRM_XS " id=%.8s", op->action, op->target, op->id); finalize_op(op, NULL, false); return op; case st_duplicate: crm_info("Requesting peer fencing (%s) targeting %s (duplicate) " CRM_XS " id=%.8s", op->action, op->target, op->id); return op; default: crm_notice("Requesting peer fencing (%s) targeting %s " CRM_XS " id=%.8s state=%s base_timeout=%d", op->action, op->target, op->id, stonith_op_state_str(op->state), op->base_timeout); } query = stonith_create_op(op->client_callid, op->id, STONITH_OP_QUERY, NULL, op->call_options); crm_xml_add(query, F_STONITH_REMOTE_OP_ID, op->id); crm_xml_add(query, F_STONITH_TARGET, op->target); crm_xml_add(query, F_STONITH_ACTION, op_requested_action(op)); crm_xml_add(query, F_STONITH_ORIGIN, op->originator); crm_xml_add(query, F_STONITH_CLIENTID, op->client_id); crm_xml_add(query, F_STONITH_CLIENTNAME, op->client_name); crm_xml_add_int(query, F_STONITH_TIMEOUT, op->base_timeout); /* In case of RELAY operation, RELAY information is added to the query to delete the original operation of RELAY. */ operation = crm_element_value(request, F_STONITH_OPERATION); if (pcmk__str_eq(operation, STONITH_OP_RELAY, pcmk__str_none)) { relay_op_id = crm_element_value(request, F_STONITH_REMOTE_OP_ID); if (relay_op_id) { crm_xml_add(query, F_STONITH_REMOTE_OP_ID_RELAY, relay_op_id); } } send_cluster_message(NULL, crm_msg_stonith_ng, query, FALSE); free_xml(query); query_timeout = op->base_timeout * TIMEOUT_MULTIPLY_FACTOR; op->query_timer = g_timeout_add((1000 * query_timeout), remote_op_query_timeout, op); return op; } enum find_best_peer_options { /*! Skip checking the target peer for capable fencing devices */ FIND_PEER_SKIP_TARGET = 0x0001, /*! Only check the target peer for capable fencing devices */ FIND_PEER_TARGET_ONLY = 0x0002, /*! Skip peers and devices that are not verified */ FIND_PEER_VERIFIED_ONLY = 0x0004, }; static peer_device_info_t * find_best_peer(const char *device, remote_fencing_op_t * op, enum find_best_peer_options options) { GList *iter = NULL; gboolean verified_devices_only = (options & FIND_PEER_VERIFIED_ONLY) ? TRUE : FALSE; if (!device && pcmk_is_set(op->call_options, st_opt_topology)) { return NULL; } for (iter = op->query_results; iter != NULL; iter = iter->next) { peer_device_info_t *peer = iter->data; crm_trace("Testing result from %s targeting %s with %d device%s: %d %x", peer->host, op->target, peer->ndevices, pcmk__plural_s(peer->ndevices), peer->tried, options); if ((options & FIND_PEER_SKIP_TARGET) && pcmk__str_eq(peer->host, op->target, pcmk__str_casei)) { continue; } if ((options & FIND_PEER_TARGET_ONLY) && !pcmk__str_eq(peer->host, op->target, pcmk__str_casei)) { continue; } if (pcmk_is_set(op->call_options, st_opt_topology)) { if (grab_peer_device(op, peer, device, verified_devices_only)) { return peer; } } else if ((peer->tried == FALSE) && count_peer_devices(op, peer, verified_devices_only)) { /* No topology: Use the current best peer */ crm_trace("Simple fencing"); return peer; } } return NULL; } static peer_device_info_t * stonith_choose_peer(remote_fencing_op_t * op) { const char *device = NULL; peer_device_info_t *peer = NULL; uint32_t active = fencing_active_peers(); do { if (op->devices) { device = op->devices->data; crm_trace("Checking for someone to fence (%s) %s using %s", op->action, op->target, device); } else { crm_trace("Checking for someone to fence (%s) %s", op->action, op->target); } /* Best choice is a peer other than the target with verified access */ peer = find_best_peer(device, op, FIND_PEER_SKIP_TARGET|FIND_PEER_VERIFIED_ONLY); if (peer) { crm_trace("Found verified peer %s for %s", peer->host, device?device:""); return peer; } if(op->query_timer != 0 && op->replies < QB_MIN(op->replies_expected, active)) { crm_trace("Waiting before looking for unverified devices to fence %s", op->target); return NULL; } /* If no other peer has verified access, next best is unverified access */ peer = find_best_peer(device, op, FIND_PEER_SKIP_TARGET); if (peer) { crm_trace("Found best unverified peer %s", peer->host); return peer; } /* If no other peer can do it, last option is self-fencing * (which is never allowed for the "on" phase of a remapped reboot) */ if (op->phase != st_phase_on) { peer = find_best_peer(device, op, FIND_PEER_TARGET_ONLY); if (peer) { crm_trace("%s will fence itself", peer->host); return peer; } } /* Try the next fencing level if there is one (unless we're in the "on" * phase of a remapped "reboot", because we ignore errors in that case) */ } while ((op->phase != st_phase_on) && pcmk_is_set(op->call_options, st_opt_topology) && (advance_topology_level(op, false) == pcmk_rc_ok)); crm_notice("Couldn't find anyone to fence (%s) %s using %s", op->action, op->target, (device? device : "any device")); return NULL; } static int get_device_timeout(const remote_fencing_op_t *op, const peer_device_info_t *peer, const char *device) { device_properties_t *props; if (!peer || !device) { return op->base_timeout; } props = g_hash_table_lookup(peer->devices, device); if (!props) { return op->base_timeout; } return (props->custom_action_timeout[op->phase]? props->custom_action_timeout[op->phase] : op->base_timeout) + props->delay_max[op->phase]; } struct timeout_data { const remote_fencing_op_t *op; const peer_device_info_t *peer; int total_timeout; }; /*! * \internal * \brief Add timeout to a total if device has not been executed yet * * \param[in] key GHashTable key (device ID) * \param[in] value GHashTable value (device properties) * \param[in] user_data Timeout data */ static void add_device_timeout(gpointer key, gpointer value, gpointer user_data) { const char *device_id = key; device_properties_t *props = value; struct timeout_data *timeout = user_data; if (!props->executed[timeout->op->phase] && !props->disallowed[timeout->op->phase]) { timeout->total_timeout += get_device_timeout(timeout->op, timeout->peer, device_id); } } static int get_peer_timeout(const remote_fencing_op_t *op, const peer_device_info_t *peer) { struct timeout_data timeout; timeout.op = op; timeout.peer = peer; timeout.total_timeout = 0; g_hash_table_foreach(peer->devices, add_device_timeout, &timeout); return (timeout.total_timeout? timeout.total_timeout : op->base_timeout); } static int get_op_total_timeout(const remote_fencing_op_t *op, const peer_device_info_t *chosen_peer) { int total_timeout = 0; stonith_topology_t *tp = find_topology_for_host(op->target); if (pcmk_is_set(op->call_options, st_opt_topology) && tp) { int i; GList *device_list = NULL; GList *iter = NULL; /* Yep, this looks scary, nested loops all over the place. * Here is what is going on. * Loop1: Iterate through fencing levels. * Loop2: If a fencing level has devices, loop through each device * Loop3: For each device in a fencing level, see what peer owns it * and what that peer has reported the timeout is for the device. */ for (i = 0; i < ST_LEVEL_MAX; i++) { if (!tp->levels[i]) { continue; } for (device_list = tp->levels[i]; device_list; device_list = device_list->next) { for (iter = op->query_results; iter != NULL; iter = iter->next) { const peer_device_info_t *peer = iter->data; if (find_peer_device(op, peer, device_list->data)) { total_timeout += get_device_timeout(op, peer, device_list->data); break; } } /* End Loop3: match device with peer that owns device, find device's timeout period */ } /* End Loop2: iterate through devices at a specific level */ } /*End Loop1: iterate through fencing levels */ } else if (chosen_peer) { total_timeout = get_peer_timeout(op, chosen_peer); } else { total_timeout = op->base_timeout; } return total_timeout ? total_timeout : op->base_timeout; } static void report_timeout_period(remote_fencing_op_t * op, int op_timeout) { GList *iter = NULL; xmlNode *update = NULL; const char *client_node = NULL; const char *client_id = NULL; const char *call_id = NULL; if (op->call_options & st_opt_sync_call) { /* There is no reason to report the timeout for a synchronous call. It * is impossible to use the reported timeout to do anything when the client * is blocking for the response. This update is only important for * async calls that require a callback to report the results in. */ return; } else if (!op->request) { return; } crm_trace("Reporting timeout for %s (id=%.8s)", op->client_name, op->id); client_node = crm_element_value(op->request, F_STONITH_CLIENTNODE); call_id = crm_element_value(op->request, F_STONITH_CALLID); client_id = crm_element_value(op->request, F_STONITH_CLIENTID); if (!client_node || !call_id || !client_id) { return; } if (pcmk__str_eq(client_node, stonith_our_uname, pcmk__str_casei)) { // Client is connected to this node, so send update directly to them do_stonith_async_timeout_update(client_id, call_id, op_timeout); return; } /* The client is connected to another node, relay this update to them */ update = stonith_create_op(op->client_callid, op->id, STONITH_OP_TIMEOUT_UPDATE, NULL, 0); crm_xml_add(update, F_STONITH_REMOTE_OP_ID, op->id); crm_xml_add(update, F_STONITH_CLIENTID, client_id); crm_xml_add(update, F_STONITH_CALLID, call_id); crm_xml_add_int(update, F_STONITH_TIMEOUT, op_timeout); send_cluster_message(crm_get_peer(0, client_node), crm_msg_stonith_ng, update, FALSE); free_xml(update); for (iter = op->duplicates; iter != NULL; iter = iter->next) { remote_fencing_op_t *dup = iter->data; crm_trace("Reporting timeout for duplicate %.8s to client %s", dup->id, dup->client_name); report_timeout_period(iter->data, op_timeout); } } /*! * \internal * \brief Advance an operation to the next device in its topology * * \param[in] op Fencer operation to advance * \param[in] device ID of device that just completed * \param[in] msg If not NULL, XML reply of last delegated fencing operation */ static void advance_topology_device_in_level(remote_fencing_op_t *op, const char *device, xmlNode *msg) { /* Advance to the next device at this topology level, if any */ if (op->devices) { op->devices = op->devices->next; } /* Handle automatic unfencing if an "on" action was requested */ if ((op->phase == st_phase_requested) && pcmk__str_eq(op->action, "on", pcmk__str_casei)) { /* If the device we just executed was required, it's not anymore */ remove_required_device(op, device); /* If there are no more devices at this topology level, run through any * remaining devices with automatic unfencing */ if (op->devices == NULL) { op->devices = op->automatic_list; } } if ((op->devices == NULL) && (op->phase == st_phase_off)) { /* We're done with this level and with required devices, but we had * remapped "reboot" to "off", so start over with "on". If any devices * need to be turned back on, op->devices will be non-NULL after this. */ op_phase_on(op); } // This function is only called if the previous device succeeded pcmk__set_result(&op->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); if (op->devices) { /* Necessary devices remain, so execute the next one */ crm_trace("Next targeting %s on behalf of %s@%s", op->target, op->client_name, op->originator); // The requested delay has been applied for the first device if (op->delay > 0) { op->delay = 0; } request_peer_fencing(op, NULL); } else { /* We're done with all devices and phases, so finalize operation */ crm_trace("Marking complex fencing op targeting %s as complete", op->target); op->state = st_done; finalize_op(op, msg, false); } } static gboolean check_watchdog_fencing_and_wait(remote_fencing_op_t * op) { if (node_does_watchdog_fencing(op->target)) { crm_notice("Waiting %lds for %s to self-fence (%s) for " "client %s " CRM_XS " id=%.8s", (stonith_watchdog_timeout_ms / 1000), op->target, op->action, op->client_name, op->id); op->op_timer_one = g_timeout_add(stonith_watchdog_timeout_ms, remote_op_watchdog_done, op); return TRUE; } else { crm_debug("Skipping fallback to watchdog-fencing as %s is " "not in host-list", op->target); } return FALSE; } /*! * \internal * \brief Ask a peer to execute a fencing operation * * \param[in] op Fencing operation to be executed * \param[in] peer If NULL or topology is in use, choose best peer to execute * the fencing, otherwise use this peer */ static void request_peer_fencing(remote_fencing_op_t *op, peer_device_info_t *peer) { const char *device = NULL; int timeout; CRM_CHECK(op != NULL, return); crm_trace("Action %.8s targeting %s for %s is %s", op->id, op->target, op->client_name, stonith_op_state_str(op->state)); timeout = op->base_timeout; if ((peer == NULL) && !pcmk_is_set(op->call_options, st_opt_topology)) { peer = stonith_choose_peer(op); } if (!op->op_timer_total) { op->total_timeout = TIMEOUT_MULTIPLY_FACTOR * get_op_total_timeout(op, peer); op->op_timer_total = g_timeout_add(1000 * op->total_timeout, remote_op_timeout, op); report_timeout_period(op, op->total_timeout); crm_info("Total timeout set to %d for peer's fencing targeting %s for %s" CRM_XS "id=%.8s", op->total_timeout, op->target, op->client_name, op->id); } if (pcmk_is_set(op->call_options, st_opt_topology) && op->devices) { /* Ignore the caller's peer preference if topology is in use, because * that peer might not have access to the required device. With * topology, stonith_choose_peer() removes the device from further * consideration, so the timeout must be calculated beforehand. * * @TODO Basing the total timeout on the caller's preferred peer (above) * is less than ideal. */ peer = stonith_choose_peer(op); device = op->devices->data; timeout = get_device_timeout(op, peer, device); } if (peer) { int timeout_one = 0; xmlNode *remote_op = stonith_create_op(op->client_callid, op->id, STONITH_OP_FENCE, NULL, 0); crm_xml_add(remote_op, F_STONITH_REMOTE_OP_ID, op->id); crm_xml_add(remote_op, F_STONITH_TARGET, op->target); crm_xml_add(remote_op, F_STONITH_ACTION, op->action); crm_xml_add(remote_op, F_STONITH_ORIGIN, op->originator); crm_xml_add(remote_op, F_STONITH_CLIENTID, op->client_id); crm_xml_add(remote_op, F_STONITH_CLIENTNAME, op->client_name); crm_xml_add_int(remote_op, F_STONITH_TIMEOUT, timeout); crm_xml_add_int(remote_op, F_STONITH_CALLOPTS, op->call_options); crm_xml_add_int(remote_op, F_STONITH_DELAY, op->delay); if (device) { timeout_one = TIMEOUT_MULTIPLY_FACTOR * get_device_timeout(op, peer, device); crm_notice("Requesting that %s perform '%s' action targeting %s " "using %s " CRM_XS " for client %s (%ds)", peer->host, op->action, op->target, device, op->client_name, timeout_one); crm_xml_add(remote_op, F_STONITH_DEVICE, device); } else { timeout_one = TIMEOUT_MULTIPLY_FACTOR * get_peer_timeout(op, peer); crm_notice("Requesting that %s perform '%s' action targeting %s " CRM_XS " for client %s (%ds, %lds)", peer->host, op->action, op->target, op->client_name, timeout_one, stonith_watchdog_timeout_ms); } op->state = st_exec; if (op->op_timer_one) { g_source_remove(op->op_timer_one); } if (!(stonith_watchdog_timeout_ms > 0 && ( (pcmk__str_eq(device, STONITH_WATCHDOG_ID, pcmk__str_none)) || (pcmk__str_eq(peer->host, op->target, pcmk__str_casei) && !pcmk__str_eq(op->action, "on", pcmk__str_casei))) && check_watchdog_fencing_and_wait(op))) { /* Some thoughts about self-fencing cases reaching this point: - Actually check in check_watchdog_fencing_and_wait shouldn't fail if STONITH_WATCHDOG_ID is chosen as fencing-device and it being present implies watchdog-fencing is enabled anyway - If watchdog-fencing is disabled either in general or for a specific target - detected in check_watchdog_fencing_and_wait - for some other kind of self-fencing we can't expect a success answer but timeout is fine if the node doesn't come back in between - Delicate might be the case where we have watchdog-fencing enabled for a node but the watchdog-fencing-device isn't explicitly chosen for suicide. Local pe-execution in sbd may detect the node as unclean and lead to timely suicide. Otherwise the selection of stonith-watchdog-timeout at least is questionable. */ op->op_timer_one = g_timeout_add((1000 * timeout_one), remote_op_timeout_one, op); } send_cluster_message(crm_get_peer(0, peer->host), crm_msg_stonith_ng, remote_op, FALSE); peer->tried = TRUE; free_xml(remote_op); return; } else if (op->phase == st_phase_on) { /* A remapped "on" cannot be executed, but the node was already * turned off successfully, so ignore the error and continue. */ crm_warn("Ignoring %s 'on' failure (no capable peers) targeting %s " "after successful 'off'", device, op->target); advance_topology_device_in_level(op, device, NULL); return; } else if (op->owner == FALSE) { crm_err("Fencing (%s) targeting %s for client %s is not ours to control", op->action, op->target, op->client_name); } else if (op->query_timer == 0) { /* We've exhausted all available peers */ crm_info("No remaining peers capable of fencing (%s) %s for client %s " CRM_XS " state=%s", op->action, op->target, op->client_name, stonith_op_state_str(op->state)); CRM_CHECK(op->state < st_done, return); finalize_timed_out_op(op, "All nodes failed, or are unable, to " "fence target"); } else if(op->replies >= op->replies_expected || op->replies >= fencing_active_peers()) { /* if the operation never left the query state, * but we have all the expected replies, then no devices * are available to execute the fencing operation. */ if(stonith_watchdog_timeout_ms > 0 && pcmk__str_eq(device, STONITH_WATCHDOG_ID, pcmk__str_null_matches)) { if (check_watchdog_fencing_and_wait(op)) { return; } } if (op->state == st_query) { crm_info("No peers (out of %d) have devices capable of fencing " "(%s) %s for client %s " CRM_XS " state=%s", op->replies, op->action, op->target, op->client_name, stonith_op_state_str(op->state)); pcmk__reset_result(&op->result); pcmk__set_result(&op->result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, NULL); } else { if (pcmk_is_set(op->call_options, st_opt_topology)) { pcmk__reset_result(&op->result); pcmk__set_result(&op->result, CRM_EX_ERROR, PCMK_EXEC_NO_FENCE_DEVICE, NULL); } /* ... else use existing result from previous failed attempt * (topology is not in use, and no devices remain to be attempted). * Overwriting the result with PCMK_EXEC_NO_FENCE_DEVICE would * prevent finalize_op() from setting the correct delegate if * needed. */ crm_info("No peers (out of %d) are capable of fencing (%s) %s " "for client %s " CRM_XS " state=%s", op->replies, op->action, op->target, op->client_name, stonith_op_state_str(op->state)); } op->state = st_failed; finalize_op(op, NULL, false); } else { crm_info("Waiting for additional peers capable of fencing (%s) %s%s%s " "for client %s " CRM_XS " id=%.8s", op->action, op->target, (device? " using " : ""), (device? device : ""), op->client_name, op->id); } } /*! * \internal * \brief Comparison function for sorting query results * * \param[in] a GList item to compare * \param[in] b GList item to compare * * \return Per the glib documentation, "a negative integer if the first value * comes before the second, 0 if they are equal, or a positive integer * if the first value comes after the second." */ static gint sort_peers(gconstpointer a, gconstpointer b) { const peer_device_info_t *peer_a = a; const peer_device_info_t *peer_b = b; return (peer_b->ndevices - peer_a->ndevices); } /*! * \internal * \brief Determine if all the devices in the topology are found or not */ static gboolean all_topology_devices_found(remote_fencing_op_t * op) { GList *device = NULL; GList *iter = NULL; device_properties_t *match = NULL; stonith_topology_t *tp = NULL; gboolean skip_target = FALSE; int i; tp = find_topology_for_host(op->target); if (!tp) { return FALSE; } if (pcmk__is_fencing_action(op->action)) { /* Don't count the devices on the target node if we are killing * the target node. */ skip_target = TRUE; } for (i = 0; i < ST_LEVEL_MAX; i++) { for (device = tp->levels[i]; device; device = device->next) { match = NULL; for (iter = op->query_results; iter && !match; iter = iter->next) { peer_device_info_t *peer = iter->data; if (skip_target && pcmk__str_eq(peer->host, op->target, pcmk__str_casei)) { continue; } match = find_peer_device(op, peer, device->data); } if (!match) { return FALSE; } } } return TRUE; } /*! * \internal * \brief Parse action-specific device properties from XML * * \param[in] msg XML element containing the properties * \param[in] peer Name of peer that sent XML (for logs) * \param[in] device Device ID (for logs) * \param[in] action Action the properties relate to (for logs) * \param[in] phase Phase the properties relate to * \param[in,out] props Device properties to update */ static void parse_action_specific(xmlNode *xml, const char *peer, const char *device, const char *action, remote_fencing_op_t *op, enum st_remap_phase phase, device_properties_t *props) { props->custom_action_timeout[phase] = 0; crm_element_value_int(xml, F_STONITH_ACTION_TIMEOUT, &props->custom_action_timeout[phase]); if (props->custom_action_timeout[phase]) { crm_trace("Peer %s with device %s returned %s action timeout %d", peer, device, action, props->custom_action_timeout[phase]); } props->delay_max[phase] = 0; crm_element_value_int(xml, F_STONITH_DELAY_MAX, &props->delay_max[phase]); if (props->delay_max[phase]) { crm_trace("Peer %s with device %s returned maximum of random delay %d for %s", peer, device, props->delay_max[phase], action); } props->delay_base[phase] = 0; crm_element_value_int(xml, F_STONITH_DELAY_BASE, &props->delay_base[phase]); if (props->delay_base[phase]) { crm_trace("Peer %s with device %s returned base delay %d for %s", peer, device, props->delay_base[phase], action); } /* Handle devices with automatic unfencing */ if (pcmk__str_eq(action, "on", pcmk__str_casei)) { int required = 0; crm_element_value_int(xml, F_STONITH_DEVICE_REQUIRED, &required); if (required) { crm_trace("Peer %s requires device %s to execute for action %s", peer, device, action); add_required_device(op, device); } } /* If a reboot is remapped to off+on, it's possible that a node is allowed * to perform one action but not another. */ if (pcmk__xe_attr_is_true(xml, F_STONITH_ACTION_DISALLOWED)) { props->disallowed[phase] = TRUE; crm_trace("Peer %s is disallowed from executing %s for device %s", peer, action, device); } } /*! * \internal * \brief Parse one device's properties from peer's XML query reply * * \param[in] xml XML node containing device properties * \param[in,out] op Operation that query and reply relate to * \param[in,out] peer Peer's device information * \param[in] device ID of device being parsed */ static void add_device_properties(xmlNode *xml, remote_fencing_op_t *op, peer_device_info_t *peer, const char *device) { xmlNode *child; int verified = 0; device_properties_t *props = calloc(1, sizeof(device_properties_t)); /* Add a new entry to this peer's devices list */ CRM_ASSERT(props != NULL); g_hash_table_insert(peer->devices, strdup(device), props); /* Peers with verified (monitored) access will be preferred */ crm_element_value_int(xml, F_STONITH_DEVICE_VERIFIED, &verified); if (verified) { crm_trace("Peer %s has confirmed a verified device %s", peer->host, device); props->verified = TRUE; } /* Parse action-specific device properties */ parse_action_specific(xml, peer->host, device, op_requested_action(op), op, st_phase_requested, props); for (child = pcmk__xml_first_child(xml); child != NULL; child = pcmk__xml_next(child)) { /* Replies for "reboot" operations will include the action-specific * values for "off" and "on" in child elements, just in case the reboot * winds up getting remapped. */ if (pcmk__str_eq(ID(child), "off", pcmk__str_casei)) { parse_action_specific(child, peer->host, device, "off", op, st_phase_off, props); } else if (pcmk__str_eq(ID(child), "on", pcmk__str_casei)) { parse_action_specific(child, peer->host, device, "on", op, st_phase_on, props); } } } /*! * \internal * \brief Parse a peer's XML query reply and add it to operation's results * * \param[in,out] op Operation that query and reply relate to * \param[in] host Name of peer that sent this reply * \param[in] ndevices Number of devices expected in reply * \param[in] xml XML node containing device list * * \return Newly allocated result structure with parsed reply */ static peer_device_info_t * add_result(remote_fencing_op_t *op, const char *host, int ndevices, xmlNode *xml) { peer_device_info_t *peer = calloc(1, sizeof(peer_device_info_t)); xmlNode *child; // cppcheck seems not to understand the abort logic in CRM_CHECK // cppcheck-suppress memleak CRM_CHECK(peer != NULL, return NULL); peer->host = strdup(host); peer->devices = pcmk__strkey_table(free, free); /* Each child element describes one capable device available to the peer */ for (child = pcmk__xml_first_child(xml); child != NULL; child = pcmk__xml_next(child)) { const char *device = ID(child); if (device) { add_device_properties(child, op, peer, device); } } peer->ndevices = g_hash_table_size(peer->devices); CRM_CHECK(ndevices == peer->ndevices, crm_err("Query claimed to have %d device%s but %d found", ndevices, pcmk__plural_s(ndevices), peer->ndevices)); op->query_results = g_list_insert_sorted(op->query_results, peer, sort_peers); return peer; } /*! * \internal * \brief Handle a peer's reply to our fencing query * * Parse a query result from XML and store it in the remote operation * table, and when enough replies have been received, issue a fencing request. * * \param[in] msg XML reply received * * \return pcmk_ok on success, -errno on error * * \note See initiate_remote_stonith_op() for how the XML query was initially * formed, and stonith_query() for how the peer formed its XML reply. */ int process_remote_stonith_query(xmlNode * msg) { int ndevices = 0; gboolean host_is_target = FALSE; gboolean have_all_replies = FALSE; const char *id = NULL; const char *host = NULL; remote_fencing_op_t *op = NULL; peer_device_info_t *peer = NULL; uint32_t replies_expected; xmlNode *dev = get_xpath_object("//@" F_STONITH_REMOTE_OP_ID, msg, LOG_ERR); CRM_CHECK(dev != NULL, return -EPROTO); id = crm_element_value(dev, F_STONITH_REMOTE_OP_ID); CRM_CHECK(id != NULL, return -EPROTO); dev = get_xpath_object("//@" F_STONITH_AVAILABLE_DEVICES, msg, LOG_ERR); CRM_CHECK(dev != NULL, return -EPROTO); crm_element_value_int(dev, F_STONITH_AVAILABLE_DEVICES, &ndevices); op = g_hash_table_lookup(stonith_remote_op_list, id); if (op == NULL) { crm_debug("Received query reply for unknown or expired operation %s", id); return -EOPNOTSUPP; } replies_expected = fencing_active_peers(); if (op->replies_expected < replies_expected) { replies_expected = op->replies_expected; } if ((++op->replies >= replies_expected) && (op->state == st_query)) { have_all_replies = TRUE; } host = crm_element_value(msg, F_ORIG); host_is_target = pcmk__str_eq(host, op->target, pcmk__str_casei); crm_info("Query result %d of %d from %s for %s/%s (%d device%s) %s", op->replies, replies_expected, host, op->target, op->action, ndevices, pcmk__plural_s(ndevices), id); if (ndevices > 0) { peer = add_result(op, host, ndevices, dev); } pcmk__set_result(&op->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); if (pcmk_is_set(op->call_options, st_opt_topology)) { /* If we start the fencing before all the topology results are in, * it is possible fencing levels will be skipped because of the missing * query results. */ if (op->state == st_query && all_topology_devices_found(op)) { /* All the query results are in for the topology, start the fencing ops. */ crm_trace("All topology devices found"); request_peer_fencing(op, peer); } else if (have_all_replies) { crm_info("All topology query replies have arrived, continuing (%d expected/%d received) ", replies_expected, op->replies); request_peer_fencing(op, NULL); } } else if (op->state == st_query) { int nverified = count_peer_devices(op, peer, TRUE); /* We have a result for a non-topology fencing op that looks promising, * go ahead and start fencing before query timeout */ if ((peer != NULL) && !host_is_target && nverified) { /* we have a verified device living on a peer that is not the target */ crm_trace("Found %d verified device%s", nverified, pcmk__plural_s(nverified)); request_peer_fencing(op, peer); } else if (have_all_replies) { crm_info("All query replies have arrived, continuing (%d expected/%d received) ", replies_expected, op->replies); request_peer_fencing(op, NULL); } else { crm_trace("Waiting for more peer results before launching fencing operation"); } } else if ((peer != NULL) && (op->state == st_done)) { crm_info("Discarding query result from %s (%d device%s): " "Operation is %s", peer->host, peer->ndevices, pcmk__plural_s(peer->ndevices), stonith_op_state_str(op->state)); } return pcmk_ok; } /*! * \internal * \brief Handle a peer's reply to a fencing request * * Parse a fencing reply from XML, and either finalize the operation * or attempt another device as appropriate. * * \param[in] msg XML reply received */ void fenced_process_fencing_reply(xmlNode *msg) { const char *id = NULL; const char *device = NULL; remote_fencing_op_t *op = NULL; xmlNode *dev = get_xpath_object("//@" F_STONITH_REMOTE_OP_ID, msg, LOG_ERR); pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; CRM_CHECK(dev != NULL, return); id = crm_element_value(dev, F_STONITH_REMOTE_OP_ID); CRM_CHECK(id != NULL, return); dev = stonith__find_xe_with_result(msg); CRM_CHECK(dev != NULL, return); stonith__xe_get_result(dev, &result); device = crm_element_value(dev, F_STONITH_DEVICE); if (stonith_remote_op_list) { op = g_hash_table_lookup(stonith_remote_op_list, id); } if ((op == NULL) && pcmk__result_ok(&result)) { /* Record successful fencing operations */ const char *client_id = crm_element_value(dev, F_STONITH_CLIENTID); op = create_remote_stonith_op(client_id, dev, TRUE); } if (op == NULL) { /* Could be for an event that began before we started */ /* TODO: Record the op for later querying */ crm_info("Received peer result of unknown or expired operation %s", id); pcmk__reset_result(&result); return; } pcmk__reset_result(&op->result); op->result = result; // The operation takes ownership of the result if (op->devices && device && !pcmk__str_eq(op->devices->data, device, pcmk__str_casei)) { crm_err("Received outdated reply for device %s (instead of %s) to " "fence (%s) %s. Operation already timed out at peer level.", device, (const char *) op->devices->data, op->action, op->target); return; } if (pcmk__str_eq(crm_element_value(msg, F_SUBTYPE), "broadcast", pcmk__str_casei)) { if (pcmk__result_ok(&op->result)) { op->state = st_done; } else { op->state = st_failed; } finalize_op(op, msg, false); return; } else if (!pcmk__str_eq(op->originator, stonith_our_uname, pcmk__str_casei)) { /* If this isn't a remote level broadcast, and we are not the * originator of the operation, we should not be receiving this msg. */ crm_err("Received non-broadcast fencing result for operation %.8s " "we do not own (device %s targeting %s)", op->id, device, op->target); return; } if (pcmk_is_set(op->call_options, st_opt_topology)) { const char *device = NULL; const char *reason = op->result.exit_reason; /* We own the op, and it is complete. broadcast the result to all nodes * and notify our local clients. */ if (op->state == st_done) { finalize_op(op, msg, false); return; } device = crm_element_value(msg, F_STONITH_DEVICE); if ((op->phase == 2) && !pcmk__result_ok(&op->result)) { /* A remapped "on" failed, but the node was already turned off * successfully, so ignore the error and continue. */ crm_warn("Ignoring %s 'on' failure (%s%s%s) targeting %s " "after successful 'off'", device, pcmk_exec_status_str(op->result.execution_status), (reason == NULL)? "" : ": ", (reason == NULL)? "" : reason, op->target); pcmk__set_result(&op->result, CRM_EX_OK, PCMK_EXEC_DONE, NULL); } else { crm_notice("Action '%s' targeting %s%s%s on behalf of %s@%s: " "%s%s%s%s", op->action, op->target, ((device == NULL)? "" : " using "), ((device == NULL)? "" : device), op->client_name, op->originator, pcmk_exec_status_str(op->result.execution_status), (reason == NULL)? "" : " (", (reason == NULL)? "" : reason, (reason == NULL)? "" : ")"); } if (pcmk__result_ok(&op->result)) { /* An operation completed successfully. Try another device if * necessary, otherwise mark the operation as done. */ advance_topology_device_in_level(op, device, msg); return; } else { /* This device failed, time to try another topology level. If no other * levels are available, mark this operation as failed and report results. */ if (advance_topology_level(op, false) != pcmk_rc_ok) { op->state = st_failed; finalize_op(op, msg, false); return; } } } else if (pcmk__result_ok(&op->result) && (op->devices == NULL)) { op->state = st_done; finalize_op(op, msg, false); return; } else if ((op->result.execution_status == PCMK_EXEC_TIMEOUT) && (op->devices == NULL)) { /* If the operation timed out don't bother retrying other peers. */ op->state = st_failed; finalize_op(op, msg, false); return; } else { /* fall-through and attempt other fencing action using another peer */ } /* Retry on failure */ crm_trace("Next for %s on behalf of %s@%s (result was: %s)", op->target, op->originator, op->client_name, pcmk_exec_status_str(op->result.execution_status)); request_peer_fencing(op, NULL); } gboolean stonith_check_fence_tolerance(int tolerance, const char *target, const char *action) { GHashTableIter iter; time_t now = time(NULL); remote_fencing_op_t *rop = NULL; if (tolerance <= 0 || !stonith_remote_op_list || target == NULL || action == NULL) { return FALSE; } g_hash_table_iter_init(&iter, stonith_remote_op_list); while (g_hash_table_iter_next(&iter, NULL, (void **)&rop)) { if (strcmp(rop->target, target) != 0) { continue; } else if (rop->state != st_done) { continue; /* We don't have to worry about remapped reboots here * because if state is done, any remapping has been undone */ } else if (strcmp(rop->action, action) != 0) { continue; } else if ((rop->completed + tolerance) < now) { continue; } crm_notice("Target %s was fenced (%s) less than %ds ago by %s on behalf of %s", target, action, tolerance, rop->delegate, rop->originator); return TRUE; } return FALSE; } diff --git a/daemons/fenced/pacemaker-fenced.c b/daemons/fenced/pacemaker-fenced.c index caab7de83e..fd4ef4ff93 100644 --- a/daemons/fenced/pacemaker-fenced.c +++ b/daemons/fenced/pacemaker-fenced.c @@ -1,1729 +1,1735 @@ /* * Copyright 2009-2022 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU General Public License version 2 * or later (GPLv2+) WITHOUT ANY WARRANTY. */ #include #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 #include char *stonith_our_uname = NULL; long stonith_watchdog_timeout_ms = 0; GList *stonith_watchdog_targets = NULL; static GMainLoop *mainloop = NULL; gboolean stand_alone = FALSE; static gboolean no_cib_connect = FALSE; static gboolean stonith_shutdown_flag = FALSE; static qb_ipcs_service_t *ipcs = NULL; static xmlNode *local_cib = NULL; static pe_working_set_t *fenced_data_set = NULL; static const unsigned long long data_set_flags = pe_flag_quick_location | pe_flag_no_compat | pe_flag_no_counts; static cib_t *cib_api = NULL; static pcmk__output_t *out = NULL; pcmk__supported_format_t formats[] = { PCMK__SUPPORTED_FORMAT_LOG, PCMK__SUPPORTED_FORMAT_NONE, PCMK__SUPPORTED_FORMAT_TEXT, { NULL, NULL, NULL } }; static void stonith_shutdown(int nsig); 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", CRM_EX_PROTOCOL); return 0; } op = crm_element_value(request, F_CRM_TASK); if(pcmk__str_eq(op, CRM_OP_RM_NODE_CACHE, pcmk__str_casei)) { crm_xml_add(request, F_TYPE, 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); 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, F_ORIG); 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, F_ORIG, from); /* crm_xml_add_int(xml, F_SEQ, wrapper->id); */ 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(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) { 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, F_SUBTYPE); 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|crm_ipc_server_error); 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, F_TYPE, 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, F_TYPE, T_STONITH_NOTIFY); crm_xml_add(update_msg, F_SUBTYPE, 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)); } static void topology_remove_helper(const char *node, int level) { char *desc = NULL; pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; xmlNode *data = create_xml_node(NULL, XML_TAG_FENCING_LEVEL); crm_xml_add(data, F_STONITH_ORIGIN, __func__); crm_xml_add_int(data, XML_ATTR_STONITH_INDEX, level); crm_xml_add(data, XML_ATTR_STONITH_TARGET, node); fenced_unregister_level(data, &desc, &result); fenced_send_level_notification(STONITH_OP_LEVEL_DEL, &result, desc); pcmk__reset_result(&result); free_xml(data); free(desc); } static void remove_cib_device(xmlXPathObjectPtr xpathObj) { int max = numXpathResults(xpathObj), lpc = 0; for (lpc = 0; lpc < max; lpc++) { const char *rsc_id = NULL; const char *standard = NULL; xmlNode *match = getXpathResult(xpathObj, lpc); CRM_LOG_ASSERT(match != NULL); if(match != NULL) { standard = crm_element_value(match, XML_AGENT_ATTR_CLASS); } if (!pcmk__str_eq(standard, PCMK_RESOURCE_CLASS_STONITH, pcmk__str_casei)) { continue; } rsc_id = crm_element_value(match, XML_ATTR_ID); stonith_device_remove(rsc_id, true); } } static void -handle_topology_change(xmlNode *match, bool remove) +remove_topology_level(xmlNode *match) { - char *desc = NULL; - pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; + int index = 0; + char *key = NULL; CRM_CHECK(match != NULL, return); - crm_trace("Updating %s", ID(match)); - if(remove) { - int index = 0; - char *key = stonith_level_key(match, fenced_target_by_unknown); + key = stonith_level_key(match, fenced_target_by_unknown); + crm_element_value_int(match, XML_ATTR_STONITH_INDEX, &index); + topology_remove_helper(key, index); + free(key); +} - crm_element_value_int(match, XML_ATTR_STONITH_INDEX, &index); - topology_remove_helper(key, index); - free(key); - } +static void +add_topology_level(xmlNode *match) +{ + char *desc = NULL; + pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; + + CRM_CHECK(match != NULL, return); fenced_register_level(match, &desc, &result); fenced_send_level_notification(STONITH_OP_LEVEL_ADD, &result, desc); pcmk__reset_result(&result); free(desc); } static void remove_fencing_topology(xmlXPathObjectPtr xpathObj) { int max = numXpathResults(xpathObj), lpc = 0; for (lpc = 0; lpc < max; lpc++) { xmlNode *match = getXpathResult(xpathObj, lpc); CRM_LOG_ASSERT(match != NULL); if (match && crm_element_value(match, XML_DIFF_MARKER)) { /* Deletion */ int index = 0; char *target = stonith_level_key(match, fenced_target_by_unknown); crm_element_value_int(match, XML_ATTR_STONITH_INDEX, &index); if (target == NULL) { crm_err("Invalid fencing target in element %s", ID(match)); } else if (index <= 0) { crm_err("Invalid level for %s in element %s", target, ID(match)); } else { topology_remove_helper(target, index); } /* } else { Deal with modifications during the 'addition' stage */ } } } static void register_fencing_topology(xmlXPathObjectPtr xpathObj) { int max = numXpathResults(xpathObj), lpc = 0; for (lpc = 0; lpc < max; lpc++) { xmlNode *match = getXpathResult(xpathObj, lpc); - handle_topology_change(match, TRUE); + remove_topology_level(match); + add_topology_level(match); } } /* Fencing */ static void fencing_topology_init(void) { xmlXPathObjectPtr xpathObj = NULL; const char *xpath = "//" XML_TAG_FENCING_LEVEL; crm_trace("Full topology refresh"); free_topology_list(); init_topology_list(); /* Grab everything */ xpathObj = xpath_search(local_cib, xpath); register_fencing_topology(xpathObj); freeXpathObject(xpathObj); } #define rsc_name(x) x->clone_name?x->clone_name:x->id /*! * \internal * \brief Check whether our uname is in a resource's allowed node list * * \param[in] rsc Resource to check * * \return Pointer to node object if found, NULL otherwise */ static pe_node_t * our_node_allowed_for(pe_resource_t *rsc) { GHashTableIter iter; pe_node_t *node = NULL; if (rsc && stonith_our_uname) { g_hash_table_iter_init(&iter, rsc->allowed_nodes); while (g_hash_table_iter_next(&iter, NULL, (void **)&node)) { if (node && strcmp(node->details->uname, stonith_our_uname) == 0) { break; } node = NULL; } } return node; } static void watchdog_device_update(void) { if (stonith_watchdog_timeout_ms > 0) { if (!g_hash_table_lookup(device_list, STONITH_WATCHDOG_ID) && !stonith_watchdog_targets) { /* getting here watchdog-fencing enabled, no device there yet and reason isn't stonith_watchdog_targets preventing that */ int rc; xmlNode *xml; xml = create_device_registration_xml( STONITH_WATCHDOG_ID, st_namespace_internal, STONITH_WATCHDOG_AGENT, NULL, /* stonith_device_register