deactivated processors on rx7620

[Sherif A. Louis]

we have an rx7620 with ONE cell and FOUR 1.5 6MB cache processors. The machine is running windows 2003 server and it crashed 2 days ago and after looking at the MP logs and the EFI info cpu we found that 3 processors were deactivated.
and the 'ps' command showed the processors as populated and no failures were detected on any hardware.
we activated the processors from EFI and they were activated successfully, the server is up and running, BUT we need an explanation to why the processors were deactivated from the first place. Thank you.

[Sameer_Nirmal]

Maybe looking at the FPL events would help.

[Sherif A. Louis]

there you go, the FPL and the SEL.

[Sherif A. Louis]

this is the SEL

[YAQUB_1]

Hi Sherif,

>>> BUT we need an explanation to why the processors were deactivated from the first place.
===> Keyword = CELL_HW_DEGRADED

Description:
A dimm or CPU has failed and is not operational for the system. This event is emitted prior to determining if the cell should be integrated into the Partition.

Action:
A deconfigured dimm or cpu has been detected. Examine earlier events to isolate the problem.

Thanks--Yaqub
I am a Customer Advocate!!!

[Sherif A. Louis]

yes i know that Yaqub. The problem is the customer wants an explanation to why they were deactivated! and the thing is that we reactivated the processors ( three of them ) and they worked fine and they still are!

[YAQUB_1]

Hello Sherif,

Apologize for delay response.

Have U check my description, BTW, If U need more technical explanation regarding this issue then U opens a case HP customer support center. Because of HP engineer provide U actual root cause analysis report, they have some internal analysis tools U know.

Thanks--Yaqub
I am a Customer Advocate!!!

[Sherif A. Louis]

you don't have to apologize for being late. i should be thanking you for trying to help me out.
i do know that HP has there tools the thing is that the servers are out of warranty and we haven't renewed them yet. i am the on site support engineer btw. the thing is that i'm not a windows specialist. i work with HP-UX so that windows environment is so obscure to me.
I just want to tell the customer a reason why the two processors were deactivated, when there's nothing wrong with them, cuz as i mentioned earlier we reactivated them successfully and they are working perfectly!

[Sandy Chen]

Hi,

I remember that feeling, going onsite handling Windows on Integrity while you know nothing about Windows :D

Anyway.. has you see anything on the windows event log? Perhaps it could give you a bit of clue.

I actually experience somthing like this because one of the CPU is defected. There are load order for the CPU that why the other is going deactivated.

Regards,
Sandy

[Sherif A. Louis]

i did follow the cpu load order on reactivation. The perplexing thing here is that they were activated and they are operating fine,in contrast with the SEL which said that a CPU has been deactivated cause it failed, and should be replaced.
i did check the windows event logs it did mention something about an error on a disk controller nothing about the processors. The thing is that they are two servers (identical) connected to an XP 12000 operating oracle DB and an oracle RAC. the two servers keep rebooting until this very moment.
NOW the customer is expecting an explanation to why were the processors deactivated when there were nothing wrong with them!