Operating System - Tru64 Unix
1827697 Members
3085 Online
109967 Solutions
New Discussion

Re: ES40 System Crash

 
SOLVED
Go to solution
Emad Omar
Regular Advisor

ES40 System Crash

Hi all,

I have a crash case with Alpha Server ES40, 2CPUs 667MHz and 4GB Memory.It is running on Tru64 V5.1 . I suspected with the CPUs and I removed them and installed a new CPU instead.But I'm still getting the same crash.Please find attached the crash file hoping this could help to diagnose the problem.

Kind regards,

Emad Omar
10 REPLIES 10
Michael Schulte zur Sur
Honored Contributor

Re: ES40 System Crash

Hi Emad,

the best advice would be to open an call with HP in case you have a maintenance contract. Otherwise can you post the output of decevent or webes from that incident?

greetings,

Michael
Ivan Ferreira
Honored Contributor

Re: ES40 System Crash

We had this panic string, and the problem was a memory module.

The panic string can be caused also by a bad power suppy.
Por que hacerlo dificil si es posible hacerlo facil? - Why do it the hard way, when you can do it the easy way?
Emad Omar
Regular Advisor

Re: ES40 System Crash

Dear Ivan,

The ES40 has redundant power supply, which it means that if we got a power supply failure, then system will still run without any interruption. So I exclude the PSU and may be I have a defective Memory.
Also I need to know if there is another command to test the Memory because when I issued the command memexer at >>> and then issued >>> show_status , I got nothing important.

Please advise . . .

Emad Omar
Michael Schulte zur Sur
Honored Contributor

Re: ES40 System Crash

Hi Emad,

the status of the power you can see with
show power
Have yo run a general
test
?

Michael
Ivan Ferreira
Honored Contributor

Re: ES40 System Crash

Redundant power suppy is not warranty, still the problem can be that, a bad power supply can damage components. Also, there are rules about power supply and the ammount of processors (That i can't remember)

We detected the bad power supply using dia -R -o full, also you can use the show power SRM command as methioned above.
Por que hacerlo dificil si es posible hacerlo facil? - Why do it the hard way, when you can do it the easy way?
amrelsayed
Frequent Advisor
Solution

Re: ES40 System Crash

Dear Emad,

your problem is memory, as it is become a common in ES40 machines, i want your binary.errlog to analze it, and i will tell you which memory medule wants to be replaced.

Best Regards,
Amr
Try To Be Smart
Vladimir Fabecic
Honored Contributor

Re: ES40 System Crash

I also think that memory may be a problem.
If you have enough time try to do memory tests:
>>> init
>>> memexer 3
I had simular problem once. After two hours of testing memexer found dagaged memory module.
In vino veritas, in VMS cluster
Adam Strobel
Frequent Advisor

Re: ES40 System Crash

I agree with having a memory problem. My ES40 was just crashing every other day or so with the same error "Processor Machine Check" and It turned out to be bad memory.

good luck

--Adam
Emad Omar
Regular Advisor

Re: ES40 System Crash

Dear all,

Thank yu for all of you. Yes I got 1 DIMM defected at one of Memory Module .

Kind regards,

Emad Omar
Emad Omar
Regular Advisor

Re: ES40 System Crash

Memory Failure as I mentioned.

Thank you again.