Please start any new threads on our new
site at https://forums.sqlteam.com. We've got lots of great SQL Server
experts to answer whatever question you can come up with.
| Author |
Topic |
|
aiken
Aged Yak Warrior
525 Posts |
 Posted - 2002-05-11 : 02:46:53
|
The transaction volume on my DB is continuing to increase. Unfortunately, I don't have any numbers on that.However, my uptime is continuing to decrease. After some amount of time/transaction volume, my SQL server is dying with tons of messages like "SQL Server could not spawn process_loginread thread." in its SQL Server log. The only fix so far is to reboot the entire server (SQL server doesn't stop cleanly once in this state).select @@version:"Microsoft SQL Server 2000 - 8.00.384 (Intel X86) May 23 2001 00:02:52 Copyright (c) 1988-2000 Microsoft Corporation Enterprise Edition on Windows NT 4.0 (Build 1381: Service Pack 6)" I'm a little concerned because I've graphed the uptime, and if my three datapoints represent an exponential curve, we'll be up for less than 6 hours at a time sometime on June 21. That's not good!Any ideas about the error message, or a better way to troubleshoot it? We've got a call in to MS, but based on past experience I'm expecting them to deny any problem, point fingers at our hardware and app, and generally stonewall until they announce some bugfix in mid July (for those following the exponential curve, we'll be up for about 8 seconds at a time by then).Cheers, and thanks-b |
|
|
chadmat
The Chadinator
1974 Posts |
Posted - 2002-05-11 : 04:50:34
|
| Well, despite your badmouthing of my company, and the organization from which I learned everything I know about SQL (MS SQL Support), I am willing to help you out on this.A couple of questions first. How much physical Memory is on the box? Do you have any calls to sp_OAxx procs?What are your Max and Min server Memory settings?Do you have any DMO code running on your server?I would get into a flame war defending MS SQL Support, but I think everyone here knows my position. -Chad |
 |
|
|
robvolk
Most Valuable Yak
15732 Posts |
Posted - 2002-05-11 : 09:42:29
|
| And I'm gonna agree with Chad on this one. My personal experience with MS tech support has been nothing less than stellar: good, hardworking, knowledgeable people on the phone who spare no effort to help. They have always been polite, and never "blamed" anyone or anything for a problem. Unfortunately you seem to want to blame them for your problems, instead of working with them towards a solution.I imagine that they give as they get, so if you are argumentative with them, so shall they be with you. If you are polite and courteous, you will receive likewise attention.There's one thing you might try that could solve the problem: apply SP2. <heavy sarcasm>It's only been available since November.</heavy sarcasm> Not to mention searching the MS Knowledge Base; I found 2 links related to your problem, and Chad has already suggested one of them as a possible cause. Both of them are fixed with SP2, you don't have to wait for the next service pack.Edited by - robvolk on 05/11/2002 09:43:47 |
|
|
|
Kevin Snow
Posting Yak Master
149 Posts |
Posted - 2002-05-13 : 14:01:40
|
| Is your memory usage growing consistently over time?Are your users accessing any OLE Automation support procedures? If so, read the folling, and check this article.http://www.sqlpass.org/news/newsletter/kb_mar01.cfmCalling OLE Automation support stored procedures, namely sp_OAMethod, sp_OAGetProperty, and sp_OASetProperty, leak a small amount of memory in the OLE Automation Object operation DLL (Odsole70.dll) file. This may cause the following errors to be reported in the SQL Server error log: 2000-11-27 11:22:59.87 server SQL Server could not spawn process_loginread thread. |
|
|
|
|
|
|
|
|
Hard crashes, getting more frequent...