Server reboot after panic

[Timothy Czarnik]

Hey all,

Could use a bit of help with this one. We had a K380 server reboot this morning, as is shown in the last entry in /etc/shutdownlog:

08:12 Mon Jul 09 2001. Reboot after panic: , isr.ior = 0'162fc00.0'7f7d2000

Last week I replaced a processor in this server. I used XSTM to check for log errors, and found none for the processor or memory. Is there a place that i can look up those error codes? How would the gurus start to tackle this problem?

Thanks in advance!

-Tim

[James R. Ferguson]

Hi Tim:

Knowledge Base document #UXDNKBRC00004522 has some explanation and some guidelines.

...JRF...

[James R. Ferguson]

Hi (again) Tim:

...and Eran actually posted the text of the document I suggested you look at ;-)

...JRF...

[A. Clay Stephenson]

Hi Tim,

There should have been a bit more in the shutdownlog. That would be helpful. ISR - Interrupt Space Register IOR - Interrupt Offset Register - these are virtual addresses.

The best way to diagnose the problem is by using the trace command of the q4 utility to examine the crash dump. This will pinpoint the problem.

[Timothy Czarnik]

Hey all,

OK, I ran the q4 thing (which is VERY cool, btw). I created the ana.txt and what.txt files that the q4 document said to create. Here is what I see:

# grep HPMC ana.txt
Crash Event 0 (HPMC, struct crash_event_table_struct at 0x22000):
crash event was an HPMC
crash event was an HPMC
#

and this shows up in the what.txt file:

+--------------------------------------------+
| Processor activity |
+--------------------------------------------+
Processor 1 started it by HPMC'ing. Here is the stack trace:
stack trace for event 0
crash event was an HPMC
can't get struct frame_marker at 0.0xffffffff'ffff13b0
The Save State registers for this level are:


Is this telling me that we have a bad processor, or that the HPMC was detected by processor 1? Please note that processor 1 was replaced last week due to failure. See:

http://forums.itrc.hp.com/cm/QuestionAnswer/1,1150,0x5beaf9beca68d511abcd0090277a778c,00.html

Am I looking at another faulty CPU? Again, any info is appreciated! :)

-Tim