Troubleshooting Reverse Resynchronization

Before preparing a preview session for a backup database instance the Reverse Resync Begin operation always creates a recovery point. If this recovery point is lost, then you cannot transit from the Reverse Resync Begin state.

The following steps can help in troubleshooting the instance and let you transit to the Reverse Resync Begin state:

  1. Make sure to stop the Oracle database instance using SQL > Shutdown immediate.
  2. Unmount all the snapshot volumes. The snapshot volume names are prefixed with a snapshot prefix as specified in the snapshot configuration file. The default value of the prefix is "SNAP_".

    For example:

    # umount /data/prodDB 
  3. Deport snapshot disk group(s). The snapshot disk group names are prefixed with a snapshot prefix as specified in the snapshot configuration file.

    For example:

    # vxdg deport SNAP_proddg
  4. Mount the primary data volumes.

    For example:

    #mount -t vxfs /dev/vx/dsk/proddg/dbvol1 /data/prodDB
  5. Restore the backup repository to switch to a previous state.

    The RRBegin operation automatically backs up the SFDB repository.

    [oracle@testbox ~]$ sfua_rept_util -S $ORACLE_SID -o list
    Listing Repository Backups for SID='RR'
    TIME                    SIZE            NAME
    2013-07-30-13-43-55     18160           rept_RR_2013-07-30-13-43-55.tgz
    2013-07-30-16-16-25     18159           rept_RR_2013-07-30-16-16-25.tgz
    2013-07-30-16-16-49     18161           rept_RR_2013-07-30-16-16-49.tgz
    2013-07-30-16-17-56     18163           rept_RR_2013-07-30-16-17-56.tgz
    2013-07-30-16-18-01     18163           rept_RR_2013-07-30-16-18-01.tgz
    2013-07-30-16-18-07     18161           rept_RR_2013-07-30-16-18-07.tgz
    
    [oracle@testbox ~]$ sfua_rept_util -S $ORACLE_SID -o restore -f
    rept_RR_2013-07-30-16-18-07.tgz
    Restoring repository for SID='RR'
    Target directory='/etc/vx/vxdba/logs/oracle/backup'
    Restore Repository for SID='RR' Completed