January 29, 2009 at 7:45 am
Need help in approaching a case of stalled I/O. Our DB is SQL Server 2000 SP4 on Windows 2000 SP4, two node cluster with 4GB RAM, 100 GB SAN drive for the cluster. Hardware is HP Proliant DL580.
We see that abruptly the performance drops during the day. Queries start hanging for several seconds. We are seeing lot of Page IO Latches. I have gone through the MS articles: http://support.microsoft.com/kb/897284 and http://msdn.microsoft.com/en-us/library/aa175396(SQL.80).aspx but these are guidelines. Has anyone actually successfully resolved cases of stalled IOs and how?
Regards,
Girish
January 29, 2009 at 8:06 am
It's hard to say, could be a multitude of things, maybe someone wrote a proc with READCOMMITTED TABLOCKX with a WAITFOR statement for 30 mins. Rebuilding your indexes during peak production?
I'd suggest running Performance Monitor on the server and analysing some relevant counters such as disk reads and writes, processor useage, locks and blocking locks (check spwho2 to).
Max
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply