Cameron:

Easy there.  Put it on its own SCSI channel, lest it hangs the bus and fouls
the another disk.  This happened to me.

JTN
----- Original Message ----- 
From: "Robert Zagarello" <bzag0@yahoo.com>
To: "NetBSD Port-i386 Users" <port-i386@NetBSD.org>
Cc: "Manuel Bouyer" <bouyer@antioche.eu.org>
Sent: Monday, June 13, 2005 1:17 PM
Subject: Re: [port-i386@NetBSD.org] SCSI drive failure.


>
> Manuel,
>
> If it is at all possible remove the drive and put it
> on another SCSI chain on another computer, with the
> same and/or different cable, just to see if you get
> any reported errors.  Use a chain you know works.  It
> sounds like you've just installed NetBSD on an old
> system, meaning you don't know if it was working
> previous to installation.  The OS you use to test with
> doesn't matter either, as long as the existing chain
> is working.
>
> BZAG
> ===========================
>
> --- Manuel Bouyer <bouyer@antioche.eu.org> wrote:
>
> > On Mon, Jun 13, 2005 at 12:13:54PM -0400, Cameron
> > Brown wrote:
> > > Hi.
> > >
> > > I am having a recurring SCSI error on an older box
> > with a new NetBSD installation. I am not finding any
> > useful answers out there in netland yet, though will
> > keep searching.  Would also appreciate any feedback
> > from this group.
> > >
> > > Thanks,
> > > Cameron
> > >
> > >
> >
> ________________________________________________________
> > >
> > > HERE IS THE ERROR
> > >
> >
> ________________________________________________________
> > >
> > > sd0 (esiop0:0:0:0): command with tag id # reset
> >      [ this line repeats with # incrementing ]
> >
> > What's interesting is what caused the SCSI bus to be
> > reset. This is probably
> > in previous messages. Could be a command timeout or
> > a parity error.
> >
> >
> > > sd0: async, 8-bit transfers, tagged queuing
> > > sd0: sync (50.00ns offset 15), 16-bit (40.000MB/s)
> > transfers, tagged queuing
> > > sd0: waiting for pack to spin up...
> >
> > This probably means the drive's motor has stopped.
> > I'd say this drive is
> > dead.
> > But command timeout can also be caused by bad SCSI
> > cables.
> >
> > -- 
> > Manuel Bouyer <bouyer@antioche.eu.org>
> >      NetBSD: 26 ans d'experience feront toujours la
> > difference
> > --
> >
>