I have a two node cluster running Ubuntu 12.04 and had everything working
just fine (STONITH, failover, etc.) until I changed up the virtual machines
and the crm configuration to match (my virtual machines get booted by
pacemaker on the survivor node when a failover occurs).  The primary node
currently has a status of "UNCLEAN (online)" as it tried to boot a VM that
no longer existed - had changed the VMs but not the crm configuration at
this point.  I have since modified the configuration and synced data with
DRBD so everything is good to go except for pacemaker.

Is there a way to remove the error and set the UNCLEAN node to just
online?  I think since it's currently seen as unclean, the new
configuration won't propagate to the secondary node.  In order to make
ensure high availability, both machines need to be clean, online, and have
the same crm configuration.

I'm hoping there's a quick way to get this cluster back on track.  I doubt
one of my servers are going to fail any time soon, but you never know.

Thanks in advance,

Josh
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Reply via email to