will add our own name as PCMK_STONITH_HOST_LIST param so we can skip that here */ NULL); - rc = stonith_device_register(xml, NULL, TRUE); + rc = stonith_device_register(xml, TRUE); free_xml(xml); if (rc != pcmk_ok) { crm_crit("Cannot register watchdog pseudo fence agent"); crm_exit(CRM_EX_FATAL); } } } else { /* be silent if no device - todo parameter to stonith_device_remove */ if (g_hash_table_lookup(device_list, STONITH_WATCHDOG_ID)) { stonith_device_remove(STONITH_WATCHDOG_ID, true); } } } static void update_stonith_watchdog_timeout_ms(xmlNode *cib) { xmlNode *stonith_enabled_xml = NULL; bool stonith_enabled = false; int rc = pcmk_rc_ok; long timeout_ms = 0; stonith_enabled_xml = get_xpath_object("//nvpair[@name='stonith-enabled']", cib, LOG_NEVER); rc = pcmk__xe_get_bool_attr(stonith_enabled_xml, XML_NVPAIR_ATTR_VALUE, &stonith_enabled); if (rc != pcmk_rc_ok || stonith_enabled) { xmlNode *stonith_watchdog_xml = NULL; const char *value = NULL; stonith_watchdog_xml = get_xpath_object("//nvpair[@name='stonith-watchdog-timeout']", cib, LOG_NEVER); if (stonith_watchdog_xml) { value = crm_element_value(stonith_watchdog_xml, XML_NVPAIR_ATTR_VALUE); } if (value) { timeout_ms = crm_get_msec(value); } if (timeout_ms < 0) { timeout_ms = pcmk__auto_watchdog_timeout(); } } stonith_watchdog_timeout_ms = timeout_ms; } /*! * \internal * \brief If a resource or any of its children are STONITH devices, update their * definitions given a cluster working set. * * \param[in] rsc Resource to check * \param[in] data_set Cluster working set with device information */ static void cib_device_update(pe_resource_t *rsc, pe_working_set_t *data_set) { pe_node_t *node = NULL; const char *value = NULL; const char *rclass = NULL; pe_node_t *parent = NULL; /* If this is a complex resource, check children rather than this resource itself. */ if(rsc->children) { GList *gIter = NULL; for (gIter = rsc->children; gIter != NULL; gIter = gIter->next) { cib_device_update(gIter->data, data_set); if(pe_rsc_is_clone(rsc)) { crm_trace("Only processing one copy of the clone %s", rsc->id); break; } } return; } /* We only care about STONITH resources. */ rclass = crm_element_value(rsc->xml, XML_AGENT_ATTR_CLASS); if (!pcmk__str_eq(rclass, PCMK_RESOURCE_CLASS_STONITH, pcmk__str_casei)) { return; } /* If this STONITH resource is disabled, remove it. */ if (pe__resource_is_disabled(rsc)) { crm_info("Device %s has been disabled", rsc->id); return; } /* if watchdog-fencing is disabled handle any watchdog-fence resource as if it was disabled */ if ((stonith_watchdog_timeout_ms <= 0) && pcmk__str_eq(rsc->id, STONITH_WATCHDOG_ID, pcmk__str_none)) { crm_info("Watchdog-fencing disabled thus handling " "device %s as disabled", rsc->id); return; } /* Check whether our node is allowed for this resource (and its parent if in a group) */ node = our_node_allowed_for(rsc); if (rsc->parent && (rsc->parent->variant == pe_group)) { parent = our_node_allowed_for(rsc->parent); } if(node == NULL) { /* Our node is disallowed, so remove the device */ GHashTableIter iter; crm_info("Device %s has been disabled on %s: unknown", rsc->id, stonith_our_uname); g_hash_table_iter_init(&iter, rsc->allowed_nodes); while (g_hash_table_iter_next(&iter, NULL, (void **)&node)) { crm_trace("Available: %s = %d", node->details->uname, node->weight); } return; } else if(node->weight < 0 || (parent && parent->weight < 0)) { /* Our node (or its group) is disallowed by score, so remove the device */ char *score = score2char((node->weight < 0) ? node->weight : parent->weight); crm_info("Device %s has been disabled on %s: score=%s", rsc->id, stonith_our_uname, score); free(score); return; } else { /* Our node is allowed, so update the device information */ int rc; xmlNode *data; GHashTable *rsc_params = NULL; GHashTableIter gIter; stonith_key_value_t *params = NULL; const char *name = NULL; const char *agent = crm_element_value(rsc->xml, XML_EXPR_ATTR_TYPE); const char *rsc_provides = NULL; crm_debug("Device %s is allowed on %s: score=%d", rsc->id, stonith_our_uname, node->weight); rsc_params = pe_rsc_params(rsc, node, data_set); get_meta_attributes(rsc->meta, rsc, node, data_set); rsc_provides = g_hash_table_lookup(rsc->meta, PCMK_STONITH_PROVIDES); g_hash_table_iter_init(&gIter, rsc_params); while (g_hash_table_iter_next(&gIter, (gpointer *) & name, (gpointer *) & value)) { if (!name || !value) { continue; } params = stonith_key_value_add(params, name, value); crm_trace(" %s=%s", name, value); } data = create_device_registration_xml(rsc_name(rsc), st_namespace_any, agent, params, rsc_provides); stonith_key_value_freeall(params, 1, 1); - rc = stonith_device_register(data, NULL, TRUE); + rc = stonith_device_register(data, TRUE); CRM_ASSERT(rc == pcmk_ok); free_xml(data); } } /*! * \internal * \brief Update all STONITH device definitions based on current CIB */ static void cib_devices_update(void) { GHashTableIter iter; stonith_device_t *device = NULL; crm_info("Updating devices to version %s.%s.%s", crm_element_value(local_cib, XML_ATTR_GENERATION_ADMIN), crm_element_value(local_cib, XML_ATTR_GENERATION), crm_element_value(local_cib, XML_ATTR_NUMUPDATES)); if (fenced_data_set->now != NULL) { crm_time_free(fenced_data_set->now); fenced_data_set->now = NULL; } fenced_data_set->localhost = stonith_our_uname; pcmk__schedule_actions(local_cib, data_set_flags, fenced_data_set); g_hash_table_iter_init(&iter, device_list); while (g_hash_table_iter_next(&iter, NULL, (void **)&device)) { if (device->cib_registered) { device->dirty = TRUE; } } /* have list repopulated if cib has a watchdog-fencing-resource TODO: keep a cached list for queries happening while we are refreshing */ g_list_free_full(stonith_watchdog_targets, free); stonith_watchdog_targets = NULL; g_list_foreach(fenced_data_set->resources, (GFunc) cib_device_update, fenced_data_set); g_hash_table_iter_init(&iter, device_list); while (g_hash_table_iter_next(&iter, NULL, (void **)&device)) { if (device->dirty) { g_hash_table_iter_remove(&iter); } } fenced_data_set->input = NULL; // Wasn't a copy, so don't let API free it pe_reset_working_set(fenced_data_set); } static void update_cib_stonith_devices_v2(const char *event, xmlNode * msg) { xmlNode *change = NULL; char *reason = NULL; bool needs_update = FALSE; xmlNode *patchset = get_message_xml(msg, F_CIB_UPDATE_RESULT); for (change = pcmk__xml_first_child(patchset); change != NULL; change = pcmk__xml_next(change)) { const char *op = crm_element_value(change, XML_DIFF_OP); const char *xpath = crm_element_value(change, XML_DIFF_PATH); const char *shortpath = NULL; if ((op == NULL) || (strcmp(op, "move") == 0) || strstr(xpath, "/"XML_CIB_TAG_STATUS)) { continue; } else if (pcmk__str_eq(op, "delete", pcmk__str_casei) && strstr(xpath, "/"XML_CIB_TAG_RESOURCE)) { const char *rsc_id = NULL; char *search = NULL; char *mutable = NULL; if (strstr(xpath, XML_TAG_ATTR_SETS) || strstr(xpath, XML_TAG_META_SETS)) { needs_update = TRUE; reason = strdup("(meta) attribute deleted from resource"); break; } mutable = strdup(xpath); rsc_id = strstr(mutable, "primitive[@id=\'"); if (rsc_id != NULL) { rsc_id += strlen("primitive[@id=\'"); search = strchr(rsc_id, '\''); } if (search != NULL) { *search = 0; stonith_device_remove(rsc_id, true); /* watchdog_device_update called afterwards to fall back to implicit definition if needed */ } else { crm_warn("Ignoring malformed CIB update (resource deletion)"); } free(mutable); } else if (strstr(xpath, "/"XML_CIB_TAG_RESOURCES) || strstr(xpath, "/"XML_CIB_TAG_CONSTRAINTS) || strstr(xpath, "/"XML_CIB_TAG_RSCCONFIG)) { shortpath = strrchr(xpath, '/'); CRM_ASSERT(shortpath); reason = crm_strdup_printf("%s %s", op, shortpath+1); needs_update = TRUE; break; } } if(needs_update) { crm_info("Updating device list from CIB: %s", reason); cib_devices_update(); } else { crm_trace("No updates for device list found in CIB"); } free(reason); } static void update_cib_stonith_devices_v1(const char *event, xmlNode * msg) { const char *reason = "none"; gboolean needs_update = FALSE; xmlXPathObjectPtr xpath_obj = NULL; /* process new constraints */ xpath_obj = xpath_search(msg, "//" F_CIB_UPDATE_RESULT "//" XML_CONS_TAG_RSC_LOCATION); if (numXpathResults(xpath_obj) > 0) { int max = numXpathResults(xpath_obj), lpc = 0; /* Safest and simplest to always recompute */ needs_update = TRUE; reason = "new location constraint"; for (lpc = 0; lpc < max; lpc++) { xmlNode *match = getXpathResult(xpath_obj, lpc); crm_log_xml_trace(match, "new constraint"); } } freeXpathObject(xpath_obj); /* process deletions */ xpath_obj = xpath_search(msg, "//" F_CIB_UPDATE_RESULT "//" XML_TAG_DIFF_REMOVED "//" XML_CIB_TAG_RESOURCE); if (numXpathResults(xpath_obj) > 0) { remove_cib_device(xpath_obj); } freeXpathObject(xpath_obj); /* process additions */ xpath_obj = xpath_search(msg, "//" F_CIB_UPDATE_RESULT "//" XML_TAG_DIFF_ADDED "//" XML_CIB_TAG_RESOURCE); if (numXpathResults(xpath_obj) > 0) { int max = numXpathResults(xpath_obj), lpc = 0; for (lpc = 0; lpc < max; lpc++) { const char *rsc_id = NULL; const char *standard = NULL; xmlNode *match = getXpathResult(xpath_obj, lpc); rsc_id = crm_element_value(match, XML_ATTR_ID); standard = crm_element_value(match, XML_AGENT_ATTR_CLASS); if (!pcmk__str_eq(standard, PCMK_RESOURCE_CLASS_STONITH, pcmk__str_casei)) { continue; } crm_trace("Fencing resource %s was added or modified", rsc_id); reason = "new resource"; needs_update = TRUE; } } freeXpathObject(xpath_obj); if(needs_update) { crm_info("Updating device list from CIB: %s", reason); cib_devices_update(); } } static void update_cib_stonith_devices(const char *event, xmlNode * msg) { int format = 1; xmlNode *patchset = get_message_xml(msg, F_CIB_UPDATE_RESULT); CRM_ASSERT(patchset); crm_element_value_int(patchset, "format", &format); switch(format) { case 1: update_cib_stonith_devices_v1(event, msg); break; case 2: update_cib_stonith_devices_v2(event, msg); break; default: crm_warn("Unknown patch format: %d", format); } } /* Needs to hold node name + attribute name + attribute value + 75 */ #define XPATH_MAX 512 /*! * \internal * \brief Check whether a node has a specific attribute name/value * * \param[in] node Name of node to check * \param[in] name Name of an attribute to look for * \param[in] value The value the named attribute needs to be set to in order to be considered a match * * \return TRUE if the locally cached CIB has the specified node attribute */ gboolean node_has_attr(const char *node, const char *name, const char *value) { char xpath[XPATH_MAX]; xmlNode *match; int n; CRM_CHECK(local_cib != NULL, return FALSE); /* Search for the node's attributes in the CIB. While the schema allows * multiple sets of instance attributes, and allows instance attributes to * use id-ref to reference values elsewhere, that is intended for resources, * so we ignore that here. */ n = snprintf(xpath, XPATH_MAX, "//" XML_CIB_TAG_NODES "/" XML_CIB_TAG_NODE "[@uname='%s']/" XML_TAG_ATTR_SETS "/" XML_CIB_TAG_NVPAIR "[@name='%s' and @value='%s']", node, name, value); match = get_xpath_object(xpath, local_cib, LOG_NEVER); CRM_CHECK(n < XPATH_MAX, return FALSE); return (match != NULL); } /*! * \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)); } static void update_fencing_topology(const char *event, xmlNode * msg) { int format = 1; const char *xpath; xmlXPathObjectPtr xpathObj = NULL; xmlNode *patchset = get_message_xml(msg, F_CIB_UPDATE_RESULT); CRM_ASSERT(patchset); crm_element_value_int(patchset, "format", &format); if(format == 1) { /* Process deletions (only) */ xpath = "//" F_CIB_UPDATE_RESULT "//" XML_TAG_DIFF_REMOVED "//" XML_TAG_FENCING_LEVEL; xpathObj = xpath_search(msg, xpath); remove_fencing_topology(xpathObj); freeXpathObject(xpathObj); /* Process additions and changes */ xpath = "//" F_CIB_UPDATE_RESULT "//" XML_TAG_DIFF_ADDED "//" XML_TAG_FENCING_LEVEL; xpathObj = xpath_search(msg, xpath); register_fencing_topology(xpathObj); freeXpathObject(xpathObj); } else if(format == 2) { xmlNode *change = NULL; int add[] = { 0, 0, 0 }; int del[] = { 0, 0, 0 }; xml_patch_versions(patchset, add, del); for (change = pcmk__xml_first_child(patchset); change != NULL; change = pcmk__xml_next(change)) { const char *op = crm_element_value(change, XML_DIFF_OP); const char *xpath = crm_element_value(change, XML_DIFF_PATH); if(op == NULL) { continue; } else if(strstr(xpath, "/" XML_TAG_FENCING_LEVEL) != NULL) { /* Change to a specific entry */ crm_trace("Handling %s operation %d.%d.%d for %s", op, add[0], add[1], add[2], xpath); if(strcmp(op, "move") == 0) { continue; } else if(strcmp(op, "create") == 0) { - handle_topology_change(change->children, FALSE); + add_topology_level(change->children); } else if(strcmp(op, "modify") == 0) { xmlNode *match = first_named_child(change, XML_DIFF_RESULT); if(match) { - handle_topology_change(match->children, TRUE); + remove_topology_level(match->children); + add_topology_level(match->children); } } else if(strcmp(op, "delete") == 0) { /* Nuclear option, all we have is the path and an id... not enough to remove a specific entry */ crm_info("Re-initializing fencing topology after %s operation %d.%d.%d for %s", op, add[0], add[1], add[2], xpath); fencing_topology_init(); return; } } else if (strstr(xpath, "/" XML_TAG_FENCING_TOPOLOGY) != NULL) { /* Change to the topology in general */ crm_info("Re-initializing fencing topology after top-level %s operation %d.%d.%d for %s", op, add[0], add[1], add[2], xpath); fencing_topology_init(); return; } else if (strstr(xpath, "/" XML_CIB_TAG_CONFIGURATION)) { /* Changes to the whole config section, possibly including the topology as a whild */ if(first_named_child(change, XML_TAG_FENCING_TOPOLOGY) == NULL) { crm_trace("Nothing for us in %s operation %d.%d.%d for %s.", op, add[0], add[1], add[2], xpath); } else if(strcmp(op, "delete") == 0 || strcmp(op, "create") == 0) { crm_info("Re-initializing fencing topology after top-level %s operation %d.%d.%d for %s.", op, add[0], add[1], add[2], xpath); fencing_topology_init(); return; } } else { crm_trace("Nothing for us in %s operation %d.%d.%d for %s", op, add[0], add[1], add[2], xpath); } } } else { crm_warn("Unknown patch format: %d", format); } } static bool have_cib_devices = FALSE; static void update_cib_cache_cb(const char *event, xmlNode * msg) { int rc = pcmk_ok; xmlNode *stonith_enabled_xml = NULL; static gboolean stonith_enabled_saved = TRUE; long timeout_ms_saved = stonith_watchdog_timeout_ms; gboolean need_full_refresh = FALSE; bool value = false; if(!have_cib_devices) { crm_trace("Skipping updates until we get a full dump"); return; } else if(msg == NULL) { crm_trace("Missing %s update", event); return; } /* Maintain a local copy of the CIB so that we have full access * to device definitions, location constraints, and node attributes */ if (local_cib != NULL) { int rc = pcmk_ok; xmlNode *patchset = NULL; crm_element_value_int(msg, F_CIB_RC, &rc); if (rc != pcmk_ok) { return; } patchset = get_message_xml(msg, F_CIB_UPDATE_RESULT); xml_log_patchset(LOG_TRACE, "Config update", patchset); rc = xml_apply_patchset(local_cib, patchset, TRUE); switch (rc) { case pcmk_ok: case -pcmk_err_old_data: break; case -pcmk_err_diff_resync: case -pcmk_err_diff_failed: crm_notice("[%s] Patch aborted: %s (%d)", event, pcmk_strerror(rc), rc); free_xml(local_cib); local_cib = NULL; break; default: crm_warn("[%s] ABORTED: %s (%d)", event, pcmk_strerror(rc), rc); free_xml(local_cib); local_cib = NULL; } } if (local_cib == NULL) { crm_trace("Re-requesting full CIB"); rc = cib_api->cmds->query(cib_api, NULL, &local_cib, cib_scope_local | cib_sync_call); if(rc != pcmk_ok) { crm_err("Couldn't retrieve the CIB: %s (%d)", pcmk_strerror(rc), rc); return; } CRM_ASSERT(local_cib != NULL); stonith_enabled_saved = FALSE; /* Trigger a full refresh below */ } pcmk__refresh_node_caches_from_cib(local_cib); update_stonith_watchdog_timeout_ms(local_cib); stonith_enabled_xml = get_xpath_object("//nvpair[@name='stonith-enabled']", local_cib, LOG_NEVER); if (pcmk__xe_get_bool_attr(stonith_enabled_xml, XML_NVPAIR_ATTR_VALUE, &value) == pcmk_rc_ok && !value) { crm_trace("Ignoring CIB updates while fencing is disabled"); stonith_enabled_saved = FALSE; } else if (stonith_enabled_saved == FALSE) { crm_info("Updating fencing device and topology lists " "now that fencing is enabled"); stonith_enabled_saved = TRUE; need_full_refresh = TRUE; } else { if (timeout_ms_saved != stonith_watchdog_timeout_ms) { need_full_refresh = TRUE; } else { update_fencing_topology(event, msg); update_cib_stonith_devices(event, msg); watchdog_device_update(); } } if (need_full_refresh) { fencing_topology_init(); cib_devices_update(); watchdog_device_update(); } } static void init_cib_cache_cb(xmlNode * msg, int call_id, int rc, xmlNode * output, void *user_data) { crm_info("Updating device list from CIB"); have_cib_devices = TRUE; local_cib = copy_xml(output); pcmk__refresh_node_caches_from_cib(local_cib); update_stonith_watchdog_timeout_ms(local_cib); fencing_topology_init(); cib_devices_update(); watchdog_device_update(); } static 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); } else { stonith_cleanup(); crm_exit(CRM_EX_OK); } } static void cib_connection_destroy(gpointer user_data) { if (stonith_shutdown_flag) { crm_info("Connection to the CIB manager closed"); return; } else { crm_crit("Lost connection to the CIB manager, shutting down"); } if (cib_api) { cib_api->cmds->signoff(cib_api); } stonith_shutdown(0); } static void stonith_cleanup(void) { if (cib_api) { cib_api->cmds->del_notify_callback(cib_api, T_CIB_DIFF_NOTIFY, update_cib_cache_cb); cib_api->cmds->signoff(cib_api); } 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; free_xml(local_cib); local_cib = NULL; } static pcmk__cli_option_t long_options[] = { // long option, argument type, storage, short option, description, flags { "stand-alone", no_argument, 0, 's', "\tDeprecated (will be removed in a future release)", pcmk__option_default }, { "stand-alone-w-cpg", no_argument, 0, 'c', "\tIntended for use in regression testing only", pcmk__option_default }, { "logfile", required_argument, 0, 'l', NULL, pcmk__option_default }, { "verbose", no_argument, 0, 'V', NULL, pcmk__option_default }, { "version", no_argument, 0, '$', NULL, pcmk__option_default }, { "help", no_argument, 0, '?', NULL, pcmk__option_default }, { 0, 0, 0, 0 } }; static void setup_cib(void) { int rc, retries = 0; cib_api = cib_new(); if (cib_api == NULL) { crm_err("No connection to the CIB manager"); return; } do { sleep(retries); rc = cib_api->cmds->signon(cib_api, CRM_SYSTEM_STONITHD, cib_command); } while (rc == -ENOTCONN && ++retries < 5); if (rc != pcmk_ok) { crm_err("Could not connect to the CIB manager: %s (%d)", pcmk_strerror(rc), rc); } else if (pcmk_ok != cib_api->cmds->add_notify_callback(cib_api, T_CIB_DIFF_NOTIFY, update_cib_cache_cb)) { crm_err("Could not set CIB notification callback"); } else { rc = cib_api->cmds->query(cib_api, NULL, NULL, cib_scope_local); cib_api->cmds->register_callback(cib_api, rc, 120, FALSE, NULL, "init_cib_cache_cb", init_cib_cache_cb); cib_api->cmds->set_connection_dnotify(cib_api, cib_connection_destroy); crm_info("Watching for fencing topology changes"); } } 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, F_XML_TAGNAME, "stonith_command"); crm_xml_add(query, F_TYPE, 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); } } int main(int argc, char **argv) { int flag; int lpc = 0; int argerr = 0; int option_index = 0; crm_cluster_t *cluster = NULL; const char *actions[] = { "reboot", "off", "on", "list", "monitor", "status" }; crm_ipc_t *old_instance = NULL; int rc = pcmk_rc_ok; crm_log_preinit(NULL, argc, argv); pcmk__set_cli_options(NULL, "[options]", long_options, "daemon for executing fencing devices in a " "Pacemaker cluster"); while (1) { flag = pcmk__next_cli_option(argc, argv, &option_index, NULL); if (flag == -1) { break; } switch (flag) { case 'V': crm_bump_log_level(argc, argv); break; case 'l': { int rc = pcmk__add_logfile(optarg); if (rc != pcmk_rc_ok) { /* Logging has not yet been initialized, so stderr is * the only way to get information out */ fprintf(stderr, "Logging to %s is disabled: %s\n", optarg, pcmk_rc_str(rc)); } } break; case 's': stand_alone = TRUE; break; case 'c': stand_alone = FALSE; no_cib_connect = TRUE; break; case '$': case '?': pcmk__cli_help(flag, CRM_EX_OK); break; default: ++argerr; break; } } if (argc - optind == 1 && pcmk__str_eq("metadata", argv[optind], pcmk__str_casei)) { printf("\n"); printf("\n"); printf(" 1.0\n"); printf(" Instance attributes available for all \"stonith\"-class resources" " and used by Pacemaker's fence daemon, formerly known as stonithd\n"); #ifdef ENABLE_NLS printf(_(" Instance attributes available for all \"stonith\"-class resources" " and used by Pacemaker's fence daemon, formerly known as stonithd\n")); #endif printf(" Instance attributes available for all \"stonith\"-class resources\n"); #ifdef ENABLE_NLS printf(_(" Instance attributes available for all \"stonith\"-class resources\n")); #endif printf(" \n"); #if 0 // priority is not implemented yet printf(" \n"); printf(" Devices that are not in a topology " "are tried in order of highest to lowest integer priority\n"); printf(" \n"); printf(" \n"); #endif printf(" \n", PCMK_STONITH_HOST_ARGUMENT); printf(" 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 " "'%s' can be used to tell the cluster not to supply any " "additional parameters.\n" " \n", PCMK__VALUE_NONE); #ifdef ENABLE_NLS printf(_(" 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 " "'%s' can be used to tell the cluster not to supply any " "additional parameters.\n" " \n"), PCMK__VALUE_NONE); #endif printf (" Advanced use only: An alternate parameter to supply instead of 'port'\n"); #ifdef ENABLE_NLS printf (_(" Advanced use only: An alternate parameter to supply instead of 'port'\n")); #endif printf(" \n"); printf(" \n"); printf(" \n", PCMK_STONITH_HOST_MAP); printf (" Eg. node1:1;node2:2,3 would tell the cluster to use port 1 for node1 and ports 2 and 3 for node2\n"); #ifdef ENABLE_NLS printf (_(" Eg. node1:1;node2:2,3 would tell the cluster to use port 1 for node1 and ports 2 and 3 for node2\n")); #endif printf (" A mapping of host names to ports numbers for devices that do not support host names.\n"); #ifdef ENABLE_NLS printf (_(" A mapping of host names to ports numbers for devices that do not support host names.\n")); #endif printf(" \n"); printf(" \n"); printf(" \n", PCMK_STONITH_HOST_LIST); printf(" Eg. node1,node2,node3\n"); printf(" A list of machines controlled by " "this device (Optional unless %s=static-list).\n", PCMK_STONITH_HOST_CHECK); printf(" \n"); printf(" \n"); printf(" \n", PCMK_STONITH_HOST_CHECK); printf(" Allowed values: dynamic-list " "(query the device via the 'list' command), static-list " "(check the " PCMK_STONITH_HOST_LIST " attribute), status " "(query the device via the 'status' command), " PCMK__VALUE_NONE " (assume every device can fence every " "machine)\n"); printf (" How to determine which machines are controlled by the device.\n"); printf(" \n"); printf(" \n"); printf(" \n", PCMK_STONITH_DELAY_MAX); printf(" This prevents double fencing when " "using slow devices such as sbd.\nUse this to enable a random " "delay for fencing actions.\nThe overall delay is derived from " "this random delay value adding a static delay so that the sum " "is kept below the maximum delay.\n"); printf(" 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_STONITH_DELAY_BASE " and adding a random delay value such " "that the sum is kept below this maximum.\n"); printf(" \n"); printf(" \n"); printf(" \n", PCMK_STONITH_DELAY_BASE); printf(" 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_STONITH_DELAY_MAX " is also used, a random delay will be " "added such that the total delay is kept below that value.\n" "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 per target.\n" " \n"); printf(" Enable a base delay for " "fencing actions and specify base delay value.\n"); printf(" \n"); printf(" \n"); printf(" \n", PCMK_STONITH_ACTION_LIMIT); printf (" Cluster property concurrent-fencing=true needs to be configured first.\n" "Then use this to specify the maximum number of actions can be performed in parallel on this device. -1 is unlimited.\n"); printf (" The maximum number of actions can be performed in parallel on this device\n"); printf(" \n"); printf(" \n"); for (lpc = 0; lpc < PCMK__NELEM(actions); lpc++) { printf(" \n", actions[lpc]); printf (" Some devices do not support the standard commands or may provide additional ones.\n" "Use this to specify an alternate, device-specific, command that implements the '%s' action.\n", actions[lpc]); printf (" Advanced use only: An alternate command to run instead of '%s'\n", actions[lpc]); printf(" \n", actions[lpc]); printf(" \n"); printf(" \n", actions[lpc]); printf (" Some devices need much more/less time to complete than normal.\n" "Use this to specify an alternate, device-specific, timeout for '%s' actions.\n", actions[lpc]); printf (" Advanced use only: Specify an alternate timeout to use for %s actions instead of stonith-timeout\n", actions[lpc]); printf(" \n"); printf(" \n"); printf(" \n", actions[lpc]); printf(" Some devices do not support multiple connections." " Operations may 'fail' if the device is busy with another task so Pacemaker will automatically retry the operation, if there is time remaining." " Use this option to alter the number of times Pacemaker retries '%s' actions before giving up." "\n", actions[lpc]); printf (" Advanced use only: The maximum number of times to retry the '%s' command within the timeout period\n", actions[lpc]); printf(" \n"); printf(" \n"); } printf(" \n"); printf("\n"); return CRM_EX_OK; } if (optind != argc) { ++argerr; } if (argerr) { pcmk__cli_help('?', CRM_EX_USAGE); } crm_log_init(NULL, LOG_INFO, TRUE, FALSE, argc, argv, FALSE); crm_notice("Starting Pacemaker fencer"); old_instance = crm_ipc_new("stonith-ng", 0); if (crm_ipc_connect(old_instance)) { /* IPC end-point already up */ crm_ipc_close(old_instance); crm_ipc_destroy(old_instance); crm_err("pacemaker-fenced is already active, aborting startup"); crm_exit(CRM_EX_OK); } 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(); fenced_data_set = pe_new_working_set(); CRM_ASSERT(fenced_data_set != NULL); cluster = calloc(1, sizeof(crm_cluster_t)); CRM_ASSERT(cluster != NULL); if (stand_alone == FALSE) { if (is_corosync_cluster()) { #if SUPPORT_COROSYNC cluster->destroy = stonith_peer_cs_destroy; cluster->cpg.cpg_deliver_fn = stonith_peer_ais_callback; cluster->cpg.cpg_confchg_fn = pcmk_cpg_membership; #endif } crm_set_status_callback(&st_peer_update_callback); if (crm_cluster_connect(cluster) == FALSE) { crm_crit("Cannot sign in to the cluster... terminating"); crm_exit(CRM_EX_FATAL); } stonith_our_uname = strdup(cluster->uname); if (no_cib_connect == FALSE) { setup_cib(); } } else { stonith_our_uname = strdup("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); pcmk__register_formats(NULL, formats); rc = pcmk__output_new(&out, "log", NULL, argv); if ((rc != pcmk_rc_ok) || (out == NULL)) { crm_err("Can't log resource details due to internal error: %s\n", pcmk_rc_str(rc)); crm_exit(CRM_EX_FATAL); } pe__register_messages(out); pcmk__register_lib_messages(out); pcmk__output_set_log_level(out, LOG_TRACE); fenced_data_set->priv = out; /* 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); stonith_cleanup(); free(cluster->uuid); free(cluster->uname); free(cluster); pe_free_working_set(fenced_data_set); out->finish(out, CRM_EX_OK, true, NULL); pcmk__output_free(out); pcmk__unregister_formats(); crm_exit(CRM_EX_OK); } diff --git a/daemons/fenced/pacemaker-fenced.h b/daemons/fenced/pacemaker-fenced.h index 63abd7c9ce..acd14fd927 100644 --- a/daemons/fenced/pacemaker-fenced.h +++ b/daemons/fenced/pacemaker-fenced.h @@ -1,290 +1,290 @@ /* * Copyright 2009-2022 the Pacemaker project contributors * * This source code is licensed under the GNU General Public License version 2 * or later (GPLv2+) WITHOUT ANY WARRANTY. */ #include // uint32_t, uint64_t #include /*! * \internal * \brief Check whether target has already been fenced recently * * \param[in] tolerance Number of seconds to look back in time * \param[in] target Name of node to search for * \param[in] action Action we want to match * * \return TRUE if an equivalent fencing operation took place in the last * \p tolerance seconds, FALSE otherwise */ gboolean stonith_check_fence_tolerance(int tolerance, const char *target, const char *action); typedef struct stonith_device_s { char *id; char *agent; char *namespace; /*! list of actions that must execute on the target node. Used for unfencing */ char *on_target_actions; GList *targets; time_t targets_age; gboolean has_attr_map; /* should nodeid parameter for victim be included in agent arguments */ gboolean include_nodeid; /* whether the cluster should automatically unfence nodes with the device */ gboolean automatic_unfencing; guint priority; uint32_t flags; // Group of enum st_device_flags GHashTable *params; GHashTable *aliases; GList *pending_ops; mainloop_timer_t *timer; crm_trigger_t *work; xmlNode *agent_metadata; /*! A verified device is one that has contacted the * agent successfully to perform a monitor operation */ gboolean verified; gboolean cib_registered; gboolean api_registered; gboolean dirty; } stonith_device_t; /* These values are used to index certain arrays by "phase". Usually an * operation has only one "phase", so phase is always zero. However, some * reboots are remapped to "off" then "on", in which case "reboot" will be * phase 0, "off" will be phase 1 and "on" will be phase 2. */ enum st_remap_phase { st_phase_requested = 0, st_phase_off = 1, st_phase_on = 2, st_phase_max = 3 }; typedef struct remote_fencing_op_s { /* The unique id associated with this operation */ char *id; /*! The node this operation will fence */ char *target; /*! The fencing action to perform on the target. (reboot, on, off) */ char *action; /*! When was the fencing action recorded (seconds since epoch) */ time_t created; /*! Marks if the final notifications have been sent to local stonith clients. */ gboolean notify_sent; /*! The number of query replies received */ guint replies; /*! The number of query replies expected */ guint replies_expected; /*! Does this node own control of this operation */ gboolean owner; /*! After query is complete, This the high level timer that expires the entire operation */ guint op_timer_total; /*! This timer expires the current fencing request. Many fencing * requests may exist in a single operation */ guint op_timer_one; /*! This timer expires the query request sent out to determine * what nodes are contain what devices, and who those devices can fence */ guint query_timer; /*! This is the default timeout to use for each fencing device if no * custom timeout is received in the query. */ gint base_timeout; /*! This is the calculated total timeout an operation can take before * expiring. This is calculated by adding together all the timeout * values associated with the devices this fencing operation may call */ gint total_timeout; /*! Requested fencing delay. * Value -1 means disable any static/random fencing delays. */ int delay; /*! Delegate is the node being asked to perform a fencing action * on behalf of the node that owns the remote operation. Some operations * will involve multiple delegates. This value represents the final delegate * that is used. */ char *delegate; /*! The point at which the remote operation completed */ time_t completed; //! Group of enum stonith_call_options associated with this operation uint32_t call_options; /*! The current state of the remote operation. This indicates * what stage the op is in, query, exec, done, duplicate, failed. */ enum op_state state; /*! The node that owns the remote operation */ char *originator; /*! The local client id that initiated the fencing request */ char *client_id; /*! The client's call_id that initiated the fencing request */ int client_callid; /*! The name of client that initiated the fencing request */ char *client_name; /*! List of the received query results for all the nodes in the cpg group */ GList *query_results; /*! The original request that initiated the remote stonith operation */ xmlNode *request; /*! The current topology level being executed */ guint level; /*! The current operation phase being executed */ enum st_remap_phase phase; /*! Devices with automatic unfencing (always run if "on" requested, never if remapped) */ GList *automatic_list; /*! List of all devices at the currently executing topology level */ GList *devices_list; /*! Current entry in the topology device list */ GList *devices; /*! List of duplicate operations attached to this operation. Once this operation * completes, the duplicate operations will be closed out as well. */ GList *duplicates; /*! The point at which the remote operation completed(nsec) */ long long completed_nsec; /*! The (potentially intermediate) result of the operation */ pcmk__action_result_t result; } remote_fencing_op_t; void fenced_broadcast_op_result(remote_fencing_op_t *op, bool op_merged); // Fencer-specific client flags enum st_client_flags { st_callback_unknown = UINT64_C(0), st_callback_notify_fence = (UINT64_C(1) << 0), st_callback_device_add = (UINT64_C(1) << 2), st_callback_device_del = (UINT64_C(1) << 4), st_callback_notify_history = (UINT64_C(1) << 5), st_callback_notify_history_synced = (UINT64_C(1) << 6) }; // How the user specified the target of a topology level enum fenced_target_by { fenced_target_by_unknown = -1, // Invalid or not yet parsed fenced_target_by_name, // By target name fenced_target_by_pattern, // By a pattern matching target names fenced_target_by_attribute, // By a node attribute/value on target }; /* * Complex fencing requirements are specified via fencing topologies. * A topology consists of levels; each level is a list of fencing devices. * Topologies are stored in a hash table by node name. When a node needs to be * fenced, if it has an entry in the topology table, the levels are tried * sequentially, and the devices in each level are tried sequentially. * Fencing is considered successful as soon as any level succeeds; * a level is considered successful if all its devices succeed. * Essentially, all devices at a given level are "and-ed" and the * levels are "or-ed". * * This structure is used for the topology table entries. * Topology levels start from 1, so levels[0] is unused and always NULL. */ typedef struct stonith_topology_s { enum fenced_target_by kind; // How target was specified /*! Node name regex or attribute name=value for which topology applies */ char *target; char *target_value; char *target_pattern; char *target_attribute; /*! Names of fencing devices at each topology level */ GList *levels[ST_LEVEL_MAX]; } stonith_topology_t; void init_device_list(void); void free_device_list(void); void init_topology_list(void); void free_topology_list(void); void free_stonith_remote_op_list(void); void init_stonith_remote_op_hash_table(GHashTable **table); void free_metadata_cache(void); void fenced_unregister_handlers(void); uint64_t get_stonith_flag(const char *name); void stonith_command(pcmk__client_t *client, uint32_t id, uint32_t flags, xmlNode *op_request, const char *remote_peer); -int stonith_device_register(xmlNode * msg, const char **desc, gboolean from_cib); +int stonith_device_register(xmlNode *msg, gboolean from_cib); void stonith_device_remove(const char *id, bool from_cib); char *stonith_level_key(xmlNode * msg, int mode); void fenced_register_level(xmlNode *msg, char **desc, pcmk__action_result_t *result); void fenced_unregister_level(xmlNode *msg, char **desc, pcmk__action_result_t *result); stonith_topology_t *find_topology_for_host(const char *host); void do_local_reply(xmlNode *notify_src, pcmk__client_t *client, int call_options); xmlNode *fenced_construct_reply(xmlNode *request, xmlNode *data, pcmk__action_result_t *result); void do_stonith_async_timeout_update(const char *client, const char *call_id, int timeout); void fenced_send_notification(const char *type, const pcmk__action_result_t *result, xmlNode *data); void fenced_send_device_notification(const char *op, const pcmk__action_result_t *result, const char *desc); void fenced_send_level_notification(const char *op, const pcmk__action_result_t *result, const char *desc); remote_fencing_op_t *initiate_remote_stonith_op(pcmk__client_t *client, xmlNode *request, gboolean manual_ack); void fenced_process_fencing_reply(xmlNode *msg); int process_remote_stonith_query(xmlNode * msg); void *create_remote_stonith_op(const char *client, xmlNode * request, gboolean peer); void stonith_fence_history(xmlNode *msg, xmlNode **output, const char *remote_peer, int options); void stonith_fence_history_trim(void); bool fencing_peer_active(crm_node_t *peer); void set_fencing_completed(remote_fencing_op_t * op); int fenced_handle_manual_confirmation(pcmk__client_t *client, xmlNode *msg); gboolean node_has_attr(const char *node, const char *name, const char *value); gboolean node_does_watchdog_fencing(const char *node); static inline void fenced_set_protocol_error(pcmk__action_result_t *result) { pcmk__set_result(result, CRM_EX_PROTOCOL, PCMK_EXEC_INVALID, "Fencer API request missing required information (bug?)"); } extern char *stonith_our_uname; extern gboolean stand_alone; extern GHashTable *device_list; extern GHashTable *topology; extern long stonith_watchdog_timeout_ms; extern GList *stonith_watchdog_targets; extern GHashTable *stonith_remote_op_list; diff --git a/include/crm/stonith-ng.h b/include/crm/stonith-ng.h index 4fe52ef397..aa2c54922b 100644 --- a/include/crm/stonith-ng.h +++ b/include/crm/stonith-ng.h @@ -1,588 +1,588 @@ /* * Copyright 2004-2022 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #ifndef PCMK__CRM_STONITH_NG__H # define PCMK__CRM_STONITH_NG__H #ifdef __cplusplus extern "C" { #endif /** * \file * \brief Fencing aka. STONITH * \ingroup fencing */ /* IMPORTANT: DLM source code includes this file directly, without having access * to other Pacemaker headers on its include path, so this file should *not* * include any other Pacemaker headers. (DLM might be updated to avoid the * issue, but we should still follow this guideline for a long time after.) */ # include # include # include // bool # include // uint32_t # include // time_t # define T_STONITH_NOTIFY_DISCONNECT "st_notify_disconnect" # define T_STONITH_NOTIFY_FENCE "st_notify_fence" # define T_STONITH_NOTIFY_HISTORY "st_notify_history" # define T_STONITH_NOTIFY_HISTORY_SYNCED "st_notify_history_synced" /* *INDENT-OFF* */ enum stonith_state { stonith_connected_command, stonith_connected_query, stonith_disconnected, }; enum stonith_call_options { st_opt_none = 0x00000000, st_opt_verbose = 0x00000001, st_opt_allow_suicide = 0x00000002, st_opt_manual_ack = 0x00000008, st_opt_discard_reply = 0x00000010, /* st_opt_all_replies = 0x00000020, */ st_opt_topology = 0x00000040, st_opt_scope_local = 0x00000100, st_opt_cs_nodeid = 0x00000200, st_opt_sync_call = 0x00001000, /*! Allow the timeout period for a callback to be adjusted * based on the time the server reports the operation will take. */ st_opt_timeout_updates = 0x00002000, /*! Only report back if operation is a success in callback */ st_opt_report_only_success = 0x00004000, /* used where ever apropriate - e.g. cleanup of history */ st_opt_cleanup = 0x000080000, /* used where ever apropriate - e.g. send out a history query to all nodes */ st_opt_broadcast = 0x000100000, }; /*! Order matters here, do not change values */ enum op_state { st_query, st_exec, st_done, st_duplicate, st_failed, }; // Supported fence agent interface standards enum stonith_namespace { st_namespace_invalid, st_namespace_any, st_namespace_internal, // Implemented internally by Pacemaker /* Neither of these projects are active any longer, but the fence agent * interfaces they created are still in use and supported by Pacemaker. */ st_namespace_rhcs, // Red Hat Cluster Suite compatible st_namespace_lha, // Linux-HA compatible }; enum stonith_namespace stonith_text2namespace(const char *namespace_s); const char *stonith_namespace2text(enum stonith_namespace st_namespace); enum stonith_namespace stonith_get_namespace(const char *agent, const char *namespace_s); typedef struct stonith_key_value_s { char *key; char *value; struct stonith_key_value_s *next; } stonith_key_value_t; typedef struct stonith_history_s { char *target; char *action; char *origin; char *delegate; char *client; int state; time_t completed; struct stonith_history_s *next; long completed_nsec; char *exit_reason; } stonith_history_t; typedef struct stonith_s stonith_t; typedef struct stonith_event_s { char *id; - char *type; - char *message; + char *type; //!