1. Get rid of all advertisements and get unlimited access to documents by upgrading to Premium Membership. Upgrade to Premium Now and also get a Premium Badge!

Oracle ASM disk issue after Veeam recovery

Discussion in 'Server Administration and Options' started by maverick964_uk, Oct 1, 2013.

  1. maverick964_uk

    maverick964_uk Active Member

    Messages:
    2
    Likes Received:
    0
    Trophy Points:
    55
    We have an Oracle 11G ASM managed database running on Redhat 5.6 via a vmware virtual machine.

    As a test we recovered the host using Veeam B&R.

    Post recovery, the VM started but ASM wouldnt start.

    The ASM rules file (/dev/udev/....) shows the UUID of the disks its supposed to see.
    scsi_id -g -s /block/sdd etc show different UUID's.

    Does anyone know how the UUID's are assigned and I'm assuming Veeam is not the best method to recover an oracle database server.... :eek:

    We do perform nightly rman backups to disk however we do need an operating database to recover from the disk backups.

    Any help would be appreciated.

    thanks :D
     
  2. zargon

    zargon Community Moderator Forum Guru

    Messages:
    2,347
    Likes Received:
    348
    Trophy Points:
    1,430
    Location:
    Aurora, CO
    No, it isn't. The environment is virtualized, the database is not. That product is designed to backup and recover the virtual configuration NOT the database.

    I would suspect you will need to re-install ASM since it won't start.
     
  3. maverick964_uk

    maverick964_uk Active Member

    Messages:
    2
    Likes Received:
    0
    Trophy Points:
    55
    yes its true that Veeam is not designed to recover an oracle database. At present Veeam backs up the host but the DBA also runs a nightly rman script which dumps the backup to disk. Veeam will then hoover that up as part of the snapshot backup.

    I'm trying to work out why the UUIDs are changing

    We are using scsi_id -g -s on the sdd sde dev files etc (on all the disks) to check the UUIDs

    The rules file shows different UUIDs. Also we recovered the VM twice and the UUIDs reported from scsi_id are different. Something is changing them. Maybe its vmware?

    So its understanding what should be the backup and recovery process for Oracle ASM on a virtual machine

    many thanks
     
  4. zargon

    zargon Community Moderator Forum Guru

    Messages:
    2,347
    Likes Received:
    348
    Trophy Points:
    1,430
    Location:
    Aurora, CO
    There are releases of Redhat where such issues exist and they exist outside of the virtualization platform. Possibly you should move to a current version of Redhat.