January 15, 2007 at 10:12 pm
I have a backup job that runs about 9 pm daily. I am getting error messages for the past few weeks. One of the messages is that the backup failed 'backup disk file : create media : backup device H:\backup failed'
In the SQL server 2000 logs, I am getting this error message also 'WARNING: 4 waited 300 sec. on latch 80913d94. Not a BUF latch'. Can any one let me know as to what's happening?
Thank you.
January 15, 2007 at 11:47 pm
what service pack you are on and is backup failure time and latch warning time matches?
Latch issues mostly related to HW....
http://support.microsoft.com/kb/310834
http://support.microsoft.com/kb/303640
MohammedU
Microsoft SQL Server MVP
January 16, 2007 at 3:46 am
Check your disks. Sounds like you may have IO throughput problems
I had latch waits on one of my databases some months back. We traced it down to very slow SAN access.
Get your hands on SQLIOSim (don't have an url offhand, search microsoft) and maybe also check the perfmon disk counters, make sure they aren't excessive (physical disk sec/read, sec/wwite, sec/transfer should be under 10ms)
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
January 16, 2007 at 8:49 am
I have seen this error when the transaction log can't grow fast enough. Are you using Maintenance Plans? Are you doing reindexing along with the backups?
Check that you allow your tlogs to autogrow and make sure it's set to a #of MB and not a percentage.
-SQLBill
January 13, 2012 at 9:23 am
Unfortunately, we are still running some servers with SQL 2000 SP4. I was recently getting this error. Other symtoms may occur such as backups being blocked, processes blocked by negative SPIDs (or even *). Basically, this is a bug in SQL Server that appears to be a rare occurance. We saw this problem early in the morning and were able to make it to the evening with little problems besides the lack of backups. To resolve the issue followed: http://support.microsoft.com/kb/898709, but since it was on an active\active cluster, we tried to shut down the SQL Server resource in cluster administrator (Server 2003). It failed, so we had to reboot the node. Once the node came back up, we moved the SQL Group back over and it started successfully. I checked the error logs and they were clean. The latch type 3 error was not present anymore and I was able to take a backup.
This is now 2 days later, and still no issues. Hope this helps!
-Slev
Viewing 5 posts - 1 through 4 (of 4 total)
You must be logged in to reply to this topic. Login to reply