[rescue] Disk going bad or other..?

[velociraptor]

On 9/12/05, Eric Webb <ttlchaos at randomc.com> wrote:
> On Monday 12 September 2005 01:52 am, velociraptor wrote:
> > On 9/12/05, Eric Webb <ttlchaos at randomc.com> wrote:
> > >
> > > Fatal SCSI error  at script address 258 Unexpected disconnect
> > > Boot device: disk:a  File and args:
> >
> > I'd check /var/adm/messages and prtdiag -v to see if anything shows
> > up.  If not, then you might have some "dangling" device references.
> > Try running devfsadm with the '-C' switch, which invokes the "clean
> > up" routines.
>
> I'm getting nothing from messages and prtdiag...  I'm not in front of the
> machine so I can't check the devfsadm cleanup, but I question anyway what
> device references in /dev have to do with OBP messages when trying to
> bootstrap off a disk...?

Point taken, I mis-read that.  Perhaps you've got a bogus devalias
in the OBP NVRAM?  If you were actually seeing HW messages
in /var/adm/messages, I'd think intermittent disk or SCSI controller.
(I just had a box that would boot fine, then start throwing disk errors--
root disk vs data disks).  But since you're not, perhaps there's some-
thing funky with the NVRAM?

=Nadine=

=Nadine=