ProLiant Servers (ML,DL,SL)
cancel
Showing results for 
Search instead for 
Did you mean: 

CPU Hog with cpqacmgr.nlm and ASR does'nt restart server

Bron_3
Occasional Contributor

CPU Hog with cpqacmgr.nlm and ASR does'nt restart server

Hi,

My Proliant ML350 G3 with Netware 6SP5 had encountered a CPU Hog with cpqacmgr.nlm, and the server blocks (hung).

The ASR, which is enabled, doesn't restart the server.

What could be the problem with this NLM and with ASR ?

Many thanks, Alain
5 REPLIES
dmowry
Occasional Visitor

Re: CPU Hog with cpqacmgr.nlm and ASR does'nt restart server

Alain--
First, can you please check that you are running version 1.06 (most recent version) of CPQACPI.PSM, CPQACA.NLM, CPQACMGR.NLM, CPQACPWR.NLM, and CPQASL.NLM modules. If not, can you please upgrade to those versions. Please note that the ProLiant ACPI drivers are a "matched set" so those drivers must all be the same version number; if you download/run the Component from hp.com ProLiant Driver webpage then you should get all of the correct versions.

If you are already running version 1.06 of those modules, then we would need to see an ABEND.LOG file to have a better idea of what the issue is.

Also: what is the value of the set parameter AUTO RESTART AFTER ABEND ? What is the value of the set parameter CPU HOG TIMEOUT AMOUNT ? When the CPU Hog Timeout ABEND occurs, is NetWare just booting up, has the server been up and running a while, or is it shutting down ? Is this a stand-alone server or is it part of a Cluster ? When the "server blocks(hung)", can you still get into the Kernel Debugger or select the option on the ABEND screen to create an ABEND.LOG and exit/reboot ?

As far as ASR goes, in order for ASR to work, ASR must be enabled in the ROM Based Setup Utility *AND* the ProLiant Health driver must be loaded. In order to capture an ABEND.LOG, you may need to disable ASR in RBSU or unload the Health driver. What versions of CPQHLTH.NLM and CPQASM.NLM are loaded ? (If you can get the ABEND.LOG, then this version information should be part of the driver information in that log file which may save you a little time.) ASR not restarting the server is probably a separate issue from CPQACMGR.NLM.

Thanks.
--Daniel
Bron_3
Occasional Contributor

Re: CPU Hog with cpqacmgr.nlm and ASR does'nt restart server

We run alreday in 1.06

The abend.log say:
<<
Server HYPATIA halted Sunday, 5 December 2004 18:14:24,706
Abend 3 on P00: Server-5.60.05-1937: CPU Hog Detected by Timer

Registers:
CS = 0008 DS = 0010 ES = 0010 FS = 0010 GS = 0010 SS = 0010
EAX = FCE17833 EBX = 00000000 ECX = CA94FDA8 EDX = 00000046
ESI = 00000000 EDI = 15E2F7B7 EBP = CD414A44 ESP = CD414A30
EIP = 00000000 FLAGS = 00000046


Running process: Interrupt service routine (nested count 1)
Interrupted process: Server 10 Process
Thread Owned by NLM: SERVER.NLM
Stack pointer: CD414CAC
OS Stack limit: CD40D040
Scheduling priority: 67371008
Wait state: 3030070 Yielded CPU
...
>>

AUTO RESTART AFTER ABEND = 1
CPU HOG TIMEOUT AMOUNT = 1 minute

This stand-alone server was up since 60 days.

When this server was halted (and not abend), all is blocked (we need to power off/on)

CPQHLTH.NLM 5.07
CPQASM.NLM 1.07

The ASR is set with CPQ Insight Manager.

Many thanks,
Alain Bron
dmowry
Occasional Visitor

Re: CPU Hog with cpqacmgr.nlm and ASR does'nt restart server

Alain--
Can you please copy/paste the full text of the ABEND.LOG file, including the stack dump for ABEND 3. If the previous ABENDs (ABEND 1, ABEND 2) occurred within ten minutes of ABEND 3, then we need to see those stack dumps also.
Thanks.
--Daniel
Bron_3
Occasional Contributor

Re: CPU Hog with cpqacmgr.nlm and ASR does'nt restart server

The precedent abend appears 6 weeks before.

Alain
dmowry
Occasional Visitor

Re: CPU Hog with cpqacmgr.nlm and ASR does'nt restart server

Alain--
One of the developers down the hall is already looking into a similar issue, but we may not know anything definitive until after the holidays.
Thanks.
--Daniel