topic Re: ml350 internal 72i scsi tape problem in Operating System - Linux

ml350 internal 72i scsi tape problem

Andrea Rossi — Mon, 19 Apr 2004 08:28:42 GMT

Hi all

I got a problem with an ML350G3 (2 processors, 2,5 GB ram, Redhat ES 2.1 kernel 2.4.9-e27, PSP 7).

performing a full backup of about 30GB I got this message (after about 1,5 hours)

kernel: scsi : aborting command due to timeout : pid 0, scsi1, channel 0, id 5, lun 0 Write (6) 00 00 28 00 00

three times, then system hang, so that I can't even reboot it.

internal scsi is seen as Adaptec AIC-7899 (driver aic7xxxx)
tape is a C7438A

Thanks a lot

Re: ml350 internal 72i scsi tape problem

Mark Grant — Mon, 19 Apr 2004 08:54:27 GMT

personally, I'd suspect the device or the cable. However, you might try the latest red hat kernel which may have an updated driver for the tape drive.

Re: ml350 internal 72i scsi tape problem

Steven E. Protter — Mon, 19 Apr 2004 10:41:00 GMT

causes in rough order of priority

1) Problem with drive.
2) problem with cable
3) problem with scsi termination
4) problem with scsi card.

You need to check all of these things with utilities or eyeballs.

SEP

Re: ml350 internal 72i scsi tape problem

Dineshkumar Surpur — Thu, 22 Apr 2004 00:34:39 GMT

The abort command here is due to the fact
that there was no reply from the device on
the status of write command and driver
has timed out on that write resulting in kernel panic. You can do the following

1. Try backup for a small GB and see if it
is successful
2. Update the tape device firmware and driver
update.

Most probably i feel its a device problem
rather then a termination or cable issue.

Re: ml350 internal 72i scsi tape problem

Andrea Rossi — Thu, 22 Apr 2004 02:31:45 GMT

they found out the jumpers on the tape were wrong.
Anyway after tape substitution the system hanged on Tuesday night, before the backup.
Tonight backup was successfull, dispite the fact I found "kernel: scsi : aborting command due to timeout" in /var/log/messages.

Re: ml350 internal 72i scsi tape problem

Andrea Rossi — Mon, 26 Apr 2004 04:55:57 GMT

the system hangs once a day!
what's happening?
the hardware test was successfull.

P.S. I found
mtrr: your CPUs had inconsistent fixed MTRR settings
mtrr: probably your BIOS does not setup all CPUs

in /var/log/messages.