[Pacemaker] CRM resource restarts on same machine

[Andrew Beekhof]

On Wed, Apr 25, 2012 at 2:27 AM, Sucheta <sucheta at ee.iitb.ac.in> wrote:
>
>
> ---------- Forwarded message ----------
> From: Sucheta <sucheta at ee.iitb.ac.in>
> Date: Tue, Apr 24, 2012 at 9:20 PM
> Subject: CRM resource restarts on same machine
> To: mailman at oss-us-1.clusterlabs.org
>
>
>
> Hello,
>
>
> HA configuration is as follows--
> 1. DRBD 8.4.1
> 2. Pacemaker 1.0
> 3. Openais
>
> 1. failover and failback are happening between the nodes upon making standby
> on/off.
>
> Following are the issues:
>
> 1. jboss crm_resource restarts on the same node upon failure or killed
> manually. The expected behavior should be, all the resources should
> fail-over onto node-2, which is not happening.

Actually this is not the expected behaviour.
You would need to configured migration-threshold=1 which tells the
cluster to stop trying to recover the resource on the failed node.

> 2. Any resource failure on node-1 should move all the resources onto node-2
> and vice-versa.
>
>
> Attached file has the CRM configuration of 2 Virtual machines and  2
> physical machines.
>
> Please guide.
>
> --
> Sucheta
>
>
> _______________________________________________
> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>
> Project Home: http://www.clusterlabs.org
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> Bugs: http://bugs.clusterlabs.org
>