HP Ultrium 960 and Backup Exec 9.2 - problems with full backup

[cechapman]

Backup Exec - Job Server Version 9.20.1391
Netware 6.5
scsi driver lsimptnw.ham v3.99.05
This same error is occurring on three different servers, all with the same configuration. It ONLY happens when doing a full backup or restore, incrementals are fine, which makes me suspect a speed mismatch of some kind when a large amount of data is being transferred.

We have tried moving the card to a faster slot, no difference, changing the block size transfer within Backup Exec, no difference. We must resolve this problem asap. If anyone has any suggestions, we would be grateful to hear them. Thank you.

Log results pasted below:

##ERR##Error on HP ULTRIUM960 ID:3 HU10539H3G.
##ERR##A hardware error has been detected during this operation. This
##ERR##media should not be used for any additional backup operations.
##ERR##Data written to this media prior to the error may still be
##ERR##restored.
##ERR##Possible corrective actions for this error are:
##ERR##1.) Use a different media
##ERR##2.) Check the SCSI cabling.
##ERR##Vendor: HP
##ERR##Product: ULTRIUM960 DRV
##ERR##ID:
##ERR##Firmware: G2AD
##ERR##Function: Write(5)
##ERR##Error: A hardware failure has occurred on drive HP ULTRIUM960
##ERR##ID:3 HU10539H3G. Please consult the job log for details.(10)
##ERR##Sense Data:
##ERR##70 00 04 00 - 00 00 00 00 - 00 00 00 00 - be 08 00 00
##ERR##00 00 00 00 - 00 00 00 00 - 00 00 00 00 - 00 00 00 00

[David Ruska]

The Sense data indicates a "hardware error" sense key (the "4" in byte 2), but the ASC/ASCQ (additional sense) data of BE 08 (bytes 12 & 13) is not one that the drive should return.

Can you install and run HP library & tape tools (www.hp.com/support/tapetools) and collect a support ticket from a drive? We can then look to see what errors the drive may have logged.

Is this scsi attached, or a fibre attached drive?

[cechapman]

It is scsi connected. We have tried to run the HP library & tape tools but it abends our servers.

[David Ruska]

> We have tried to run the HP library & tape tools but it abends our servers.

Since you mention this on multiple servers, it would be good to email the ltt_team@hp.com, and they can look at the abend issue. The cause might shed some light on your backup problem as well.

[Erik Withaar]

Hi,

I get exactly the same errors, using Backup Exec 9.0 (with newest drivers) and running NetWare 6.5 SP5. Is this problem yet resolved? and if so, what's the solution?
Regards,
Erik Withaar
Imtech Projects Zuid B.V.
The Netherlands

[David Lincheck]

Hello,

I am also seeing the exact same problem on a new server/new tape drive that I am installing.

Ultrium 960 attached to a HP/LSI scsi card in a HP ML370 running Netware 6.5 SP5 with Backup Exec 9.1.

[David Lincheck]

Unloading the CPQSCSA.NLM file seems to have fixed the problem. Two successful full backups and a third test in progress. I'm not sure what functionality this nlm provides, but it appears to be in conflict.

Dave

[Erik Withaar]

The resulotion of Dave works for me also:
at 22:55 I unload CPQSCSA.NLM
at 23:00 the backup starts
at approx. 01:45 the backup stops (succesfully)
at 07:00 I load CPQSCSA.NLM

unloading & loading is managed by cron.nlm
Entries in sys:etc\crontab:

#Unload & load cpqscsa (backup ULTRIUM960 drive)
55 22 * * * unload cpqscsa
0 7 * * * load cpqscsa

Thanks,
Erik

[mikalp]

We did have same error. We have HP 360G4p server with autoloader 1/8 960. CPQSCSA.NLM when unloaded backup is working fine.

What is CPQSCSA.NLM? is it important? do we need to load this?

[Marino Meloni_1]

have a look at this doc

http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?lang=en&cc=us&objectID=PSD_EN011106_CW01&jumpid=reg_R1002_USEN