System dump

[Hoang Minh Tuan]

my system run OPS Oracle database and use MC/ServiceGuard. I have logical volume /dev/vg01Ra/lvarch created in AutoRaid.
After 3 minutes interval database create archieve file in this file system.
My system be dump and the message show
"/dev/vg01/lvarch file system full
Unexpected HPMC/TOC. Processor HPA FFFFFFFF'FFFA0000"
When system startup I check by bdf command and it show used 54%. I don't know why this is happened. Anyone can clearly explain this problem.
Thank a lots.

[Marco Santerre]

If you had a crash and you got a dump in /var/adm/crash, I would certainly get Hewlett-Packard Response Center involved in this so that they can help me figure out where the problem lies. Though at first view, I doubt your file system is the culprit. You had a Transfer-of-Control that happened, and right after it talks about Processor. I'm no expert when it comes to crashes, but this is certainly where I would start.

[Michael Steele_2]

HPMC or High Priority Machine Check is a message not sysadmin ever wants to see for it indicates a panic, usually hardware related.

These logs are needed to determine what has happened, you can post them here or start a software call with the HP response center and email in the logs after you get a case number.

/etc/shutdownlog (* note the last entry and panic string *)
/var/tombstones/ts99 (* note the time of HPMC *)

I've attached a 'q4' document specifically written just for HPMC's that provide more details.

Check /var/adm/crash/core.# -or- /var/adm/crash/crash.# for an automatically system saved dump. If you don't have crash dump then this command might capture one still.

savecrash -rf /tmp/crash_file