/*
see if the event scripts think we are healthy
*/
static void ctdb_check_health(struct event_context *ev, struct timed_event *te,
struct timeval t, void *private_data)
{
struct ctdb_context *ctdb = talloc_get_type(private_data, struct ctdb_context);
int ret = 0;
if (ctdb->recovery_mode != CTDB_RECOVERY_NORMAL ||
(ctdb->monitor->monitoring_mode == CTDB_MONITORING_DISABLED && ctdb->done_startup)) {
event_add_timed(ctdb->ev, ctdb->monitor->monitor_context,
timeval_current_ofs(ctdb->monitor->next_interval, 0),
ctdb_check_health, ctdb);
return;
}
if (!ctdb->done_startup) {
ret = ctdb_event_script_callback(ctdb,
ctdb->monitor->monitor_context, ctdb_startup_callback,
ctdb, false,
CTDB_EVENT_STARTUP, "%s", "");
} else {
int i;
int skip_monitoring = 0;
if (ctdb->recovery_mode != CTDB_RECOVERY_NORMAL) {
skip_monitoring = 1;
DEBUG(DEBUG_ERR,("Skip monitoring during recovery\n"));
}
for (i=1; i<=NUM_DB_PRIORITIES; i++) {
if (ctdb->freeze_handles[i] != NULL) {
DEBUG(DEBUG_ERR,("Skip monitoring since databases are frozen\n"));
skip_monitoring = 1;
break;
}
}
if (skip_monitoring != 0) {
event_add_timed(ctdb->ev, ctdb->monitor->monitor_context,
timeval_current_ofs(ctdb->monitor->next_interval, 0),
ctdb_check_health, ctdb);
return;
} else {
ret = ctdb_event_script_callback(ctdb,
ctdb->monitor->monitor_context, ctdb_health_callback,
ctdb, false,
CTDB_EVENT_MONITOR, "%s", "");
}
}
if (ret != 0) {
DEBUG(DEBUG_ERR,("Unable to launch monitor event script\n"));
ctdb->monitor->next_interval = 5;
event_add_timed(ctdb->ev, ctdb->monitor->monitor_context,
timeval_current_ofs(5, 0),
ctdb_check_health, ctdb);
}
}
static void ctdb_run_startup(struct event_context *ev, struct timed_event *te,
struct timeval t, void *private_data)
{
struct ctdb_context *ctdb = talloc_get_type(private_data,
struct ctdb_context);
int ret;
/* This is necessary to avoid the "startup" event colliding
* with the "ipreallocated" event from the takeover run
* following the first recovery. We might as well serialise
* these things if we can.
*/
if (ctdb->runstate < CTDB_RUNSTATE_STARTUP) {
DEBUG(DEBUG_NOTICE,
("Not yet in startup runstate. Wait one more second\n"));
event_add_timed(ctdb->ev, ctdb->monitor->monitor_context,
timeval_current_ofs(1, 0),
ctdb_run_startup, ctdb);
return;
}
DEBUG(DEBUG_NOTICE,("Running the \"startup\" event.\n"));
ret = ctdb_event_script_callback(ctdb,
ctdb->monitor->monitor_context,
ctdb_startup_callback,
ctdb, CTDB_EVENT_STARTUP, "%s", "");
if (ret != 0) {
DEBUG(DEBUG_ERR,("Unable to launch startup event script\n"));
event_add_timed(ctdb->ev, ctdb->monitor->monitor_context,
timeval_current_ofs(5, 0),
ctdb_run_startup, ctdb);
}
}
/*
see if the event scripts think we are healthy
*/
static void ctdb_check_health(struct tevent_context *ev,
struct tevent_timer *te,
struct timeval t, void *private_data)
{
struct ctdb_context *ctdb = talloc_get_type(private_data, struct ctdb_context);
bool skip_monitoring = false;
int ret = 0;
if (ctdb->recovery_mode != CTDB_RECOVERY_NORMAL ||
ctdb->monitor->monitoring_mode == CTDB_MONITORING_DISABLED) {
skip_monitoring = true;
} else {
if (ctdb_db_all_frozen(ctdb)) {
DEBUG(DEBUG_ERR,
("Skip monitoring since databases are frozen\n"));
skip_monitoring = true;
}
}
if (skip_monitoring) {
tevent_add_timer(ctdb->ev, ctdb->monitor->monitor_context,
timeval_current_ofs(ctdb->monitor->next_interval, 0),
ctdb_check_health, ctdb);
return;
}
ret = ctdb_event_script_callback(ctdb,
ctdb->monitor->monitor_context,
ctdb_health_callback,
ctdb, CTDB_EVENT_MONITOR, "%s", "");
if (ret != 0) {
DEBUG(DEBUG_ERR,("Unable to launch monitor event script\n"));
ctdb->monitor->next_interval = 5;
tevent_add_timer(ctdb->ev, ctdb->monitor->monitor_context,
timeval_current_ofs(5, 0),
ctdb_check_health, ctdb);
}
}
