SQL Server Performance

845 Error System Outage During Critical Time

Discussion in 'Performance Tuning for DBAs' started by ptimmerm, Apr 7, 2008.

  1. ptimmerm New Member

    On Friday night I had a very important server go down as a result of a: Error: 845, Severity: 17, State: 1 Time-out occurred while waiting for buffer latch type 4 for page (1:281), database ID 1. I know these errors are usually a result of an I/O bandwidth problem. The SQL services were not responding, so I downed the box and it came back up just fine. It has functioned great ever since. I can find NO evidence of that the problem was. Even when the services were "hung" I did not see any excessive IO activity. The only lead I have to go on is I am told that the file system has some errors when I do a chkdsk on the C: drive. The only thing SQL Server related on the C: drive is the binaries. I guess that could be it? Has anyone ever come across this and had it be a "one time thing". I don't know where else to go. All hardware diagnostics are fine and everything at the DB level is checking out. Thoughts? Paul
  2. Luis Martin Moderator

    After chkdsk the problem never came back?
  3. ptimmerm New Member

    Actually I did not do the CHKDSK until this morning. THe problem never came back after the initial reboot. I am of course paranoid that it may come back as I have NO indications of the cause.
  4. Luis Martin Moderator

    I believe you already did everything to find out the problem.
    Unfortunately you have to wait (hope never) for the next event.
  5. ptimmerm New Member

    Another tidbit of information..... that page referenced belongs to sysxlogins.
    This does nothing to clarify the issue for me. Anyone else have any thoughts?
  6. ptimmerm New Member

    Am I correct in that you would expect to see this error thrown from a page/object that experiences a high level of I/O? SYSXLOGINS does not fit that bill. Maybe it was just a fluke? I don't like that answer, but I guess I may have to live with it.

Share This Page