April 29, 2002 at 5:30 am
I'm just checking out filegroups and am not sure how the restore methodology works - perhaps someone can explain it to me. Suppose you have a full database backup, log1 backup, filegroup1 backup, log2 backup and filegroup2 backup: full->log1->fg1->log2->fg2. In this case the system goes wrong before fg2 and you want to restore. When you restore log2, the transactions may apply to the tables and indexes of fg1 or fg2, however fg1 has already been restored, while fg2 needs all the transactions to be applied. Does the restore option take care of this internally? Or possibly filegroups can only be restored as a whole , ie fg1 and fg2 are both needed? Hope this is clear!
Paul Ibison
Paul Ibison
Paul.Ibison@replicationanswers.com
April 29, 2002 at 5:57 am
I do believe the log files verify commited transactions so that a file can be restored idenpendently. However I have not tested and just cannot do so right now. Test on another machine and see is the best way to be sure.
"Don't roll your eyes at me. I will tape them in place." (Teacher on Boston Public)
May 9, 2002 at 12:09 am
There is a kb article explaining the use and limitations of filegroup backups.
Look at Q281122 on MS Technet.
You would be better off using Differential backups between Full backups.
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply