[Pacemaker] coronosyc 1.2.1 with pacemaker and openais is suse11sp1

Timi aliaj00 at gmail.com
Wed Apr 15 12:06:59 UTC 2015


Hi guys,

we have a cluster setup with: coronosyc 1.2.1 with pacemaker and openais is
suse11sp1 on two nodes connected via direct cable for heartbeat, we checked
the connection and its ok.

we are having this on the logs:

12:23] <[1]TT> Apr 03 03:40:52 flexy-db-1 lrmd: [9285]: info:
perform_op:2894: postponing all ops on resource floating-ip by 1000 ms
[12:23] <[1]TT> Apr 03 03:40:52 flexy-db-1 lrmd: [9285]: debug:
rsc:floating-ip:112: monitor
[12:23] <[1]TT> Apr 03 03:40:52 flexy-db-1 lrmd: [23676]: debug:
perform_ra_op: resetting scheduler class to SCHED_OTHER
[12:23] <[1]TT> Apr 03 03:40:53 flexy-db-1 lrmd: [9285]: debug:
rsc:internal-floating-ip:115: monitor
[12:23] <[1]TT> Apr 03 03:40:53 flexy-db-1 lrmd: [23684]: debug:
perform_ra_op: resetting scheduler class to SCHED_OTHER
[12:23] <[1]TT> Apr 03 03:40:56 flexy-db-1 lrmd: [9285]: debug:
rsc:floating-ip:35: monitor
[12:23] <[1]TT> Apr 03 03:40:56 flexy-db-1 lrmd: [23698]: debug:
perform_ra_op: resetting scheduler class to SCHED_OTHER
 

this happens frequently and we believe it causes instability issues as we
sometimes after these logs the:

corosync [TOTEM ] Process pause detected for 5370 ms, flushing membership
messages.



and there it starts to panic and causes the msql to restart each 5 min.

Has anyone any idea what might be causing this.





More information about the Pacemaker mailing list