OBP - POST endless loop

Here's my story. I've stumbled through these tests a long time ago,
managed to
make it work but now have a problem, cockpit error.

I have tried to make a summary of what happened. If you don't know this
one, can
you forward to someone or recommend a good place to post?

Think the problem might be because I failed to do setenv auto-boot?
false ( stop Solaris booting )
Now test keeps rebooting.

I also read where a CTL-D locks you out. No statement of what to do
about it.

......

I managed to lock myself out of my V880 doing a POST test via the OK
prompt.
by setting
ok> setenv diag-switch? true
ok> setenv diag-level max

I don't think I changed setenv auto-boot? false but think it is true
now. I think
everything else was default?

I have a dumb term VT510 emulating a VT100 and a PC-like keyboard hooked
up to TTYA
to view POST output during test. Now nothing comes up on the normal Sun
Console
and keyboard input does nothing. L1-A or Stop-A on that keyboard and
console does nothing.

When powered on, POST runs with output on the VT100 display.

I managed to find the lowest memory DIMM was bad from the POST output I
saw,
I replaced the DIMM and reran and everything seems all right. I had to
power down
to get out of the diagnostics.

Now I have created another problem for myself. I can't return to the OK
prompt or
get out of this test, it just keeps repeating. At the end of the POST
it just starts again.

Typing Break, Ctr-B, or pressing the space bar interrupts the test and
says - "POST is
running". This is on the dumb term, could be the wrong keyboard (PC).
No response from
the console and keyboard.

I have checked connections with main keyboard and video, seems all right.

Is there a jumper on the mother board to get the prom in the default
state?

If you have ever encountered this, can you tell me what might work or
where would be
a good place to post the same question?

Thanks in advance,

klue

On 11/27/11 08:00 PM, Steven Klooster wrote:
> Here's my story. I've stumbled through these tests a long time ago,
> managed to
> make it work but now have a problem, cockpit error.
>
> I have tried to make a summary of what happened. If you don't know this
> one, can
> you forward to someone or recommend a good place to post?
>
> Think the problem might be because I failed to do setenv auto-boot?
> false ( stop Solaris booting )
> Now test keeps rebooting.

Does Solaris actually try and boot ?


> I also read where a CTL-D locks you out. No statement of what to do
> about it.
>
> .....
>
> I managed to lock myself out of my V880 doing a POST test via the OK

Do you have an RSC ?

IIRC, you should be able to abort the test from there.

http://www.sunshack.org/data/sh/2.1....9-45504-1.html


> prompt.
> by setting
> ok> setenv diag-switch? true
> ok> setenv diag-level max
>
> I don't think I changed setenv auto-boot? false but think it is true
> now. I think
> everything else was default?
>
> I have a dumb term VT510 emulating a VT100 and a PC-like keyboard hooked
> up to TTYA
> to view POST output during test. Now nothing comes up on the normal Sun
> Console
> and keyboard input does nothing. L1-A or Stop-A on that keyboard and
> console does nothing.

You are better off using a PC to connect via the serial port.

>
> When powered on, POST runs with output on the VT100 display.
>
> I managed to find the lowest memory DIMM was bad from the POST output I
> saw,
> I replaced the DIMM and reran and everything seems all right. I had to
> power down
> to get out of the diagnostics.
>
> Now I have created another problem for myself. I can't return to the OK
> prompt or
> get out of this test, it just keeps repeating. At the end of the POST
> it just starts again.
>
> Typing Break, Ctr-B, or pressing the space bar interrupts the test and
> says - "POST is
> running". This is on the dumb term, could be the wrong keyboard (PC).
> No response from
> the console and keyboard.
>
> I have checked connections with main keyboard and video, seems all right.
>
> Is there a jumper on the mother board to get the prom in the default
> state?

Take the key out of diag mode ?

