ML110G4 Xeon3040 throws parity errors every month or so

[Stefan Huebner]

Hi all,

we have a ML110G4 with the Dual Core Xeon 3040 running Windows Server 2000 and MS SQL Server 2000.
The machine locks up with blue screen and the message "NMI: Parity Check / Memory Parity Error".
Sure enough we have swapped memory after each of those incidents, we had the server running on the original 512MB HP stick (one and two of them), a 1GB HP stick, Elpida 2x1GB and Kingston 4x1GB, none of them kept the system up.
I did not manage to simulate the failure by applying heavy stress on the OS or the SQL server.
I will swap the raid disks with the system into a new machine with the same configuration this week, but I don't feel very comfortable this would help anything.
Is there a known compatibility issue with Windows 2000 on this machines by the way?

regards
Stefan Huebner

[Blazhev_1]

Hi Stefan,

there is no known issue I know about, except that windows 2000 is not officially supported...w2k is not tested and there are no HP drivers for the components...

Do you have any optional components installed, any PCI cards or something like that? Does the system crash during normal operation or while loading windows?
Is the BSOD occuring while there is heavy load, or there is always such load :)
You can try to run the Offline insight diagnostics to check for hardware error in memory and CPU :

http://h20000.www2.hp.com/bizsupport/TechSupport/SoftwareDescription.jsp?lang=en&cc=us&prodTypeId=15351&prodSeriesId=3186395&swItem=MTX-0dca7662b68647a09c8129213b&prodNameId=3279681&swEnvOID=1005&swLang=8&taskId=135&mode=4&idx=3

Other thing is to check the memory dump to see if this is a driver-related issue(you will find that too when moving the HDDs to the other server)

Pac

[Stefan Huebner]

Thanks for your reply,
no, I've no cards installed, just a Surestore DDS4 drive on internal USB2.0.
It does appear randomly, last time on saturday, with almost no activity on the machine.
I already downloaded the test tool and will check the other 3040 based ML110 sitting here which will be swapped for the troublesome one when it passes.

[KarloChacon · Accepted solution]

hi

Stefan

check if the error you got is similar the one you have in this advisory

http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?lang=en&cc=us&objectID=c01177414

regards

[Stefan Huebner]

Hi,
it's exactely that message, but we don't have the storport.sys and/or the hotfix installed.
But it confirms that we most probably have a driver problem. But hey, how should we rule out the culprit when it takes 4-6 weeks for the error to appear and not leaving any trace (like minidump or memory dump) except from a dumb "the system has been shut down unexpectedly" (free translation from the german log entry)

I think I will change the machine and try to get updates for the BIOS and all installed hardware for now.

[KarloChacon]

hi

I had the same issue with Dl380 G2

a call to Microsoft support, applying SP 2, some other patches and hot fix solved the issue

regards

[KarloChacon]

hi again

I see you have W2000 but try call MS

regards