From owner-freebsd-current  Fri Mar 22 00:16:04 1996
Return-Path: owner-current
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id AAA22914
          for current-outgoing; Fri, 22 Mar 1996 00:16:04 -0800 (PST)
Received: from GndRsh.aac.dev.com (GndRsh.aac.dev.com [198.145.92.241])
          by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id AAA22869
          Fri, 22 Mar 1996 00:15:54 -0800 (PST)
Received: (from rgrimes@localhost) by GndRsh.aac.dev.com (8.6.12/8.6.12) id AAA25003; Fri, 22 Mar 1996 00:14:38 -0800
From: "Rodney W. Grimes" <rgrimes@GndRsh.aac.dev.com>
Message-Id: <199603220814.AAA25003@GndRsh.aac.dev.com>
Subject: Re: Crash advice needed (long) APPENDIX A
To: George.Scott@cc.monash.edu.au (George Scott)
Date: Fri, 22 Mar 1996 00:14:38 -0800 (PST)
Cc: current@freebsd.org, hackers@freebsd.org, uhclem@nemesis.lonestar.org
In-Reply-To: <199603220627.QAA17304@moa.cc.monash.edu.au> from "George Scott" at Mar 22, 96 04:27:50 pm
X-Mailer: ELM [version 2.4 PL24]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-current@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

> 
> > sd0(aha0:0:0:) timed out:
> > adapter not taking commands.. frozen?!
> > Debugger("aha1542") called		(this system has a 1540B)
> > AGAIN
> > aha0: MBO 02 and not 00 (free)
> > These last few lines repeat every ten seconds or so until I got
> > tired of looking at them and pressed RESET.
> > 
> > (I have seen this MBO 02 and not 00 (free) message on a 1542C
> >  controller at work when it is talking to a WangDat tape drive.
> >  Sometimes this system also gets a "NO SENSE" message and
> >  will claim the tape drive is not ready (it is ready) or busy (it isn't).
> >  Solution on that system: turn everything off and back on and everyone
> >  is happy.  Unfortunately, cycling power does not improve things here.)
> 
> I had this frequently under 2.1.0-RELEASE on a '486 with an AHA-1542CP (with
> the P stuff disabled).
> 
> I had 5 drives on the external connector and 2 on the internal and noticed
> that the time outs stopped happening when I disconnected either the external
> or internal drives, so I figured bus length as the cause.  It was over 20'
> so I shortened it as much as I could (to 16' if I remember) but the problems
> still occured.
> 
> I ended up putting the internal drives on a spare AVA-1515 I had waiting for
> another project and haven't had one problem since.

Thought the first person here is running a 1540B which is not a fast scsi-ii
controller, I can fully see why George was having a problem, and why it
went away.  The 1542CP _is_ a fast scsi-ii device and I suspect you had
some fast devices (>5MHz) on there.  When running fast scsi-ii it is very
hard to get away with a cable chain longer than 3 meters (~10 feet) and
you had better be using active termination and _high_ quality cables.

I can just see the problems comming when Ultra SCSI 20Mhz starts becoming
popular as that will often require a total cable length on the order of
1.5 meter tops unless you pay extra special attention to the fact that
this becomes a high frequency transmission line and must be treated us such.

-- 
Rod Grimes                                      rgrimes@gndrsh.aac.dev.com
Accurate Automation Company                 Reliable computers for FreeBSD