URGENT: System Hang Detected via timer popping

[Guettache]

Hello,

My system hanged arounr 10:32 (CET) and the following console message appreared:
System Alert
System name: gsp
Date: 05/07/2002 Time 08:29:49
Alert Level 13: System Hang Detected via timer popping

Does anyone know what could be the problem?

My system is an HP9000/rp5400 OS: hp-ux 11.0

Thanks for your hel

Narimane

[melvyn burnard]

This appears to be a hardware related issue, possibly firmware. I suggest you log a call with your HP Hardware Support.

[Steve Steel]

Hi

See

http://bizforums.itrc.hp.com/cm/QuestionAnswer/1,,0x85e8ff77de2bd611abd50090277a778c,00.html


Steve Steel

[Johannes_K]

The message is generated from the kernel routine pat_heartbeat() where it calls chassis_log_timeout() to create the information.
So if the OS hangs or is being reset, we could see such events. Would be interesting to search other GSP messages at this time for more information. In case of a system crash the crash files like ts99 or /var/adm/crash will be more helpful.

[Guettache]

Hi,
Because the system did not log anything in the syslog.log, the only way to get information about the hang was to use the toc.
This is the info from the dump and note from HP:
==============================
What I see in this dump it is a HPMC, but it looks like a arpa or btlan
problem on your network card:

Stack Trace for Crash event 0
=============================

============== EVENT ============================
= Event #0 is HPMC on CPU #0
= p crash_event_t 0x9e7000
= p rpb_t 0x9e0ae0
= Using pc from pim.wide.rp_pcoq_head_hi = 0x3bd374
============== EVENT ============================
SR4=0x00000000
SR.SP RP
SR4.0x000000000099b980 0x003bd374 dino3_read_32+0x4c
SR4.0x000000000099b8e0 0x003bcfa0 dino3_rd_uint32+0x18
SR4.0x000000000099b860 0x0055c048 btlan_offline_isr+0x1b8
SR4.0x000000000099b7c0 0x0055c2fc btlan_isr+0x18c
SR4.0x000000000099b720 0x003ba474 dino_isr+0xb4
SR4.0x000000000099b650 0x00259490 up_ext_interrupt+0x250
SR4.0x000000000099b520 0x0015ec9c ihandler+0x8cc
+------------- TRAP ----------------------------
| Trap type 4 in KERNEL mode at 0x2093f8 (boot+0x900)
| p struct save_state 0.0x99b050
+------------- TRAP ----------------------------
I have updated all the network patches, but it did not solve the problem.
We have changed the network card, but still the same problem.
I found out later taht the system hang always happed when an in house application was being used.
I have to tell you that we just have upgraded from a D210 system running 10.20 (32 bit processor) to an L1000 running 11.0 (64 bit processor)
The application has always worked on the old system and ported to the new one without any problem.
The day before the crash happend, I recompiled the application (in house developped) on the new system (using ansi c compiler :
LINT B.11.11.04 CXREF B.11.11.04
HP92453-01 B.11.11.04 HP C Compiler
$ Sep 8 2000 23:13:51 $)

Question: Why the new binary recompiledwith the new system has caused the server hang when using it.
This application is used in user mode.

Thanks for any help
Nariamn

[Guettache]

Hi,
I missed the following in my previous reply:

I monitored with TOP command
the running application
It was the highest process.
The %CPU reached 24 and then the system hanged.
May be it can help to understand the problem

Thanks again

Narimane

[David Meidam]

Did your problem ever get resolved?
I'm building an L-3000, and in the middle of loading the QPK patches I recieved this error.

We removed 217 filesets of the QPK, and we will try search for and hardware enablement patches, perhaps for the lan card, to install. Then give it another go.

Please let me know if you ever got a firm answer as to the cause of the error.

thanks,