Solaris and HP Tape Drive Problems

[CraigM]

Hi,

I have a problem with a tape drive, and would like to know how I can diagnose if it is a probelm with the SCSI on the our V440 box.

Over the last couple of days, our tape unit has failed to backup with various errors. Being an HP tape drive, we thought we would swap it and try another before going to Sun and it also failed. We also swapped SCSI cables, but nothing happened.

Unfortunately, Sun support seem very slow to get back, and when they do they ask me to run varous commands to check the status, all of which seem to give errors. For example

iostat -En

Media Error: 0 Device Not Ready: 0 No Device: 0 Recoverable: 0
Illegal Request: 0 Predictive Failure Analysis: 0
rmt/0 Soft Errors: 0 Hard Errors: 0 Transport Errors: 39
Vendor: HP Product: Ultrium 1-SCSI Revision: E32D Serial No:

or mt -f /dev/rmt/0 status
/dev/rmt/0: no tape loaded or drive offline

there are also errors in the messages file, and I have run explore and sent the info off to them.

Is there anything else I can do to check this? I am reluctant to take the system down again, as I have had to throw our 60+ users off a number of times to run the diagnostics they require.

Thanks and sorry for the long post.

Craig.

 

[cndcadams]

CraigM;

Is this a single tape unit or library. Post a few of the errors from messages file. What is the tape drive attached to, onboard scsi port or hba? What model tape drives are you working with.

mt -f /dev/rmt/0 status
/dev/rmt/0: no tape loaded or drive offline

was there a tape in the unit?


Thanks

CA

 

[CraigM]

Hi CA,

the Tape unit is a single tape device HP Ultrium 230. It did have a tape in it, and we have swapped it a couple of times to eliminate it being a tape problem. It is connected to an onboard scsi and we've changed the cable to check that too.

I've included an extract from messages below. Any help would be appreciated.

Thanks,

Craig

Oct 18 05:00:00 helio scsi: [ID 365881 kern.info] /pci@1f,700000/scsi@2,1/st@0,0 (st21):
Oct 18 05:00:00 helio <HP Ultrium LTO>
Oct 18 05:00:00 helio scsi: [ID 193665 kern.info] st21 at mpt3: target 0 lun 0 Oct 18 05:00:00 helio genunix: [ID 936769 kern.info] st21 is /pci@1f,700000/scsi@2,1/st@0,0 Oct 18 05:01:07 helio genunix: [ID 408822 kern.info] NOTICE: mpt3: fault detected in device; service still available Oct 18 05:01:07 helio genunix: [ID 611667 kern.info] NOTICE: mpt3: Disconnected command timeout for Target 0 Oct 18 05:02:07 helio genunix: [ID 408822 kern.info] NOTICE: mpt3: fault detected in device; service still available Oct 18 05:02:07 helio genunix: [ID 611667 kern.info] NOTICE: mpt3: lost interrupt during polling - resetting controller Oct 18 05:02:07 helio scsi: [ID 365881 kern.info] /pci@1f,700000/scsi@2,1 (mpt3):

 

[cndcadams]

CraigM;


What OS is this?

Also run this for me

prtconf -vp | grep firmware-version


or

# dmesg | grep 'Firmware version'


and prtconf -V


post the output.


Also a quick look around is pointing me to a possible driver issue with mpt or OBP issue.

Did sun have you check on the OBP or mpt firmware?


Thanks

CA

 

[CraigM]

Thanks,

I have run the commands you asked and get

firmware-version: '1.03.11.01'

OBP 4.18.10 2006/03/03 13:57.

Sun support have said they will be sending a driver upgrade so I guess I'll give that a go and see what happens first.

Thanks,

Craig.

 

[cndcadams]

CraigM;

If you are running Solaris 9 the kernel patch 118558-34 is suppossed to fix mpt issues. They used to have an mpt driver patch but it was obsoleted by the kernel patch.

Just a thought while you are waiting on sun to ship you the upgrade.


Thanks

CA

 

[CraigM]

Thanks for your help,

just spent the morning with a Sun engineer, who replaced the board. Everything is up and running, so I'm now doing a full backup.

Having said that, it does seem to be taking a lot longer than the normal backup, but I'll have a look around the system first and maybe try another scsi cable first.

Craig.