SQL Server 2005 SP4 32-Bit

I have a DBCC CHECKDB job running nightly. Last night, soon after the job started, I got the errors below. The database is NOT in suspect mode, and CHECKDB comes back clean when I run it now. The database is fully accessible. Should I be concerned? I'd hate to go back to a backup at this point.

 2013-04-02 02:10:55.53 spid56      Error: 1203, Severity: 20, State: 1.
 2013-04-02 02:10:55.53 spid56      Process ID 56 attempted to unlock a resource it
                                    does not own: PAGE: 34:1:388664. Retry the 
                                    transaction, because this error may be caused 
                                    by a timing condition. If the problem persists, 
                                    contact the database administrator.
 2013-04-02 02:10:55.58 spid56      Error: 3314, Severity: 17, State: 3.
 2013-04-02 02:10:55.58 spid56      During undoing of a logged operation in 
                                    database 'MY_DATABASE_NAME', an error occurred 
                                    at log record ID (1342973:12519:37). Typically, 
                                    the specific failure is logged previously as 
                                    an error in the Windows Event Log service. 
                                    Restore the database or file from a backup, 
                                    or repair the database.
  • Do you know the statement that caused the error to occur? If so, can you recreate the issue? Perhaps restore the database to a test machine and see if you can repeat the problem there. Also, what does the Windows Error log say at the time in question? – Max Vernon Apr 2 '13 at 14:26
  • Have you seen this Connect item which details a hotfix for SP4? connect.microsoft.com/SQLServer/feedback/details/241265/… – Max Vernon Apr 2 '13 at 14:27
  • The Windows logs have nothing else at that time. There was an SSIS package running at the time which finished after the error message with a SUCCESS. I'd have to do a lot of digging and work with developers to try to replicate this somewhere else, but I can do it if needed. I did see that connect item, but we're at SP4 already, so I don't think that will help. – SomeGuy Apr 2 '13 at 14:36
  • We were getting similar error when DB server was bounced after patching and one of the database was marked as suspect with the same error as you. What we found someone wrote a crappy code that kills open transaction in tempdb after one hour. We removed that code and never seen that issue again. That connect item is misleading you need to find root cause that will be hard but run SQLDiag and get what was run in last few hours on that server. – SQL Learner Apr 2 '13 at 14:52
  • 2
    Contact product support, this error may indicate SQL Server bugs. CSS may know a workaround. – Remus Rusanu Apr 2 '13 at 19:36

If this is just one time issue and CHECKDB run without any errors and you have good backups, do not bother to investigate it (unless you want to spend time and resources to what is not reproducible).

If you are able to reproduce the issue, then I would recommend you to

  1. Replicate on a test environment and capture whats going using a Profiler.
  2. As a proactive measure, you can probably have an agent job or powershell job to scan the error log (note that you can put a logic inside Powershell job to only scan from the last time it ended) to look for specific errors like Error: 1203 or Error: 3314 and then fire SQLDIAG to capture what was actually going on when the issue happened.

Let me know if you need any such script and I will be happy to share it.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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