PCI BUS Error

[Byron Zhu]

For the past weeks, We have a problem on a Proliant DL580 Gen 1.
We get the following error:

1) Blue Screen
Hardware Malfunction
NMI: Parity Check/Memeory Parity Error
The system has halted.

2) In IML
- PCI Bus Error(Slot 4, Bus 7, Device 6, Function 0)
- ASR Detected by system ROM
- POST Error: Acritical Error occurred prior to this power-up

3) In Event Log
Event ID:4196
The compaq System Management Driver has detected that the system encountered a Non-Maskable Interrupt (NMI) prior to this boot. The NMI source was: PCI bus error, slot4.

Slot 4 is one 5302 raid array controller, I have replaced new card, but the same problem exist still.

Thanks for any help.

[Christian Schwartz-Søre]

If you're using Insight Manager, a hardware error should show up on the server, making it easy to identify where the fault is.

Sounds like a memory problem, or problem with a pci slot.

Check Insight Manager for errors. Move Array controller to different slot.

Regards
christian

[Eugeny Brychkov]

If you replace (or already replaced) card in slot 4 with known good one, if you checked that it's seated properly, if you checked PCI slot pins with magnifier that they are ok, then I think you need to replace system board.
Although (as soon as you see errors during POST already) you may try to clear CMOS or use system config utility to clear NVSRAM config and rebuild it from the scratch
Eugeny

[Brian DelPizzo]

That sure sounds like a memory parity error. You may want to download a memory checker and test your memory. If you encounter a single error on parity memory you are OK. If you encounter multiple errors on a failing module, your system will blue screen. The PCI error is probably be a result of the system halting.