Return to BSD News archive
Newsgroups: comp.unix.bsd.freebsd.misc
Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!munnari.OZ.AU!news.ecn.uoknor.edu!qns3.qns.net!news-out.internetmci.com!newsfeed.internetmci.com!howland.erols.net!bloom-beacon.mit.edu!apollo.hp.com!lf.hp.com!news.dtc.hp.com!hplntx!hplb!hpcpb!jocko.bri.hp.com!aslater
From: aslater@jocko.bri.hp.com (Al Slater)
Subject: Re: HP C1533A DAT tape ?
Sender: news@bri.hp.com (News User)
Message-ID: <E0pE8u.2H1@bri.hp.com>
Date: Mon, 11 Nov 1996 11:22:06 GMT
Reply-To: aslater@jocko.bri.hp.com
References: <stanbDzLuJv.CEo@netcom.com> <Dzrx1z.9xv@bri.hp.com> <564u6k$ev@klemm.gtn.com>
Nntp-Posting-Host: jocko.bri.hp.com
Organization: Hewlett-Packard
X-Newsreader: TIN [version 1.2 PL2]
Lines: 57
Andreas Klemm (andreas@klemm.gtn.com) wrote:
: In article <54rce7$9ta@uriah.heep.sax.de>,
: j@uriah.heep.sax.de (J Wunsch) writes:
: > Try the `mt comp 1' command. I've just checked, it's around there at
: > least since FreeBSD 2.0.5. The man page still lies that it were not
: > implemented though... but i cannot remember a single feedback whether
: > it's actually working now. It has been implemented blindly after the
: > SCSI-2 specs by Julian Elischer.
: With my TDC 4222:
: root{1007} / mt status
: Present Mode: Density = QIC-154(2GBALDC) Blocksize variable
: ---------available modes---------
: Mode 0: Density = 0x00 Blocksize variable
: Mode 1: Density = 0x00 Blocksize variable
: Mode 2: Density = QIC-150 Blocksize variable
: Mode 3: Density = QIC-120 Blocksize variable
: root{1008} / mt -f /dev/rst0 comp 1
: mt: /dev/rst0: comp: Invalid argument
: root{1009} / mt -f /dev/rst0.ctl comp 1
: mt: /dev/rst0.ctl: comp: Inappropriate ioctl for device
: root{1011} / mt -f /dev/nrst0 comp 1
: mt: /dev/nrst0: comp: Invalid argument
: My TDC 4222 doesn't have jumpers, to enable or disable data
: compression. So ... can you guess, why it doesn't accept the command ?
: Or is there a magic scsi command, that would show the devices exact
: error code ?
Hmm. ISTR you either get the SCSI errors on console or you can rebuild
your kernel for verbose scsi error reports (though that now seems to be
more like Linux). Anyways, you should get a wodge of sense data if it has
gone that badly wrong; i.e the device has check conditioned on being sent
something it can't cope with.
It also depends how the drive implements compression, there are, for
example, at least three ways of turning it on/off. Density code, two
different mode pages, etc.
Inappropriate ioctl looks like it just simply isn't implemented via
that interface.
If that really is a QIC device, are you _sure_ it does DC?
And if it does... you might have to hack about with what the driver
does at st_mount time, as ISTR that there's one mode select that
essentially asks the drive to do variable mode which interacts
badly with some drives' firmware...(OTOH, it sounds like you can talk to
the drive to read/write to which it wouldn't even get that far if this
was a problem).
If you've got the request sense data popping up on console, post it
and we'll see if it turns up anything obvious..
regards,
al
(not speaking for HP..)