db server locking up and cant kill user processes

  • Hey all

    I am getting the following errors in the error logs on our test box...

    Insufficient memory available..

    LazyWriter: warning, no free buffers found.

    Buffer Distribution: Stolen=7561 Free=0 Procedures=57

    Inram=21953 Dirty=19 Kept=0

    I/O=0, Latched=106, Other=0

    Buffer Counts: Commited=29696 Target=29696 Hashed=22078

    InternalReservation=222 ExternalReservation=3240 Min Free=464

    Procedure Cache: TotalProcs=1 TotalPages=57 InUsePages=57

    Dynamic Memory Manager: Stolen=5890 OS Reserved=312

    OS Committed=134

    OS In Use=275

    Query Plan=162 Optimizer=0

    General=5917

    Utilities=34 Connection=52

    Global Memory Objects: Resource=449 Locks=5277

    SQLCache=16 Replication=2

    LockBytes=2 ServerGlobal=43

    Xact=12

    Query Memory Manager: Grants=1 Waiting=0 Maximum=19873 Available=14905

    which I forceably stopped via:

    SQL Server is terminating due to 'stop' request from Service Control Manager.

    This happens to large, rouge SQL queries running against the box. I cant login to the server at all, all connects timeout and the sql statement runs for a miniumm of 30mins before completing and returning control, unfortunatly, the instance seems to be in a state where any smaller queries return the same errors above.

    The box has 512Mb ram (bugger all I know), and ive allocated 400 to the instance.

    Anyone experienced this before and worked out a cleaner way or addressing the issue?

    Cheers

    Ck


    Chris Kempster
    www.chriskempster.com
    Author of "SQL Server Backup, Recovery & Troubleshooting"
    Author of "SQL Server 2k for the Oracle DBA"

  • Hi all

    OK, managed to repeat it on our support server, but its very spiratic. We have a large query, the programmer was running it via query analyser and was chewing around 70% cpu and memory usage raised to the fixed max quickly. We cancelled the query and waited for around 5mins, the job was still trying to cancel and the above mentioned errors occured, in this particular case stopping the service also failed and required a server reboot.

    Cheers

    Ck


    Chris Kempster
    www.chriskempster.com
    Author of "SQL Server Backup, Recovery & Troubleshooting"
    Author of "SQL Server 2k for the Oracle DBA"

  • Tend to agree with Andy. Large queries are problematic in that you cannot limit the CPU devoted to the process and this can "lock" the box. It will come back when the query completes or the timeout is reached.

    You might want to limit the time a query can execute (Server properties, connections tab). This will at least limit the effects slightly until you can get more memory. 400MB is not very much. Should go to 1GB at least.

    Steve Jones

    sjones@sqlservercentral.com

    http://www.sqlservercentral.com/columnists/sjones

Viewing 4 posts - 1 through 3 (of 3 total)

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