ML530 G2 Server regular bugchecks with 0x00000077

[Peter Johnstone]

Our ML530 G2 server has recently started bugchecking regularly at 9:40am each day with the following error "BugCheck, STOP: 0x00000077 (0xC000000E, 0xC000000E, 0x00000000, 0x00D96000)". Microsoft article 228753 indicates this is likely to be disk, array or controller card problem - but it seems strange to be happening so consistently at the same time each day?
We have upgraded the firmware on the 6404 controller to the latest version (2.26) which has cleared an error message about the accelerator cache battery; however the server is still blue-screening.
Config is: ML530 G2, dual Xeon 3.06, 2GB RAM,6404 RAID Ctrlr,12 x 36GB HP Disks, Windows Server 2003, SQL Server 2000.

Any help appreciated.

[Peter Johnstone]

Correction to earlier posting: have updated 6404 firmware to 2.32, and updated all drivers with latest Support Pack (7.10)

[Victor Pendlebury]

Hi Peter

Anything scheduled to run at the time ? ( Antivirus, backup ... ? )

Regards
Victor

[Brian_Murdoch]

Hi Peter,

Have a look at the Integrated Management Log (IML) via START-PROGRAMS-COMPAQ/HP SYSTEM TOOLS-IML viewer.

Check to see if you have any noticeable hardware errors reported on the system. You can usually ignore any network port errors.

Also check the system event log and application log for any build up of errors prior to 09:40 on the system. Do you have a user with a particularly process hungry application which runs when he logs in?

Some further hardware/firmware information can also be obtained by looking at the survey report at C:\COMPAQ\SURVEY\SURVEY.TXT.

In addition, if you run the Array Diagnostic Utility, you can generate a report which can be fed into an analysis tool to simplify the results and check if there is indeed an underlying issue with the array controller.

Download the ADU from here.

http://h18023.www1.hp.com/support/files/server/us/download/21867.html

If you wish I can process this for you if you e-mail it to me together with your survey report (survey.txt) to brian.murdoch@hp.com

Regards,

Brian

[Peter Johnstone]

Thanks for your replies. The IML log did report a failed accelerator cache battery however this issue was resolved by upgrading the 6404 firmware from 1.60 to 2.32. The only other item in the IML log is "Drive array not configured"; this is because we don't have any drives connected to the 2nd half (EM) of the 6404.

The first sign that anything is wrong when the system crashes is a "Windows System Error" popup with "Unknown Hard Error" as it's text, and an OK button. This reappears if you click OK, and then after about 1 munute the system blue-screens, with the text "STOP: c00021a Unknown Hard Error, Unknown Hard Error" and then reboots.

Just before (30-40 seconds) the system crashed a process "cqmgserv.exe" started in task manager. This appears to be the HP/Compaq Insight Manager Server Agent; I have disabled this service and it's associated services for now to see whether the problem recurs tomorrow.

We also changed the 6404 controller yesterday for a spare, however this did not fix the problem.

[Peter Johnstone]

Just to update this thread the server did not crash today with the Insight Manager Server Agent service disabled; so it looks like that might be the culprit. I'm leaving it disabled for now until we find out why it might be causing the problem, but would like to re-enable it as soon as possible.

As the server was crashing with an older version of the agent, and continued to crash even after the agent was updated to 7.10 I don't think it is a version specific issue. The only change to the server environment we made at around the time was to approve and install a number of Microsoft critical updates and patches; so there may be a conflict there. These were also installed on a number of our other servers at the same time (also running the Insight agent) without any problem, however the ML530 is the only SQL server.

As always, any advice will be appreciated!