< \deprecated Will be removed in future release + char *message; //!< \deprecated Will be removed in future release char *operation; int result; char *origin; char *target; char *action; char *executioner; char *device; /*! The name of the client that initiated the action. */ char *client_origin; //! \internal This field should be treated as internal to Pacemaker void *opaque; } stonith_event_t; typedef struct stonith_callback_data_s { int rc; int call_id; void *userdata; //! \internal This field should be treated as internal to Pacemaker void *opaque; } stonith_callback_data_t; typedef struct stonith_api_operations_s { /*! * \brief Destroy the stonith api structure. */ int (*free) (stonith_t *st); /*! * \brief Connect to the local stonith daemon. * * \return Legacy Pacemaker return code */ int (*connect) (stonith_t *st, const char *name, int *stonith_fd); /*! * \brief Disconnect from the local stonith daemon. * * \return Legacy Pacemaker return code */ int (*disconnect)(stonith_t *st); /*! * \brief Unregister a fence device with the local fencer * * \return pcmk_ok (if synchronous) or positive call ID (if asynchronous) * on success, otherwise a negative legacy Pacemaker return code */ int (*remove_device)( stonith_t *st, int options, const char *name); /*! * \brief Register a fence device with the local fencer * * \return pcmk_ok (if synchronous) or positive call ID (if asynchronous) * on success, otherwise a negative legacy Pacemaker return code */ int (*register_device)( stonith_t *st, int options, const char *id, const char *provider, const char *agent, stonith_key_value_t *params); /*! * \brief Unregister a fencing level for specified node with local fencer * * \return pcmk_ok (if synchronous) or positive call ID (if asynchronous) * on success, otherwise a negative legacy Pacemaker return code */ int (*remove_level)( stonith_t *st, int options, const char *node, int level); /*! * \brief Register a fencing level for specified node with local fencer * * \return pcmk_ok (if synchronous) or positive call ID (if asynchronous) * on success, otherwise a negative legacy Pacemaker return code */ int (*register_level)( stonith_t *st, int options, const char *node, int level, stonith_key_value_t *device_list); /*! * \brief Get the metadata documentation for a resource. * * \note Value is returned in output. Output must be freed when set. * * \return Legacy Pacemaker return code */ int (*metadata)(stonith_t *st, int options, const char *device, const char *provider, char **output, int timeout); /*! * \brief Retrieve a list of installed stonith agents * * \note if provider is not provided, all known agents will be returned * \note list must be freed using stonith_key_value_freeall() * \note call_options parameter is not used, it is reserved for future use. * * \return Number of items in list on success, or negative errno otherwise */ int (*list_agents)(stonith_t *stonith, int call_options, const char *provider, stonith_key_value_t **devices, int timeout); /*! * \brief Retrieve string listing hosts and port assignments from a local stonith device. * * \return pcmk_ok (if synchronous) or positive call ID (if asynchronous) * on success, otherwise a negative legacy Pacemaker return code */ int (*list)(stonith_t *st, int options, const char *id, char **list_output, int timeout); /*! * \brief Check to see if a local stonith device is reachable * * \return pcmk_ok (if synchronous) or positive call ID (if asynchronous) * on success, otherwise a negative legacy Pacemaker return code */ int (*monitor)(stonith_t *st, int options, const char *id, int timeout); /*! * \brief Check to see if a local stonith device's port is reachable * * \return pcmk_ok (if synchronous) or positive call ID (if asynchronous) * on success, otherwise a negative legacy Pacemaker return code */ int (*status)(stonith_t *st, int options, const char *id, const char *port, int timeout); /*! * \brief Retrieve a list of registered stonith devices. * * \note If node is provided, only devices that can fence the node id * will be returned. * * \return Number of items in list on success, or negative errno otherwise */ int (*query)(stonith_t *st, int options, const char *node, stonith_key_value_t **devices, int timeout); /*! * \brief Issue a fencing action against a node. * * \note Possible actions are, 'on', 'off', and 'reboot'. * * \param st, stonith connection * \param options, call options * \param node, The target node to fence * \param action, The fencing action to take * \param timeout, The default per device timeout to use with each device * capable of fencing the target. * * \return pcmk_ok (if synchronous) or positive call ID (if asynchronous) * on success, otherwise a negative legacy Pacemaker return code */ int (*fence)(stonith_t *st, int options, const char *node, const char *action, int timeout, int tolerance); /*! * \brief Manually confirm that a node is down. * * \return pcmk_ok (if synchronous) or positive call ID (if asynchronous) * on success, otherwise a negative legacy Pacemaker return code */ int (*confirm)(stonith_t *st, int options, const char *node); /*! * \brief Retrieve a list of fencing operations that have occurred for a specific node. * * \return Legacy Pacemaker return code */ int (*history)(stonith_t *st, int options, const char *node, stonith_history_t **output, int timeout); int (*register_notification)( stonith_t *st, const char *event, void (*notify)(stonith_t *st, stonith_event_t *e)); /*! * \brief Remove a previously registered notification for \c event, or all * notifications if NULL. * * \param[in] st Fencer connection to use * \param[in] event The event to remove notifications for (may be NULL). * * \return Legacy Pacemaker return code */ int (*remove_notification)(stonith_t *st, const char *event); /*! * \brief Register a callback to receive the result of an asynchronous call * * \param[in] call_id The call ID to register callback for * \param[in] timeout Default time to wait until callback expires * \param[in] options Bitmask of \c stonith_call_options (respects * \c st_opt_timeout_updates and * \c st_opt_report_only_success) * \param[in] userdata Pointer that will be given to callback * \param[in] callback_name Unique name to identify callback * \param[in] callback The callback function to register * * \return \c TRUE on success, \c FALSE if call_id is negative, -errno otherwise */ int (*register_callback)(stonith_t *st, int call_id, int timeout, int options, void *userdata, const char *callback_name, void (*callback)(stonith_t *st, stonith_callback_data_t *data)); /*! * \brief Remove a registered callback for a given call id * * \return pcmk_ok */ int (*remove_callback)(stonith_t *st, int call_id, bool all_callbacks); /*! * \brief Unregister fencing level for specified node, pattern or attribute * * \param[in] st Fencer connection to use * \param[in] options Bitmask of stonith_call_options to pass to the fencer * \param[in] node If not NULL, target level by this node name * \param[in] pattern If not NULL, target by node name using this regex * \param[in] attr If not NULL, target by this node attribute * \param[in] value If not NULL, target by this node attribute value * \param[in] level Index number of level to remove * * \return pcmk_ok (if synchronous) or positive call ID (if asynchronous) * on success, otherwise a negative legacy Pacemaker return code * * \note The caller should set only one of node, pattern or attr/value. */ int (*remove_level_full)(stonith_t *st, int options, const char *node, const char *pattern, const char *attr, const char *value, int level); /*! * \brief Register fencing level for specified node, pattern or attribute * * \param[in] st Fencer connection to use * \param[in] options Bitmask of stonith_call_options to pass to fencer * \param[in] node If not NULL, target level by this node name * \param[in] pattern If not NULL, target by node name using this regex * \param[in] attr If not NULL, target by this node attribute * \param[in] value If not NULL, target by this node attribute value * \param[in] level Index number of level to add * \param[in] device_list Devices to use in level * * \return pcmk_ok (if synchronous) or positive call ID (if asynchronous) * on success, otherwise a negative legacy Pacemaker return code * * \note The caller should set only one of node, pattern or attr/value. */ int (*register_level_full)(stonith_t *st, int options, const char *node, const char *pattern, const char *attr, const char *value, int level, stonith_key_value_t *device_list); /*! * \brief Validate an arbitrary stonith device configuration * * \param[in] st Stonithd connection to use * \param[in] call_options Bitmask of stonith_call_options to use with fencer * \param[in] rsc_id ID used to replace CIB secrets in params * \param[in] namespace_s Namespace of fence agent to validate (optional) * \param[in] agent Fence agent to validate * \param[in] params Configuration parameters to pass to fence agent * \param[in] timeout Fail if no response within this many seconds * \param[out] output If non-NULL, where to store any agent output * \param[out] error_output If non-NULL, where to store agent error output * * \return pcmk_ok if validation succeeds, -errno otherwise * * \note If pcmk_ok is returned, the caller is responsible for freeing * the output (if requested). */ int (*validate)(stonith_t *st, int call_options, const char *rsc_id, const char *namespace_s, const char *agent, stonith_key_value_t *params, int timeout, char **output, char **error_output); /*! * \brief Issue a fencing action against a node with requested fencing delay. * * \note Possible actions are, 'on', 'off', and 'reboot'. * * \param st, stonith connection * \param options, call options * \param node, The target node to fence * \param action, The fencing action to take * \param timeout, The default per device timeout to use with each device * capable of fencing the target. * \param delay, Apply a fencing delay. Value -1 means disable also any * static/random fencing delays from pcmk_delay_base/max * * \return pcmk_ok (if synchronous) or positive call ID (if asynchronous) * on success, otherwise a negative legacy Pacemaker return code */ int (*fence_with_delay)(stonith_t *st, int options, const char *node, const char *action, int timeout, int tolerance, int delay); } stonith_api_operations_t; struct stonith_s { enum stonith_state state; int call_id; int call_timeout; //!< \deprecated Unused void *st_private; stonith_api_operations_t *cmds; }; /* *INDENT-ON* */ /* Core functions */ stonith_t *stonith_api_new(void); void stonith_api_delete(stonith_t * st); void stonith_dump_pending_callbacks(stonith_t * st); bool stonith_dispatch(stonith_t * st); stonith_key_value_t *stonith_key_value_add(stonith_key_value_t * kvp, const char *key, const char *value); void stonith_key_value_freeall(stonith_key_value_t * kvp, int keys, int values); void stonith_history_free(stonith_history_t *history); // Convenience functions int stonith_api_connect_retry(stonith_t *st, const char *name, int max_attempts); const char *stonith_op_state_str(enum op_state state); /* Basic helpers that allows nodes to be fenced and the history to be * queried without mainloop or the caller understanding the full API * * At least one of nodeid and uname are required */ int stonith_api_kick(uint32_t nodeid, const char *uname, int timeout, bool off); time_t stonith_api_time(uint32_t nodeid, const char *uname, bool in_progress); /* * Helpers for using the above functions without install-time dependencies * * Usage: * #include * * To turn a node off by corosync nodeid: * stonith_api_kick_helper(nodeid, 120, 1); * * To check the last fence date/time (also by nodeid): * last = stonith_api_time_helper(nodeid, 0); * * To check if fencing is in progress: * if(stonith_api_time_helper(nodeid, 1) > 0) { ... } * * eg. #include #include #include int main(int argc, char ** argv) { int rc = 0; int nodeid = 102; rc = stonith_api_time_helper(nodeid, 0); printf("%d last fenced at %s\n", nodeid, ctime(rc)); rc = stonith_api_kick_helper(nodeid, 120, 1); printf("%d fence result: %d\n", nodeid, rc); rc = stonith_api_time_helper(nodeid, 0); printf("%d last fenced at %s\n", nodeid, ctime(rc)); return 0; } */ # define STONITH_LIBRARY "libstonithd.so.26" typedef int (*st_api_kick_fn) (int nodeid, const char *uname, int timeout, bool off); typedef time_t (*st_api_time_fn) (int nodeid, const char *uname, bool in_progress); static inline int stonith_api_kick_helper(uint32_t nodeid, int timeout, bool off) { static void *st_library = NULL; static st_api_kick_fn st_kick_fn; if (st_library == NULL) { st_library = dlopen(STONITH_LIBRARY, RTLD_LAZY); } if (st_library && st_kick_fn == NULL) { st_kick_fn = (st_api_kick_fn) dlsym(st_library, "stonith_api_kick"); } if (st_kick_fn == NULL) { #ifdef ELIBACC return -ELIBACC; #else return -ENOSYS; #endif } return (*st_kick_fn) (nodeid, NULL, timeout, off); } static inline time_t stonith_api_time_helper(uint32_t nodeid, bool in_progress) { static void *st_library = NULL; static st_api_time_fn st_time_fn; if (st_library == NULL) { st_library = dlopen(STONITH_LIBRARY, RTLD_LAZY); } if (st_library && st_time_fn == NULL) { st_time_fn = (st_api_time_fn) dlsym(st_library, "stonith_api_time"); } if (st_time_fn == NULL) { return 0; } return (*st_time_fn) (nodeid, NULL, in_progress); } /** * Does the given agent describe a stonith resource that can exist? * * \param[in] agent What is the name of the agent? * \param[in] timeout Timeout to use when querying. If 0 is given, * use a default of 120. * * \return A boolean */ bool stonith_agent_exists(const char *agent, int timeout); /*! * \brief Turn stonith action into a more readable string. * * \param action Stonith action */ const char *stonith_action_str(const char *action); #if !defined(PCMK_ALLOW_DEPRECATED) || (PCMK_ALLOW_DEPRECATED == 1) /* Normally we'd put this section in a separate file (crm/fencing/compat.h), but * we can't do that for the reason noted at the top of this file. That does mean * we have to duplicate these declarations where they're implemented. */ //! \deprecated Use stonith_get_namespace() instead const char *get_stonith_provider(const char *agent, const char *provider); #endif #ifdef __cplusplus } #endif #endif diff --git a/lib/fencing/st_client.c b/lib/fencing/st_client.c index 0620b8a218..aea59944f4 100644 --- a/lib/fencing/st_client.c +++ b/lib/fencing/st_client.c @@ -1,2545 +1,2599 @@ /* * Copyright 2004-2022 the Pacemaker project contributors * * The version control history for this file may have further details. * * This source code is licensed under the GNU Lesser General Public License * version 2.1 or later (LGPLv2.1+) WITHOUT ANY WARRANTY. */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include "fencing_private.h" CRM_TRACE_INIT_DATA(stonith); // Used as stonith_t:st_private typedef struct stonith_private_s { char *token; crm_ipc_t *ipc; mainloop_io_t *source; GHashTable *stonith_op_callback_table; GList *notify_list; int notify_refcnt; bool notify_deletes; void (*op_callback) (stonith_t * st, stonith_callback_data_t * data); } stonith_private_t; // Used as stonith_event_t:opaque struct event_private { pcmk__action_result_t result; }; typedef struct stonith_notify_client_s { const char *event; const char *obj_id; /* implement one day */ const char *obj_type; /* implement one day */ void (*notify) (stonith_t * st, stonith_event_t * e); bool delete; } stonith_notify_client_t; typedef struct stonith_callback_client_s { void (*callback) (stonith_t * st, stonith_callback_data_t * data); const char *id; void *user_data; gboolean only_success; gboolean allow_timeout_updates; struct timer_rec_s *timer; } stonith_callback_client_t; struct notify_blob_s { stonith_t *stonith; xmlNode *xml; }; struct timer_rec_s { int call_id; int timeout; guint ref; stonith_t *stonith; }; typedef int (*stonith_op_t) (const char *, int, const char *, xmlNode *, xmlNode *, xmlNode *, xmlNode **, xmlNode **); bool stonith_dispatch(stonith_t * st); xmlNode *stonith_create_op(int call_id, const char *token, const char *op, xmlNode * data, int call_options); static int stonith_send_command(stonith_t *stonith, const char *op, xmlNode *data, xmlNode **output_data, int call_options, int timeout); static void stonith_connection_destroy(gpointer user_data); static void stonith_send_notification(gpointer data, gpointer user_data); static int stonith_api_del_notification(stonith_t *stonith, const char *event); /*! * \brief Get agent namespace by name * * \param[in] namespace_s Name of namespace as string * * \return Namespace as enum value */ enum stonith_namespace stonith_text2namespace(const char *namespace_s) { if (pcmk__str_eq(namespace_s, "any", pcmk__str_null_matches)) { return st_namespace_any; } else if (!strcmp(namespace_s, "redhat") || !strcmp(namespace_s, "stonith-ng")) { return st_namespace_rhcs; } else if (!strcmp(namespace_s, "internal")) { return st_namespace_internal; } else if (!strcmp(namespace_s, "heartbeat")) { return st_namespace_lha; } return st_namespace_invalid; } /*! * \brief Get agent namespace name * * \param[in] namespace Namespace as enum value * * \return Namespace name as string */ const char * stonith_namespace2text(enum stonith_namespace st_namespace) { switch (st_namespace) { case st_namespace_any: return "any"; case st_namespace_rhcs: return "stonith-ng"; case st_namespace_internal: return "internal"; case st_namespace_lha: return "heartbeat"; default: break; } return "unsupported"; } /*! * \brief Determine namespace of a fence agent * * \param[in] agent Fence agent type * \param[in] namespace_s Name of agent namespace as string, if known * * \return Namespace of specified agent, as enum value */ enum stonith_namespace stonith_get_namespace(const char *agent, const char *namespace_s) { - if (pcmk__str_eq(namespace_s, "internal", pcmk__str_casei)) { + if (pcmk__str_eq(namespace_s, "internal", pcmk__str_none)) { return st_namespace_internal; } if (stonith__agent_is_rhcs(agent)) { return st_namespace_rhcs; } #if HAVE_STONITH_STONITH_H if (stonith__agent_is_lha(agent)) { return st_namespace_lha; } #endif crm_err("Unknown fence agent: %s", agent); return st_namespace_invalid; } gboolean stonith__watchdog_fencing_enabled_for_node_api(stonith_t *st, const char *node) { gboolean rv = FALSE; stonith_t *stonith_api = st?st:stonith_api_new(); char *list = NULL; if(stonith_api) { if (stonith_api->state == stonith_disconnected) { int rc = stonith_api->cmds->connect(stonith_api, "stonith-api", NULL); if (rc != pcmk_ok) { crm_err("Failed connecting to Stonith-API for watchdog-fencing-query."); } } if (stonith_api->state != stonith_disconnected) { /* caveat!!! * this might fail when when stonithd is just updating the device-list * probably something we should fix as well for other api-calls */ int rc = stonith_api->cmds->list(stonith_api, st_opt_sync_call, STONITH_WATCHDOG_ID, &list, 0); if ((rc != pcmk_ok) || (list == NULL)) { /* due to the race described above it can happen that * we drop in here - so as not to make remote nodes * panic on that answer */ if (rc == -ENODEV) { crm_notice("Cluster does not have watchdog fencing device"); } else { crm_warn("Could not check for watchdog fencing device: %s", pcmk_strerror(rc)); } } else if (list[0] == '\0') { rv = TRUE; } else { GList *targets = stonith__parse_targets(list); rv = pcmk__str_in_list(node, targets, pcmk__str_casei); g_list_free_full(targets, free); } free(list); if (!st) { /* if we're provided the api we still might have done the * connection - but let's assume the caller won't bother */ stonith_api->cmds->disconnect(stonith_api); } } if (!st) { stonith_api_delete(stonith_api); } } else { crm_err("Stonith-API for watchdog-fencing-query couldn't be created."); } crm_trace("Pacemaker assumes node %s %sto do watchdog-fencing.", node, rv?"":"not "); return rv; } gboolean stonith__watchdog_fencing_enabled_for_node(const char *node) { return stonith__watchdog_fencing_enabled_for_node_api(NULL, node); } /* when cycling through the list we don't want to delete items so just