The restore selections that you choose in Backup Exec are converted to a script. RMAN uses the script to determine what to restore from the Backup Exec media. After the data has been restored to the Oracle server, RMAN completes any requested recovery and restore operations. These recovery and restore operations are determined by the options that you select.
See Restoring from a legacy GRFS Oracle Agent database backup.
Some recovery operations may not require media from the media server. For example, the redo logs may still be on the Oracle server. During a restore operation, the amount of data that is restored may not be equal to the amount of data that is backed up. In some cases, the amount of data that is restored is listed as 0 bytes. This behavior is normal because Oracle might skip data files that are already up-to-date on the disk.
If you perform a complete recovery on the whole database, or on a tablespace or datafile, you must restore a backup of the database or files that you want to recover. Then you must apply online or archived redo logs, or both. For jobs that are initiated both from the media server and from a DBA, RMAN determines the specific data that it requires from Backup Exec to complete the restore and the recovery that you request.
|
Note: |
Backup Exec does not support Oracle tablespace point-in-time restore (TSPITR) through server-initiated operations. |
You can only choose Oracle restore selections from the View by Resource tab on the Restore Job Properties dialog box. The View by Media tab displays backup sets, but you cannot browse or select the contents.
On the View by Resource tab, you can make restore selections from the online database or from control files.
Table: Restore selections for Oracle resources