Re: Setting up High Availability Group of PostgreSQL instances using RedHat Cluster Suite

Hi,

On Sun, 2012-12-16 at 03:25 +0500, Terry Khatri wrote:

> I am having an issue with postgres service on redhat cluster suite
> (Rehl6), the service gets started on 2 nodes out of 3, is that the
> right behaviour ?

Depends.

> I would like it to be restricted to 1 node at a time becuase I am
> also doing replication and when the service gets started on 2 nodes
> the slaves get stuck as the masters on both nodes generate their own
> wall segments.in the same folder on shared san disk on GFS2

This is a bit strange. If they are using the same $PGDATA, how can they
start at the same time?

I'd start with setting ordered=1 and restricted=0 in cluster.conf. Also
make sure that you have disabled both postgresql init scripts to start
on boot.

Let me know if these don't help.

Regards,
--
Devrim GÜNDÜZ
Principal Systems Engineer @ EnterpriseDB: http://www.enterprisedb.com
PostgreSQL Danışmanı/Consultant, Red Hat Certified Engineer
Community: devrim~PostgreSQL.org, devrim.gunduz~linux.org.tr
http://www.gunduz.org Twitter: http://twitter.com/devrimgunduz