mark them and when we know nobody is using the list loop over it to remove the marked items */ static void foreach_notify_entry (stonith_private_t *private, GFunc func, gpointer user_data) { private->notify_refcnt++; g_list_foreach(private->notify_list, func, user_data); private->notify_refcnt--; if ((private->notify_refcnt == 0) && private->notify_deletes) { GList *list_item = private->notify_list; private->notify_deletes = FALSE; while (list_item != NULL) { stonith_notify_client_t *list_client = list_item->data; GList *next = g_list_next(list_item); if (list_client->delete) { free(list_client); private->notify_list = g_list_delete_link(private->notify_list, list_item); } list_item = next; } } } static void stonith_connection_destroy(gpointer user_data) { stonith_t *stonith = user_data; stonith_private_t *native = NULL; struct notify_blob_s blob; crm_trace("Sending destroyed notification"); blob.stonith = stonith; blob.xml = create_xml_node(NULL, "notify"); native = stonith->st_private; native->ipc = NULL; native->source = NULL; free(native->token); native->token = NULL; stonith->state = stonith_disconnected; crm_xml_add(blob.xml, F_TYPE, T_STONITH_NOTIFY); crm_xml_add(blob.xml, F_SUBTYPE, T_STONITH_NOTIFY_DISCONNECT); foreach_notify_entry(native, stonith_send_notification, &blob); free_xml(blob.xml); } xmlNode * create_device_registration_xml(const char *id, enum stonith_namespace namespace, const char *agent, stonith_key_value_t *params, const char *rsc_provides) { xmlNode *data = create_xml_node(NULL, F_STONITH_DEVICE); xmlNode *args = create_xml_node(data, XML_TAG_ATTRS); #if HAVE_STONITH_STONITH_H if (namespace == st_namespace_any) { namespace = stonith_get_namespace(agent, NULL); } if (namespace == st_namespace_lha) { hash2field((gpointer) "plugin", (gpointer) agent, args); agent = "fence_legacy"; } #endif crm_xml_add(data, XML_ATTR_ID, id); crm_xml_add(data, F_STONITH_ORIGIN, __func__); crm_xml_add(data, "agent", agent); if ((namespace != st_namespace_any) && (namespace != st_namespace_invalid)) { crm_xml_add(data, "namespace", stonith_namespace2text(namespace)); } if (rsc_provides) { crm_xml_add(data, "rsc_provides", rsc_provides); } for (; params; params = params->next) { hash2field((gpointer) params->key, (gpointer) params->value, args); } return data; } static int stonith_api_register_device(stonith_t * st, int call_options, const char *id, const char *namespace, const char *agent, stonith_key_value_t * params) { int rc = 0; xmlNode *data = NULL; data = create_device_registration_xml(id, stonith_text2namespace(namespace), agent, params, NULL); rc = stonith_send_command(st, STONITH_OP_DEVICE_ADD, data, NULL, call_options, 0); free_xml(data); return rc; } static int stonith_api_remove_device(stonith_t * st, int call_options, const char *name) { int rc = 0; xmlNode *data = NULL; data = create_xml_node(NULL, F_STONITH_DEVICE); crm_xml_add(data, F_STONITH_ORIGIN, __func__); crm_xml_add(data, XML_ATTR_ID, name); rc = stonith_send_command(st, STONITH_OP_DEVICE_DEL, data, NULL, call_options, 0); free_xml(data); return rc; } static int stonith_api_remove_level_full(stonith_t *st, int options, const char *node, const char *pattern, const char *attr, const char *value, int level) { int rc = 0; xmlNode *data = NULL; CRM_CHECK(node || pattern || (attr && value), return -EINVAL); data = create_xml_node(NULL, XML_TAG_FENCING_LEVEL); crm_xml_add(data, F_STONITH_ORIGIN, __func__); if (node) { crm_xml_add(data, XML_ATTR_STONITH_TARGET, node); } else if (pattern) { crm_xml_add(data, XML_ATTR_STONITH_TARGET_PATTERN, pattern); } else { crm_xml_add(data, XML_ATTR_STONITH_TARGET_ATTRIBUTE, attr); crm_xml_add(data, XML_ATTR_STONITH_TARGET_VALUE, value); } crm_xml_add_int(data, XML_ATTR_STONITH_INDEX, level); rc = stonith_send_command(st, STONITH_OP_LEVEL_DEL, data, NULL, options, 0); free_xml(data); return rc; } static int stonith_api_remove_level(stonith_t * st, int options, const char *node, int level) { return stonith_api_remove_level_full(st, options, node, NULL, NULL, NULL, level); } /*! * \internal * \brief Create XML for fence topology level registration request * * \param[in] node If not NULL, target level by this node name * \param[in] pattern If not NULL, target by node name using this regex * \param[in] attr If not NULL, target by this node attribute * \param[in] value If not NULL, target by this node attribute value * \param[in] level Index number of level to register * \param[in] device_list List of devices in level * * \return Newly allocated XML tree on success, NULL otherwise * * \note The caller should set only one of node, pattern or attr/value. */ xmlNode * create_level_registration_xml(const char *node, const char *pattern, const char *attr, const char *value, int level, stonith_key_value_t *device_list) { size_t len = 0; char *list = NULL; xmlNode *data; CRM_CHECK(node || pattern || (attr && value), return NULL); data = create_xml_node(NULL, XML_TAG_FENCING_LEVEL); CRM_CHECK(data, return NULL); crm_xml_add(data, F_STONITH_ORIGIN, __func__); crm_xml_add_int(data, XML_ATTR_ID, level); crm_xml_add_int(data, XML_ATTR_STONITH_INDEX, level); if (node) { crm_xml_add(data, XML_ATTR_STONITH_TARGET, node); } else if (pattern) { crm_xml_add(data, XML_ATTR_STONITH_TARGET_PATTERN, pattern); } else { crm_xml_add(data, XML_ATTR_STONITH_TARGET_ATTRIBUTE, attr); crm_xml_add(data, XML_ATTR_STONITH_TARGET_VALUE, value); } // cppcheck seems not to understand the abort logic behind pcmk__realloc // cppcheck-suppress memleak for (; device_list; device_list = device_list->next) { pcmk__add_separated_word(&list, &len, device_list->value, ","); } crm_xml_add(data, XML_ATTR_STONITH_DEVICES, list); free(list); return data; } static int stonith_api_register_level_full(stonith_t * st, int options, const char *node, const char *pattern, const char *attr, const char *value, int level, stonith_key_value_t *device_list) { int rc = 0; xmlNode *data = create_level_registration_xml(node, pattern, attr, value, level, device_list); CRM_CHECK(data != NULL, return -EINVAL); rc = stonith_send_command(st, STONITH_OP_LEVEL_ADD, data, NULL, options, 0); free_xml(data); return rc; } static int stonith_api_register_level(stonith_t * st, int options, const char *node, int level, stonith_key_value_t * device_list) { return stonith_api_register_level_full(st, options, node, NULL, NULL, NULL, level, device_list); } static int stonith_api_device_list(stonith_t * stonith, int call_options, const char *namespace, stonith_key_value_t ** devices, int timeout) { int count = 0; enum stonith_namespace ns = stonith_text2namespace(namespace); if (devices == NULL) { crm_err("Parameter error: stonith_api_device_list"); return -EFAULT; } #if HAVE_STONITH_STONITH_H // Include Linux-HA agents if requested if ((ns == st_namespace_any) || (ns == st_namespace_lha)) { count += stonith__list_lha_agents(devices); } #endif // Include Red Hat agents if requested if ((ns == st_namespace_any) || (ns == st_namespace_rhcs)) { count += stonith__list_rhcs_agents(devices); } return count; } static int stonith_api_device_metadata(stonith_t * stonith, int call_options, const char *agent, const char *namespace, char **output, int timeout) { /* By executing meta-data directly, we can get it from stonith_admin when * the cluster is not running, which is important for higher-level tools. */ enum stonith_namespace ns = stonith_get_namespace(agent, namespace); crm_trace("Looking up metadata for %s agent %s", stonith_namespace2text(ns), agent); switch (ns) { case st_namespace_rhcs: return stonith__rhcs_metadata(agent, timeout, output); #if HAVE_STONITH_STONITH_H case st_namespace_lha: return stonith__lha_metadata(agent, timeout, output); #endif default: crm_err("Can't get fence agent '%s' meta-data: No such agent", agent); break; } return -ENODEV; } static int stonith_api_query(stonith_t * stonith, int call_options, const char *target, stonith_key_value_t ** devices, int timeout) { int rc = 0, lpc = 0, max = 0; xmlNode *data = NULL; xmlNode *output = NULL; xmlXPathObjectPtr xpathObj = NULL; CRM_CHECK(devices != NULL, return -EINVAL); data = create_xml_node(NULL, F_STONITH_DEVICE); crm_xml_add(data, F_STONITH_ORIGIN, __func__); crm_xml_add(data, F_STONITH_TARGET, target); crm_xml_add(data, F_STONITH_ACTION, "off"); rc = stonith_send_command(stonith, STONITH_OP_QUERY, data, &output, call_options, timeout); if (rc < 0) { return rc; } xpathObj = xpath_search(output, "//@agent"); if (xpathObj) { max = numXpathResults(xpathObj); for (lpc = 0; lpc < max; lpc++) { xmlNode *match = getXpathResult(xpathObj, lpc); CRM_LOG_ASSERT(match != NULL); if(match != NULL) { xmlChar *match_path = xmlGetNodePath(match); crm_info("%s[%d] = %s", "//@agent", lpc, match_path); free(match_path); *devices = stonith_key_value_add(*devices, NULL, crm_element_value(match, XML_ATTR_ID)); } } freeXpathObject(xpathObj); } free_xml(output); free_xml(data); return max; } static int stonith_api_call(stonith_t * stonith, int call_options, const char *id, const char *action, const char *victim, int timeout, xmlNode ** output) { int rc = 0; xmlNode *data = NULL; data = create_xml_node(NULL, F_STONITH_DEVICE); crm_xml_add(data, F_STONITH_ORIGIN, __func__); crm_xml_add(data, F_STONITH_DEVICE, id); crm_xml_add(data, F_STONITH_ACTION, action); crm_xml_add(data, F_STONITH_TARGET, victim); rc = stonith_send_command(stonith, STONITH_OP_EXEC, data, output, call_options, timeout); free_xml(data); return rc; } static int stonith_api_list(stonith_t * stonith, int call_options, const char *id, char **list_info, int timeout) { int rc; xmlNode *output = NULL; rc = stonith_api_call(stonith, call_options, id, "list", NULL, timeout, &output); if (output && list_info) { const char *list_str; list_str = crm_element_value(output, F_STONITH_OUTPUT); if (list_str) { *list_info = strdup(list_str); } } if (output) { free_xml(output); } return rc; } static int stonith_api_monitor(stonith_t * stonith, int call_options, const char *id, int timeout) { return stonith_api_call(stonith, call_options, id, "monitor", NULL, timeout, NULL); } static int stonith_api_status(stonith_t * stonith, int call_options, const char *id, const char *port, int timeout) { return stonith_api_call(stonith, call_options, id, "status", port, timeout, NULL); } static int stonith_api_fence_with_delay(stonith_t * stonith, int call_options, const char *node, const char *action, int timeout, int tolerance, int delay) { int rc = 0; xmlNode *data = NULL; data = create_xml_node(NULL, __func__); crm_xml_add(data, F_STONITH_TARGET, node); crm_xml_add(data, F_STONITH_ACTION, action); crm_xml_add_int(data, F_STONITH_TIMEOUT, timeout); crm_xml_add_int(data, F_STONITH_TOLERANCE, tolerance); crm_xml_add_int(data, F_STONITH_DELAY, delay); rc = stonith_send_command(stonith, STONITH_OP_FENCE, data, NULL, call_options, timeout); free_xml(data); return rc; } static int stonith_api_fence(stonith_t * stonith, int call_options, const char *node, const char *action, int timeout, int tolerance) { return stonith_api_fence_with_delay(stonith, call_options, node, action, timeout, tolerance, 0); } static int stonith_api_confirm(stonith_t * stonith, int call_options, const char *target) { stonith__set_call_options(call_options, target, st_opt_manual_ack); return stonith_api_fence(stonith, call_options, target, "off", 0, 0); } static int stonith_api_history(stonith_t * stonith, int call_options, const char *node, stonith_history_t ** history, int timeout) { int rc = 0; xmlNode *data = NULL; xmlNode *output = NULL; stonith_history_t *last = NULL; *history = NULL; if (node) { data = create_xml_node(NULL, __func__); crm_xml_add(data, F_STONITH_TARGET, node); } stonith__set_call_options(call_options, node, st_opt_sync_call); rc = stonith_send_command(stonith, STONITH_OP_FENCE_HISTORY, data, &output, call_options, timeout); free_xml(data); if (rc == 0) { xmlNode *op = NULL; xmlNode *reply = get_xpath_object("//" F_STONITH_HISTORY_LIST, output, LOG_NEVER); for (op = pcmk__xml_first_child(reply); op != NULL; op = pcmk__xml_next(op)) { stonith_history_t *kvp; long long completed; long long completed_nsec = 0L; pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; kvp = calloc(1, sizeof(stonith_history_t)); kvp->target = crm_element_value_copy(op, F_STONITH_TARGET); kvp->action = crm_element_value_copy(op, F_STONITH_ACTION); kvp->origin = crm_element_value_copy(op, F_STONITH_ORIGIN); kvp->delegate = crm_element_value_copy(op, F_STONITH_DELEGATE); kvp->client = crm_element_value_copy(op, F_STONITH_CLIENTNAME); crm_element_value_ll(op, F_STONITH_DATE, &completed); kvp->completed = (time_t) completed; crm_element_value_ll(op, F_STONITH_DATE_NSEC, &completed_nsec); kvp->completed_nsec = completed_nsec; crm_element_value_int(op, F_STONITH_STATE, &kvp->state); stonith__xe_get_result(op, &result); kvp->exit_reason = result.exit_reason; result.exit_reason = NULL; pcmk__reset_result(&result); if (last) { last->next = kvp; } else { *history = kvp; } last = kvp; } } free_xml(output); return rc; } void stonith_history_free(stonith_history_t *history) { stonith_history_t *hp, *hp_old; for (hp = history; hp; hp_old = hp, hp = hp->next, free(hp_old)) { free(hp->target); free(hp->action); free(hp->origin); free(hp->delegate); free(hp->client); free(hp->exit_reason); } } static gint stonithlib_GCompareFunc(gconstpointer a, gconstpointer b) { int rc = 0; const stonith_notify_client_t *a_client = a; const stonith_notify_client_t *b_client = b; if (a_client->delete || b_client->delete) { /* make entries marked for deletion not findable */ return -1; } CRM_CHECK(a_client->event != NULL && b_client->event != NULL, return 0); rc = strcmp(a_client->event, b_client->event); if (rc == 0) { if (a_client->notify == NULL || b_client->notify == NULL) { return 0; } else if (a_client->notify == b_client->notify) { return 0; } else if (((long)a_client->notify) < ((long)b_client->notify)) { crm_err("callbacks for %s are not equal: %p vs. %p", a_client->event, a_client->notify, b_client->notify); return -1; } crm_err("callbacks for %s are not equal: %p vs. %p", a_client->event, a_client->notify, b_client->notify); return 1; } return rc; } xmlNode * stonith_create_op(int call_id, const char *token, const char *op, xmlNode * data, int call_options) { xmlNode *op_msg = create_xml_node(NULL, "stonith_command"); CRM_CHECK(op_msg != NULL, return NULL); CRM_CHECK(token != NULL, return NULL); crm_xml_add(op_msg, F_XML_TAGNAME, "stonith_command"); crm_xml_add(op_msg, F_TYPE, T_STONITH_NG); crm_xml_add(op_msg, F_STONITH_CALLBACK_TOKEN, token); crm_xml_add(op_msg, F_STONITH_OPERATION, op); crm_xml_add_int(op_msg, F_STONITH_CALLID, call_id); crm_trace("Sending call options: %.8lx, %d", (long)call_options, call_options); crm_xml_add_int(op_msg, F_STONITH_CALLOPTS, call_options); if (data != NULL) { add_message_xml(op_msg, F_STONITH_CALLDATA, data); } return op_msg; } static void stonith_destroy_op_callback(gpointer data) { stonith_callback_client_t *blob = data; if (blob->timer && blob->timer->ref > 0) { g_source_remove(blob->timer->ref); } free(blob->timer); free(blob); } static int stonith_api_signoff(stonith_t * stonith) { stonith_private_t *native = stonith->st_private; crm_debug("Disconnecting from the fencer"); if (native->source != NULL) { /* Attached to mainloop */ mainloop_del_ipc_client(native->source); native->source = NULL; native->ipc = NULL; } else if (native->ipc) { /* Not attached to mainloop */ crm_ipc_t *ipc = native->ipc; native->ipc = NULL; crm_ipc_close(ipc); crm_ipc_destroy(ipc); } free(native->token); native->token = NULL; stonith->state = stonith_disconnected; return pcmk_ok; } static int stonith_api_del_callback(stonith_t * stonith, int call_id, bool all_callbacks) { stonith_private_t *private = stonith->st_private; if (all_callbacks) { private->op_callback = NULL; g_hash_table_destroy(private->stonith_op_callback_table); private->stonith_op_callback_table = pcmk__intkey_table(stonith_destroy_op_callback); } else if (call_id == 0) { private->op_callback = NULL; } else { pcmk__intkey_table_remove(private->stonith_op_callback_table, call_id); } return pcmk_ok; } /*! * \internal * \brief Invoke a (single) specified fence action callback * * \param[in] st Fencer API connection * \param[in] call_id If positive, call ID of completed fence action, otherwise * legacy return code for early action failure * \param[in] result Full result for action * \param[in] userdata User data to pass to callback * \param[in] callback Fence action callback to invoke */ static void invoke_fence_action_callback(stonith_t *st, int call_id, pcmk__action_result_t *result, void *userdata, void (*callback) (stonith_t *st, stonith_callback_data_t *data)) { stonith_callback_data_t data = { 0, }; data.call_id = call_id; data.rc = pcmk_rc2legacy(stonith__result2rc(result)); data.userdata = userdata; data.opaque = (void *) result; callback(st, &data); } /*! * \internal * \brief Invoke any callbacks registered for a specified fence action result * * Given a fence action result from the fencer, invoke any callback registered * for that action, as well as any global callback registered. * * \param[in] st Fencer API connection * \param[in] msg If non-NULL, fencer reply * \param[in] call_id If \p msg is NULL, call ID of action that timed out */ static void invoke_registered_callbacks(stonith_t *stonith, xmlNode *msg, int call_id) { stonith_private_t *private = NULL; stonith_callback_client_t *cb_info = NULL; pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; CRM_CHECK(stonith != NULL, return); CRM_CHECK(stonith->st_private != NULL, return); private = stonith->st_private; if (msg == NULL) { // Fencer didn't reply in time pcmk__set_result(&result, CRM_EX_ERROR, PCMK_EXEC_TIMEOUT, "Fencer accepted request but did not reply in time"); CRM_LOG_ASSERT(call_id > 0); } else { // We have the fencer reply if ((crm_element_value_int(msg, F_STONITH_CALLID, &call_id) != 0) || (call_id <= 0)) { crm_log_xml_warn(msg, "Bad fencer reply"); } stonith__xe_get_result(msg, &result); } if (call_id > 0) { cb_info = pcmk__intkey_table_lookup(private->stonith_op_callback_table, call_id); } if ((cb_info != NULL) && (cb_info->callback != NULL) && (pcmk__result_ok(&result) || !