PFN_LIST_CORRUPT BSOD on Proliant DL380G4

[George Salmaniw]

Have an Adaptec 29160 installed to connect to an external NEXSAN ATABOY1 array in a DL380G4.

On an intermittent basis will receive a BSOD message

PFN_LIST_CORRUPT
0x0000004e (0x0000009a,0x000005659, 0x00000006, 0x00000002)
core dump then reboot

In the event log it seems that prior to the BSOD I get an Event ID 11 error from source adpu160m - The driver detected a controller error on \Device\Scsi\adpu160m1. The event may occur anywhere from 20m to 12 hrs before the BSOD.

I also have 6i used for internal mirror of two 36GB drives, and the LSI Ultra320 connectec to HP SDLT 320 drive.

I don't believe the HP HBAs have anything to do with the BSOD. Based on what I've read about this error, it seems to be more of a driver issue and tracking of physical pages of memory corruption, or incorrect MDL handling. As one would expect a BSOD after a controller, the delay in BSOD makes me believe it to be a MDL handling issue. I am currently running verifier to try and isolate the adpu160m driver. Verifier shows it as v6.4.645.100 built by WinDDK.

Any suggestions as how to proceed; or similar problems?

It may just be a simple matter of a controller HW problem?

[Oleg Koroz]

would you have option to try instead 64-Bit/66MHz an 64-Bit/133MHz adapters ?

http://h18000.www1.hp.com/products/quickspecs/12028_na/12028_na.HTML

Storage Controllers

[George Salmaniw]

Not really ... NEXSAN recommends the Adaptec 29160 or 39160.

[Oleg Koroz]

29160 is not listed under optional
if I'm not wrong G4 requires 64/133 Cards in the Riser

[George Salmaniw]

The 29160 is 64bit PCI card ... and I have to think that there are others using Adaptec cards in Proliant machines?