|
|
Posted by Erland Sommarskog on 10/02/21 11:48
Marco van de Voort (marcov@stack.nl) writes:
> It occurs less than once a month (unfortunately on a painful moment).
> IOW, I can't reproduce it at will. Which is why I asked if sb recognized
> the problems.
I'm afraid that without access to the real situation, it is difficult
to say that much intelligent. The fact that it occurs only rarely,
indicates that it is not a general problem with unhandled query timeouts.
But it still smells of transactions that are not committed/rolled back
properly.
The only thing I can suggest is that when it sets try to collect as
much data you can before the reboot panic sets in. I have a stored
procedure on my web site which is good for this purpose:
http://www.sommarskog.se/sqlutil/aba_lockinfo.html.
>> Also, I have you examined the SQL Server error log?
>
> Yes, nothing. But maybe my logging settings are wrong.
It's always good to have trace flags 1204 and 3605 enabled to get
deadlock information written to the error log, but that was not I
had in mind. I was thinking of error 17883, which indicates that
SQL Server is in bad shape at the moment. This diagnostic message
was added in SQL 2000 SP3, and was augmented in some hotfixes soon
thereafter. SP4 has an even wider set of these messages.
The fact that you don't have message 17883 in the log amplifies the
impression that the problem is related to the application.
--
Erland Sommarskog, SQL Server MVP, esquel@sommarskog.se
Books Online for SQL Server 2005 at
http://www.microsoft.com/technet/prodtechnol/sql/2005/downloads/books.mspx
Books Online for SQL Server 2000 at
http://www.microsoft.com/sql/prodinfo/previousversions/books.mspx
Navigation:
[Reply to this message]
|