System Rebooted Automatically (Digital Unix V4.0)

[Anil_29]

Hi All
I have a Digital unix box with version 4, the system all of a sudden got rebooted. When I check the /var/adm/messages file it has recorded the below errors from 25th of June 04:
----------------------------------------------
Jun 25 21:22:15 suboz017 vmunix: WARNING: too many System corrected errors detec
ted on cpu 0. Reporting suspended.

Jun 25 21:24:15 suboz017 vmunix: WARNING: too many System corrected errors detec
ted on cpu 1. Reporting suspended.

Jun 25 21:27:14 suboz017 vmunix: WARNING: too many System corrected errors detec
ted on cpu 0. Reporting suspended.
----------------------------------------------

Now I received a single error on cpu1 also:

Jun 28 06:36:04 suboz017 vmunix: WARNING: too many Processor corrected errors de
tected on cpu 1. Reporting suspended.
-----------------------------------------------

What does this error message mean ?
Will the system be rebooted again ?

Please help....

Thanks & Regards,
Anil.

[Mohanasundaram_1]

Anil,

Sounds like an LPMC equivalent in digital UNIX. The message is coming for CPU0 and CPU1 more than once.

However I have not worked with digital UNIX to comment further on this.

You may try this question at the appropriate forum.

Cheers,
Mohan.

[Sanjay Kumar Suri]

Not sure if the following links will help:

http://dbforums.com/t629496.html


sks

[Bharat Katkar]

Anil,
Not very sure about this since it is Digital unix but looking at the problem it loooks like cpu0 and cpu1 are in problem. You may try shutting down the system (if possible, better to get experts advice) and analysize system for any H/w problem like bad CPU or any supporting H/w.
Check whether all internal FANS are working and Heat sinks are not getting too Hot. Just a thought. :)

Hope that helps.
Regards,

[Ian Miller.]

you may get more help in
http://forums1.itrc.hp.com/service/forums/familyhome.do?familyId=280

[Michael Schulte zur Sur]

Hi,

look into binary errorlog with
dia -R | more
for more info on that.

greetings,

Michael

[Anil_29]

Hi ALL,

Thanks for all the replies.

Michael can you please give the dia command in detail, when I type "dia -R | more " the system says cant find dia.
Is dia the executable or .......

Thanks & Regards,
Anil.

[Michael Schulte zur Sur]

Hi,

dia is the command for decevent. You will have to install it.
Can you tell us more about the box, hardware and os release?

I would think it is either cpu or memory.
Do you have support?

greetings,

Michael

[Anil_29]

Hi Michael,

We do not have this tool installed, is there any other way to find out the root cause for the server reboot.
We are still getting the errors every 8seconds, and worried if the server will reboot again.
Pls help.
Will this help:

pr001srv:/ > uname -a
OSF1 pr001srv V4.0 878 alpha



Reg
Anil.

[Ralf Puchner]

there are other postings in this forum always ask for help about the same problem.
Solution: open a call within HP and let's analyze the binary.errlog. If using dia or uerf you must be an Alpha specialist or do you know every bit/register and their meaning?

Mostly your cpu, memory or cache is defect - but HP support center will tell you exactly what must be replaced.