STOP error after PSP 7.8

[Jim Mulvey]

Why isn't there someone from HP on this thread to say authoritatively where they are on this issue?

I've been lurking for a few day, but I'm also a sufferer of this problem. I'm sure there are many more like me.

Is this a priority for HP?

[Bruce Holzrichter]

For everyone, I am currently working with HP escalation engineer on this issue and they acknowledged the issue as real.

There are two distinct BSOD issues, however, one is NIC related. I have NOT run into that one on our equipment so can't speak to that particular issue.

The other is Agent related, but could be hardware,driver,agent as HP has NOT identified root cause for it yet, but HAS been able to duplicate it.

The only solutions are workarounds currently, I just got off the phone with HP engineering and the latest and easiest workaround if you want to run 7.8 agents still is to:

Under "Control-Panel -> HP Management Agents" from the Services tab, select "Drive Array Information" from the "Active Agents" and "Remove" Ensure it is now listed in the "Inactive Agents"

I am testing the above now, so far so good.

[Bruce Holzrichter]

ok, so my machine has blue screened twice under testing with the newer fix offered by HP of disabling the "Drive Array" agent.

I am going back to removing the whole "HP Insight Management Agents" as I don't expect an fix from HP until next week at the earliest.

[Chris Young_6]

All,

To date, I can report that down-grading the NIC drivers did not solve the BSOD issues with my four DL380 G5s.

My experience has been similar to Bruce H. and Robert Mader. I thought the problem was related to storage drivers.

These machines were built on 5/16/07, along with numerous other models as part of a new distributed SQL DB environment. All the systems have the same OS, service packs, and applications, along with PSP 7.80a. These hosts ran fine with SAN LUNs attached.

I experienced my first BSODs with the DL380 G5s when I began presenting snap-clones and snap-shots of SAN LUNs on 6/12/07.

After reviewing the varied BCCs that were reported, I contacted HP Storage Support, who suggested that the problem was related to the Microsoft Storport driver.

However, nothing we did related to storage drivers resolved the problem.

Then I stumbled accross this thread two days ago, and for me the only fix that worked was completely removing PSP 7.80a. I removed that and replaced it with PSP 7.70a as suggested by Joshua Small and Jesse Zellmer on 6/13, and re-inforced by Robert Mader on 6/15.

I did some detailed testing of the TOE issues as reported by Jonathan Rees (and others), as our company has experience fatal errors in the past attempting to integrate TOE products into our enterprise infrastructure.

As I related previously, we particulary need the teaming functionality of HPs NC373i and Intel Pro/1000 PT Quad port cards in these systems.

No changes to teaming, NIC drivers, or firmware resolved the BSODs we were experiencing.

Summary:

Use appwiz.cpl and remove all HP titled objects (except for specialty items like MPIO, etc)

Reboot, and install PSP 7.70a with the force over-write existing and newer option.


Thanks again to all those who posted problems and fixes for this issue.

[Rita Salim]

Arian, i did the shutdown or reboot, no, the server doesn't have BSOD but it prompted pmp.exe - application error instead, I just clicked ok to continue the shutdown or reboot.

[Joshua Small_2]

Hi Bruce,

As of my last post, HP's "backlevel engineers" were still telling me that the only issue in existence is a confirmation that I have failed to install the latest drivers.

Whilst I appreciate you acknowledging some issue here, I would desperately urge you to get some official tech note out there or something.

[Joshua Small_2]

To clarify my last post, it should have better read "urge the engineers you are talking to" to release an official notice.

[Jesse Zellmer]

Microsoft just came out with a KB article about: You receive a "Stop 0x000000D1" error message, and the computer restarts automatically in Windows Server 2003.

< http://support.microsoft.com/kb/938648/en-us >

I would be curious if this helps any of the people who are having this issue. I am going on almost 10 days without a BSOD since I downgraded to PSP 7.7A.

[Joshua Small_2]

That does look potentially promising with relation to the NIC issue.
I aren't able to pull down our production server to find out unfortunately, particularly after all the downtime we've just had.

I'll be very curious to see how anyone else goes with it though.

[P.Ritaluoma]

Hi!
Here is another setup that has BSOD problem:
Proliant DL 365 (amd)
13 gb memory
3 x 72 Gb / 15k Sas Drives
512 mb BBWC
2x fc1142 hbas linked to eva8000
W2003R2 X64 enterpise w SP2 + patches installed week ago.
Microsoft Cluster Service.

One Nic reserved for public lan and the other for Cluster heartbeat i.e no teaming.

Applying this hotfix http://support.microsoft.com/kb/932755 did not resolve the issue.

After read this whole thread, I will try to fix it with uninstalling Insight Management Agents like Bruce Holzrichter did. I post the results later.