(cb_info->only_success))) { crm_trace("Invoking callback %s for call %d", crm_str(cb_info->id), call_id); invoke_fence_action_callback(stonith, call_id, &result, cb_info->user_data, cb_info->callback); } else if ((private->op_callback == NULL) && !pcmk__result_ok(&result)) { crm_warn("Fencing action without registered callback failed: %d (%s%s%s)", result.exit_status, pcmk_exec_status_str(result.execution_status), ((result.exit_reason == NULL)? "" : ": "), ((result.exit_reason == NULL)? "" : result.exit_reason)); crm_log_xml_debug(msg, "Failed fence update"); } if (private->op_callback != NULL) { crm_trace("Invoking global callback for call %d", call_id); invoke_fence_action_callback(stonith, call_id, &result, NULL, private->op_callback); } if (cb_info != NULL) { stonith_api_del_callback(stonith, call_id, FALSE); } pcmk__reset_result(&result); } static gboolean stonith_async_timeout_handler(gpointer data) { struct timer_rec_s *timer = data; crm_err("Async call %d timed out after %dms", timer->call_id, timer->timeout); invoke_registered_callbacks(timer->stonith, NULL, timer->call_id); /* Always return TRUE, never remove the handler * We do that in stonith_del_callback() */ return TRUE; } static void set_callback_timeout(stonith_callback_client_t * callback, stonith_t * stonith, int call_id, int timeout) { struct timer_rec_s *async_timer = callback->timer; if (timeout <= 0) { return; } if (!async_timer) { async_timer = calloc(1, sizeof(struct timer_rec_s)); callback->timer = async_timer; } async_timer->stonith = stonith; async_timer->call_id = call_id; /* Allow a fair bit of grace to allow the server to tell us of a timeout * This is only a fallback */ async_timer->timeout = (timeout + 60) * 1000; if (async_timer->ref) { g_source_remove(async_timer->ref); } async_timer->ref = g_timeout_add(async_timer->timeout, stonith_async_timeout_handler, async_timer); } static void update_callback_timeout(int call_id, int timeout, stonith_t * st) { stonith_callback_client_t *callback = NULL; stonith_private_t *private = st->st_private; callback = pcmk__intkey_table_lookup(private->stonith_op_callback_table, call_id); if (!callback || !callback->allow_timeout_updates) { return; } set_callback_timeout(callback, st, call_id, timeout); } static int stonith_dispatch_internal(const char *buffer, ssize_t length, gpointer userdata) { const char *type = NULL; struct notify_blob_s blob; stonith_t *st = userdata; stonith_private_t *private = NULL; CRM_ASSERT(st != NULL); private = st->st_private; blob.stonith = st; blob.xml = string2xml(buffer); if (blob.xml == NULL) { crm_warn("Received malformed message from fencer: %s", buffer); return 0; } /* do callbacks */ type = crm_element_value(blob.xml, F_TYPE); crm_trace("Activating %s callbacks...", type); - if (pcmk__str_eq(type, T_STONITH_NG, pcmk__str_casei)) { + if (pcmk__str_eq(type, T_STONITH_NG, pcmk__str_none)) { invoke_registered_callbacks(st, blob.xml, 0); - } else if (pcmk__str_eq(type, T_STONITH_NOTIFY, pcmk__str_casei)) { + } else if (pcmk__str_eq(type, T_STONITH_NOTIFY, pcmk__str_none)) { foreach_notify_entry(private, stonith_send_notification, &blob); - } else if (pcmk__str_eq(type, T_STONITH_TIMEOUT_VALUE, pcmk__str_casei)) { + } else if (pcmk__str_eq(type, T_STONITH_TIMEOUT_VALUE, pcmk__str_none)) { int call_id = 0; int timeout = 0; crm_element_value_int(blob.xml, F_STONITH_TIMEOUT, &timeout); crm_element_value_int(blob.xml, F_STONITH_CALLID, &call_id); update_callback_timeout(call_id, timeout, st); } else { crm_err("Unknown message type: %s", type); crm_log_xml_warn(blob.xml, "BadReply"); } free_xml(blob.xml); return 1; } static int stonith_api_signon(stonith_t * stonith, const char *name, int *stonith_fd) { int rc = pcmk_ok; stonith_private_t *native = NULL; const char *display_name = name? name : "client"; struct ipc_client_callbacks st_callbacks = { .dispatch = stonith_dispatch_internal, .destroy = stonith_connection_destroy }; CRM_CHECK(stonith != NULL, return -EINVAL); native = stonith->st_private; CRM_ASSERT(native != NULL); crm_debug("Attempting fencer connection by %s with%s mainloop", display_name, (stonith_fd? "out" : "")); stonith->state = stonith_connected_command; if (stonith_fd) { /* No mainloop */ native->ipc = crm_ipc_new("stonith-ng", 0); if (native->ipc && crm_ipc_connect(native->ipc)) { *stonith_fd = crm_ipc_get_fd(native->ipc); } else if (native->ipc) { crm_ipc_close(native->ipc); crm_ipc_destroy(native->ipc); native->ipc = NULL; } } else { /* With mainloop */ native->source = mainloop_add_ipc_client("stonith-ng", G_PRIORITY_MEDIUM, 0, stonith, &st_callbacks); native->ipc = mainloop_get_ipc_client(native->source); } if (native->ipc == NULL) { rc = -ENOTCONN; } else { xmlNode *reply = NULL; xmlNode *hello = create_xml_node(NULL, "stonith_command"); crm_xml_add(hello, F_TYPE, T_STONITH_NG); crm_xml_add(hello, F_STONITH_OPERATION, CRM_OP_REGISTER); crm_xml_add(hello, F_STONITH_CLIENTNAME, name); rc = crm_ipc_send(native->ipc, hello, crm_ipc_client_response, -1, &reply); if (rc < 0) { crm_debug("Couldn't register with the fencer: %s " CRM_XS " rc=%d", pcmk_strerror(rc), rc); rc = -ECOMM; } else if (reply == NULL) { crm_debug("Couldn't register with the fencer: no reply"); rc = -EPROTO; } else { const char *msg_type = crm_element_value(reply, F_STONITH_OPERATION); native->token = crm_element_value_copy(reply, F_STONITH_CLIENTID); - if (!pcmk__str_eq(msg_type, CRM_OP_REGISTER, pcmk__str_casei)) { + if (!pcmk__str_eq(msg_type, CRM_OP_REGISTER, pcmk__str_none)) { crm_debug("Couldn't register with the fencer: invalid reply type '%s'", (msg_type? msg_type : "(missing)")); crm_log_xml_debug(reply, "Invalid fencer reply"); rc = -EPROTO; } else if (native->token == NULL) { crm_debug("Couldn't register with the fencer: no token in reply"); crm_log_xml_debug(reply, "Invalid fencer reply"); rc = -EPROTO; } else { crm_debug("Connection to fencer by %s succeeded (registration token: %s)", display_name, native->token); rc = pcmk_ok; } } free_xml(reply); free_xml(hello); } if (rc != pcmk_ok) { crm_debug("Connection attempt to fencer by %s failed: %s " CRM_XS " rc=%d", display_name, pcmk_strerror(rc), rc); stonith->cmds->disconnect(stonith); } return rc; } static int stonith_set_notification(stonith_t * stonith, const char *callback, int enabled) { int rc = pcmk_ok; xmlNode *notify_msg = create_xml_node(NULL, __func__); stonith_private_t *native = stonith->st_private; if (stonith->state != stonith_disconnected) { crm_xml_add(notify_msg, F_STONITH_OPERATION, T_STONITH_NOTIFY); if (enabled) { crm_xml_add(notify_msg, F_STONITH_NOTIFY_ACTIVATE, callback); } else { crm_xml_add(notify_msg, F_STONITH_NOTIFY_DEACTIVATE, callback); } rc = crm_ipc_send(native->ipc, notify_msg, crm_ipc_client_response, -1, NULL); if (rc < 0) { crm_perror(LOG_DEBUG, "Couldn't register for fencing notifications: %d", rc); rc = -ECOMM; } else { rc = pcmk_ok; } } free_xml(notify_msg); return rc; } static int stonith_api_add_notification(stonith_t * stonith, const char *event, void (*callback) (stonith_t * stonith, stonith_event_t * e)) { GList *list_item = NULL; stonith_notify_client_t *new_client = NULL; stonith_private_t *private = NULL; private = stonith->st_private; crm_trace("Adding callback for %s events (%d)", event, g_list_length(private->notify_list)); new_client = calloc(1, sizeof(stonith_notify_client_t)); new_client->event = event; new_client->notify = callback; list_item = g_list_find_custom(private->notify_list, new_client, stonithlib_GCompareFunc); if (list_item != NULL) { crm_warn("Callback already present"); free(new_client); return -ENOTUNIQ; } else { private->notify_list = g_list_append(private->notify_list, new_client); stonith_set_notification(stonith, event, 1); crm_trace("Callback added (%d)", g_list_length(private->notify_list)); } return pcmk_ok; } static void del_notify_entry(gpointer data, gpointer user_data) { stonith_notify_client_t *entry = data; stonith_t * stonith = user_data; if (!entry->delete) { crm_debug("Removing callback for %s events", entry->event); stonith_api_del_notification(stonith, entry->event); } } static int stonith_api_del_notification(stonith_t * stonith, const char *event) { GList *list_item = NULL; stonith_notify_client_t *new_client = NULL; stonith_private_t *private = stonith->st_private; if (event == NULL) { foreach_notify_entry(private, del_notify_entry, stonith); crm_trace("Removed callback"); return pcmk_ok; } crm_debug("Removing callback for %s events", event); new_client = calloc(1, sizeof(stonith_notify_client_t)); new_client->event = event; new_client->notify = NULL; list_item = g_list_find_custom(private->notify_list, new_client, stonithlib_GCompareFunc); stonith_set_notification(stonith, event, 0); if (list_item != NULL) { stonith_notify_client_t *list_client = list_item->data; if (private->notify_refcnt) { list_client->delete = TRUE; private->notify_deletes = TRUE; } else { private->notify_list = g_list_remove(private->notify_list, list_client); free(list_client); } crm_trace("Removed callback"); } else { crm_trace("Callback not present"); } free(new_client); return pcmk_ok; } static int stonith_api_add_callback(stonith_t * stonith, int call_id, int timeout, int options, void *user_data, const char *callback_name, void (*callback) (stonith_t * st, stonith_callback_data_t * data)) { stonith_callback_client_t *blob = NULL; stonith_private_t *private = NULL; CRM_CHECK(stonith != NULL, return -EINVAL); CRM_CHECK(stonith->st_private != NULL, return -EINVAL); private = stonith->st_private; if (call_id == 0) { // Add global callback private->op_callback = callback; } else if (call_id < 0) { // Call failed immediately, so call callback now if (!(options & st_opt_report_only_success)) { pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; crm_trace("Call failed, calling %s: %s", callback_name, pcmk_strerror(call_id)); pcmk__set_result(&result, CRM_EX_ERROR, stonith__legacy2status(call_id), NULL); invoke_fence_action_callback(stonith, call_id, &result, user_data, callback); } else { crm_warn("Fencer call failed: %s", pcmk_strerror(call_id)); } return FALSE; } blob = calloc(1, sizeof(stonith_callback_client_t)); blob->id = callback_name; blob->only_success = (options & st_opt_report_only_success) ? TRUE : FALSE; blob->user_data = user_data; blob->callback = callback; blob->allow_timeout_updates = (options & st_opt_timeout_updates) ? TRUE : FALSE; if (timeout > 0) { set_callback_timeout(blob, stonith, call_id, timeout); } pcmk__intkey_table_insert(private->stonith_op_callback_table, call_id, blob); crm_trace("Added callback to %s for call %d", callback_name, call_id); return TRUE; } static void stonith_dump_pending_op(gpointer key, gpointer value, gpointer user_data) { int call = GPOINTER_TO_INT(key); stonith_callback_client_t *blob = value; crm_debug("Call %d (%s): pending", call, crm_str(blob->id)); } void stonith_dump_pending_callbacks(stonith_t * stonith) { stonith_private_t *private = stonith->st_private; if (private->stonith_op_callback_table == NULL) { return; } return g_hash_table_foreach(private->stonith_op_callback_table, stonith_dump_pending_op, NULL); } /*! * \internal * \brief Get the data section of a fencer notification * * \param[in] msg Notification XML * \param[in] ntype Notification type */ static xmlNode * get_event_data_xml(xmlNode *msg, const char *ntype) { char *data_addr = crm_strdup_printf("//%s", ntype); xmlNode *data = get_xpath_object(data_addr, msg, LOG_DEBUG); free(data_addr); return data; } /* */ static stonith_event_t * xml_to_event(xmlNode *msg) { stonith_event_t *event = calloc(1, sizeof(stonith_event_t)); - const char *ntype = crm_element_value(msg, F_SUBTYPE); struct event_private *event_private = NULL; CRM_ASSERT(event != NULL); event->opaque = calloc(1, sizeof(struct event_private)); CRM_ASSERT(event->opaque != NULL); event_private = (struct event_private *) event->opaque; crm_log_xml_trace(msg, "stonith_notify"); - // All notification types have the operation result + // All notification types have the operation result and notification subtype stonith__xe_get_result(msg, &event_private->result); + event->operation = crm_element_value_copy(msg, F_STONITH_OPERATION); // @COMPAT The API originally provided the result as a legacy return code event->result = pcmk_rc2legacy(stonith__result2rc(&event_private->result)); - // Fence notifications have additional information - if (pcmk__str_eq(ntype, T_STONITH_NOTIFY_FENCE, pcmk__str_casei)) { - xmlNode *data = get_event_data_xml(msg, ntype); + // Some notification subtypes have additional information + + if (pcmk__str_eq(event->operation, T_STONITH_NOTIFY_FENCE, + pcmk__str_none)) { + xmlNode *data = get_event_data_xml(msg, event->operation); if (data == NULL) { - crm_err("No data for %s event", ntype); + crm_err("No data for %s event", event->operation); crm_log_xml_notice(msg, "BadEvent"); } else { event->origin = crm_element_value_copy(data, F_STONITH_ORIGIN); event->action = crm_element_value_copy(data, F_STONITH_ACTION); event->target = crm_element_value_copy(data, F_STONITH_TARGET); event->executioner = crm_element_value_copy(data, F_STONITH_DELEGATE); event->id = crm_element_value_copy(data, F_STONITH_REMOTE_OP_ID); event->client_origin = crm_element_value_copy(data, F_STONITH_CLIENTNAME); event->device = crm_element_value_copy(data, F_STONITH_DEVICE); } - event->operation = crm_element_value_copy(msg, F_STONITH_OPERATION); + + } else if (pcmk__str_any_of(event->operation, + STONITH_OP_DEVICE_ADD, STONITH_OP_DEVICE_DEL, + STONITH_OP_LEVEL_ADD, STONITH_OP_LEVEL_DEL, + NULL)) { + xmlNode *data = get_event_data_xml(msg, event->operation); + + if (data == NULL) { + crm_err("No data for %s event", event->operation); + crm_log_xml_notice(msg, "BadEvent"); + } else { + event->device = crm_element_value_copy(data, F_STONITH_DEVICE); + } } return event; } static void event_free(stonith_event_t * event) { struct event_private *event_private = event->opaque; free(event->id); free(event->type); free(event->message); free(event->operation); free(event->origin); free(event->action); free(event->target); free(event->executioner); free(event->device); free(event->client_origin); pcmk__reset_result(&event_private->result); free(event->opaque); free(event); } static void stonith_send_notification(gpointer data, gpointer user_data) { struct notify_blob_s *blob = user_data; stonith_notify_client_t *entry = data; stonith_event_t *st_event = NULL; const char *event = NULL; if (blob->xml == NULL) { crm_warn("Skipping callback - NULL message"); return; } event = crm_element_value(blob->xml, F_SUBTYPE); if (entry == NULL) { crm_warn("Skipping callback - NULL callback client"); return; } else if (entry->delete) { crm_trace("Skipping callback - marked for deletion"); return; } else if (entry->notify == NULL) { crm_warn("Skipping callback - NULL callback"); return; - } else if (!pcmk__str_eq(entry->event, event, pcmk__str_casei)) { + } else if (!pcmk__str_eq(entry->event, event, pcmk__str_none)) { crm_trace("Skipping callback - event mismatch %p/%s vs. %s", entry, entry->event, event); return; } st_event = xml_to_event(blob->xml); crm_trace("Invoking callback for %p/%s event...", entry, event); entry->notify(blob->stonith, st_event); crm_trace("Callback invoked..."); event_free(st_event); } /*! * \internal * \brief Create and send an API request * * \param[in] stonith Stonith connection * \param[in] op API operation to request * \param[in] data Data to attach to request * \param[out] output_data If not NULL, will be set to reply if synchronous * \param[in] call_options Bitmask of stonith_call_options to use * \param[in] timeout Error if not completed within this many seconds * * \return pcmk_ok (for synchronous requests) or positive call ID * (for asynchronous requests) on success, -errno otherwise */ static int stonith_send_command(stonith_t * stonith, const char *op, xmlNode * data, xmlNode ** output_data, int call_options, int timeout) { int rc = 0; int reply_id = -1; xmlNode *op_msg = NULL; xmlNode *op_reply = NULL; stonith_private_t *native = NULL; CRM_ASSERT(stonith && stonith->st_private && op); native = stonith->st_private; if (output_data != NULL) { *output_data = NULL; } if ((stonith->state == stonith_disconnected) || (native->token == NULL)) { return -ENOTCONN; } /* Increment the call ID, which must be positive to avoid conflicting with * error codes. This shouldn't be a problem unless the client mucked with * it or the counter wrapped around. */ stonith->call_id++; if (stonith->call_id < 1) { stonith->call_id = 1; } op_msg = stonith_create_op(stonith->call_id, native->token, op, data, call_options); if (op_msg == NULL) { return -EINVAL; } crm_xml_add_int(op_msg, F_STONITH_TIMEOUT, timeout); crm_trace("Sending %s message to fencer with timeout %ds", op, timeout); if (data) { const char *delay_s = crm_element_value(data, F_STONITH_DELAY); if (delay_s) { crm_xml_add(op_msg, F_STONITH_DELAY, delay_s); } } { enum crm_ipc_flags ipc_flags = crm_ipc_flags_none; if (call_options & st_opt_sync_call) { pcmk__set_ipc_flags(ipc_flags, "stonith command", crm_ipc_client_response); } rc = crm_ipc_send(native->ipc, op_msg, ipc_flags, 1000 * (timeout + 60), &op_reply); } free_xml(op_msg); if (rc < 0) { crm_perror(LOG_ERR, "Couldn't perform %s operation (timeout=%ds): %d", op, timeout, rc); rc = -ECOMM; goto done; } crm_log_xml_trace(op_reply, "Reply"); if (!(call_options & st_opt_sync_call)) { crm_trace("Async call %d, returning", stonith->call_id); free_xml(op_reply); return stonith->call_id; } rc = pcmk_ok; crm_element_value_int(op_reply, F_STONITH_CALLID, &reply_id); if (reply_id == stonith->call_id) { pcmk__action_result_t result = PCMK__UNKNOWN_RESULT; crm_trace("Synchronous reply %d received", reply_id); stonith__xe_get_result(op_reply, &result); rc = pcmk_rc2legacy(stonith__result2rc(&result)); pcmk__reset_result(&result); if ((call_options & st_opt_discard_reply) || output_data == NULL) { crm_trace("Discarding reply"); } else { *output_data = op_reply; op_reply = NULL; /* Prevent subsequent free */ } } else if (reply_id <= 0) { crm_err("Received bad reply: No id set"); crm_log_xml_err(op_reply, "Bad reply"); free_xml(op_reply); rc = -ENOMSG; } else { crm_err("Received bad reply: %d (wanted %d)", reply_id, stonith->call_id); crm_log_xml_err(op_reply, "Old reply"); free_xml(op_reply); rc = -ENOMSG; } done: if (crm_ipc_connected(native->ipc) == FALSE) { crm_err("Fencer disconnected"); free(native->token); native->token = NULL; stonith->state = stonith_disconnected; } free_xml(op_reply); return rc; } /* Not used with mainloop */ bool stonith_dispatch(stonith_t * st) { gboolean stay_connected = TRUE; stonith_private_t *private = NULL; CRM_ASSERT(st != NULL); private = st->st_private; while (crm_ipc_ready(private->ipc)) { if (crm_ipc_read(private->ipc) > 0) { const char *msg = crm_ipc_buffer(private->ipc); stonith_dispatch_internal(msg, strlen(msg), st); } if (crm_ipc_connected(private->ipc) == FALSE) { crm_err("Connection closed"); stay_connected = FALSE; } } return stay_connected; } static int stonith_api_free(stonith_t * stonith) { int rc = pcmk_ok; crm_trace("Destroying %p", stonith); if (stonith->state != stonith_disconnected) { crm_trace("Disconnecting %p first", stonith); rc = stonith->cmds->disconnect(stonith); } if (stonith->state == stonith_disconnected) { stonith_private_t *private = stonith->st_private; crm_trace("Removing %d callbacks", g_hash_table_size(private->stonith_op_callback_table)); g_hash_table_destroy(private->stonith_op_callback_table); crm_trace("Destroying %d notification clients", g_list_length(private->notify_list)); g_list_free_full(private->notify_list, free); free(stonith->st_private); free(stonith->cmds); free(stonith); } else { crm_err("Not free'ing active connection: %s (%d)", pcmk_strerror(rc), rc); } return rc; } void stonith_api_delete(stonith_t * stonith) { crm_trace("Destroying %p", stonith); if(stonith) { stonith->cmds->free(stonith); } } static int stonith_api_validate(stonith_t *st, int call_options, const char *rsc_id, const char *namespace_s, const char *agent, stonith_key_value_t *params, int timeout, char **output, char **error_output) { /* Validation should be done directly via the agent, so we can get it from * stonith_admin when the cluster is not running, which is important for * higher-level tools. */ int rc = pcmk_ok; /* Use a dummy node name in case the agent requires a target. We assume the * actual target doesn't matter for validation purposes (if in practice, * that is incorrect, we will need to allow the caller to pass the target). */ const char *target = "node1"; const char *host_arg = NULL; GHashTable *params_table = pcmk__strkey_table(free, free); // Convert parameter list to a hash table for (; params; params = params->next) { if (pcmk__str_eq(params->key, PCMK_STONITH_HOST_ARGUMENT, - pcmk__str_casei)) { + pcmk__str_none)) { host_arg = params->value; } if (!pcmk_stonith_param(params->key)) { g_hash_table_insert(params_table, strdup(params->key), strdup(params->value)); } } #if SUPPORT_CIBSECRETS rc = pcmk__substitute_secrets(rsc_id, params_table); if (rc != pcmk_rc_ok) { crm_warn("Could not replace secret parameters for validation of %s: %s", agent, pcmk_rc_str(rc)); // rc is standard return value, don't return it in this function } #endif if (output) { *output = NULL; } if (error_output) { *error_output = NULL; } switch (stonith_get_namespace(agent, namespace_s)) { case st_namespace_rhcs: rc = stonith__rhcs_validate(st, call_options, target, agent, params_table, host_arg, timeout, output, error_output); break; #if HAVE_STONITH_STONITH_H case st_namespace_lha: rc = stonith__lha_validate(st, call_options, target, agent, params_table, timeout, output, error_output); break; #endif default: rc = -EINVAL; errno = EINVAL; crm_perror(LOG_ERR, "Agent %s not found or does not support validation", agent); break; } g_hash_table_destroy(params_table); return rc; } stonith_t * stonith_api_new(void) { stonith_t *new_stonith = NULL; stonith_private_t *private = NULL; new_stonith = calloc(1, sizeof(stonith_t)); if (new_stonith == NULL) { return NULL; } private = calloc(1, sizeof(stonith_private_t)); if (private == NULL) { free(new_stonith); return NULL; } new_stonith->st_private = private; private->stonith_op_callback_table = pcmk__intkey_table(stonith_destroy_op_callback); private->notify_list = NULL; private->notify_refcnt = 0; private->notify_deletes = FALSE; new_stonith->call_id = 1; new_stonith->state = stonith_disconnected; new_stonith->cmds = calloc(1, sizeof(stonith_api_operations_t)); if (new_stonith->cmds == NULL) { free(new_stonith->st_private); free(new_stonith); return NULL; } /* *INDENT-OFF* */ new_stonith->cmds->free = stonith_api_free; new_stonith->cmds->connect = stonith_api_signon; new_stonith->cmds->disconnect = stonith_api_signoff; new_stonith->cmds->list = stonith_api_list; new_stonith->cmds->monitor = stonith_api_monitor; new_stonith->cmds->status = stonith_api_status; new_stonith->cmds->fence = stonith_api_fence; new_stonith->cmds->fence_with_delay = stonith_api_fence_with_delay; new_stonith->cmds->confirm = stonith_api_confirm; new_stonith->cmds->history = stonith_api_history; new_stonith->cmds->list_agents = stonith_api_device_list; new_stonith->cmds->metadata = stonith_api_device_metadata; new_stonith->cmds->query = stonith_api_query; new_stonith->cmds->remove_device = stonith_api_remove_device; new_stonith->cmds->register_device = stonith_api_register_device; new_stonith->cmds->remove_level = stonith_api_remove_level; new_stonith->cmds->remove_level_full = stonith_api_remove_level_full; new_stonith->cmds->register_level = stonith_api_register_level; new_stonith->cmds->register_level_full = stonith_api_register_level_full; new_stonith->cmds->remove_callback = stonith_api_del_callback; new_stonith->cmds->register_callback = stonith_api_add_callback; new_stonith->cmds->remove_notification = stonith_api_del_notification; new_stonith->cmds->register_notification = stonith_api_add_notification; new_stonith->cmds->validate = stonith_api_validate; /* *INDENT-ON* */ return new_stonith; } /*! * \brief Make a blocking connection attempt to the fencer * * \param[in,out] st Fencer API object * \param[in] name Client name to use with fencer * \param[in] max_attempts Return error if this many attempts fail * * \return pcmk_ok on success, result of last attempt otherwise */ int stonith_api_connect_retry(stonith_t *st, const char *name, int max_attempts) { int rc = -EINVAL; // if max_attempts is not positive for (int attempt = 1; attempt <= max_attempts; attempt++) { rc = st->cmds->connect(st, name, NULL); if (rc == pcmk_ok) { return pcmk_ok; } else if (attempt < max_attempts) { crm_notice("Fencer connection attempt %d of %d failed (retrying in 2s): %s " CRM_XS " rc=%d", attempt, max_attempts, pcmk_strerror(rc), rc); sleep(2); } } crm_notice("Could not connect to fencer: %s " CRM_XS " rc=%d", pcmk_strerror(rc), rc); return rc; } stonith_key_value_t * stonith_key_value_add(stonith_key_value_t * head, const char *key, const char *value) { stonith_key_value_t *p, *end; p = calloc(1, sizeof(stonith_key_value_t)); pcmk__str_update(&p->key, key); pcmk__str_update(&p->value, value); end = head; while (end && end->next) { end = end->next; } if (end) { end->next = p; } else { head = p; } return head; } void stonith_key_value_freeall(stonith_key_value_t * head, int keys, int values) { stonith_key_value_t *p; while (head) { p = head->next; if (keys) { free(head->key); } if (values) { free(head->value); } free(head); head = p; } } #define api_log_open() openlog("stonith-api", LOG_CONS | LOG_NDELAY | LOG_PID, LOG_DAEMON) #define api_log(level, fmt, args...) syslog(level, "%s: "fmt, __func__, args) int stonith_api_kick(uint32_t nodeid, const char *uname, int timeout, bool off) { int rc = pcmk_ok; stonith_t *st = stonith_api_new(); const char *action = off? "off" : "reboot"; api_log_open(); if (st == NULL) { api_log(LOG_ERR, "API initialization failed, could not kick (%s) node %u/%s", action, nodeid, uname); return -EPROTO; } rc = st->cmds->connect(st, "stonith-api", NULL); if (rc != pcmk_ok) { api_log(LOG_ERR, "Connection failed, could not kick (%s) node %u/%s : %s (%d)", action, nodeid, uname, pcmk_strerror(rc), rc); } else { char *name = (uname == NULL)? pcmk__itoa(nodeid) : strdup(uname); int opts = 0; stonith__set_call_options(opts, name, st_opt_sync_call|st_opt_allow_suicide); if ((uname == NULL) && (nodeid > 0)) { stonith__set_call_options(opts, name, st_opt_cs_nodeid); } rc = st->cmds->fence(st, opts, name, action, timeout, 0); free(name); if (rc != pcmk_ok) { api_log(LOG_ERR, "Could not kick (%s) node %u/%s : %s (%d)", action, nodeid, uname, pcmk_strerror(rc), rc); } else { api_log(LOG_NOTICE, "Node %u/%s kicked: %s", nodeid, uname, action); } } stonith_api_delete(st); return rc; } time_t stonith_api_time(uint32_t nodeid, const char *uname, bool in_progress) { int rc = pcmk_ok; time_t when = 0; stonith_t *st = stonith_api_new(); stonith_history_t *history = NULL, *hp = NULL; if (st == NULL) { api_log(LOG_ERR, "Could not retrieve fence history for %u/%s: " "API initialization failed", nodeid, uname); return when; } rc = st->cmds->connect(st, "stonith-api", NULL); if (rc != pcmk_ok) { api_log(LOG_NOTICE, "Connection failed: %s (%d)", pcmk_strerror(rc), rc); } else { int entries = 0; int progress = 0; int completed = 0; int opts = 0; char *name = (uname == NULL)? pcmk__itoa(nodeid) : strdup(uname); stonith__set_call_options(opts, name, st_opt_sync_call); if ((uname == NULL) && (nodeid > 0)) { stonith__set_call_options(opts, name, st_opt_cs_nodeid); } rc = st->cmds->history(st, opts, name, &history, 120); free(name); for (hp = history; hp; hp = hp->next) { entries++; if (in_progress) { progress++; if (hp->state != st_done && hp->state != st_failed) { when = time(NULL); } } else if (hp->state == st_done) { completed++; if (hp->completed > when) { when = hp->completed; } } } stonith_history_free(history); if(rc == pcmk_ok) { api_log(LOG_INFO, "Found %d entries for %u/%s: %d in progress, %d completed", entries, nodeid, uname, progress, completed); } else { api_log(LOG_ERR, "Could not retrieve fence history for %u/%s: %s (%d)", nodeid, uname, pcmk_strerror(rc), rc); } } stonith_api_delete(st); if(when) { api_log(LOG_INFO, "Node %u/%s last kicked at: %ld", nodeid, uname, (long int)when); } return when; } bool stonith_agent_exists(const char *agent, int timeout) { stonith_t *st = NULL; stonith_key_value_t *devices = NULL; stonith_key_value_t *dIter = NULL; bool rc = FALSE; if (agent == NULL) { return rc; } st = stonith_api_new(); if (st == NULL) { crm_err("Could not list fence agents: API memory allocation failed"); return FALSE; } st->cmds->list_agents(st, st_opt_sync_call, NULL, &devices, timeout == 0 ? 120 : timeout); for (dIter = devices; dIter != NULL; dIter = dIter->next) { if (pcmk__str_eq(dIter->value, agent, pcmk__str_none)) { rc = TRUE; break; } } stonith_key_value_freeall(devices, 1, 1); stonith_api_delete(st); return rc; } const char * stonith_action_str(const char *action) { if (action == NULL) { return "fencing"; } else if (!strcmp(action, "on")) { return "unfencing"; } else if (!strcmp(action, "off")) { return "turning off"; } else { return action; } } /*! * \internal * \brief Parse a target name from one line of a target list string * * \param[in] line One line of a target list string * \param[in] len String length of line * \param[in,out] output List to add newly allocated target name to */ static void parse_list_line(const char *line, int len, GList **output) { size_t i = 0; size_t entry_start = 0; /* Skip complaints about additional parameters device doesn't understand * * @TODO Document or eliminate the implied restriction of target names */ if (strstr(line, "invalid") || strstr(line, "variable")) { crm_debug("Skipping list output line: %s", line); return; } // Process line content, character by character for (i = 0; i <= len; i++) { if (isspace(line[i]) || (line[i] == ',') || (line[i] == ';') || (line[i] == '\0')) { // We've found a separator (i.e. the end of an entry) int rc = 0; char *entry = NULL; if (i == entry_start) { // Skip leading and sequential separators entry_start = i + 1; continue; } entry = calloc(i - entry_start + 1, sizeof(char)); CRM_ASSERT(entry != NULL); /* Read entry, stopping at first separator * * @TODO Document or eliminate these character restrictions */ rc = sscanf(line + entry_start, "%[a-zA-Z0-9_-.]", entry); if (rc != 1) { crm_warn("Could not parse list output entry: %s " CRM_XS " entry_start=%d position=%d", line + entry_start, entry_start, i); free(entry); } else if (pcmk__strcase_any_of(entry, "on", "off", NULL)) { /* Some agents print the target status in the list output, * though none are known now (the separate list-status command * is used for this, but it can also print "UNKNOWN"). To handle * this possibility, skip such entries. * * @TODO Document or eliminate the implied restriction of target * names. */ free(entry); } else { // We have a valid entry *output = g_list_append(*output, entry); } entry_start = i + 1; } } } /*! * \internal * \brief Parse a list of targets from a string * * \param[in] list_output Target list as a string * * \return List of target names * \note The target list string format is flexible, to allow for user-specified * lists such pcmk_host_list and the output of an agent's list action * (whether direct or via the API, which escapes newlines). There may be * multiple lines, separated by either a newline or an escaped newline * (backslash n). Each line may have one or more target names, separated * by any combination of whitespace, commas, and semi-colons. Lines * containing "invalid" or "variable" will be ignored entirely. Target * names "on" or "off" (case-insensitive) will be ignored. Target names * may contain only alphanumeric characters, underbars (_), dashes (-), * and dots (.) (if any other character occurs in the name, it and all * subsequent characters in the name will be ignored). * \note The caller is responsible for freeing the result with * g_list_free_full(result, free). */ GList * stonith__parse_targets(const char *target_spec) { GList *targets = NULL; if (target_spec != NULL) { size_t out_len = strlen(target_spec); size_t line_start = 0; // Starting index of line being processed for (size_t i = 0; i <= out_len; ++i) { if ((target_spec[i] == '\n') || (target_spec[i] == '\0') || ((target_spec[i] == '\\') && (target_spec[i + 1] == 'n'))) { // We've reached the end of one line of output int len = i - line_start; if (len > 0) { char *line = strndup(target_spec + line_start, len); line[len] = '\0'; // Because it might be a newline parse_list_line(line, len, &targets); free(line); } if (target_spec[i] == '\\') { ++i; // backslash-n takes up two positions } line_start = i + 1; } } } return targets; } /*! * \internal * \brief Check whether a fencing failure was followed by an equivalent success * * \param[in] event Fencing failure * \param[in] top_history Complete fencing history (must be sorted by * stonith__sort_history() beforehand) * * \return The name of the node that executed the fencing if a later successful * event exists, or NULL if no such event exists */ const char * stonith__later_succeeded(stonith_history_t *event, stonith_history_t *top_history) { const char *other = NULL; for (stonith_history_t *prev_hp = top_history; prev_hp; prev_hp = prev_hp->next) { if (prev_hp == event) { break; } if ((prev_hp->state == st_done) && pcmk__str_eq(event->target, prev_hp->target, pcmk__str_casei) && - pcmk__str_eq(event->action, prev_hp->action, pcmk__str_casei) && + pcmk__str_eq(event->action, prev_hp->action, pcmk__str_none) && ((event->completed < prev_hp->completed) || ((event->completed == prev_hp->completed) && (event->completed_nsec < prev_hp->completed_nsec)))) { if ((event->delegate == NULL) || pcmk__str_eq(event->delegate, prev_hp->delegate, pcmk__str_casei)) { // Prefer equivalent fencing by same executioner return prev_hp->delegate; } else if (other == NULL) { // Otherwise remember first successful executioner other = (prev_hp->delegate == NULL)? "some node" : prev_hp->delegate; } } } return other; } /*! * \internal * \brief Sort the stonith-history * sort by competed most current on the top * pending actions lacking a completed-stamp are gathered at the top * * \param[in] history List of stonith actions * */ stonith_history_t * stonith__sort_history(stonith_history_t *history) { stonith_history_t *new = NULL, *pending = NULL, *hp, *np, *tmp; for (hp = history; hp; ) { tmp = hp->next; if ((hp->state == st_done) || (hp->state == st_failed)) { /* sort into new */ if ((!new) || (hp->completed > new->completed) || ((hp->completed == new->completed) && (hp->completed_nsec > new->completed_nsec))) { hp->next = new; new = hp; } else { np = new; do { if ((!np->next) || (hp->completed > np->next->completed) || ((hp->completed == np->next->completed) && (hp->completed_nsec > np->next->completed_nsec))) { hp->next = np->next; np->next = hp; break; } np = np->next; } while (1); } } else { /* put into pending */ hp->next = pending; pending = hp; } hp = tmp; } /* pending actions don't have a completed-stamp so make them go front */ if (pending) { stonith_history_t *last_pending = pending; while (last_pending->next) { last_pending = last_pending->next; } last_pending->next = new; new = pending; } return new; } /*! * \brief Return string equivalent of an operation state value * * \param[in] state Fencing operation state value * * \return Human-friendly string equivalent of state */ const char * stonith_op_state_str(enum op_state state) { switch (state) { case st_query: return "querying"; case st_exec: return "executing"; case st_done: return "completed"; case st_duplicate: return "duplicate"; case st_failed: return "failed"; } return "unknown"; } stonith_history_t * stonith__first_matching_event(stonith_history_t *history, bool (*matching_fn)(stonith_history_t *, void *), void *user_data) { for (stonith_history_t *hp = history; hp; hp = hp->next) { if (matching_fn(hp, user_data)) { return hp; } } return NULL; } bool stonith__event_state_pending(stonith_history_t *history, void *user_data) { return history->state != st_failed && history->state != st_done; } bool stonith__event_state_eq(stonith_history_t *history, void *user_data) { return history->state == GPOINTER_TO_INT(user_data); } bool stonith__event_state_neq(stonith_history_t *history, void *user_data) { return history->state != GPOINTER_TO_INT(user_data); } void stonith__device_parameter_flags(uint32_t *device_flags, const char *device_name, xmlNode *metadata) { xmlXPathObjectPtr xpath = NULL; int max = 0; int lpc = 0; CRM_CHECK((device_flags != NULL) && (metadata != NULL), return); xpath = xpath_search(metadata, "//parameter"); max = numXpathResults(xpath); if (max <= 0) { freeXpathObject(xpath); return; } for (lpc = 0; lpc < max; lpc++) { const char *parameter = NULL; xmlNode *match = getXpathResult(xpath, lpc); CRM_LOG_ASSERT(match != NULL); if (match == NULL) { continue; } parameter = crm_element_value(match, "name"); if (pcmk__str_eq(parameter, "plug", pcmk__str_casei)) { stonith__set_device_flags(*device_flags, device_name, st_device_supports_parameter_plug); } else if (pcmk__str_eq(parameter, "port", pcmk__str_casei)) { stonith__set_device_flags(*device_flags, device_name, st_device_supports_parameter_port); } } freeXpathObject(xpath); } /*! * \internal * \brief Return the exit status from an async action callback * * \param[in] data Callback data * * \return Exit status from callback data */ int stonith__exit_status(stonith_callback_data_t *data) { if ((data == NULL) || (data->opaque == NULL)) { return CRM_EX_ERROR; } return ((pcmk__action_result_t *) data->opaque)->exit_status; } /*! * \internal * \brief Return the execution status from an async action callback * * \param[in] data Callback data * * \return Execution status from callback data */ int stonith__execution_status(stonith_callback_data_t *data) { if ((data == NULL) || (data->opaque == NULL)) { return PCMK_EXEC_UNKNOWN; } return ((pcmk__action_result_t *) data->opaque)->execution_status; } /*! * \internal * \brief Return the exit reason from an async action callback * * \param[in] data Callback data * * \return Exit reason from callback data */ const char * stonith__exit_reason(stonith_callback_data_t *data) { if ((data == NULL) || (data->opaque == NULL)) { return NULL; } return ((pcmk__action_result_t *) data->opaque)->exit_reason; } /*! * \internal * \brief Return the exit status from an event notification * * \param[in] event Event * * \return Exit status from event */ int stonith__event_exit_status(stonith_event_t *event) { if ((event == NULL) || (event->opaque == NULL)) { return CRM_EX_ERROR; } else { struct event_private *event_private = event->opaque; return event_private->result.exit_status; } } /*! * \internal * \brief Return the execution status from an event notification * * \param[in] event Event * * \return Execution status from event */ int stonith__event_execution_status(stonith_event_t *event) { if ((event == NULL) || (event->opaque == NULL)) { return PCMK_EXEC_UNKNOWN; } else { struct event_private *event_private = event->opaque; return event_private->result.execution_status; } } /*! * \internal * \brief Return the exit reason from an event notification * * \param[in] event Event * * \return Exit reason from event */ const char * stonith__event_exit_reason(stonith_event_t *event) { if ((event == NULL) || (event->opaque == NULL)) { return NULL; } else { struct event_private *event_private = event->opaque; return event_private->result.exit_reason; } } /*! * \internal * \brief Return a human-friendly description of a fencing event * * \param[in] event Event to describe * * \return Newly allocated string with description of \p event * \note The caller is responsible for freeing the return value. * This function asserts on memory errors and never returns NULL. - * \note This currently is useful only for events of type - * T_STONITH_NOTIFY_FENCE. */ char * stonith__event_description(stonith_event_t *event) { const char *origin = event->client_origin; + const char *origin_node = event->origin; const char *executioner = event->executioner; + const char *device = event->device; + const char *action = event->action; + const char *target = event->target; const char *reason = stonith__event_exit_reason(event); const char *status; + // Use somewhat readable defaults if (origin == NULL) { origin = "a client"; } + if (origin_node == NULL) { + origin_node = "a node"; + } if (executioner == NULL) { executioner = "the cluster"; } + if (device == NULL) { + device = "unknown"; + } + if (action == NULL) { + action = (event->operation == NULL)? "(unknown)" : event->operation; + } + if (target == NULL) { + target = "no node"; + } if (stonith__event_execution_status(event) != PCMK_EXEC_DONE) { status = pcmk_exec_status_str(stonith__event_execution_status(event)); } else if (stonith__event_exit_status(event) != CRM_EX_OK) { status = pcmk_exec_status_str(PCMK_EXEC_ERROR); } else { status = crm_exit_str(CRM_EX_OK); } + if (pcmk__str_eq(event->operation, T_STONITH_NOTIFY_HISTORY, + pcmk__str_none)) { + return crm_strdup_printf("Fencing history may have changed"); + + } else if (pcmk__str_eq(event->operation, STONITH_OP_DEVICE_ADD, + pcmk__str_none)) { + return crm_strdup_printf("A fencing device (%s) was added", device); + + } else if (pcmk__str_eq(event->operation, STONITH_OP_DEVICE_DEL, + pcmk__str_none)) { + return crm_strdup_printf("A fencing device (%s) was removed", device); + + } else if (pcmk__str_eq(event->operation, STONITH_OP_LEVEL_ADD, + pcmk__str_none)) { + return crm_strdup_printf("A fencing topology level (%s) was added", + device); + + } else if (pcmk__str_eq(event->operation, STONITH_OP_LEVEL_DEL, + pcmk__str_none)) { + return crm_strdup_printf("A fencing topology level (%s) was removed", + device); + } + + // event->operation should be T_STONITH_NOTIFY_FENCE at this point + return crm_strdup_printf("Operation %s of %s by %s for %s@%s: %s%s%s%s (ref=%s)", - event->action, event->target, executioner, origin, - event->origin, status, + action, target, executioner, origin, origin_node, + status, ((reason == NULL)? "" : " ("), ((reason == NULL)? "" : reason), ((reason == NULL)? "" : ")"), - event->id); + ((event->id == NULL)? "(none)" : event->id)); } // Deprecated functions kept only for backward API compatibility // LCOV_EXCL_START const char *get_stonith_provider(const char *agent, const char *provider); const char * get_stonith_provider(const char *agent, const char *provider) { return stonith_namespace2text(stonith_get_namespace(agent, provider)); } // LCOV_EXCL_STOP // End deprecated API