along with this program; if not, see <http://www.gnu.org/licenses/>.
*/
#include "includes.h"
-#include "lib/events/events.h"
+#include "lib/tevent/tevent.h"
#include "lib/tdb/include/tdb.h"
#include "system/time.h"
#include "system/network.h"
}
params->pulldata = talloc_realloc_size(NULL, params->pulldata, rec->length + params->len);
if (params->pulldata == NULL) {
- DEBUG(DEBUG_ERR,(__location__ " Failed to expand pulldb_data to %u (%u records)\n",
- rec->length + params->len, params->pulldata->count));
- params->failed = true;
- return -1;
+ DEBUG(DEBUG_CRIT,(__location__ " Failed to expand pulldb_data to %u\n", rec->length + params->len));
+ ctdb_fatal(params->ctdb, "failed to allocate memory for recovery. shutting down\n");
}
params->pulldata->count++;
memcpy(params->len+(uint8_t *)params->pulldata, rec, rec->length);
params.len = offsetof(struct ctdb_marshall_buffer, data);
params.failed = false;
+ if (ctdb_db->unhealthy_reason) {
+ /* this is just a warning, as the tdb should be empty anyway */
+ DEBUG(DEBUG_WARNING,("db(%s) unhealty in ctdb_control_pull_db: %s\n",
+ ctdb_db->db_name, ctdb_db->unhealthy_reason));
+ }
+
if (ctdb_lock_all_databases_mark(ctdb, ctdb_db->priority) != 0) {
DEBUG(DEBUG_ERR,(__location__ " Failed to get lock on entired db - failing\n"));
return -1;
{
double l = timeval_elapsed(&state->start_time);
- ctdb_reclock_latency(state->ctdb, "daemon reclock", &state->ctdb->statistics.reclock.ctdbd, l);
+ CTDB_UPDATE_RECLOCK_LATENCY(state->ctdb, "daemon reclock", reclock.ctdbd, l);
if (state->fd[0] != -1) {
state->fd[0] = -1;
ctdb_release_all_ips(ctdb);
}
+/*
+ * Set up an event to drop all public ips if we remain in recovery for too
+ * long
+ */
+int ctdb_deferred_drop_all_ips(struct ctdb_context *ctdb)
+{
+ if (ctdb->release_ips_ctx != NULL) {
+ talloc_free(ctdb->release_ips_ctx);
+ }
+ ctdb->release_ips_ctx = talloc_new(ctdb);
+ CTDB_NO_MEMORY(ctdb, ctdb->release_ips_ctx);
+
+ event_add_timed(ctdb->ev, ctdb->release_ips_ctx, timeval_current_ofs(ctdb->tunable.recovery_drop_all_ips, 0), ctdb_drop_all_ips_event, ctdb);
+ return 0;
+}
+
/*
set the recovery mode
*/
talloc_free(ctdb->release_ips_ctx);
ctdb->release_ips_ctx = NULL;
} else {
- talloc_free(ctdb->release_ips_ctx);
- ctdb->release_ips_ctx = talloc_new(ctdb);
- CTDB_NO_MEMORY(ctdb, ctdb->release_ips_ctx);
-
- event_add_timed(ctdb->ev, ctdb->release_ips_ctx, timeval_current_ofs(ctdb->tunable.recovery_drop_all_ips, 0), ctdb_drop_all_ips_event, ctdb);
+ if (ctdb_deferred_drop_all_ips(ctdb) != 0) {
+ DEBUG(DEBUG_ERR,("Failed to set up deferred drop all ips\n"));
+ }
}
if (recmode != ctdb->recovery_mode) {
char cc = 0;
close(state->fd[0]);
+ debug_extra = talloc_asprintf(NULL, "set_recmode:");
/* we should not be able to get the lock on the reclock file,
as it should be held by the recovery master
*/
talloc_set_destructor(state, set_recmode_destructor);
- DEBUG(DEBUG_NOTICE, (__location__ " Created PIPE FD:%d for setrecmode\n", state->fd[0]));
