Morning folks (morning where I am), I just found out that during the same time frame, mysql reported via monit that there were too many connections and could not connect to MySQL server over the VIP.
Will the scenario of having the mysql clogging up the VIP, cause the heartbeat to repetitively keep failing? Though.. in the logs.. I only get this entry as a sign its shutting down: heartbeat[30080]: 2010/04/22_12:21:36 info: Heartbeat shutdown in progress. (30080) The entries after this are normal shutting down procedures. Thanks! Daren On Wed, May 19, 2010 at 2:46 AM, DarenTay <[email protected]> wrote: > Hi there, > > am new here, so please pardon me if I said anything wrong :) > > I have 2 nodes (RHEL 4.5). > Running MySQL 5.0.45, Heartbeat 2.1.2 (but using v1 config) > node #1 is primary, node #2 is secondary. auto failback is off. > > > *Part of ha.cf* > ucast eth1 192.168.32.111 > ucast eth1 192.168.32.112 > ucast eth2 10.10.10.1 > ucast eth2 10.10.10.2 > ucast eth3 10.10.20.1 > ucast eth3 10.10.20.2 > auto_failback off > deadping 30 > ping 192.168.32.254 > respawn hacluster /usr/lib64/heartbeat/ipfail > respawn hacluster /usr/lib64/heartbeat/dopd > apiauth dopd gid=haclient uid=hacluster > > *resources* > me.domain.com IPaddr::192.168.32.113/24 \ > drbddisk::all \ > Filesystem::/dev/drbd0::/mysql/bin/logs::ext3 \ > Filesystem::/dev/drbd1::/mysql/redo::ext3 \ > Filesystem::/dev/drbd2::/mysql/data::ext3 \ > mysql > > *!!Problem*!! > > A while back, heartbeat detected a shutdown in node1 and initiate a > failover. > Shortly after (10-20min), node2 detects a shutdown and node1 takes over > again. > This goes on for a few days... the VIP keeps rolling over, irritating my > users as their backup is dependent on the assigned node1. > > What should I provide to get better help? > The logs show nothing useful except log entries saying the other party has > shutdown down and it is taking over. But nothing leading up to the rapid > exchange of primary host role. > Am at my wits end. > > If anyone got a suggestion as to how I can start trouble shooting it... I'm > all ears. > > Thanks! > Daren > _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
