will create a latch and hold onto it indefinitely. I have no explanation
for why the offending process fails to complete. Since the latch is on a
frequently used table, all subsequent insert operations are blocked until
the stubborn process is manually killed.
I've run across it about 4 times in the last month on two instances of a
similar database (SQL2000 latest SP, Simple Recovery Model). The type of
latch and the type of process that created it has varied.
Latch Type Process that caused it
============== ======================================
LATCH_EX -- From a stored proc. with a single INSERT
statement
PAGELATCH_EX -- From a stored proc. with a single INSERT statement
NETWORKIO -- From a SELECT query
Again, I haven't tracked down a legitimate reason for why the offending
process gets stuck. The table where the problems are occurring is
essentially a message log receiving ~400,000 records/day. It has 4 indices
and I've seen some postings regarding timeout problems on heavily indexed
tables with lots of activity. There are no operations that take place in
the database that are suspect for deadlocking. The vast majority of the
operations are simple one record inserts into a single table. All other
operations are occasional SELECT queries coming from a web report. Under
one occasion, the problem was caused by a select query but the others have
been caused by the insert stored procedure.
As a temporary work-around, I'm periodically polling the sysprocesses table
for waiting processes and performing a kill on the process if it is one of
the three latch types above and it has been waiting more than 10 seconds
(for my situation, accidentally killing a innocent process has little
impact).
Since it is not occurring in frequently, I have not been able to catch it
under a SQL Profiler log but I'll try to capture the behavior in the future.
Hi Paul,
Thanks for using MSDN Newsgroup!
From your descriptions, I understood that some latches will not release 4
times in two instance. Have I understood you? If there is anything I
misunderstood, please feel free to let me know
First of all, I think compared with your large amount of data processing, 4
times is relatively a small number, isn't it? So it would be hard for us to
troubleshooting.
Secondly, we will have to collect the following information to make further
troubleshooting regarding blocking issue. The following documents will
show you how to collect the information when blocking
INF: How to Monitor SQL Server 2000 Blocking
http://support.microsoft.com/?id=271509
You will find we need the result from sp_blocker_pss80, performance log
from performance monitor and SQL Server error log.
I fully understood you will have to wait until the block happened again and
it may be hard to get the error messsage from KB: 271509. However, we need
all these information to do troubleshooting.
Unfortunately, according to our Newsgroup policy I was not able to monitor
this post thread too long and looking the nature of this issue, it would
require intensive troubleshooting which would be done quickly and
effectively with direct assistance from a Microsoft Support Professional
through Microsoft Product Support Services.
BTW, You can contact Microsoft Product Support directly to discuss
additional support options you may have available, by contacting us at
1-(800)936-5800 or by choosing one of the options listed at
http://support.microsoft.com/default...=sz;en-us;top. If this is not
an urgent issue and your would like us to create an incident for you and
have Microsoft Customer Service Representative contact you directly, please
send email to (remove "online." from this no Spam email address):
mailto:dscommhf@.online.microsoft.com with the following information,
Anyway, you could continue post all the information listed in the KB:
271509 here. I will be glad to serve you until it is resolved
Thank you for your patience and cooperation. If you have any questions or
concerns, don't hesitate to let me know. We are here to be of assistance!
Sincerely yours,
Mingqing Cheng
Microsoft Developer Community Support
Introduction to Yukon! - http://www.microsoft.com/sql/yukon
This posting is provided "as is" with no warranties and confers no rights.
Please reply to newsgroups only, many thanks!
|||Hi Mingqing,
Thanks for the info. I am running the additional logging and trace
mechanisms described in the KB article you refered to. When the problem
occurs again, I will hopefully have enough info to identify the cause.
From the list of the 6 common categories of blocks listed in KB: 271509 I
suspect #6 Blocking Caused by Orphaned Connection. Although the article (or
SQL BO) doesn't describe the scenario in detail, my guess is that a client
getting a dropped connection due to a network failure or Server performance
bottleneck and that connection is never has its locks released. If this is
the case, the artcle suggests that the only way to deal with it is to kill
the process (I guess my temporary workaround may become permanent).
I will repost if the logs turn up new info when the problem occurs next.
--Paul
""Mingqing Cheng [MSFT]"" <v-mingqc@.online.microsoft.com> wrote in message
news:kfwgvbCWEHA.3440@.cpmsftngxa10.phx.gbl...
> Hi Paul,
> Thanks for using MSDN Newsgroup!
> From your descriptions, I understood that some latches will not release 4
> times in two instance. Have I understood you? If there is anything I
> misunderstood, please feel free to let me know
> First of all, I think compared with your large amount of data processing,
4
> times is relatively a small number, isn't it? So it would be hard for us
to
> troubleshooting.
> Secondly, we will have to collect the following information to make
further
> troubleshooting regarding blocking issue. The following documents will
> show you how to collect the information when blocking
> INF: How to Monitor SQL Server 2000 Blocking
> http://support.microsoft.com/?id=271509
> You will find we need the result from sp_blocker_pss80, performance log
> from performance monitor and SQL Server error log.
> I fully understood you will have to wait until the block happened again
and
> it may be hard to get the error messsage from KB: 271509. However, we need
> all these information to do troubleshooting.
> Unfortunately, according to our Newsgroup policy I was not able to monitor
> this post thread too long and looking the nature of this issue, it would
> require intensive troubleshooting which would be done quickly and
> effectively with direct assistance from a Microsoft Support Professional
> through Microsoft Product Support Services.
> BTW, You can contact Microsoft Product Support directly to discuss
> additional support options you may have available, by contacting us at
> 1-(800)936-5800 or by choosing one of the options listed at
> http://support.microsoft.com/default...=sz;en-us;top. If this is
not
> an urgent issue and your would like us to create an incident for you and
> have Microsoft Customer Service Representative contact you directly,
please
> send email to (remove "online." from this no Spam email address):
> mailto:dscommhf@.online.microsoft.com with the following information,
> Anyway, you could continue post all the information listed in the KB:
> 271509 here. I will be glad to serve you until it is resolved
>
> Thank you for your patience and cooperation. If you have any questions or
> concerns, don't hesitate to let me know. We are here to be of assistance!
>
> Sincerely yours,
> Mingqing Cheng
> Microsoft Developer Community Support
> Introduction to Yukon! - http://www.microsoft.com/sql/yukon
> This posting is provided "as is" with no warranties and confers no rights.
> Please reply to newsgroups only, many thanks!
>
No comments:
Post a Comment