I just attempted to update the OBP on an Ultra-2 system (actually a Sparc-20
with an Ultra-2 motherboard upgrade) and now it refuses to boot. The old
version was:

OBP 3.7.8 1998/03/26 19:13 POST 3.3.1 1997/07/09

I downloaded Patch ID# 104169-08 and attempted to install:

OBP 3.25.0 199/12/03 11:35 POST 3.3.8 1998/11/16

Everything seemed to go smoothly (no errors were reported during erasing,
programming, and verification) until the first reset. The last thing I saw
displayed on the screen was:

Programming was successful.
Resetting ...

Next I am staring at a dark screen. The monitor led is is flashing every few
seconds, as if there is no video signal. Nothing seems to work. Typing in
the blind does not seem to have any effect. Power cycling seems to produce
the same conditions.

I dug out a NULL modem cable, set up a terminal session on serial port A
and cycled the power. After a few seconds, I received the following Fatal
Reset message and nothing further:


Fatal Reset
SC Control 0000.0000.0200.0000
SysIO status 0000.0000.0100.0000
P0 Status 0000.0000.04c0.0000
P1 Status 0000.0000.04c0.0000
MC CTL0 0000.0000.04c0.0000
CC Fault 0000.0000.04c0.0000
SysIO UE AFSR 0000.ffe0.0000.0000
SysIO CE AFSR 00ff.ffe0.0000.0000
CPU AFSR 0000.0001.a0f7.0048
SDBH AFSR 0000.0000.0000.0000
SDBL AFSR 0000.0000.0000.0000

On subsequent reboot attempts, the same info is displayed. Only the CPU AFSR
seems to change.

What does all this mean? And how can I fix it?

Thanks for any insights or assistance,
Tom Taylor

Tom dot Taylor at unisys dot com

Tom Taylor wrote:
> I just attempted to update the OBP on an Ultra-2 system (actually a Sparc-20
> with an Ultra-2 motherboard upgrade) and now it refuses to boot. The old
> version was:
>
> OBP 3.7.8 1998/03/26 19:13 POST 3.3.1 1997/07/09
>
> I downloaded Patch ID# 104169-08 and attempted to install:
>
> OBP 3.25.0 199/12/03 11:35 POST 3.3.8 1998/11/16
>
> Everything seemed to go smoothly (no errors were reported during erasing,
> programming, and verification) until the first reset. The last thing I saw
> displayed on the screen was:
>
> Programming was successful.
> Resetting ...
>
> Next I am staring at a dark screen. The monitor led is is flashing every few
> seconds, as if there is no video signal. Nothing seems to work. Typing in
> the blind does not seem to have any effect. Power cycling seems to produce
> the same conditions.
>
> I dug out a NULL modem cable, set up a terminal session on serial port A
> and cycled the power. After a few seconds, I received the following Fatal
> Reset message and nothing further:
>
>
> Fatal Reset
> SC Control 0000.0000.0200.0000
> SysIO status 0000.0000.0100.0000
> P0 Status 0000.0000.04c0.0000
> P1 Status 0000.0000.04c0.0000
> MC CTL0 0000.0000.04c0.0000
> CC Fault 0000.0000.04c0.0000
> SysIO UE AFSR 0000.ffe0.0000.0000
> SysIO CE AFSR 00ff.ffe0.0000.0000
> CPU AFSR 0000.0001.a0f7.0048
> SDBH AFSR 0000.0000.0000.0000
> SDBL AFSR 0000.0000.0000.0000
>
> On subsequent reboot attempts, the same info is displayed. Only the CPU AFSR
> seems to change.
>
> What does all this mean? And how can I fix it?
>
> Thanks for any insights or assistance,
> Tom Taylor
>
> Tom dot Taylor at unisys dot com
>
>
>
I applied that patch successfully only a few weeks ago, so I can confirm
that it works on at least one system (mine).

Did you need to remove a board to access the jumper? Is it possible
that when you reinstalled the board that it did not seat properly?

