Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 13 Apr 1999 15:12:05 -0700 (PDT)
From:      patl@phoenix.volant.org
To:        se@freebsd.org
Cc:        freebsd-scsi@freebsd.org, Stefan Esser <se@freebsd.org>
Subject:   Re: Seagate ST15230 wedging
Message-ID:  <ML-3.4.924041525.5566.patl@asimov>
In-Reply-To: <19990413135209.B309@dialup124.mi.uni-koeln.de>

next in thread | previous in thread | raw e-mail | index | archive | help
On 13-Apr-99 at 13:21, Stefan Esser (se@mi.uni-koeln.de) wrote:
> On 1999-04-12 09:04 -0700, patl@phoenix.volant.org wrote:
> > Do any of you have any experience with Seagate ST15230 drives?
> > I have one that occasionally wedges when backing up to tape.  The
> > wedge spews:
> > 
> >      sd1(ncr1:1:0): Vendor Specific asc:80,0 Vendor Specific ASC
> >      field replaceable unit: 1
> 
> Well, that's a vendor specific return code.
> You may have to ask your vendor.

The first time this happened, I went grovelling through their Web site;
but didn't find anything relevant.  I have a cynical expectation that
attempts to get the info from a tech support person will be rather
frustrating.  But it's beginning to look like I'll have to try.  But
first, another check of their Web site...  Ah, it looks like they have
quite a lot more available on-line now.  But I still can't find anything
that looks like 'ASC' or 'Error Code' - are there any other terms they
might be using for this?

> > I've no idea what that message means, except that I can't access the
> > disk and have to power-cycle to reset it.  (Well, actually, it might
> > reset with a simple reboot.  But I figure I may as well power-cycle
> > just to be sure.)
> 
> Are there other error messages from the SCSI
> sub-system ?

Nope.

> Do other SCSI devices continue to work ?

Yep.  The other disk on that controller still responds normally.
(The tape and an ancient CD-ROM changer are on a separate controller.)

> > It's on a system that, for various reasons, is still running FreeBSD
> > 2.2.2, and will be for a little while yet unless I find some urgent
> > reason to upgrade.  (Such as someone assuring me that later drivers
> > for the NCR 53c585 fix it.)
> 
> No, sorry, that's not a driver problem, I'm sure.
> (The SCSI command was executed to the end, and the 
> ASC/ASCQ values obtained through a separate SCSI
> command, after the drive returned an error status.)

Hmm.  I was kind of hoping that maybe the drivers weren't properly
handling the error.  Which implied the hope that there would be some
sort of soft reset that would clear the problem...

> But you may try to disable tagged commands, or to
> lower the synchronous data rate, just to see whether
> that makes a difference.
> 
> The problem you report could even be caused by the
> power supply, which may not be able to deliver 
> sufficient current at 12V (assuming your tape drive
> motor runs on 12V, as do the mechanical parts of the
> disk drive).

The tape drive is an external unit with its own power supply.
I don't think it plays an integral part in the problem.  But
the backup is about the only time that this system experiences
extended periods of heavy disk activity.  Note that this drive
holds the Amanda buffer area for the backups of the partitions
on the other drive; so there would be both heavy read and write
activity.  And it sometimes overlaps with the find in the nightly
security script.

Note that these wedges don't happen every time.  It can be months
between them, or only days.  I don't believe it has ever happened
twice in a row; but if it is related to the (duration of the) load,
that may be an effect of Amanda's choice of full or incremental
backup.


-Pat


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-scsi" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?ML-3.4.924041525.5566.patl>