3 May
2015
3 May
'15
12:50 p.m.
The PROM battery appears to have died in my Sun U1 (Ethernet address and
host ID all ff's etc.).
I can hack a new battery into it at some point - but can anyone confirm
whether the openboot environment should be responsive even while the
battery is hosed? Via a serial console, I can send a break and get an OK
prompt back, but then it appears to go unresponsive and won't accept
commands (no key presses aren't echoed back to the screen).
Using the same terminal emulator / cabling is fine with my other Sun (a
U2), so I'm reasonably confident that the environment is OK, and that the
problem is local to the U1 machine.
cheers
Jules
3 May
3 May
4:48 p.m.
I have seen almost anything once on Sun machines :O Are you testing with
all cables disconnected just power and serial terminal? I have seriously
seen a bad network cable cause mysterious hangs on a SS10 once...
You've tried some of the other usual troubleshooting stuff like swapping
out/around some of the memory, removing Sbus cards, etc?
Under normal circumstances, it /should/ drop to OF with a dead IDPROM per
my experience ... I always think of my Macs and my NeXTs as being the most
troublesome for not powering on when the battery is flat ... Suns are
almost always ... 99.99% of the time ... good about it.
Best,
Sean
On Sun, May 3, 2015 at 12:50 PM, Jules Richardson <
jules.richardson99 at gmail.com> wrote:
The PROM battery appears to have died in my Sun U1 (Ethernet address and
host ID all ff's etc.).
I can hack a new battery into it at some point - but can anyone confirm
whether the openboot environment should be responsive even while the
battery is hosed? Via a serial console, I can send a break and get an OK
prompt back, but then it appears to go unresponsive and won't accept
commands (no key presses aren't echoed back to the screen).
Using the same terminal emulator / cabling is fine with my other Sun (a
U2), so I'm reasonably confident that the environment is OK, and that the
problem is local to the U1 machine.
cheers
Jules
6:50 p.m.
On 05/03/2015 03:48 PM, Sean Caron wrote:
I have seen almost anything once on Sun machines :O
Are you testing with
all cables disconnected just power and serial terminal? I have seriously
seen a bad network cable cause mysterious hangs on a SS10 once...
Yeah, initially I just had serial and power hooked up, but because it seems
to be defaulting to a network boot I did try connecting it up to my LAN
too, but with no luck.
You've tried some of the other usual
troubleshooting stuff like swapping
out/around some of the memory, removing Sbus cards, etc?
Not yet, but will do so.
I did just attack the PROM module with a dremel and so am now running with
good battery voltage via an external battery (but at this stage it still
throws up an "The IDPROM contents are invalid" message, of course)
I'm suddenly wondering if the serial port is fried. It's possible that I've
only ever run this machine with keyboard/rodent/monitor attached, not via
serial console, and a hardware fault on the rx side would explain the
symptoms. I'll drag the necessary bits out of the basement and see if that
changes anything (this room is rapidly filling up with vintage crap ;-)
cheers
Jules
6:51 p.m.
Have you tried sending a break?
On May 3, 2015, at 5:50 PM, Jules Richardson
<jules.richardson99 at gmail.com> wrote:
On 05/03/2015 03:48 PM, Sean Caron wrote:
I have seen almost anything once on Sun machines
:O Are you testing with
all cables disconnected just power and serial terminal? I have seriously
seen a bad network cable cause mysterious hangs on a SS10 once...
Yeah, initially I just had serial and power hooked up, but because it seems to be
defaulting to a network boot I did try connecting it up to my LAN too, but with no luck.
You've tried some of the other usual
troubleshooting stuff like swapping
out/around some of the memory, removing Sbus cards, etc?
Not yet, but will do so.
I did just attack the PROM module with a dremel and so am now running with good battery
voltage via an external battery (but at this stage it still throws up an "The IDPROM
contents are invalid" message, of course)
I'm suddenly wondering if the serial port is fried. It's possible that I've
only ever run this machine with keyboard/rodent/monitor attached, not via serial console,
and a hardware fault on the rx side would explain the symptoms. I'll drag the
necessary bits out of the basement and see if that changes anything (this room is rapidly
filling up with vintage crap ;-)
cheers
Jules
7:23 p.m.
On 05/03/2015 05:51 PM, Ryan K. Brooks wrote:
Have you tried sending a break?
Yes, I had....
Turns out that it *is* the receive side of the console serial port that's
toast - using a real keyboard/mouse/screen is fine, so I've been able to
reprogram the NVRAM and the machine is booting happily again.
I'm not sure what, if anything, I can do about the serial console - is
there a way of defaulting to console on port B rather than on port A? It's
not a massive problem or anything, but I can see that there might be times
when I want to just quickly use a headless Solaris box (such as today!)
without needing a full GUI.
cheers
Jules
9:16 p.m.
On 05/03/2015 04:23 PM, Jules Richardson wrote:
I'm not sure what, if anything, I can do about the serial console - is
there a way of defaulting to console on port B rather than on port A?
Not quite what you might be looking for, but: Run "printenv" at the "ok
" prompt, you should see two variables to set the console input and
output. "ttyb" should do the trick IIRC.
Remember, that'll get lost if the battery dies or gets disconnected...
--S.
7 May
7 May
12:38 a.m.
I haven't had a chance to catch up on email for a few days, so I am hoping this
is old stuff and you have your box up and running.
I have two comments for you.
1. in addition to Ryan's break comment, you may also want to try a STOP-N ,
which should reset the OBP back to defaults.
2. Back in the early 2000's, I was at a small company, and one day my boss
brought me a stack of Ultra-1's and Ultra-2, which we used small servers,
serving up DNS, FTP, etc.
Several had dead OBP batteries, but I was still able to boot them, although I
still received the error. I was too lazy to do anything (physical) about that,
but what I did, was to run a script, early in the boot process, to assign a mac
address to the ethernet card, before any daemons came up, and those boxes served
us well for many years. The script was a really simple one liner of something
like
ifconfig hme0 ethers 08:00:20:c0:ff:ee
or something similar.
Again, hope you have this all figured out and running.
Jerry
On 05/ 3/15 05:51 PM, Ryan K. Brooks wrote:
Have you tried sending a break?
On May 3, 2015, at 5:50 PM, Jules Richardson
<jules.richardson99 at gmail.com> wrote:
On 05/03/2015 03:48 PM, Sean Caron wrote:
I have seen almost anything once on Sun machines
:O Are you testing with
all cables disconnected just power and serial terminal? I have seriously
seen a bad network cable cause mysterious hangs on a SS10 once...
Yeah, initially I just had serial and power hooked up, but because it seems to be
defaulting to a network boot I did try connecting it up to my LAN too, but with no luck.
You've tried some of the other usual
troubleshooting stuff like swapping
out/around some of the memory, removing Sbus cards, etc?
Not yet, but will do so.
I did just attack the PROM module with a dremel and so am now running with good battery
voltage via an external battery (but at this stage it still throws up an "The IDPROM
contents are invalid" message, of course)
I'm suddenly wondering if the serial port is fried. It's possible that I've
only ever run this machine with keyboard/rodent/monitor attached, not via serial console,
and a hardware fault on the rx side would explain the symptoms. I'll drag the
necessary bits out of the basement and see if that changes anything (this room is rapidly
filling up with vintage crap ;-)
cheers
Jules
3397
days inactive
3401
days old
6 comments
5 participants
participants (5)
-
classiccmp＠crash.com -
jules.richardson99＠gmail.com -
other＠oryx.us -
ryan＠hack.net -
scaron＠umich.edu