K460 DEAD! - INIT C30C Hang

[Tim Krego_1]

Our K460 died today. I initally had a FLT CBF0 error on the LCD. When I search for docs I found out about HPMC (Something new every day!). I put the machine into service mode and I cannot get past the test phase. Our other K460 I can get into service menu, but there isn't anything wrong with that machine.

When I press 'SL' I can see the last errors, the one it stops on is INIT C30C. I cannot find out what that is.

My problem is that we do not have an HP service contract yet, all I have is this forum and an extra K460 to pull parts from if needed. Oh, I also have a make_recovery tape that I did last week. :)

Any help is appreciated.
Thanks

[Bill McNAMARA_1]

When you get to BCH prompt (10 secs to press a key) type
SER PIM HPMC
and post the output up here, in particular the responder field..

The CBF0 I believe is just indicating that a pmc occured.. you need the list from sl of what happend up to then..

Later,
Bill

[paul courry]

Tim, turn that key to the right all the way (service mode) and type in SL (from the ctrl B prompt of course) every ten seconds. This should catch the FLT codes. Now if you post the FLT codes we will decipher them for you and tell you what to do. FYI, there is a document on ITRC that has all the flt codes in them if you wish to dig around.

BTW there are HPMC's and LPMC's. High Priority Machine Checks cause a system halt and can not be recovered from (bad juju). Low Priority Machine Checks can be recovered from and do not cause a system halt, but do cause error messages to be put on the screen and the system log files (not so bad juju).

[James R. Ferguson · Accepted solution]

Hi Tim:

'C30C' is a "Monarch slave check, making sure slaves responded ". The "Monarch" is the lead processor during bootup. It sounds like he is waiting for other processors to join ("RENDEZ") but hasn't heard from them.

You might want to place a hardware service call.

BTW: K-class fault codes can be found in Knowledge Base document #DFQKBCA00000115.

...JRF...

[Tim Krego_1]

We did get the machine back up. All I did was reseat all the boards and it fixed the problem. I am concerned as to why this may have happened, but now that the machine is up so I am breathing a little better.

Not sure what the INIT C30C error was caused by.

[paul courry]

"I am concerned as to why this may have happened"

It happens because the CPU (and memory) boards are extremely suseptible to small amounts of corrosion on the contacts. This is why reseating the boards (sometimes several times) may solve the problem. Did you happen to move or replace a board?

[paul courry]

I forgot to add that leaving the cover off of the CPU's is also a major reason for failure. Those rubber baby buggy bumpers on the back of the plate that goes over the CPU's serves the purpose of holding those CPU's in TIGHT to the backplane. Without it you can get all sorts of spurious results.