I note that the readme file for the patch says:

"* PLEASE REVIEW NOT ONLY THIS README BUT ALSO ADDITIONAL DOCUMENTATION
GIVEN.
The file "802-3233-xx.ps" where xx stands for the latest numeric
revision is a PostScript copy of the "Sun Flash PROM Guide for
Workstations
and Workgroup Servers - Standalone Version", which provides the
following;
(i) A board-level diagram that gives the approximate location of the
hardware jumpers (that will need to be changed).
(ii) Emergency instructions in the unlikely event that the Flash
update should fail.
Note, however, that the Flash PROM Guide is written for a user who is
running the Flash PROM update utility from a bootable CDROM, and thus
it will not contain the proper boot commands for installation of this
patch. Also note that the instructions in the document for installing
packages are not valid for this patch."

<end>

Appendix A of that document contains recovery procedures for the case
in which the system does not return to the ok prompt after power is
restored. It involves moving J2204, the "Boot control" jumper, but
the full procedure is too long to reproduce here. It would be best
(if you can) to print that document off; if you can't, I can try to
help you work through it off the group.

HTH.

BTW, my e-mail address in the "reply" line is reversed; it actually
starts "ch..."

Thanks CJT,

I can't determine which jumper should be set where. These are some upgrade
boards, apparently manufactured by Sun. I don't have any pertinent
information like part numbers or model numbers. They are not the same as
Sparc 20s or Ultra-2's, even though prtdiag claimed its's an Ultra-2. At
least the jumpers are named differently, located somewhere else, or buried
under something where I can't see it.

I did not even open the box when I attempted the update, just tried it. I
have opened the box now, but don't see any jumper. I have browsed the
motherboard docs at sun and don't see any that resemble this one.

I will try on Monday to see if anyone has any of the old purchase records
for these motherboards.

Thanks,
Tom

"CJT" <abujlehc@prodigy.net> wrote in message
news:401B2DF9.2080901@prodigy.net...

> Tom Taylor wrote:
> > Snip ....


> I applied that patch successfully only a few weeks ago, so I can confirm
> that it works on at least one system (mine).
>
> Did you need to remove a board to access the jumper? Is it possible
> that when you reinstalled the board that it did not seat properly?
>
> I note that the readme file for the patch says:
>
> "* PLEASE REVIEW NOT ONLY THIS README BUT ALSO ADDITIONAL DOCUMENTATION
> GIVEN.
> The file "802-3233-xx.ps" where xx stands for the latest numeric
> revision is a PostScript copy of the "Sun Flash PROM Guide for
> Workstations
> and Workgroup Servers - Standalone Version", which provides the
> following;
> (i) A board-level diagram that gives the approximate location of the
> hardware jumpers (that will need to be changed).
> (ii) Emergency instructions in the unlikely event that the Flash
> update should fail.
> Note, however, that the Flash PROM Guide is written for a user who is
> running the Flash PROM update utility from a bootable CDROM, and thus
> it will not contain the proper boot commands for installation of this
> patch. Also note that the instructions in the document for installing
> packages are not valid for this patch."
>
> <end>
>
> Appendix A of that document contains recovery procedures for the case
> in which the system does not return to the ok prompt after power is
> restored. It involves moving J2204, the "Boot control" jumper, but
> the full procedure is too long to reproduce here. It would be best
> (if you can) to print that document off; if you can't, I can try to
> help you work through it off the group.
>
> HTH.
>
> BTW, my e-mail address in the "reply" line is reversed; it actually
> starts "ch..."
>

Tom Taylor wrote:
> Thanks CJT,
>
> I can't determine which jumper should be set where. These are some upgrade
> boards, apparently manufactured by Sun. I don't have any pertinent
> information like part numbers or model numbers. They are not the same as
> Sparc 20s or Ultra-2's, even though prtdiag claimed its's an Ultra-2.

Ouch. If they're not true Ultra-2 boards, it's unclear whether that
patch is appropriate. The README file says it's for Ultra 2 only.

A true Ultra-2 would have required that you shift jumper J2003 in order
to update the firmware. Otherwise it's write-protected. Because of the
position of that jumper, it's often necessary to remove peripheral
boards first to gain access; I assumed you had done so, which accounts
for the question I asked.


