Determining Backup Pieces needed for Point-in-Time Recovery
586595Oct 29 2008 — edited Oct 29 2008I'm running on v10.2.0.4 on a Sun Solaris system. I have a backup strategy that takes a incremental level 0 on Sunday's and level 1's for the remainder of the week. The problem that we run into quite often, is trying to determine all the backup pieces that are needed in order to restore the database to a particular point-in-time. We have lots and lots of databases of varying sizes. This particular is fairly big, about 400 GB.
Normally, we have about 12 pieces created during a level 0 and somewhere around 8 - 12 pieces during each level 1, depending on activity.
We just did a restore for Tuesday at 7:15 AM. In order to accomplish this we needed to bring over all the level 0 backup pieces from Sunday, all the level 1 pieces from Monday, all the level 1 pieces from Tuesday, and the necessary archive log up until 7:15 AM on Tuesday. So for this particular recovery we needed to identify 32 individual backup pieces and three archive logs, create a script, and move them to our development server.
We normally handle this by reviewing the LIST BACKUP output and the log files generated during the backups to get the names of these backup pieces.
What I would like to know is if there is a better way to identified what pieces are needed for a Point-in-Time Recovery. The RESTORE DATABASE VALIDATE will go through all the pieces but I'm really not interested in waiting for it to go through all the pieces. I just want a complete list quickly.
Any suggestions???