On Sun, 2012-12-02 at 01:45 +0500, Terry Khatri wrote: > > I need to do the subject setup using PostgreSQL, where when one node > goes down another takes over without manual intervention, the way it > is done in Oracle using grid infrastructure and in SQL Server 2012 > using their Alway-On feature
This is done using Red Hat Cluster Suite (RHCS). It is the "High Availability Addon" in Red Hat Enterprise Linux, and I recommend you (though not strictly required) to use GFS2 along with RHCS
has the overview of the infrastructure. Step-by-step installation might be too long for a mailing list, but the idea is the same as installing anything under RHCS -- so any tutorial will help you that describes how to configure a clustered service in RHCS. Please note that this is not an active-active system as you wrote above, so those clusters won't apply this. Another tip would be keeping lock file and pid in the shared storage, not on the local disks. You will need to adjust init scripts for that, too. There is not too much manual work btw.
Oh, that reminds me -- RHCS is currently using cman+corosync+rgmanager, so any solutions with pacemaker will work, but won't be supported by Red Hat (though AFAIK RHEL 7 will use pacemaker by default)
Feel free to use luci and ricci for configuring the services, they will help you a lot. Luci is the web interface for configuring cluster, ricci is the agent on each node.
