4

We are getting a large number of the below errors on one of our production databases. CheckDB is completely clean.

We thought it was related to replication so we shut it down using sp_removedbreplication and the problem went away. For other reasons, we did a restore from last night backup which resurrected replication again. Immediately the errors returned and have persisted despite my removing replication again.

  • Error: 9003, Severity: 20, State: 1.
  • The log scan number (390576:5234:44) passed to log scan in database 'XXXX' is not valid. This error may indicate data corruption or that the log file (.ldf) does not match the data file (.mdf). If this error occurred during replication, re-create the publication. Otherwise, restore from backup if the problem results in a failure during startup.

I'm not finding much on the web related to this except articles for:

  1. The master database and SQL Server not starting
  2. Issues with replication

Any advice would be tremendously appreciated!

Improve this question

1 Answer 1

Reset to default
3

With the help of an outside source, the issue was nailed down to a corrupted transaction log and was fixed by reinitializing replication from scratch.

It was strongly recommended that we evaluate our I/O subsystems that house the log file to make sure there isn't a problem there which caused it.

Improve this answer
2
  • 1
    Consider running SQLIOSim to stress your I/O subsystem. See support.microsoft.com/kb/231619/en-us
    – Dan Guzman
    Jan 31, 2015 at 23:36
  • Thanks Dan! I saw this when you posted it but forgot to say something. We are planning to do this over the weekend.
    – PseudoToad
    Feb 9, 2015 at 20:04

Your Answer

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

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