At
> least the jumpers are named differently, located somewhere else, or buried
> under something where I can't see it.
>
> I did not even open the box when I attempted the update, just tried it. I
> have opened the box now, but don't see any jumper. I have browsed the
> motherboard docs at sun and don't see any that resemble this one.
>
> I will try on Monday to see if anyone has any of the old purchase records
> for these motherboards.
>
> Thanks,
> Tom
>
> "CJT" <abujlehc@prodigy.net> wrote in message
> news:401B2DF9.2080901@prodigy.net...
>
>
>>Tom Taylor wrote:
>>
>>> Snip ....
>
>
>
>>I applied that patch successfully only a few weeks ago, so I can confirm
>>that it works on at least one system (mine).
>>
>>Did you need to remove a board to access the jumper? Is it possible
>>that when you reinstalled the board that it did not seat properly?
>>
>>I note that the readme file for the patch says:
>>
>>"* PLEASE REVIEW NOT ONLY THIS README BUT ALSO ADDITIONAL DOCUMENTATION
>> GIVEN.
>> The file "802-3233-xx.ps" where xx stands for the latest numeric
>> revision is a PostScript copy of the "Sun Flash PROM Guide for
>>Workstations
>> and Workgroup Servers - Standalone Version", which provides the
>>following;
>> (i) A board-level diagram that gives the approximate location of the
>> hardware jumpers (that will need to be changed).
>> (ii) Emergency instructions in the unlikely event that the Flash
>> update should fail.
>> Note, however, that the Flash PROM Guide is written for a user who is
>> running the Flash PROM update utility from a bootable CDROM, and thus
>> it will not contain the proper boot commands for installation of this
>> patch. Also note that the instructions in the document for installing
>> packages are not valid for this patch."
>>
>><end>
>>
>>Appendix A of that document contains recovery procedures for the case
>>in which the system does not return to the ok prompt after power is
>>restored. It involves moving J2204, the "Boot control" jumper, but
>>the full procedure is too long to reproduce here. It would be best
>>(if you can) to print that document off; if you can't, I can try to
>>help you work through it off the group.
>>
>>HTH.
>>
>>BTW, my e-mail address in the "reply" line is reversed; it actually
>>starts "ch..."
>>
>
>
>

In <bvf7ll$9cq$1@si05.rsvl.unisys.com> "Tom Taylor" <Tom.Taylor@nospam.unisys.com> writes:
>CPU AFSR 0000.0001.a0f7.0048
>
>On subsequent reboot attempts, the same info is displayed. Only the CPU AFSR
>seems to change.
>
>What does all this mean? And how can I fix it?

it means the processor is very upset with its level 2 cache and is
encountering various tag and data parity errors. is there any chance
you might have inadvertently caused a hardware problem? did you have
to open the machine and change the write-protect jumper before updating
the prom?

In comp.unix.solaris Tom Taylor <Tom.Taylor@nospam.unisys.com> wrote:
> I just attempted to update the OBP on an Ultra-2 system (actually a Sparc-20
> with an Ultra-2 motherboard upgrade) and now it refuses to boot. The old

I've never heard of such an "upgrade". What speed do the CPUs run at?

> I dug out a NULL modem cable, set up a terminal session on serial port A
> and cycled the power. After a few seconds, I received the following Fatal
> Reset message and nothing further:
>
>
> Fatal Reset
> SC Control 0000.0000.0200.0000
> SysIO status 0000.0000.0100.0000
> P0 Status 0000.0000.04c0.0000
> P1 Status 0000.0000.04c0.0000
> MC CTL0 0000.0000.04c0.0000
> CC Fault 0000.0000.04c0.0000
> SysIO UE AFSR 0000.ffe0.0000.0000
> SysIO CE AFSR 00ff.ffe0.0000.0000
> CPU AFSR 0000.0001.a0f7.0048
> SDBH AFSR 0000.0000.0000.0000
> SDBL AFSR 0000.0000.0000.0000

That's one very strange Fatal Reset report. It's actually reporting a
number of errors, which generally wouldn't make sense occuring all at once.
In isolation these would probably be hardware problems, but with so many
of them together this is fairly unlikely.

