SCSI Ultra320 Warning/Error on RX4640

[amandanew]

I am seeing the following errors on one of my HP Integrity RX4640 server running under HPUX 11.23.

Can anyone pls help to provide some clues or explaination on the attach error/warning and advise me if I need to take any action?Appreciate your help.

Jul 31 04:00:35 sinsx021 vmunix: SCSI Ultra320 0/5/1/0 instance 4: IO Type : SCSI IO has timed-out. Target ID: 1, LUN ID: 0. Inquiry Command - CDB: 12 00 00 00 80 00

Jul 31 04:00:41 sinsx021 vmunix: SCSI Ultra320 0/5/1/0 instance 4: An IO timeout condition was detected. Condition cleared, no intervention required.

Jul 31 04:00:45 sinsx021 vmunix: SCSI Ultra320 0/5/1/1 instance 5: Driver initiating SCSI bus reset. Condition cleared, no intervention required.

Jul 31 04:00:45 sinsx021 vmunix: SCSI Ultra320 0/5/1/0 instance 4: Controller reset has been successfully completed.

Jul 31 04:00:46 sinsx021 vmunix: SCSI Ultra320 0/5/1/0 instance 4: Driver initiating SCSI bus reset. Condition cleared, no intervention required.

[Fabio Ettore]

Hi,

I've never seen that error, but trying to interpret it I'd say that there could be an HW problem on paths 0/5/1/0 and 0/5/1/1 or a driver problem for SCSI Ultra320 driver.

Have you further messages about those paths into syslog.log and/or /var/opt/resmon/log/event.log?

What from

swlist -l product |grep U320

?

HTH.

Best regards,
Fabio

[Steven E. Protter]

Shalom,

I would inspect the system for a SCSI Id conflict or an unterminated scsi connection.

SEP

[Torsten.]

H/W path 0/5/1 is slot 8 in this server.

0/5/1/0
0/5/1/1

looks like a dual port SCSI card - sincwe you have problems on both pathes the problem is likely the HBA itself or even the system board.

I would try to re-seat the card first.

What devices are connected?

Can you provide an

"ioscan -fn"?

From your "ioscan" you will find /dev/mpt__ devices.

Run

# mptutil /dev/mpt__
and
# mptconfig /dev/mpt__

for both and post results.

[amandanew]

Hi all,

Thanks for all your advices. I had provided all in the attach. The only one that I did not provide is the event.log as in this log, the last log was on 2 July and this error only happend today, 31 July.

[Torsten.]

Strange - the first port has only a DDS drive, the second port has nothing attached.

You did the mptconfig/util commands for all the other controllers, but NOT for this one!

It is

/dev/mpt4
/dev/mpt5

Anyway, I would re-seat the card first or even try another free slot with this HBA first.

Consider to update the SCSI driver too.

[amandanew]

Sorry I miss that out yesterday.
I had done the mptutil and mptconfig for mpt4 and mpt5. I'm attaching it for your references and advises. Thanks again.

[Torsten.]

Your earlier ioscan shows a tape drive, but now there are no scsi targets.

As said earlier, you have either a problem with the scsi card or with the tape drive / cables / termination.

[amandanew]

Sorry for not closing this thread as I'm monitoring to see if there is a reoccurrence after replacing the SCSI cable on 6 Aug 2008. After replacing the SCSI cable, the error was gone. Thanks for all the valuable advises.