September 4, 2012 at 5:20 pm
I have seen many references to performing test restores as part of a good backup and recovery plan. What I do at this point is copy my latest .bak to a testing server and then run DBCC CHECKDB.
Assuming DBCC CHECKDB returns no errors, what else can be done to ensure the restore is successful? Am I missing anything?
September 4, 2012 at 5:40 pm
They're 2 different things, really. Restoring a prod backup on a test machine, if the DB comes back online, will tell you that you have backed up the DB as it was in prod. Running DBCC on that copied database will tell you if the database you backed up was corrupted in prod when you backed it up. Not all corruption will result in a failed backup or a failed restore. You can have good backup/restore processes and still be backing up/restoring garbage.
As you are doing, you need to validate both. If you can restore the backup and DBCC comes back clean, you should be pretty confident that your database has been preserved and would come back to life in a usable state.
September 4, 2012 at 8:05 pm
David,
Thank you for your answer. So, I guess this is a case where best practice doesn't necessarily mean smooth sailing.
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply