Re: 11gR2 San Migration - OCR Disk Group

  • From: Andy Colvin <acolvin@xxxxxxxxxxx>
  • To: oracle-l <oracle-l@xxxxxxxxxxxxx>
  • Date: Fri, 6 Apr 2012 10:25:21 -0500

I haven't tried with the OCR component, but I don't see why it wouldn't work.  
I have checked the behavior WRT voting disks.  Assuming that you have normal or 
high redundancy on your OCR/voting diskgroup, when one fail group containing a 
voting disk is lost and there is another fail group in the diskgroup that 
doesn't contain a voting disk, the missing voting disk will be created in that 
unused fail group.
For example, if you have a diskgroup with 4 fail groups (FG1, FG2, FG3, and 
FG4) in normal redundancy, assume that voting disks are in FG1, FG2, and FG3.  
If fail group 3 was dropped, a voting disk will automatically be created in 
FG4.  ASM will not wait for the disk_repair_time or a rebalance.  

Here's the description from Oracle's documentation ( - "If 
voting disks are stored on Oracle ASM with normal or high redundancy, and the 
storage hardware in one failure group suffers a failure, then if there is 
another disk available in a disk group in an unaffected failure group, Oracle 
ASM recovers the voting disk in the unaffected failure group."

Here's a post that I wrote about voting disk redundancy in ASM -

Andy Colvin

Principal Consultant

On Apr 6, 2012, at 9:35 AM, Jeff Thomas wrote:

> Running RAC on Linux x86.     We are migrating to a new SAN.
> With respect to the OCR diskgroup, which is defined with 3 disks with
> normal redundancy,  I do know how to migrate the OCR/voting files/ASM
> spfile to a new OCR diskgroup if need be.
> What I wonder if anyone has migrated the OCR diskgroup by simply adding the
> new disks and dropping the old ones?    Is Oracle smart enough
> to copy the 3 voting files over to the new disks, so that's there no outage
> whatsoever when migrating to the new SAN?
> Best,
> Jeff
> --


Other related posts: