1

I have a database that's size is 100GB. When I run CHECKDB on it, it fails to due a lack of space for tempdb, which has over 1TB of space. I can restart my SQL Server, flush out tempdb, the drive has 1TB of available space, and run it again, and it will use everything and fail due to a lack of space.

I suspect the database is corrupt because this seems like odd behavior, but is there a scenario where a 100GB database would use more than 1TB for running CHECKDB?

UPDATE

The table that didn't pass DBCC CHECKTABLE, I recreated its indexes and will see if it verifies faster.

7
  • 2
    That seems very unlikely that a 100GB database would overflow an empty 1 TB drive. What else is using tempdb, if anything?
    – RLF
    Commented Nov 17, 2015 at 18:51
  • @RLF Nothing; I put it on its own server just for this. Seems very odd.
    – DoubleVu
    Commented Nov 17, 2015 at 18:54
  • 2
    The fact that you were able to attach the DB on its own server is at least a sign that the boot page is intact. Can you try running DBCC CHECKDB WITH ESTIMATEONLY to see the estimated amount of tempdb usage?
    – Greg
    Commented Nov 17, 2015 at 19:05
  • Is it checking tempDB? I thought that when you ran checkDB the snapshot database was created on the same drive as the original database data file. Is there sufficient space on that drive? Can you post the actual error message? Commented Nov 17, 2015 at 19:10
  • @Greg Good idea, right now, I'm running a DBCC CHECKTABLE on each table.
    – DoubleVu
    Commented Nov 17, 2015 at 19:22

1 Answer 1

1

Normally a few events of 823 and 824 are warnings that there is possible corruption on the database, also if you have backups of this database that are failing and or jobs on it that fail to complete.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.