Hi, I’m diagnosing a problem where monitors fall out of quorum now and then. It seems that when two monitors do a new election, one answer is not received until 5 minutes later. I checked ntpd on the servers, and all of them are spot on, no sync problems. This is happening a couple of time every day now, with all of the mons being the one not answering in due time. This is from the osd12-logs, the same pattern is repeated on the others: 2016-05-17 08:20:55.851276 mon.1 10.168.7.32:6789/0 157491 : cluster [INF] mon.osd12 calling new monitor election 2016-05-17 08:20:56.750915 mon.0 10.168.7.31:6789/0 4179082 : cluster [INF] mon.osd11 calling new monitor election 2016-05-17 08:21:02.709111 mon.0 10.168.7.31:6789/0 4179083 : cluster [INF] mon.osd11@0 won leader election with quorum 0,1 2016-05-17 08:20:58.916940 mon.2 10.168.7.33:6789/0 157323 : cluster [INF] mon.osd13 calling new monitor election 2016-05-17 08:21:03.931656 mon.0 10.168.7.31:6789/0 4179090 : cluster [INF] mon.osd11 calling new monitor election 2016-05-17 08:21:03.933038 mon.1 10.168.7.32:6789/0 157495 : cluster [INF] mon.osd12 calling new monitor election 2016-05-17 08:21:03.940032 mon.0 10.168.7.31:6789/0 4179091 : cluster [INF] mon.osd11@0 won leader election with quorum 0,1,2 Anyone who have had similar problems? We’re about to upgrade to latest jewel, this is ceph version 9.2.1 (752b6a3020c3de74e07d2a8b4c5e48dab5a6b6fd). Thanks for any help, Josef _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |