[Pacemaker] Resource attempting to start on unauthorized node causing running master/slave to shut down

Mike Roest msroest at gmail.com
Thu Jun 21 16:57:35 CEST 2012


> WARN: unpack_rsc_op: Processing failed op Postgres-Server-1:0_monitor_0

> > on dbquorum.example.com: unknown error (1)
> > WARN: unpack_rsc_op: Processing failed op Postgres-IP-1_monitor_0
> > on dbquorum.example.com: unknown error (1)
>
> Under such conditions, pacemaker must assume that the resources are
> active and initiates recovery.
> Your real question should be, why did the monitor op fail with rc=1
> (instead of rc=7) for those two resources on dbquorum?
>
> Interesting our custom monitor script for the Postgres-Server-1 is not
installed on dbquorum as it doesn't have postgres so we never put it on
there.  I'm surprised that a missing monitor script would return rc=1
rather then rc=2 or rc=5.

As for the Postgres-IP we're using the standard IPaddr2 so it's really
strange that it would return the generic error rather then rc=2.

--Mike
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://oss.clusterlabs.org/pipermail/pacemaker/attachments/20120621/09e0c885/attachment.html>


More information about the Pacemaker mailing list