Log file becomes ungrowable every morning.

Have a site with a wierd problem. A 3rd party application is accessing a SQL
2000 database. Each morning, the event log contains a heap of 17055:5144
errors, "Autogrow of file 'data_log" in Database 'data' cancelled or timed
out in 30547 ms. Use ALTER DATABASE to set a smaller FILEGROWTH or to set a
new size" Users report errors that seem directly related to this problem.
The solution is to cycle the SQL service, this will see things work again
until the next morning. I have seen this error before, but there has always
been a good reason. In this case, I cannot find one. There is ample disk
space, and the log is set to grow by 10% as needed. AV is excluded for this
data, and there are no quotas in place. CHKDSK reports no errors, and
fragmentation is not excessive. Any ideas where to look next?
Machine is running SBS-2003, SQL2000 & everything else is fully updated as
of a week ago.
Thanks.On 23.04.2007 08:12, Mal Osborne wrote:
> Have a site with a wierd problem. A 3rd party application is accessing a SQL
> 2000 database. Each morning, the event log contains a heap of 17055:5144
> errors, "Autogrow of file 'data_log" in Database 'data' cancelled or timed
> out in 30547 ms. Use ALTER DATABASE to set a smaller FILEGROWTH or to set a
> new size" Users report errors that seem directly related to this problem.
> The solution is to cycle the SQL service, this will see things work again
> until the next morning. I have seen this error before, but there has always
> been a good reason. In this case, I cannot find one. There is ample disk
> space, and the log is set to grow by 10% as needed. AV is excluded for this
Well, then change it to a fixed size. With an increase of 10% every
resize needs more disk space - and thus takes longer or runs the risk of
running out of disk space. Btw, did you check space on the device? Do
you backup or shrink TX logs?
robert|||"Mal Osborne" <Mal Osborne@.discussions.microsoft.com> wrote in message
news:37B56334-A6E6-4628-8BFF-E51BE4F1C7CD@.microsoft.com...
> Have a site with a wierd problem. A 3rd party application is accessing a
> SQL
> 2000 database. Each morning, the event log contains a heap of 17055:5144
> errors, "Autogrow of file 'data_log" in Database 'data' cancelled or timed
> out in 30547 ms. Use ALTER DATABASE to set a smaller FILEGROWTH or to set
> a
> new size" Users report errors that seem directly related to this problem.
> The solution is to cycle the SQL service, this will see things work again
> until the next morning. I have seen this error before, but there has
> always
> been a good reason. In this case, I cannot find one. There is ample disk
> space, and the log is set to grow by 10% as needed. AV is excluded for
> this
> data, and there are no quotas in place. CHKDSK reports no errors, and
> fragmentation is not excessive. Any ideas where to look next?
Not sure why it's not permitted to grow.
But in any event, two questions:
Are you doing any form of index rebuild over night?
Are you doing transaction log backups on a regular basis?
> Machine is running SBS-2003, SQL2000 & everything else is fully updated as
> of a week ago.
> Thanks.
Greg Moore
SQL Server DBA Consulting Remote and Onsite available!
Email: sql (at) greenms.com http://www.greenms.com/sqlserver.html