🚀 go-pugleaf

I have a pair of VAXstation II/GPX systems that are acting
inconsistently.  I have tried as many things as I can think of and
have not been able to solve the problem.

The problem in a nutshell:  An Emulex diagnostic RX50 boots and runs
just fine on a BA23-packaged VSII/GPX with the RQDX3 at 760334 (the
second MSCP address).  The BA123-packaged system doesn't even recognize
the RQDX3 at 760334 from the boot prompt (">>>") -- it returns error
41, "DEVASSIGN".

System configurations:

	BA23-pkg system			BA123-pkg system
	---------------			----------------
CPU	KA630, M7606-EF			KA630, M7606-EF
Mem	2 x 4MB				8MB + 4MB
Disk	Emu QD32 + Fuji 2322 @ 772150	Emu QD21 + Maxt 4380E @ 772150
	RQDX3 + RX50 @ 760334		RQDX3 + RD52, RD53, RX50 @ 760334
Tape	none				TQK/TK50
Video	8-plane QDSS			8-plane QDSS


On the BA23 system, I can say "b dub0" and it boots the Emulex
diagnostic RX50 fine.  On the BA123, "b dub2" returns the DEVASSIGN
error.  Thinking that there might be a problem with just the floppy,
I tried to boot the RD53 with "b dub0" but I get the same error. 
When I boot Ultrix, I can access all of the drives on the RQDX3
without any problem.

I have tried the following:

- I switched CPUs thinking that the boot PROMs may be different.
  They aren't -- no change.

- I switched RQDX3s.  No change.

- I switched RX50s.  No change.

- Noting that the drive signal distribution is different in the BA123,
  I took a known-good BA23-style signal distribution board and jury-rigged
  it into the BA123.  No change.

- I swapped the addresses on the QD21 and the RQDX3.  Now, when I do a
  "b dua2", the floppy diskette boots!  The Emulex diagnostic monitor
  loads and gives me the EVM> prompt.  I type "L FVDMS" to load the
  MSCP diagnostic and I get "read error" from the floppy.

The next thing I am going to try is substituting all the drive cables,
but I have been fooling with this problem off and on for many months
and I think I may have already done that and forgotten about it.

If anyone can shed some light on this problem I would be most appreciative.

	Thanks,
	---Bob.

-- 
Bob Hoffman, N3CVL       pitt!hoffman
Pitt Computer Science    hoffman@cs.pitt.edu    FAX: +1 412 624 8854