I'd say it's more likely that the new firmware is trying to get the hardware
to do something it doesn't support - which brings us back again to the
question of exactly what the hardware is...

Scott.

"Tom Taylor" <Tom.Taylor@nospam.unisys.com> writes:

>I can't determine which jumper should be set where. These are some upgrade
>boards, apparently manufactured by Sun. I don't have any pertinent
>information like part numbers or model numbers. They are not the same as
>Sparc 20s or Ultra-2's, even though prtdiag claimed its's an Ultra-2. At
>least the jumpers are named differently, located somewhere else, or buried
>under something where I can't see it.


If it's an Ultra-2 then the board would not fit; what type of board is
it? Is it perhaps very much like an Ultra-2 but not sufficiently so?


If it's not exactly like an Ultra-2 then chances are that the PROM
will not work,.

Casper

HI,
Casper H.S. Dik wrote:
> "Tom Taylor" <Tom.Taylor@nospam.unisys.com> writes:
>
>
>>I can't determine which jumper should be set where. These are some upgrade
>>boards, apparently manufactured by Sun. I don't have any pertinent
>>information like part numbers or model numbers. They are not the same as
>>Sparc 20s or Ultra-2's, even though prtdiag claimed its's an Ultra-2. At
>>least the jumpers are named differently, located somewhere else, or buried
>>under something where I can't see it.
>
>
>
> If it's an Ultra-2 then the board would not fit; what type of board is
> it? Is it perhaps very much like an Ultra-2 but not sufficiently so?
I trying to remember the name of the board, it is not from Sun, could be
from Axil if I am not wrong.

I remember the one that we did buy Sun from in late 90's did try to sell
us this board, I could as said fit in a SS20.

At the same time there where these SS5 upgrades with a CPU and cache on
one board that did fit in a SS5 when the CPU was on a socket(not all had
socket for CPU)

..
Check this link for the SS5 upgrade from that time.

http://web.archive.org/web/199703300...les/manad.html

/michael

Michael Laajanen <michael.laajanen.no-spam.@telia.com> writes:

>I trying to remember the name of the board, it is not from Sun, could be
>from Axil if I am not wrong.

Axil sounds plausible; none of these boards are "flash" compatibile
with Sun, I think. They're single CPU Ultra-2 like systems with the
CPU mounted (the Ultra 2 case is higher because the CPUs stand on an
edge)

>I remember the one that we did buy Sun from in late 90's did try to sell
>us this board, I could as said fit in a SS20.

The only boards we sold directly were ATX formfactor boards.

>At the same time there where these SS5 upgrades with a CPU and cache on
>one board that did fit in a SS5 when the CPU was on a socket(not all had
>socket for CPU)

That's the pre-Ultra 170MHz TurboSPARC>

Casper
--
Expressed in this posting are my opinions. They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.

Casper H.S. Dik wrote:

> Michael Laajanen <michael.laajanen.no-spam.@telia.com> writes:
>
>>I trying to remember the name of the board, it is not from Sun, could be
>>from Axil if I am not wrong.
>
> Axil sounds plausible; none of these boards are "flash" compatibile
> with Sun, I think. They're single CPU Ultra-2 like systems with the
> CPU mounted (the Ultra 2 case is higher because the CPUs stand on an
> edge)
>
>>I remember the one that we did buy Sun from in late 90's did try to sell
>>us this board, I could as said fit in a SS20.
>
> The only boards we sold directly were ATX formfactor boards.

Sun Microelectronics definitely sold OEM Ultra 2 boards in a non-ATX
formfactor. I bought five machines based on these in 1997 from Sunrise
Systems in the UK. This may be one of those.

Dunno about Axil, but this may be a Cycle board as they had several which
accepted the SunFire CPU modules and were good for upgrading just about
anything to anything. I had one that took an SS4 to the equivalent of an
Ultra2 1/200 - it was using the SunFire module rather than the soldered CPU
of the Ultra 1.

P.