+ DEBUG(DEBUG_DEBUG, (__location__ " Created PIPE FD:%d for setrecmode\n", state->fd[0]));
state->te = event_add_timed(ctdb->ev, state, timeval_current_ofs(5, 0),
ctdb_set_recmode_timeout, state);
state->fde = event_add_fd(ctdb->ev, state, state->fd[0],
- EVENT_FD_READ|EVENT_FD_AUTOCLOSE,
+ EVENT_FD_READ,
set_recmode_handler,
(void *)state);
talloc_free(state);
return -1;
}
+ tevent_fd_set_auto_close(state->fde);
state->ctdb = ctdb;
state->recmode = recmode;
}
if (keep) {
- DEBUG(DEBUG_ERR, ("Recovery lock taken successfully\n"));
+ DEBUG(DEBUG_NOTICE, ("Recovery lock taken successfully\n"));
}
DEBUG(DEBUG_NOTICE,("ctdb_recovery_lock: Got recovery lock on '%s'\n", ctdb->recovery_lock_file));
struct recovery_callback_state *state = talloc_get_type(p, struct recovery_callback_state);
ctdb_enable_monitoring(ctdb);
+ CTDB_INCREMENT_STAT(ctdb, num_recoveries);
if (status != 0) {
DEBUG(DEBUG_ERR,(__location__ " recovered event script failed (status %d)\n", status));
+ if (status == -ETIME) {
+ ctdb_ban_self(ctdb);
+ }
}
ctdb_request_control_reply(ctdb, state->c, NULL, status, NULL);
state = talloc(ctdb, struct recovery_callback_state);
CTDB_NO_MEMORY(ctdb, state);
- state->c = talloc_steal(state, c);
+ state->c = c;
ctdb_disable_monitoring(ctdb);
- ret = ctdb_event_script_callback(ctdb,
- timeval_current_ofs(ctdb->tunable.script_timeout, 0),
- state,
+ ret = ctdb_event_script_callback(ctdb, state,
ctdb_end_recovery_callback,
- state, "recovered");
+ state,
+ false,
+ CTDB_EVENT_RECOVERED, "%s", "");
if (ret != 0) {
ctdb_enable_monitoring(ctdb);
}
/* tell the control that we will be reply asynchronously */
+ state->c = talloc_steal(state, c);
*async_reply = true;
return 0;
}
ctdb_disable_monitoring(ctdb);
- ret = ctdb_event_script_callback(ctdb,
- timeval_current_ofs(ctdb->tunable.script_timeout, 0),
- state,
+ ret = ctdb_event_script_callback(ctdb, state,
ctdb_start_recovery_callback,
- state, "startrecovery");
+ state, false,
+ CTDB_EVENT_START_RECOVERY,
+ "%s", "");
if (ret != 0) {
DEBUG(DEBUG_ERR,(__location__ " Failed to start recovery\n"));
if (ctdb->methods != NULL) {
ctdb->methods->shutdown(ctdb);
}
- ctdb_event_script(ctdb, "shutdown");
+ ctdb_event_script(ctdb, CTDB_EVENT_SHUTDOWN);
DEBUG(DEBUG_ERR, ("Recovery daemon ping timeout. Daemon has been shut down.\n"));
exit(0);
}
if (status != 0) {
DEBUG(DEBUG_ERR,(__location__ " stopped event script failed (status %d)\n", status));
ctdb->nodes[ctdb->pnn]->flags &= ~NODE_FLAGS_STOPPED;
+ if (status == -ETIME) {
+ ctdb_ban_self(ctdb);
+ }
}
ctdb_request_control_reply(ctdb, state->c, NULL, status, NULL);
ctdb_disable_monitoring(ctdb);
- ret = ctdb_event_script_callback(ctdb,
- timeval_current_ofs(ctdb->tunable.script_timeout, 0),
- state,
+ ret = ctdb_event_script_callback(ctdb, state,
ctdb_stop_node_callback,
- state, "stopped");
+ state, false,
+ CTDB_EVENT_STOPPED, "%s", "");
if (ret != 0) {
ctdb_enable_monitoring(ctdb);