very busy disk

  • Hi guys - i'm wondering if someone can help me out here. i reindexed serveral dbs on my server last night at 10pm. Jobs finished at 10:02pm. At 10:40 'Disk Busy Time" jumped from 10% to 35% and has been there ever since (is is now 10:30am). i mention the reindexes as i can think of nothing else that might have caused this (even though it seems weird that the jump in %busy would happen 40 minutes later). cpu usage is low (3% or so) as is memory and FullTableScans per sec.

    Does anyone have any idea why my disk would be so busy or how i can track down the cause? Any help would be much appreciated! Thank you!

  • I would look for some long running transactions or other jobs.

    An alternative could be related to a heavily fragmented file system (unrelated to indexes or database fragmentation).

    Jason...AKA CirqueDeSQLeil
    _______________________________________________
    I have given a name to my pain...MCM SQL Server, MVP
    SQL RNNR
    Posting Performance Based Questions - Gail Shaw[/url]
    Learn Extended Events

  • Do you have a virus scanner running on the system?

    --Jeff Moden


    RBAR is pronounced "ree-bar" and is a "Modenism" for Row-By-Agonizing-Row.
    First step towards the paradigm shift of writing Set Based code:
    ________Stop thinking about what you want to do to a ROW... think, instead, of what you want to do to a COLUMN.

    Change is inevitable... Change for the better is not.


    Helpful Links:
    How to post code problems
    How to Post Performance Problems
    Create a Tally Function (fnTally)

  • Virus scanner or some other process. You could grab the Sysinternals utilities to track down the process that is hitting the disk.

  • If it is SQL Server-related, you would expect the guilty process to jump out at you from Activity Monitor (IO latch waits/Data File I/O list depending on version).

    You could also check the DMVs: sys.dm_io_pending_io_requests and sys.dm_io_virtual_file_stats.

  • But I don't suspect SQL Server 😉

  • Steve Jones - Editor (4/3/2010)


    But I don't suspect SQL Server 😉

    :laugh:

    Ok, ok, I was talking to 'bag of birds' not you lot!

    Jusy saying...if SQL Server is in anyway responsible (autoshrink?) it will show up in those DMVs 😛

  • Steve Jones - Editor (4/2/2010)


    Virus scanner or some other process. You could grab the Sysinternals utilities to track down the process that is hitting the disk.

    I'm glad you guys suggested this. I think this is the most probable candidate.

    Jason...AKA CirqueDeSQLeil
    _______________________________________________
    I have given a name to my pain...MCM SQL Server, MVP
    SQL RNNR
    Posting Performance Based Questions - Gail Shaw[/url]
    Learn Extended Events

  • CirquedeSQLeil (4/3/2010)


    I'm glad you guys suggested this. I think this is the most probable candidate.

    Did you catch the culprit ?

    -------Bhuvnesh----------
    I work only to learn Sql Server...though my company pays me for getting their stuff done;-)

  • wow. guys, thanks for all the advice. I am feeling pretty stupid right now telling you what actually happened was due to a bug (i am almost positive) in the monitoring software (zenoss anyone?) not actual disk activity. why i did not go to the machine itself and open perfmon earlier i don't know but when i did i saw that perfmon reported no major disk activity at all. What's more, the zenoss disk activity dashboard, minutes later, plummeted. closed perfmon, and zenoss shoots back up. each time i opened perfmon and looked at the disk, zenoss shot down, then shot back up again when i closed perfmon. it was like zenoss realized it was being watched and only then started doing it's job. very weird. anyway, thanks again for the advise guys.

  • bagofbirds-767347 (4/5/2010)


    wow. guys, thanks for all the advice. I am feeling pretty stupid right now telling you what actually happened was due to a bug (i am almost positive) in the monitoring software (zenoss anyone?) not actual disk activity. why i did not go to the machine itself and open perfmon earlier i don't know but when i did i saw that perfmon reported no major disk activity at all. What's more, the zenoss disk activity dashboard, minutes later, plummeted. closed perfmon, and zenoss shoots back up. each time i opened perfmon and looked at the disk, zenoss shot down, then shot back up again when i closed perfmon. it was like zenoss realized it was being watched and only then started doing it's job. very weird. anyway, thanks again for the advise guys.

    Awesome. The culprit is the Monitoring solution.

    Thanks for letting us know what you found.

    BTW, I've never heard of Zenoss.

    Jason...AKA CirqueDeSQLeil
    _______________________________________________
    I have given a name to my pain...MCM SQL Server, MVP
    SQL RNNR
    Posting Performance Based Questions - Gail Shaw[/url]
    Learn Extended Events

  • Thanks for the update. It happens, and this thread might help someone else figure out their issue earlier because of your post.

  • I still think it was the virus scanner :rolleyes:

    (joking!)

  • Paul White NZ (4/5/2010)


    I still think it was the virus scanner :rolleyes:

    (joking!)

    Maybe Zenoss had an identity crisis and tried performing virus scans:Whistling:

    Jason...AKA CirqueDeSQLeil
    _______________________________________________
    I have given a name to my pain...MCM SQL Server, MVP
    SQL RNNR
    Posting Performance Based Questions - Gail Shaw[/url]
    Learn Extended Events

  • CirquedeSQLeil (4/5/2010)


    Paul White NZ (4/5/2010)


    I still think it was the virus scanner :rolleyes:

    (joking!)

    Maybe Zenoss had an identity crisis and tried performing virus scans:Whistling:

    Heh...maybe 😛

Viewing 15 posts - 1 through 15 (of 18 total)

You must be logged in to reply to this topic. Login to reply