*BSD News Article 67440


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.rmit.EDU.AU!news.unimelb.EDU.AU!munnari.OZ.AU!news.hawaii.edu!news.uoregon.edu!arclight.uoregon.edu!enews.sgi.com!decwrl!sunsite.doc.ic.ac.uk!yama.mcc.ac.uk!winu.fa.umist.ac.uk!bill
From: bill@winu.fa.umist.ac.uk (Bill Nutt)
Newsgroups: comp.unix.bsd.freebsd.misc
Subject: Re: Does Archive Viper 150 SCSI tape work reliably?
Date: 2 May 1996 04:39:11 GMT
Organization: Sirius Cybernetics Corporation
Lines: 44
Message-ID: <4m9e9f$3cs@yama.mcc.ac.uk>
References: <4m5tjq$gq3@yama.mcc.ac.uk> <4m8mie$jgm@gw.dgii.com>
NNTP-Posting-Host: winu.fa.umist.ac.uk
X-Newsreader: TIN [version 1.2 PL2]

Mike Pritchard (mpp@digibd.com) wrote:
: Yes, I have one on my system, and it works just fine.
: The 27/284x driver has gone through a lot of development
: since 2.1 was released.  You might try running 2.1-STABLE
: instead and see if that makes the problem go away.

hi again.    I'm SURE its not a config or termination thing - linux and OS/2 
(naturally DOS/win too) work just fine.

I've just recompiled the kernel to 2.1-stable, and it seems to have 
helped a lot.   Previously, the machine actually locked and messed the
filesystem....now...well...the below is what happened right at the end of
my first test tar (20Megs) to the tape.   It also happened immediately
after I tried to restore too...

sd1(ahc0:1:0):timed out in command phase, SCSISIGI==0x84
st0(ahc0:6:0):asserted ATN - device reset in message buffer
st0(ahc0:6:0):timed out in command phase, SCSISIGI==0x94
ahc0: Issued Channel A Bus Reset #1, 3SCBs aborted
sd1(ahc0:1:0):UNIT ATTENTION asc:29,0
sd1(ahc0:1:0):Power on reset, or bus device reset occurred field replaceable
unit: 29
,retries:3

after that, the machine seems to recover (thankfully!) - the SCSI bus doesnt
lock up like it did before.

so - looks like the slow tape doesn't relinquish the bus quick enough, and
clobbers a command intended for the disk..??????

ideas anyone?  the tar TO the tape was so close... it must be something
pretty small - as I said, 2.1-stable is a radical improvement.

the 2742AT is an EISA controller.   I also have a DX4/100 overdrive in a 
33MHz motherboard.

if anyone wants any more info to help with a possible solution, feel free to 
mail.

thanks again

--Bill Nutt
bill@winu.fa.umist.ac.uk