February 18, 2005 at 12:13 pm
I have a medium sized SQL Database, with tables of a few hundred records to one with 330,000
It is running MS SQL 2000 on a Windows 2000 advanced server, and is requiring to be rebuilt almost daily
The corruption we are getting is records out of align errors. I strongly suspect that there is a network admin who is rebooting the server whenever it gets dogged down.
However, I am also using the database with a web app that writes to and reads from it.
I know this is general, but where should I start looking to get to the next step on determining if this is programming or hardware?
Reality pales in the light of common perception.
February 21, 2005 at 8:00 am
This was removed by the editor as SPAM
February 24, 2005 at 3:53 am
Look at your server event viewer, check out your application logs.
Look at your SQL server logs - look for unexpected termination, etc.
Do you have scheduled maintenance jobs doing DBCC CHECKDB, Backups etc. ?
February 24, 2005 at 6:26 am
The application in the Event Viewer is MSSQL Error 3624, Severity 20
I tried to look this up but don't see anything on it - I am probably doing the look up wrong
Reality pales in the light of common perception.
February 24, 2005 at 7:20 am
Found the following links in reference:
http://support.microsoft.com/default.aspx/kb/885290?
http://www.tech-archive.net/Archive/SQL-Server/microsoft.public.sqlserver.server/2004-07/3596.html
See if these help ...
February 24, 2005 at 7:36 am
Look up BOL for severity levels. Level 20 is explained there.
Next I would search the Google Groups for this error along with the name of Paul Randal. He participates frequently in the newsgroups and is dev lead of the SQL Server storage engine. So, something like this:
http://groups.google.de/groups?q=error+3624+paul+randal&hl=de
--
Frank Kalis
Microsoft SQL Server MVP
Webmaster: http://www.insidesql.org/blogs
My blog: http://www.insidesql.org/blogs/frankkalis/[/url]
February 24, 2005 at 8:08 am
Thanks for all of the help. I am getting my mind aroudn this now.
It is looking like the problem could lie in the way we write to the tables in the database.
However, is there a problem with running nightly maintenance on the database that could be contributing to this, do you think?
Reality pales in the light of common perception.
Viewing 7 posts - 1 through 6 (of 6 total)
You must be logged in to reply to this topic. Login to reply