Results 1 to 9 of 9
  1. #1
    Join Date
    Aug 2010
    Posts
    6

    Unanswered: Data base corruption

    I am having an argument with the IT guy, we are having corruption mostly in 1 table that is accessed from this one workstation. I think it is network related, bad packets or something he says it must be the DB. We have dozens of different offices running this DB and this is the only one having problems. They were able to add 1,200 rows over a month since the last corruption.

    Thanks for any suggestions

  2. #2
    Join Date
    Jan 2003
    Location
    Massachusetts
    Posts
    5,800
    Provided Answers: 11
    Define "Corruption". Do you mean that the data is wrong, or that DBCC CHECKDB returns errors?

  3. #3
    Join Date
    Aug 2010
    Posts
    6
    Broken pointer

  4. #4
    Join Date
    Jan 2003
    Location
    Massachusetts
    Posts
    5,800
    Provided Answers: 11
    Check for disk errors in the database machine's system event log. The number one cause of database corruption is disk errors.

    Can you post the output for the following:
    Code:
    DBCC CHECKDB(your database name here) with no_infomsgs

  5. #5
    Join Date
    Aug 2010
    Posts
    6
    Thanks, I always thought networks and bad packets was the major problem?

  6. #6
    Join Date
    Jan 2003
    Location
    Massachusetts
    Posts
    5,800
    Provided Answers: 11
    I don't think I have ever met a corruption problem based on a network factor. In fact, I have met very few network problems at all. The reason disk problems are so far in the lead, is that no client process has access to any of the internal pointers or other disk structures that SQL Server (or any other DBMS for that matter) uses to maintain data files. All that comes from the client are directives (delete, insert, etc.) and data values.

  7. #7
    Join Date
    Aug 2010
    Posts
    6
    So what is the best way to prevent these disk problems? Sorry to be a newbie. thanks for the education.

  8. #8
    Join Date
    Jan 2003
    Location
    Massachusetts
    Posts
    5,800
    Provided Answers: 11
    Most server systems are good at making sure writes happen. If you see disk errors in the event log, you should check with your hardware vendor, and see if they think you need to upgrade firmware (they may have a bug list), or replace a few parts. It depends on what is actually dropping the data. There are a few I/O systems that SQL Server does not technically support. NAS devices, and IDE drives come to mind. I am not sure about SATA drives. Any SCSI drive or SAN system is fine.

  9. #9
    Join Date
    Aug 2010
    Posts
    6
    Thanks for the help

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •