Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!simtel!oleane!tank.news.pipex.net!pipex!in2.uu.net!noc.near.net!news3.near.net!news-server.bos.locus.com!orchard.la.locus.com!janus.la.locus.com!not-for-mail
From: fmayhar@janus.la.locus.com (Frank Mayhar)
Newsgroups: comp.unix.bsd.freebsd.misc
Subject: Re: SCSI Timeout with Adaptec 2940 and DAT drive - Help!
Date: 18 Aug 1995 11:13:15 -0700
Organization: Locus Computing Corporation
Lines: 37
Message-ID: <412l7r$2ilp@janus.la.locus.com>
References: <408m4r$7vu@news.euro.net> <40vffs$mdc@wup-gate.wup.de>
Reply-To: fmayhar@locus.com
NNTP-Posting-Host: janus.la.locus.com
In article <40vffs$mdc@wup-gate.wup.de>, Andreas Klemm <andreas@wup.de> wrote:
>gary@kampai.euronet.nl wrote:
>: I've been having problems restoring DAT backups on my system.
>: I'm running an Adaptec 2940 with a SCSI disk, HP DAT, and a CDROM.
>: The setup works fine under windoze. I've checked all cables, and
>: disconnected the CDROM, but still I keep getting SCSI timeouts
>: which cause my system to hang. The installation worked fine from
>: DAT.
>All devices internal or some externel ?
>Which cable, how long ? Suitable for Fast SCSI II ?
>What about Termination ? Really correctly terminated ? Only both ends ?
>Parity enabled or not ? If enabled, then on all devices.
>What about Controller settings ? Synchr. negotiation 'yes' or 'no'
>for all devices. I'd try 'no' if there are problems.
>What about the machines BIOS settings ? TUNED like hell or not ?
>Try standard values ...
I've seen something like this once, myself. I have an external Archive
Viper 2150S drive, I was using a 250-meg tape, and tried to do an fsf 2.
The first filemark was very far down the tape, and the driver eventually
timed out the operation, subsequent to which point sd0 and sd1 were timed
out (despite not being associated with the operation) and the SCSI bus
hung. I had to hard-reset (yukko!) the box to fix things.
I'm using a 2940W. Cabling is a 50-pin ribbon to three drives internal to
the main box, culminating in a 50-pin Centronics connector to which is
connected a standard, fairly good-quality shielded external cable to an
old Sun shoebox with new cabling, that contains the tape drive. The daisy
chain ends on the out side of the shoebox in active termination.
I really don't think it's a problem with the cabling, I think that the
problems after the initial timeout are due to bugs in the driver itself.
Needless to say, I won't be doing any more fsf operations.
--
Frank Mayhar fmayhar@locus.com
Locus Computing Corporation
9800 La Cienega Blvd., Inglewood, CA 90301 (310) 337-5007