Date: Fri, 29 Nov 1996 08:47:15 +0100 (MET) From: J Wunsch <j@uriah.heep.sax.de> To: freebsd-hackers@freebsd.org (FreeBSD hackers) Subject: Re: eject for JAZ drives Message-ID: <199611290747.IAA29645@uriah.heep.sax.de> In-Reply-To: <199611290054.LAA18652@genesis.atrad.adelaide.edu.au> from Michael Smith at "Nov 29, 96 11:24:30 am"
next in thread | previous in thread | raw e-mail | index | archive | help
As Michael Smith wrote:
> > scsi -f /dev/rsd1.ctl -c "1b 0 0 0 2 0"
>
> Unless there's a significant difference between the rsd1.ctl and rsd1c
> device for scsi(8)'s purposes, this doesn't work.
That's surprising. It's the same that all MO drives or CD drives are
using.
So somebody has to get their documentation then.
> Some interesting observations based on my testing this morning :
>
> - setting the 'immediate' bit in the above command crashes the 'ncr'
> driver ("1b 1 0 0 2 0"); I get timeouts and system death.
Expected, i'd say.
> - sending a short command causes the system to eventually grind to a
> halt ("1b 0 0 0 2") (was under X so I didn't see any console messages).
Also not unexpected. Basically, it kills that SCSI bus for quite some
time. Of course, it only grinds your system to a halt if the system
relies on that SCSI bus. :-/
Both are good reasons why the direct SCSI commands are only allowed
for the superuser.
--
cheers, J"org
joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE
Never trust an operating system you don't have sources for. ;-)
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199611290747.IAA29645>