My apologies for the late summary. Perhaps this will, at the least,
be of some use in the archives.
It turns out that our particular CPU had to have an EPROM upgrade
before 4Meg SIMMs would work. This is not true of all 4300 CPU's.
I've heard from people that just popped the SIMMs in, set the jumper,
and they worked.
You have to contact your local sales and "service" office to get the
CPU EPROM upgrade. I was told it is not available (and sure enough it
was not on our spares price list) as an FRU (Field Replaceable Unit).
I was laboring under the misconception that if I could just get the
part number, I could order it. That's not how it works -- you have to
get someone in your local Sun office to order it for you. The time and
materials problem (see my original query) turned out to be misinformation
that was resolved by talking with a field service manager.
Another thing you could do is swap your CPU through Sun Express in
order to get a different CPU card. If you can get the EPROM upgrade
kit I don't know of any reason to do this. The CPU swap costs more $'s
and increases your downtime. But I also don't have any way of knowing
if there are CPU's in the field that are incompatible with the EPROM
upgrade. It might be a good idea to make your purchase of the EPROM
upgrade kit contingent upon it solving the problem.
Many folks that replied suggested that I simply shunt the 4Meg jumper
on the CPU card. I meant to imply that we did have a shunt across the
4Meg jumper when I said we had tried the 4Meg SIMMs in the CPU card.
Sorry if that wasn't clear from the start.
Thanks to:
Harry Edmon <harry@atmos.washington.edu>
aldrich@sunrise.stanford.edu (Jeff Aldrich)
ats@prosun.first.gmd.de (Andreas Schulz)
arne@gest20.sinet.slb.com
Alan J. Rothstein <merccap!alan@uunet.uu.net>
Roy Richter PH/32 <rrichter@ph.gmr.com>
ept@eptsun1.ctd.ornl.gov (E P Tinnel)
kraussW@moravian.edu
danielle@systems.caltech.edu (Danielle Sanine)
Jim Lick <jim@ferkel.ucsb.edu>
poffen@sj.ate.slb.com (Russ Poffenberger)
JanBerger.Henriksen@ii.uib.no
don@mars.dgrc.doc.ca (Donald McLachlan)
Christian Lawrence <cal@soac.bellcore.com>
"Richard Wong" <rnw@math.princeton.edu>
John Valdes <valdes@geosun.uchicago.edu>
Special thanks to John Valdes <valdes@geosun.uchicago.edu> who wrote:
> We had 3.0.1 proms in our 4/330, and after replacing the 1MB SIMMs on an
> external memory board w/ 4MB ones, we got this same memory error upon
> booting. A prom upgrade to 3.0.3 fixed the problem. Note that there
> is also a jumper on the CPU board labeled J1900 which needs to be closed
> when using 4MB SIMMs on the CPU board (the jumper is located just to the
> left of the SIMM banks, near socket U1304, when looking down at the board
> with the connectors away from you.
>
> What you want to order from Sun is a "Boot Prom Kit", part number 595-1961.
> This will come with four IC's, whose individual part numbers are:
> 525-1035, 525-1036, 525-1037 and 525-1038.
>
> Good luck!
>
> John Valdes
My original query follows:
> We've got a 4300 CPU in a 4/330 with 1MB SIMMs and CPU EPROM version
> 3.0.1. We would like to replace all 8 1MB SIMMs with 8 4MB SIMMs.
>
> The Sun "Field Engineer Handbook" indicates that 4MB SIMMs are not
> supported on the Sun-4/330 CPU. It later goes on to say that CPU
> EPROM 3.0.3 or greater is required with 4MB SIMMs. A 3rd-party memory
> vendor has told me that the 4MB SIMMs will work in the 4300 with 3.0.3
> or greater - it has been done at customer sites.
>
> I've tried the 4MB SIMMs in the CPU card. They don't work. The diag
> led's light up with: 11010001, main memory timeout.
>
> It would appear that what I need is a set of the latest CPU EPROMs for
> a 4300 CPU card. The problem is that the 3rd party memory supplier
> doesn't have the EPROM part number and Sun won't give it to me without
> charging time and materials. It's also not on the spares price list.
>
> My questions are:
>
> 1) Is it possible that I need more than just the EPROMS, e.g. a CPU
> card upgrade?
>
> 2) What is the part number for the latest 4300 CPU EPROMs?
>
> Thanks in advance,
>
> -- Ed
Thanks again,
-- Ed
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:06:53 CDT