Hi, On 11.05.2012, at 15:39, Steven Dake wrote: > On 05/11/2012 01:10 AM, Jan Friesse wrote: >> >> Ok, >> then question is, why write blocks? > > Honza, > > I can't think of any bugzillas that were addressed related to a problem > like this. I had to restart both cluster nodes last night and currently the problem does not occur. Since the monitoring check (i.e. the "corosync-cfgtool -s" call) is run via cron on both machines at the same time (= every three minutes), could there be some kind of race condition? It still feels very strange that this didn't happen in the last ~12 months and then suddenly multiple times a day. Also, the few times the problem occured, were all on the same machine, so I wouldn't rule out some kind of (strange) hardware defect yet. If there's anything I should look out for when the problem appears again: I'm open to suggestions. -- Sebastian Kaps _______________________________________________ discuss mailing list discuss@xxxxxxxxxxxx http://lists.corosync.org/mailman/listinfo/discuss
