Hi,

It would help to write a subject too.

On Tue, Oct 09, 2007 at 01:59:34PM +0530, Sripathi, Roopa (Roopa) wrote:
> Hi,
> 
> I see two uuids in cib.xml for the same node.
> 
> And heartbeat is treating them as 3 nodes, as seen in ptest.
> 
> How did heartbeat end up in this situation and how to avoid this.
> 
> 
> ptest[6836]: 2007/09/30_12:53:22 debug: native_assign_node: Color
> IPaddr_cluster, Node[0] roopa2: 0
> ptest[6836]: 2007/09/30_12:53:22 debug: native_assign_node: Color
> IPaddr_cluster, Node[1] roopa1: 0
> ptest[6836]: 2007/09/30_12:53:22 debug: native_assign_node: Color
> IPaddr_cluster, Node[2] roopa1: 0
> ptest[6836]: 2007/09/30_12:53:22 debug: native_assign_node: Assigning
> roopa2 to IPaddr_cluster
> 
> 
> 
> Part of cib.xml is below >>>>>>>>>>>>>>
> 
> <nodes>
>        <node id="9bb16fe3-bc24-4e85-9abf-865605cb37fc" uname="roopa2"
> type="normal"/>
>        <node id="30f7e2e3-a704-403b-a9f9-9b8c62740271" uname="roopa1"
> type="normal"/>
>        <node id="993fed4a-cf1c-4a45-b561-2668b7baff05" uname="roopa1"
> type="normal"/>
>      </nodes>

Hmpf. Yes, the node id is what is actually used by the cluster,
that's why it reports three nodes. I guess that it happened in
this way: /var/lib/heartbeat/hb_uuid on roopa1 was removed so it
got a new id (that's generated using png).

Stop the cluster and empty the nodes section (reduce it to
<nodes/>). Remove /var/lib/heartbeat/hostcache on both nodes.
Then start the cluster.

Thanks,

Dejan


> 
> Thanks,
> 
> 
> Roopa Sripathi
> _______________________________________________
> Linux-HA mailing list
> [email protected]
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
_______________________________________________
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