VERITAS Storage Foundation 4.1 Oracle Administrator's Guide

Chapter 10, Using Database FlashSnap for Backup and Off-Host Processing
Prerelease 8 September 2005, 8:55am Resynchronizing Your Database to the Snapshot
253
▼
To begin reverse resynchronization
The -o reverse_resync_begin option of the dbed_vmsnap command imports the disk
group that was deported from the secondary host (in a two-host configuration) and joins it back to
the original disk group. The command unmounts the original volumes, mounts the snapshot
volumes with the file systems that are configured for the primary database, and brings up the
database snapshot image as the primary database. This operation requires the primary database to
be offline so that its contents remain unchanged.
To begin reverse resynchronization, use the -o reverse_resync_begin option as follows:
$ /opt/VRTS/bin/dbed_vmsnap -S ORACLE_SID -f SNAPPLAN \
-o reverse_resync_begin
Note Any mounted storage checkpoints must be unmounted before running dbed_vmsnap -o
reverse_resync.
After executing reverse_resync_commit, checkpoints created on the original database or
clone database will be deleted.
Limitation
Reverse resynchronization requires the primary database to be down. However, in an HA
environment, when VCS detects that the primary database is down, it starts the fail over process,
VxDBA repository gets unmounted, dbed_vmsnap dies.
Use the following workaround:
1. To temporarily freeze the VCS Resource Group for the database, enter:
# hagrp -freeze ResourceGroup
2. Shutdown the primary database.
3. Run reverse_resync.
4. When reverse_resync is completed, start up the database. Make sure it is in
archivelog mode.
5. To unfreeze the Resource Group, enter:
# hagrp -unfreeze ResourceGroup