Re: ASM "disk" replacement

Depending upon the criticality / paranoia, comfort level and the current (disk discovery) stage, you could use either some or all of the following steps :

1. Saving the mapping between your ASM devices and associated diskgroups OR all possible asm metadata information.
2. Taking a backup of objects/tablespaces (if at all applicable) that belong to the particular diskgroup where devices will be dropped/added.
3. Have action plan (from SAs/StorageAdmins) ready in case visibility of the disk(s) becomes an issue. Any udev/scsi commands/asmlib/san/lun/rescan etc commands.
4. If critical/hightly accessed objects are residing in the diskgroup where some devices will be reshuffled, have another diskgroup ready with the necessary objects where application can be quickly pointed to.
5. Be ready for (in extreme circumstances) one or two rounds of asm instance bounce in case of disk not being in proper state even after being discovered.

If I think of something else, I will let you know.

Best Regards,
Rajeev

On Thu, May 7, 2009 at 3:57 PM, Don Granaman <DonGranaman_at_solutionary.com> wrote:
> We are using ASM for 10.2.0.4 RAC on Linux x86-64 and will need to replace a
> number of disks in the storage array.
>
> I am wondering if anyone here has had any “interesting” experience with this
> sort of thing – good or bad.

--
http://www.freelists.org/webpage/oracle-l