Subject: Re: pc164/ncr scsi problems
To: None <,>
From: Ross Harvey <>
List: tech-kern
Date: 08/29/1999 17:27:54
> From: (Brett Lymn)
> According to Darren Reed:
> >
> >I'll try turning of SYNC, but I don't like having to do that.
> >
> I didn't like doing it either but in my case I was just trying to get
> data off the old disk so I lived with it :-/  What I did was just turn
> off sync negotiation on the old disk I had on the chain which seemed
> to do the trick.
> >For reference, this drive was working, synchronous, under SunOS 4.1.4
> >in a Sparc LX (Fast SCSI, just not Ultra) at 10Mbit/s.
> >
> Also for reference, my disks were working fine _for_years_ until I put
> the IBM DCHS09Z and Plextor 820T on the chain then things got tetchy.
> All the cables were the same ones I have been using for ages (twisted
> pair ribbon cable FWIW).  I could sometimes boot up and sometimes not,
> other times I could get things running by power cycling the old disk
> during boot, I did try juggling around the terminating device to no
> avail.  The old disks were a Seagate 1G and a Fujitsu 2.9G.  IIRC the
> Fuji was the one that gave me the grief.  I run fast, narrow scsi.

I hate to sound like a broken record, but 4 drives is really too many for
a singled-ended chain, especially if any of them are external.  "Sometimes
it works fine." :-) (How long was the cable, all total?) What's also
important, for pushing the limits like this, is to avoid PVC and other
hi-dielectric-constant cables.  You want an e(r) < 3 and preferably around
2, and low-skew cable construction.  Of course, if you went differential
you could use the cheapest cable and random garbage terminators and everything
would be fine.

Fortunately, all the new SCSI transport layers are differential, so these
problems should begin to fade away, though low e(r) cables may be important
in differential for the first time, given the low-V, high-f waveforms.