keep getting these messages on sql server error log

  • We have a development server that keeps getting the following errors in the sql server log:

    2005-10-12 00:19:19.99 spid1 Buffer Distribution: Stolen=661 Free=0 Procedures=73

    Inram=3253 Dirty=26 Kept=0

    I/O=0, Latched=52, Other=0

    2005-10-12 00:19:19.99 spid1 Buffer Counts: Commited=4065 Target=4065 Hashed=3331

    InternalReservation=178 ExternalReservation=346 Min Free=248

    2005-10-12 00:19:19.99 spid1 Procedure Cache: TotalProcs=1 TotalPages=73 InUsePages=73

    2005-10-12 00:19:19.99 spid1 Dynamic Memory Manager: Stolen=644 OS Reserved=1664

    OS Committed=1629

    OS In Use=1609

    Query Plan=77 Optimizer=0

    General=1913

    Utilities=27 Connection=74

    2005-10-12 00:19:19.99 spid1 Global Memory Objects: Resource=1690 Locks=47

    SQLCache=87 Replication=2

    LockBytes=2 ServerGlobal=43

    Xact=35

    2005-10-12 00:19:19.99 spid1 Query Memory Manager: Grants=1 Waiting=0 Maximum=1739 Available=1305

    It happened occassionally in the past. But last night they showed up a lot and consecutively. The weird thing is there is another instance of SQL server on the same machine, and the errors never appeared on that instance. Also, the errors only popped up mid night around 12 when there is no development going on. The only job running at that time is a log backup job, which is scheduled to run every half an hour. BTW, its SQL Server 2000.

    I have done some research, but did not find anything.

    I would appreciate any input.

    Thanks very much!

    Kathleen

  • I can't say that I am 100 percent sure but looks like output from a trace flag to me as a lot of them write to the log by default and not an error.

    Have you tried DBCC TraceStatus(-1) to see if there any set.

    hth

    David

  • Thanks for the reply.

    Sorry I missed one last message that says: 'LazyWriter: warning, no free buffers found.' So its like all seven messages keep showing up one by one.

    I will try what you suggested and see how it goes.

    Thanks a lot!

    Kathleen

  • I ran

    DBCC TRACESTATUS(-1)

    GO

    on each of the database on the server, none of them have any trace flags turned on.

    Any more thoughts on these messages?

    Thanks,

    Kathleen

  • I've not seen this on any of my servers before but given the final message about the the lazywriter I just wonder if for some reason there is not enough spare memory for your development instance as the whole point of the lazy writer is to make sure that ther are free pages in the buffer otherwise threads will have to stall and await pages becoming free before they can proceed with their work.

    To see if this is the case you could set a performanec monitor trace on the server and monitor the number of free pages etc and other memeory performance counters.

    hth

    David

  • Looks like a memory issue to me too.  Run DBCC MEMORYSTATUS then see http://support.microsoft.com/default.aspx?scid=kb;en-us;271624.  This will help explain what you are seeing.

    Linda

  • Thanks for responding.

    I will give it a try.

    Kathleen

Viewing 7 posts - 1 through 6 (of 6 total)

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