cancel
Showing results for 
Search instead for 
Did you mean: 

RH AS 2.1 and ASRs

Zate
Occasional Advisor

RH AS 2.1 and ASRs

I've had 3 different servers ASR Lockup /ASR this month and last month and I need to find out why. CIM just tells me the time and the fact that it happened. How can i find out exactly WHAT happened ? Can i configure AS 2.1 to core dump or something like that? Does it by default ? Any logs I should check? I've checked dmesg and messages, nothing just the reboot. No errors.

Anything at all ?
3 REPLIES
Don_89
Trusted Contributor

Re: RH AS 2.1 and ASRs

I had the same problem, turns out it was the new 7.0 Insight Manager agents causing the problem.. I went back to ver. 6.40 and haven't had a problem since..

Also, this is a good article on troubleshooting server crashes..

http://www.linuxdevcenter.com/pub/a/linux/2001/11/01/postmortem.html
Zate
Occasional Advisor

Re: RH AS 2.1 and ASRs

Thanks Don,

I am still using CIM 6.40 Agents though.

Compaq has suggested i enter the BIOS and disable the software ASR(ASM) to see whether the reboot is hardware or software related. I cant mess with a production box like that and just wait for it to fall over again. I expected a much better answer than that. Is there no way to determine what is happening on the box? Whats the point of running the agents if they cannot tell me anything, I can tell when my box rebooted by typing "uptime" when i login, i dont need a fancy web page and ages to tell me that. I need them to tell me WHY the box rebooted, or at least give me a clue.

Anyone from HP/Compaq assist in determining whats rebooting my boxes? I dont mind enabling some software to trap things etc etc, but i cant afford down time/reboot to mess with the bios right now.
Bill Wood_3
Respected Contributor

Re: RH AS 2.1 and ASRs

Please use the utility below to collect the necessary HW and SW details on the servers to which you are referring:

http://h18023.www1.hp.com/support/files/serve
r/us/download/16241.html

Please run with verbosity level 5. Please send the output to enterprise.ecs@hp.com and indicate your name and itrc in the subject.

About the agents:

The health driver and agents are not designed to troubleshoot complex operation system issues. They are specifically designed to warn of or prevent against server downtime by notifying the Systems Admin of hardware issues, such as, inevitable hard drive and memory failures. They also enable HP to provide its customers with a pre-failure warranty (replacing parts before they actually fail). The ASR feature in particular, will reboot the server if a lockup period exceeds the timer set in the BIOS instead of leaving the machine in a permanently hung state. Sometimes, however, it is necessary to disable this feature to capture information from the screen before the server is rebooted. In addition, some of our servers have NMI buttons that can cause the HP health driver to dump the stack on which it is running. Also, the SysReq keys can be used to collect information while the server is locked, but before the ASR feature reboots it.