views:

380

answers:

2

With SQL Server 2005 Express (obeserved on XP and Server 2003), I get sometimes huge Error logs files in production: The file C:\Program Files\Microsoft SQL Server\MSSQL.1\MSSQL\LOG\ERRORLOG grows to fill the disk (file size becomes more than 15 GB).

This file is not the transaction log, just the error log : a text log for SQL Server.

The error log starts like this: (seems to be normal)

2009-01-11 09:16:57.04 spid51      Starting up database 'SDomain'. 
2009-01-11 10:04:34.21 spid21s     SQL Server has encountered 1 occurrence(s) of cachestore flush for the 'Object Plans' cachestore (part of plan cache) due  
to some database maintenance or reconfigure operations. 
2009-01-11 10:04:34.23 spid21s     SQL Server has encountered 1 occurrence(s) of cachestore flush for the 'SQL Plans' cachestore (part of plan cache) due to  
some database maintenance or reconfigure operations. 
2009-01-11 10:04:34.23 spid21s     SQL Server has encountered 1 occurrence(s) of cachestore flush for the 'Bound Trees' cachestore (part of plan cache) due t 
o some database maintenance or reconfigure operations. 
2009-01-11 10:08:37.32 spid51      Starting up database 'SDomain'. 
2009-01-11 10:56:55.48 spid22s     SQL Server has encountered 1 occurrence(s) of cachestore flush for the 'Object Plans' cachestore (part of plan cache) due  
to some database maintenance or reconfigure operations. 
2009-01-11 10:56:55.49 spid22s     SQL Server has encountered 1 occurrence(s) of cachestore flush for the 'SQL Plans' cachestore (part of plan cache) due to  
some database maintenance or reconfigure operations. 
2009-01-11 10:56:55.49 spid22s     SQL Server has encountered 1 occurrence(s) of cachestore flush for the 'Bound Trees' cachestore (part of plan cache) due t 
o some database maintenance or reconfigure operations. 
2009-01-11 11:00:07.51 spid51      Starting up database 'SDomain'. 
2009-01-11 11:47:44.73 spid15s     SQL Server has encountered 1 occurrence(s) of cachestore flush for the 'Object Plans' cachestore (part of plan cache) due  
to some database maintenance or reconfigure operations. 
2009-01-11 11:47:44.74 spid15s     SQL Server has encountered 1 occurrence(s) of cachestore flush for the 'SQL Plans' cachestore (part of plan cache) due to  
some database maintenance or reconfigure operations. 
2009-01-11 11:47:44.74 spid15s     SQL Server has encountered 1 occurrence(s) of cachestore flush for the 'Bound Trees' cachestore (part of plan cache) due t 
o some database maintenance or reconfigure operations.

Then the file seems to contain endlessly repeating lines like this:

2008-12-17 00:12:24.03 spid13s     The log for database 'SDomain' is not available. Check the event log for related error messages. Resolve any errors and restart the database**

FYI, the windows eventlog contains exactly the same messages.

Any idea of the reason why this problem occurs? Could it be a particular issue of configuration of SQL Server? Or an issue in code causing this?

A: 

Maybe this can help? http://support.microsoft.com/kb/917828

Mr. Smith
The problem is not a performance problem (described in link to MS KB) : the result in a non functional SQL Server, and filling the disk with a huge useless file, possibly taking down the OS too.
olorin
+1  A: 

You have a database named SDomain that is set to auto-close. Whenever is accessed, it is 'started'. Recently you, or someone near you, deleted or moved the LDF file of the database. When the process that is accessing the SDomdain database is trying to open it, SQL Server will complain about the problem in the ERRORLOG. Give the database back its LDF and will stop complaining. Execute sp_cycle_errorlog to start a new ERRORLOG file so you can delete the old one.

Remus Rusanu