ProLiant DL380 G3 PCI Bus Error source?

[digimonx]

Hi,
The server reboots every night logging "PCI Bus Error (Slot Unknown, Bus 0, Device 16, Function 0)" and "ASR Detected by System ROM" in Integrated Management Log.

The System Management does not show any PCI device numbered 16.
There is an Adaptec 29320LP SCSI card in PCI-X slot 1 with an up to date driver and HP LTO4 tape drive connected to it.
It seems the server reboots when Backup Exec 12 starts verifying the backup job written to tape.

One of the hard drives is in predictive failure mode, but I doubt it has anything to do with the error.

Both, the card and tape drive, work fine on another server.

What could cause the issue?

Thank you.

[digimonx]

I reseated the PCI board and SCSI card, and the server did not reboot or log any errors since then.

I hope this took care of the issue. I will post back if it happens again.

[digimonx]

There was something else I forgot about and did not mention: I also modified the BackupExec job settings so that it does not verify backup. And now I think that it was not reseating of the cards that helped, but not verifying the backup job.

I am going to enable backup verify and see how it goes.

[digimonx]

I confirmed that it is Backup Exec that causes ASR when starting Verify operation.

Any suggestions?

It seems like I talk to myself here :) Is there anyone alive?

Thank you.

[digimonx]

I disabled ASR and when doing a restore using Backup Exec got a blue screen:

*** Hardware Malfunction

Call your hardware vendor for support

*** The system has halted ***

Then I moved Adaptec card from 133MHz slot to one of 100MHz slots and restore seems to be going just fine now, but probably slower than it would go using 133MHz slot?

I assume the 133MHz PCI-X slot is bad or not compatible with Adaptec card, although it is 133MHz card?

I will exchange the PCI board from another Proliant that does not have expansion cards and see how it does.

[digimonx]

I tried the PCI board from another Proliant server, but it did not help. The same error and reboot when using 133MHz slot, and works fine on 100MHz slot.

There must be some compatibility issue on this system, because  the card worked fine on another, not HP, server using 133MHz slot. I will check if there is any information on that on the web.

[digimonx]

It appears when the 133MHz slot is used for Adaptec card then it shares IRQ with built in SmartArray 5i SCSI controller. When 100MHz is used then Adapted does not share the IRQ.

I think this could cause the blue screen when Backup Exec uses the tape drive connected to Adaptec card on 133MHz slot.

Can anyone confirm?