Tempdb Full After Reboot

  • We recently swapped a server out of active-active cluster and replaced it with a large box. Since then, I've had issues with the server failing over. If the box is rebooted, the instance is correctly trying to fail over to the other node, but SQL Server and the Agent reports tempdb is full and then it attempt to restart on the original node. There it also reports tempdb full as the reason it fails to start.

    When I check the mdf and ldf for the server, size for both is 6GB and I have 50 GB available on the drive. Also, the initial size for creation of tempdb is 48000 MB.

    After deleting both tempdb files, SQL Server will start.

    Anyone have an idea of what might be going on and how to resolve this? Right now, the only thing I can think of is we have a bad installation of SQL server but would like confirmation if possible before I go through a weekend of uninstalling and reinstalling.

    Dave

  • Tempdb has an initial size of 48 Gig on a drive that only has 50 Gig of available space? If I'm misreading that, that's one thing, but if I'm not, that's probably your problem right there. Adjust the initial size on tempdb and the problem will probably go away.

    - Gus "GSquared", RSVP, OODA, MAP, NMVP, FAQ, SAT, SQL, DNA, RNA, UOI, IOU, AM, PM, AD, BC, BCE, USA, UN, CF, ROFL, LOL, ETC
    Property of The Thread

    "Nobody knows the age of the human race, but everyone agrees it's old enough to know better." - Anon

  • Initial size is 48,000 MB, not GB.

  • Dave Klug (4/20/2009)


    Initial size is 48,000 MB, not GB.

    48,000 MB is approximately 48 GB. If the disk you are trying to create tempdb on is only 50 GB in size, this may be your problem (as said earlier). Either shrink the initial size of tempdb or move it to a disk with more available space.

  • 48000 MB = 48 GB

  • Been a long weekend and obviously I missed that in my math. The odd thing is, once I delete tempdb, the database will start normally and the file size is 5GB. I'll try to reduce this and then see what effect it has. This was originally setup by a consultant and we're working through the issues now.

  • I've reset the size to 10GB and the cluster is now failing over correctly.

  • Excellent. Glad we could help.

    - Gus "GSquared", RSVP, OODA, MAP, NMVP, FAQ, SAT, SQL, DNA, RNA, UOI, IOU, AM, PM, AD, BC, BCE, USA, UN, CF, ROFL, LOL, ETC
    Property of The Thread

    "Nobody knows the age of the human race, but everyone agrees it's old enough to know better." - Anon

Viewing 8 posts - 1 through 7 (of 7 total)

You must be logged in to reply to this topic. Login to reply