too many Processor corrected errors detected on cpu

[Kirill Titievsky]

An alphaserver ES45, running Tru64, started crashing after a few months of normal operation. The following keeps appearing in the /var/adm/messagesJul

5 03:59:32 node2 vmunix: WARNING: too many Processor corrected errors detected on cpu 2. Reporting suspended. This message appears

The cpu # varies.

Please help.

[Mobeen_1]

Kirril,
Take a look at the discussion thread below.

http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=507209

That should answer your question.

regards
Mobeen

[Evert Jan van Ramselaar]

This might indicate a hardware problem. Try "uerf -R | more" to get some more specific information.

If you have hardware support, it would be wise to log a call with your vendor and send them the binary.errlog.

EJ

[Ralf Puchner]

please use the search function prior to post such a question. It was answered several times - if you are not aware of the search function please let us know.

And in case of your problem:
open a call within the HP support center or do you know a person knowing the alpha registers and hardware addresses from scratch?

[Kirill Titievsky]

Many thanks Mobeen, Evert, and Ralph. An HP expert used the binary log to conclude that this is a problem with the memory DIMMs and will have the DIMMs replaced. He said all errors reported the same memory register, so it sounds like Ralph knew what this was about.

Ralph, given the quality of disussions, I will most certainly take better advantage of the search function here next time.