HPE 9000 and HPE e3000 Servers
Showing results for 
Search instead for 
Did you mean: 

GSP event error - need help diagnosing problem

Randy Martin_2
Occasional Contributor

GSP event error - need help diagnosing problem

I have a couple GSP events, but I can't find anywhere that explains what it means. Online docs refer to chassis_code.codes file, but I can't find it.

HP9000 / L1000 hp-ux 11.0

Multiple times getting Source = 0 errors,
but nowhere can I find what Source = 0 represents. Docs list sources 1-8 only.

The other error is:
Source 1=processor
Source Detail 1=processor general
Source ID 0
Problem Detail 0=no problem detail
Caller Activity E=system warning
Status 0
Caller Subactivity F0=implementation dependent
Reporting Entity type E=hp-ux
Reporting Edntity id 03
0xF8E030301100EF00 00000000 0000EF00 = type 31
0x58E038301100EF00 0000670A 1C151317 = type 11

The best online doc I've been able to find is
http://docs.hp.com/hpux/pdf/A5191-96018.pdf but doesn't give me what I need.

Can anyone tell me what the events mean, or even better where I can find the information myself. (we don't have HP support on this box).

Thanks for any help!!!
Stefan Stechemesser
Honored Contributor

Re: GSP event error - need help diagnosing problem

Hi Randy,

Source=0 means that no hardware part in the server was the source ("no source"). Errors like this are often firmware or OS ("Reporting Entity Type" is "hp-ux" or "system firmware")triggered errors. This matches the other error you posted. It is a simple chassis code "EF00" (last four digits of the first hex word). EF00 is printed from HP-UX after a successful Transfer of Control (TOC). I bet that there is a E000 Chassis Code (Begin of TOC) before this message.
Possible causes:
- System Panic (look for Alert Level 12 logs, dumps in /var/adm/crash)
- Machine Check Abort (MCA / HPMC)
- manually initiated TOC
- TOC due to missing Service Guard heartbeat in a cluster

I hope this helps

Randy Martin_2
Occasional Contributor

Re: GSP event error - need help diagnosing problem

Thanks Stefan.

Found dump files which told me the problem
(Spinlock Deadlock). Found patch that references problem. This is the first crash in 3 years so I'll hold off installing patch until it happens again. The box isn't going to be running too much longer and I'd prefer to not touch OS if I don't have to.

Thanks and consider the problem closed.