CirquedeSQLeil (4/29/2010)
The part that gets me with these questions is if the hardware failure was enough to corrupt the database and the full backup - it would seem to me that you would be unable to perform a tail log backup. After all the failure had to be severe enough that it was able to corrupt files on different drives / luns. Thus I always have to step back and think a little more about it. It must not have been too severe of a hardware failure because the database was able to be brought back online.
I don't think we can infer from the question that the hardware failure was responsible for corrupting both the live database and the backup file. It's certainly possible, but I don't think it can be assumed.
It is quite possible to backup the tail of the log even if the database cannot be started - all that is required is that the log is undamaged, there are no bulk logged changes, and the database state supports the operation. So, I don't think we can assume that the database could be "brought back online", either.
Paul White
SQLPerformance.com
SQLkiwi blog
@SQL_Kiwi