How to manage OCR and voting disk redundancy in prod env.

Questions relating to Oracle Real Application Clusters (RAC) and Clusterware.

Moderator: Tim...

How to manage OCR and voting disk redundancy in prod env.

Postby patrick196901 » Mon May 14, 2012 3:14 pm

Hi Tim,

I am contacting you as I'd like to get the advice of a RAC guru and for me you are one.

I am in the process to start the installation of Oracle Grid Infrastructure 11g R2 and RAC databases on two new Oracle Sun T4 server, each server located in different sites. I want to use ASM in order to host the OCR and voting disk but I'd like to know if it is better to use external redundancy instead of high redundancy when creation the diskgroup that will host the OCR and voting disk.

To explain you, on each site we have one T4, and also one SAN. So each T4 will use LUNs on its own SAN, and both SAN will be interconnected together as our storage admin explained to me last week.

So when installing the GI infrastructure disks candidates will be presented to the ASM. Would you advise me to use the external redundancy in order to manage the OCR and voting disk, and so to let the SAN look after the ASM diskgroup in case of failure, or would you advise me to use the high redundancy and so to add more security to the system in case of failure.

In fact when installing the GI release 11gR2 in production what is the best solution when managing the OCR and voting disk, external redundancy or high redundancy?

Thanks in advance for your answer.

Regards

Patrick
patrick196901
Member
 
Posts: 13
Joined: Thu Oct 27, 2011 10:53 am

Re: How to manage OCR and voting disk redundancy in prod env

Postby Tim... » Mon May 14, 2012 3:56 pm

Hi.

1) The success of this installation will be totally down to the speed of the network between the two SANs. If that is not fast enough the setup will suck.

2) You should check your stretch cluster configuration is supported by Oracle.

3) You will need a third voting disk, on an independent storage system. Preferably on something like an NFS mount and a different server.

4) In general, Oracle suggest using external redundancy wherever possible, since hardware RAID is typically more efficient than ASM mirroring, so from that perspective, I would use external redundancy, but this doesn't stop you having multiple voting disks and OCR locations, to protect against failure associated with losing a file. Let's say you have two disk groups (FRA and DATA), then you can have an OCR location and a voting disk on each. If you plan to have a normal or high redundancy disk group just for OCR and voting disk, then that's fine also.

Check these out:

http://docs.oracle.com/cd/E11882_01/rac ... m#BGBJEBDJ
http://kevinclosson.wordpress.com/2008/ ... tup-right/

Cheers

Tim...
Tim...
Oracle ACE Director
Oracle ACE of the Year 2006 - Oracle Magazine Editors Choice Awards
OakTable Member
OCP DBA 7.3, 8, 8i, 9i, 10g, 11g
OCP Advanced PL/SQL Developer
Oracle Database: SQL Certified Expert
My website: http://www.oracle-base.com
My blog: http://www.oracle-base.com/blog
Tim...
Site Admin
 
Posts: 17966
Joined: Mon Nov 01, 2004 5:56 pm
Location: England, UK


Return to Oracle RAC and Clusterware

Who is online

Users browsing this forum: No registered users and 1 guest

cron