>
> If you have ever encountered this, can you tell me what might work or
> where would be
> a good place to post the same question?
>
> Thanks in advance,
>
> klue



--
Bruce Porter
XJR1300SP, XJ900F, Pegaso 650 Trail (x2) one red one grey
POTM#1(KoTL), WUSS#1 , YTC#1(bar), OSOS#2(KoTL) , DS#3 , IbW#18 ,Apostle#8
"The internet is a huge and diverse community but mainly friendly"
http://blog.maui.co.uk/index.php/ytc/
There *is* an alternative! http://www.openoffice.org/

In article <jb3an7$juu$>, YTC#1 <>
wrote:

> On 11/27/11 08:00 PM, Steven Klooster wrote:
> > Here's my story. I've stumbled through these tests a long time ago,
> > managed to
> > make it work but now have a problem, cockpit error.
> >
> > I have tried to make a summary of what happened. If you don't know this
> > one, can
> > you forward to someone or recommend a good place to post?
> >
> > Think the problem might be because I failed to do setenv auto-boot?
> > false ( stop Solaris booting )
> > Now test keeps rebooting.
>
> Does Solaris actually try and boot ?
>
I found a solaris spectum letter from sun that says on this machine
you can push the power button twice, about a second apart to reset
defaults. Missing it turns the machine off. It did not say how fast
that blinking was and I was trying almost immediately. Never worked.
Searching on the net says it may take 20-30 minutes to get to that stage.
It only lasts for a few seconds. Think I got it reset but the bad news
I power cycled it many times, perhaps creating another problem. It
passed all the diagnostics with the lengthy or high report setting. I
was not using the key position for diagnostics.

It now gets to the boot stage after running diagnostics, passes and
now fails which checking I-O with a fatal hardward error, then prints
out register info for each cpu. Then it starts over. I now suspect the
mother board which maybe didn't like all the power cycling on and off.
This machine has been up most of its life.

Any more ideas would be appreciated. Not sure this posted.
Thanks,
klue
>
> > I also read where a CTL-D locks you out. No statement of what to do
> > about it.
> >
> > .....
> >
> > I managed to lock myself out of my V880 doing a POST test via the OK
>
> Do you have an RSC ?
RSC is not configured.
>
> IIRC, you should be able to abort the test from there.
>
> http://www.sunshack.org/data/sh/2.1....shbk/collectio
> ns/INFODOC/1-9-45504-1.html
>
>
> > prompt.
> > by setting
> > ok> setenv diag-switch? true
> > ok> setenv diag-level max
> >
> > I don't think I changed setenv auto-boot? false but think it is true
> > now. I think
> > everything else was default?
> >
> > I have a dumb term VT510 emulating a VT100 and a PC-like keyboard hooked
> > up to TTYA
> > to view POST output during test. Now nothing comes up on the normal Sun
> > Console
> > and keyboard input does nothing. L1-A or Stop-A on that keyboard and
> > console does nothing.
>
> You are better off using a PC to connect via the serial port.
>
> >
> > When powered on, POST runs with output on the VT100 display.
> >
> > I managed to find the lowest memory DIMM was bad from the POST output I
> > saw,
> > I replaced the DIMM and reran and everything seems all right. I had to
> > power down
> > to get out of the diagnostics.
> >
> > Now I have created another problem for myself. I can't return to the OK
> > prompt or
> > get out of this test, it just keeps repeating. At the end of the POST
> > it just starts again.
> >
> > Typing Break, Ctr-B, or pressing the space bar interrupts the test and
> > says - "POST is
> > running". This is on the dumb term, could be the wrong keyboard (PC).
> > No response from
> > the console and keyboard.
> >
> > I have checked connections with main keyboard and video, seems all right.
> >
> > Is there a jumper on the mother board to get the prom in the default
> > state?
>
> Take the key out of diag mode ?
>
> >
> > If you have ever encountered this, can you tell me what might work or
> > where would be
> > a good place to post the same question?
> >
> > Thanks in advance,
> >
> > klue

On 04/12/2011 18:23, Steven Klooster wrote:
> In article<jb3an7$juu$>, YTC#1<>
> wrote:
>
>> On 11/27/11 08:00 PM, Steven Klooster wrote:
>>> Here's my story. I've stumbled through these tests a long time ago,
>>> managed to
>>> make it work but now have a problem, cockpit error.
>>>
>>> I have tried to make a summary of what happened. If you don't know this
>>> one, can
>>> you forward to someone or recommend a good place to post?
>>>
>>> Think the problem might be because I failed to do setenv auto-boot?
>>> false ( stop Solaris booting )
>>> Now test keeps rebooting.
>>
>> Does Solaris actually try and boot ?
>>
> I found a solaris spectum letter from sun that says on this machine
> you can push the power button twice, about a second apart to reset
> defaults. Missing it turns the machine off. It did not say how fast
> that blinking was and I was trying almost immediately. Never worked.
> Searching on the net says it may take 20-30 minutes to get to that stage.
> It only lasts for a few seconds. Think I got it reset but the bad news
> I power cycled it many times, perhaps creating another problem. It
> passed all the diagnostics with the lengthy or high report setting. I
> was not using the key position for diagnostics.
>
> It now gets to the boot stage after running diagnostics, passes and
> now fails which checking I-O with a fatal hardward error, then prints
> out register info for each cpu. Then it starts over. I now suspect the
> mother board which maybe didn't like all the power cycling on and off.
> This machine has been up most of its life.
>
> Any more ideas would be appreciated. Not sure this posted.

