Hi, let me share a frightening story with you that I lived
through while my family had naptime.
I decided to try out Isildur's idea of booting Ultrix using
a KA62A processor in the hope that that would be supported
right off the 4.5 boot tape. And I just happened to have one
such T2011 board in a box (taken off a lonely 6310 in a
barn full of bird-poop.) So, I took all 6 KA64As out and
stacked them on some papers that I use to pack most
of my boards (with papers between the boards so they would not
stick together :-).
Put the KA62A in and fired up. A lonely light lit and no
real function was to be observed. I moved all 4 MS65A-DA boards
out too, because I figured that may be a KA62A doesn't know
how to deal with MS65A-DA. I swapped in some of those T2014
that seem to multiply while stored in the box. Anyway, it still
wouldn't work.
The self test progress report shows that the CPU detects no
memory and then it reports that problem explicitly and stops.
The CPU error status lights say hexadecimal C0. I could not
find any table explaining what C0 status code is. It could
be boring (like "no memory found", which I know) or it
could indicate some condition that I could fix, like "backplane
wiring error" :-). I stuffed the whole XMI bus full with those
32 MB cards in the hopes that one would work, but to no
avail. So, perhaps the memory access path in that CPU board
is broken? Is there any rewiring to be done when converting
a 6400 down to a 6200?
Anyway, my daughter came downstairs indicating naptime was
over and so it was time to finish up. KA62A and all MS62A back
in the box and all 6 KA64As and 4 MS65A-DA back into the
bus fired up and all processors were indicating they were
somehow not agreeing who would get to be the primary CPU.
Hit reset and they figured it out. But now they were
bitching about memory and the CPUs showed minuses in the
selftest monitor. Rearranged the CPUs and the memory, still
no good luck. Everything was screwed up! I was afraid I
had killed all my boards by careless handling (I did this
before, may be not with that polyester pullover that might
have sent static sparks, like they do.)
Took all but one CPU out and that did do the self test OK.
But one MS65A board seemed broken. It failed regardless
which slot I put it. Finally I took it all out. Then the
CPUs back in and again it didn't work. I then put CPUs
back one by one powering up each time to see progress.
Slowly, one by one, the number of board which I thought
I had killed reduced down to 1, the memory board. The order
in which I put in the CPUs and the restarting every time
seemed to be important. Then, finally I even found one
configuration of the memory boards where all 4 were detected
and tested fine. Hey I was so scared I had killed half
of my VAX through carelessness. But I didn't. I am quite
confused why it is all so finnicky about the order of
installing the hardware. I won't mess with this machine's
XMI configuration any time soon.
regards
-Gunther
--
Gunther Schadow, M.D., Ph.D. gschadow_at_regenstrief.org
Medical Information Scientist Regenstrief Institute for Health Care
Adjunct Assistant Professor Indiana University School of Medicine
tel:1(317)630-7960 http://aurora.regenstrief.org
Received on Sun Mar 10 2002 - 19:56:51 GMT