SUMMARY: Fujitsu M2263SA & SCSI pin #26

From: Brett Ballantyne (brett@spiff.den.mmc.com)
Date: Wed Apr 24 1991 - 12:24:23 CDT


A while back I posted the following question:

>I'm posting this for a friend who doesn't have access to this list right now.
>He just bought a Fujitsu M2263SA (6?? Mb) drive to add to a 327 Mb drive
>on on of his Sun 3/60's. When he tries to format it, format won't even
>recognize it. He's tried it before and after the other scsi devices (a disk
>drive and a tape drive), and can't get anything to work. Any ideas?

>Also, I've heard some rumors about problems with SCSI pin #26 on 3/60's.
>Supposedly pin 26 is a power pin to power external terminators, but on the
>3/60 is connected to ground (or something like that), so if you use an
>external terminator all sorts of bad things can happen (like blowing
>your drive, it's power supply, etc.). Can anyone shed any light on this?

The problem with his hard drive turned out to be a hardware problem (new
drive, problem solved!). There seemed to be some interest in the pin
26 question, so I thought I should summarize. The 2 best answers are
included at the end of this message.

Thanks to all who responded, especially:
george nassiopoulos
Kevin Sheehan
Bill Selig
Jim Guyton
Dan Butzer
Eric A Pearce
Christopher Provenzano
Nick: Hugibaz

Thanks again!

Brett Ballantyne
Martin Marietta
brett@spiff.den.mmc.com

----- Begin Included Message -----

>From proven@eng.umd.edu Thu Mar 21 17:20:46 1991

I can't answer the first question, but the second one I know a little about,
having lost many drives to this very problem. Yes, Sun grounded pin 26
on their 3/50's and 3/60's but only on early revisions of the motherboard.
Anything bought in the last two years is probably safe, but if your concerned
ask sun for the specific revision numbers for the bad motherboards. You'll
have to open the machine up to check your machines. I don't remember the
actual revision number since it's been several years, but I can tell you
yet another system manager horror story (YASMHS) about the whole incedent.

  Christopher Provenzano | The views expressed here a strictly my own
     Software Engineer | and in no way do they represent the view
   N/Hance Systems, Inc. | of my employer or master.
uucp: uunet!eng.umd.edu!proven | Writing kernels is as easy an writing any
email: proven@eng.umd.edu | other piece of software. Debugging is what's
voice: (617) 461-1970 | different.

----- End Included Message -----

----- Begin Included Message -----

>From sundev!fletch!kevin@Sun.COM Sat Mar 16 13:21:29 1991

Actually, 3/50's were the real culprit, this was fixed very early in
the 3/60 line. To my knowledge, only a few of these got out, and I
thought we got them all.

The problem is indeed that pin 26 TERMPWR is grounded. WHen we first
made SCSI interfaces, 26 was undefined, so we grounded it. When it
became TERMPWR, it wasn't till I tryed hooking up a Quantum that used
it that we found this problem.

Like I said, 3/50's, and *very ancient* 3/60's are the only offender.

                l & h,
                kev

Kevin Sheehan
synergy!kevin@Sun.COM
Happily self employed.

----- End Included Message -----



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:06:13 CDT