Server Hangs since upgrade to SQL2k Enterprise from 7.0.
Server Hangs since upgrade to SQL2k Enterprise from 7.0 forcing a hard reboot. Ping still responds, but no RPC requests are handled.
We've been trying to stabilize our SQL2k environment ever since the upgrade from 7.0. At first, we thought we narrowed the problem down to the use of the maintenance wizard, as the hang took place right at the time of day that the backups were kicking off. Now the server hangs at random times throughout the day (more than five times on Friday!). Then it was stable through the weekend and Monday, and hung early this morning.
There seems to be a pattern of entries in the SQL Agent log prior to the server hanging, but it doesn't necessarily hang after every occurrence, so it may be ANOTHER wild goose chase.
Here are the log entries:
2001-03-14 14:36:00 - ! [LOG] Unable to read local eventlog (reason: The handle is invalid)
2001-03-14 14:36:00 - ? [LOG] Attempting to to re-open the local eventlog...
2001-03-14 14:36:00 - ? [201] Attempting to resize the eventlog read-buffer to 18284 bytes...
2001-03-14 14:36:00 - ? [201] Attempting to resize the eventlog read-buffer to 18292 bytes...
2001-03-14 14:36:00 - ? [201] Attempting to resize the eventlog read-buffer to 18452 bytes...
2001-03-14 14:36:00 - ? [201] Attempting to resize the eventlog read-buffer to 16384 bytes...
2001-03-14 14:36:00 - + [LOG] Successfully re-opened the local eventlog - NOTE: Some events may have been missed
This suspicious Error Log entry may be unrelated, but I include it here anyway:
2001-03-14 09:37:50.47 spid65 BobMgr::GetBuf: Bob write not complete after 60 seconds.
The server is a Dell Quad 450 with a meg of secondary cache and 4Gb of RAM and is connected to a Dell Power Vault Fiber Channel via two QLogic controllers. It's running Win2k sp1 with the two latest hot fixes applied (the hot fixes didn't fix it).
Ideas anyone???
Server Hangs since upgrade to SQL2k Enterprise from 7.0. (reply)
------------
Ken Haff at 3/14/01 3:32:16 PM
Server Hangs since upgrade to SQL2k Enterprise from 7.0 forcing a hard reboot. Ping still responds, but no RPC requests are handled.
We've been trying to stabilize our SQL2k environment ever since the upgrade from 7.0. At first, we thought we narrowed the problem down to the use of the maintenance wizard, as the hang took place right at the time of day that the backups were kicking off. Now the server hangs at random times throughout the day (more than five times on Friday!). Then it was stable through the weekend and Monday, and hung early this morning.
There seems to be a pattern of entries in the SQL Agent log prior to the server hanging, but it doesn't necessarily hang after every occurrence, so it may be ANOTHER wild goose chase.
Here are the log entries:
2001-03-14 14:36:00 - ! [LOG] Unable to read local eventlog (reason: The handle is invalid)
2001-03-14 14:36:00 - ? [LOG] Attempting to to re-open the local eventlog...
2001-03-14 14:36:00 - ? [201] Attempting to resize the eventlog read-buffer to 18284 bytes...
2001-03-14 14:36:00 - ? [201] Attempting to resize the eventlog read-buffer to 18292 bytes...
2001-03-14 14:36:00 - ? [201] Attempting to resize the eventlog read-buffer to 18452 bytes...
2001-03-14 14:36:00 - ? [201] Attempting to resize the eventlog read-buffer to 16384 bytes...
2001-03-14 14:36:00 - + [LOG] Successfully re-opened the local eventlog - NOTE: Some events may have been missed
This suspicious Error Log entry may be unrelated, but I include it here anyway:
2001-03-14 09:37:50.47 spid65 BobMgr::GetBuf: Bob write not complete after 60 seconds.
The server is a Dell Quad 450 with a meg of secondary cache and 4Gb of RAM and is connected to a Dell Power Vault Fiber Channel via two QLogic controllers. It's running Win2k sp1 with the two latest hot fixes applied (the hot fixes didn't fix it).
Ideas anyone???
Lack of disk space or poorly hard disk?
You should check all the logs - SQL Server & NT ones.
Has the workload changed?
Has anything else changed recently
Best Wishes,
Andrew
