Please help to put 2 Linux/Oracle VMs into a H/A cluster

Before providing new release of our product to a customer I would like to test it in the conditions it could face on a production environment. So, I'm thinking how to create/configure a highavailability  cluster of 2 Linux/Oracle VMs.

Thus, as a source we have:
1) VmWare VirtualMachine with CentOS Linux 5.4 x86 and Oracle 11.2 Enterprise installed on it. Static IP=192.168.18.91, hostname=centosvm01.
2) the same VirtualMachine created as a copy of 1st one with only the diff = Static IP=192.168.18.92, hostname=centosvm02. Note: if need more than 2 VMs for such task - ok, I'm ready to create one more. :)

Both VMs are working fine - ping, tnsping, connectivity from client computers - all working ok. I have checked that Oracle on both VMs is up and running after 2nd VM was created as a clone of 1st VM.

The question is - what is next?!
How to put these 2 VMs into a high-availability cluster?

I mean - the Oracle database high-availability cluster, so I expect that client computers (also - some application on cluster nodes) connect Oracle database using the clustrer IP address, if one node shutdown client applications continue working and do not know anything about problems with particular node, when node is up and running again it has up-to-date data in a database.

I have browsed lot of resources in the internet but still have not found any description of concrete steps what to do setup/configure to have a high-availability cluster.
According to a resources I have read the Oracle 11.2 should be able to do it, seems it should have all required components included but .. how?! :-)

Please help me.
Would be nice to see something like a "Oracle H/A Cluster QuickStart Guide", etc.

Regards,
Dmitry.

PS. In couple of month customer going to buy a new Linux/Oracle H/W. But "couple of month" that is quite a long and undefined term and we cannot wait. We need some H/A cluster to test our software just now! Received on Sat Jul 03 2010 - 02:02:41 CDT