It would help if you could be more specific about the IO error.

It is still not clear if Solaris has booted and if that is a disk read
error.

Have you got CD/DVD you can b boot from ? Or a JumpStart server?

You could try seeing if there is some here that could help

https://forums.oracle.com/forums/cat...categoryID=301

> Thanks,
> klue
>>
>>> I also read where a CTL-D locks you out. No statement of what to do
>>> about it.
>>>
>>> .....
>>>
>>> I managed to lock myself out of my V880 doing a POST test via the OK
>>
>> Do you have an RSC ?
> RSC is not configured.
>>
>> IIRC, you should be able to abort the test from there.
>>
>> http://www.sunshack.org/data/sh/2.1....shbk/collectio
>> ns/INFODOC/1-9-45504-1.html
>>
>>
>>> prompt.
>>> by setting
>>> ok> setenv diag-switch? true
>>> ok> setenv diag-level max
>>>
>>> I don't think I changed setenv auto-boot? false but think it is true
>>> now. I think
>>> everything else was default?
>>>
>>> I have a dumb term VT510 emulating a VT100 and a PC-like keyboard hooked
>>> up to TTYA
>>> to view POST output during test. Now nothing comes up on the normal Sun
>>> Console
>>> and keyboard input does nothing. L1-A or Stop-A on that keyboard and
>>> console does nothing.
>>
>> You are better off using a PC to connect via the serial port.
>>
>>>
>>> When powered on, POST runs with output on the VT100 display.
>>>
>>> I managed to find the lowest memory DIMM was bad from the POST output I
>>> saw,
>>> I replaced the DIMM and reran and everything seems all right. I had to
>>> power down
>>> to get out of the diagnostics.
>>>
>>> Now I have created another problem for myself. I can't return to the OK
>>> prompt or
>>> get out of this test, it just keeps repeating. At the end of the POST
>>> it just starts again.
>>>
>>> Typing Break, Ctr-B, or pressing the space bar interrupts the test and
>>> says - "POST is
>>> running". This is on the dumb term, could be the wrong keyboard (PC).
>>> No response from
>>> the console and keyboard.
>>>
>>> I have checked connections with main keyboard and video, seems all right.
>>>
>>> Is there a jumper on the mother board to get the prom in the default
>>> state?
>>
>> Take the key out of diag mode ?
>>
>>>
>>> If you have ever encountered this, can you tell me what might work or
>>> where would be
>>> a good place to post the same question?
>>>
>>> Thanks in advance,
>>>
>>> klue

On 12/05/11 10:31, YTC#1 wrote:

>>>>
>>>> If you have ever encountered this, can you tell me what might work or
>>>> where would be
>>>> a good place to post the same question?
>>>>
>>>> Thanks in advance,
>>>>
>>>> klue
>

It's not clear if you are accessing the machine via the lom port, which is
an rj45 connector on the back panel marked "serial management, or similar.
This should be usable even with power off (but connected) and uses a
separate
hardware / processor from the main cpu(s).

The lom port has two modes: lom management and obp console. Select lom
by typing #. (hash dot) <return> at the terminal. To get back to obp
console, type
"console" <return>. This also assumes a normal keyboard is not plugged in.

At the sc> / lom prompt type "help" to get all the commands, one of
which is:

"bootmode [normal:reset_nvram:diag...] etc

select "reset_nvram", then power disconnect, connect, then hard reset at
the sc>
prompt. Once all the terminal messages subside, reset boot mode back to
normal
or diags as required and another power disconnect / connect. The machine
should
then hopefully be unbricked.

I had a similar problem on a v240 recently. It took a couple of hours
and google
to work out how to fix it, but you should always be able to access the
machine
via lom, even if the main cpu subsystem is temporarily awol...

Regards,

Chris