HP-UX HB v13.00 Ch-08 - Crash Dumps

HP-UX Handbook – Rev 13.00 Page 21 (of 38)
Chapter 08 Crash Dumps
October 29, 2013
Provide the following:
swlist -l product >swlist.out (currently installed software & patches)
/var/adm/syslog/OLDsyslog.log (the syslog from the previous boot)
Additionally in case of a TOC, i.e. system hang answer these questions:
”Did you try a telnet connection to the system? How exactly did it fail?”
”Did you try a rlogin connection to the system? How exactly did it fail?”
”Did you try a console connection to the system? How exactly did it fail?”
”Did the system respond to ping?”
“What was the value shown on the hex display?”
Additionally in case of a ServiceGuard TOC:
/var/adm/syslog/[OLD]syslog.log (appropriate syslogs of all nodes in the cluster)
Additionally in case of a HPMC:
/var/tombstones/ts99 (tombstone file containing chassis logs and PIM data)
system’s serial number (obtained from MP/GSP)
Answering the following questions is very important, too:
“Did the system hang or panic more than once recently?”
“Did anything change recently?” (e.g. kernel patches installed, 3
rd
party software installed,
configuration changes or simply a reboot.
NOTE: A system that panics/hangs multiple times altough no changes have been performed
is likely to suffer from a hardware problem. Whereas hardware failures can happen all of a
sudden, software failures are usually caused by configuration changes.
Please log a hardware case when your system crashed due to HPMC, otherwise log a software
case.
About the stack trace
Before we come to panic() we execute a few other functions that are always the same. Searching
for one of these functions will too turn up lots of hits. How does this typical part of the stack
trace look like?
for UX 10.x and 11.x (PA-RISC):
panic+0x14
report_trap_or_int_and_panic+0x80
trap+0x6dc
thandler+0xd20