Simpler Navigation for Servers and Operating Systems
Completed: a much simpler Servers and Operating Systems section of the Community. We combined many of the older boards, so you won't have to click through so many levels to get at the information you need. Check the consolidated boards here as many sub-forums are now single boards.
Showing results for 
Search instead for 
Did you mean: 

Oracle RAC w/MC-SG RAC Edition.

Insu Kim
Honored Contributor

Oracle RAC w/MC-SG RAC Edition.

Hello experts,

We're planing on configuring Oracle 9i RAC with MC-SG RAC Edition.
What we're thinking is that,
In the 2 node cluster,
we're going to load 2 instances on each server and each instance on each server will access a different database, which means there'll be two different database on the two node cluster.
Is this a valid configuration from database perspective ?
Looking for any reference.

Any input would be appreciated.
Never say "no" first.
G. Vrijhoeven
Honored Contributor

Re: Oracle RAC w/MC-SG RAC Edition.


Looks fine to me. We have clusters running oracle Per package we have a oracle installation and they all use theire own loc disks ( striped ). In this case you can fail over a complete package including software and data. Mind the oratab and tnsnames.ora it needs to be in the package not on vg00.

Honored Contributor

Re: Oracle RAC w/MC-SG RAC Edition.

My references HP is the doc B5158-900044 and 11iRACBM2, and the oracle doc for install in the attachment.

We have try to install oracle with HMP, but we have some trouble with the failover, so we use UDP instead, and it's good now.
Brian Crabtree
Honored Contributor

Re: Oracle RAC w/MC-SG RAC Edition.


The scenario that you are talking about does not sound like a RAC cluster, but a 4 database system. For a RAC database, each instance is part of the same database, with the database existing on a shared disk that is available from each node. You can then access the data from the database through any of the instances that are running on the system.

It sounds like what you are looking for is a two-node cluster, with 2 databases, and an instance running on each of the nodes for both databases (2 databases, 4 instances). This is a valid solution. You should be careful and make sure that each instance runs in its own volume groups, so you don't have to worry about bringing down a database accidentally.