locked
Log files are running out of space - SQL MP RRS feed

  • Question

  • OpsMgr 2007 R2, SQL MP. I'm getting a slew of alerts similar to this

    ==========================================

    Alert: Log files are running out of space

    Source: tempdb

    Path: Server.mydomain.com;SQLEXPRESS

    Last modified by: System

    Last modified time: 6/7/2012 10:34:43 AM Alert description: The database tempdb in SQL instance SQLEXPRESS on computer Server.mydomain.com is running out of log file space. See knowledge section for more details.

    ==========================================

    I checked the database and the physical disk has plenty of space on it. The current database is 8.69 MB and the Space Available is 5.15 MB. We've tried changeing the Autogrowth on both database and log files to grow by 30 percent, unrestricted growth. We're still getting the alert. Has anyone dealt successfully with these?


    Orange County District Attorney

    Thursday, June 7, 2012 5:57 PM

Answers

  • Maybe, your DBA should be able to increase the white space of that log.  After doing so the alert should go away.  And or you can override disable this monitor for that particular log file on that server.

    Regards, Blake Email: mengotto<at>hotmail.com Blog: http://discussitnow.wordpress.com/

    • Proposed as answer by Nicholas Li Monday, June 11, 2012 8:18 AM
    • Marked as answer by Nicholas Li Monday, June 25, 2012 10:02 AM
    Thursday, June 7, 2012 6:20 PM

All replies

  • The TEMPDB log file is running out of space.  Did you check that?  This is for the LOG file, not the DB.

    Regards, Blake Email: mengotto<at>hotmail.com Blog: http://discussitnow.wordpress.com/

    Thursday, June 7, 2012 6:12 PM
  • Sorry for that. I should have worded it better. The log file is currently at 1 MB and the initial size of the log file is 1 MB however it's set to AutoGrow. Maybe the issue is that SQL is not 'growing' the log file.

    Orange County District Attorney

    Thursday, June 7, 2012 6:18 PM
  • Maybe, your DBA should be able to increase the white space of that log.  After doing so the alert should go away.  And or you can override disable this monitor for that particular log file on that server.

    Regards, Blake Email: mengotto<at>hotmail.com Blog: http://discussitnow.wordpress.com/

    • Proposed as answer by Nicholas Li Monday, June 11, 2012 8:18 AM
    • Marked as answer by Nicholas Li Monday, June 25, 2012 10:02 AM
    Thursday, June 7, 2012 6:20 PM
  • That may be the best course of action. I want to try and avoid overriding on this server if I can.

    Orange County District Attorney

    Thursday, June 7, 2012 6:28 PM

  • Hi,

    Firstly, please ensure you have the latest version of the SQL MP:

    System Center Monitoring pack for SQL Server
    http://www.microsoft.com/en-us/download/details.aspx?id=10631

    Please clear the HealthService queue on the SQL Server which meets this issue:

    1) Stop System Center Management service.
    2) Go to C:\Program Files\System Center Operations Manager 2007\, and rename the “Health Service State” folder.
    3) Restart System Center Management service.

    Meanwhile, please check the Event Log on the SQL Server and see if there are any related errors.

    Thanks.


    Nicholas Li

    TechNet Community Support


    Monday, June 11, 2012 8:24 AM
  • Thanks for the note. Yes, I do have the latest MP. I've been discovering that if I increase the database file sizes, the MP is happy and doesn't send alerts.

    Orange County District Attorney

    Monday, June 11, 2012 3:03 PM
  • Hi

    I've found SCOM to be sensitive to lower case drive letters used in the database file paths. This has not been exclusive to Lync. It seems that Microsoft have failed to notice their Wizards causing problems by automatically using lower case letters when creating the databases!

    Query the offending database, in this case ‘tempdb’ with the following;

    select* from sysfiles

    Check the results for any lower case drive letters.

    Now modify and run this ALTER DATABASE command against each log/database file to replace the lower case drive letter.

    ALTERDATABASE tempdb

    MODIFYFILE (NAME = tempdev, FILENAME= ‘C:\**\tempdb.mdf’)

    *Note the upper case 'C:\'

    Confirm the update by running the following ommand on the offending database.

    select* from sysfiles

    I hope this helps.

    Please let me know how you get on.

    Many thanks

    • Proposed as answer by HelpDaz Tuesday, July 31, 2012 11:40 AM
    Tuesday, July 31, 2012 11:40 AM
  • Thanks for the suggestion - I'll give it a try and report back on the results.

    Orange County District Attorney

    Tuesday, July 31, 2012 2:26 PM
  • Hi Sandy

    How did you get on with changing the case of your drive letters? I have had some positive feedback on the same answer on another post.

    Please do propose my reply as an answer and vote as helpful to get my Points underway.

    Many thanks

    Thursday, August 2, 2012 7:09 AM
  • Sorry for the late response. Looks like I'm still getting the alerts after making the case change.

    Orange County District Attorney

    Friday, August 3, 2012 2:01 PM
  • Hello,

    I had the same problem. You could try to check the log file physical and logical names in sys.database_files and sys.master_files.

    One of the reason of this monitor alert, could be that the different file names stored in database_files and master_files. 

    For my situation it was like that and resolved the by updating the different log file name.

    Hope it hepls.




    • Proposed as answer by Ferhat Saatci Friday, November 4, 2016 2:37 PM
    • Unproposed as answer by Ferhat Saatci Friday, November 4, 2016 2:37 PM
    • Edited by Ferhat Saatci Friday, November 4, 2016 2:42 PM
    Friday, November 4, 2016 2:35 PM
  • This worked for me. Clearing the cache (deleting the healthservicestate folder) on the DB server, did the trick.
    Monday, July 24, 2017 3:22 PM