From 76ee5f604c9adaaa7392842cfa2753bccb071e0f Mon Sep 17 00:00:00 2001 From: Ronnie Sahlberg Date: Fri, 4 Mar 2011 06:55:24 +1100 Subject: [PATCH] Restart recovery dameon if it looks like it hung. Dont shutdown ctdbd completely, that only makes the problem worse. --- server/ctdb_recover.c | 12 ++---------- 1 file changed, 2 insertions(+), 10 deletions(-) diff --git a/server/ctdb_recover.c b/server/ctdb_recover.c index 0cbd7dc3..eb3bf0a5 100644 --- a/server/ctdb_recover.c +++ b/server/ctdb_recover.c @@ -1197,18 +1197,10 @@ static void ctdb_recd_ping_timeout(struct event_context *ev, struct timed_event return; } - DEBUG(DEBUG_ERR, ("Final timeout for recovery daemon ping. Shutting down ctdb daemon. (This can be caused if the cluster filesystem has hung)\n")); + DEBUG(DEBUG_ERR, ("Final timeout for recovery daemon ping. Restarting recovery daemon. (This can be caused if the cluster filesystem has hung)\n")); ctdb_stop_recoverd(ctdb); - ctdb_stop_keepalive(ctdb); - ctdb_stop_monitoring(ctdb); - ctdb_release_all_ips(ctdb); - if (ctdb->methods != NULL) { - ctdb->methods->shutdown(ctdb); - } - ctdb_event_script(ctdb, CTDB_EVENT_SHUTDOWN); - DEBUG(DEBUG_ERR, ("Recovery daemon ping timeout. Daemon has been shut down.\n")); - exit(0); + ctdb_start_recoverd(ctdb); } /* The recovery daemon will ping us at regular intervals. -- 2.34.1