ProLiant Servers (ML,DL,SL)
cancel
Showing results for 
Search instead for 
Did you mean: 

DL380G6 Corrected Machine Check error Win2008R2

Simon Peak
Occasional Advisor

DL380G6 Corrected Machine Check error Win2008R2

I'm seeing the following on an HP DL 380G6 after freshly installing Windows Server 2008 R2 Standard Ed. x64 OS

(The processor is an Intel E5540)

Level: Warning
Source: WHEA-Logger

A corrected hardware error has occurred.

Reported by component: Processor Core
Error Source: Corrected Machine Check
Error Type: Cache Hierarchy Error
Processor ID: 4

This is logged every few minutes in the system event log and eventually the system threshold for correctable errors is exceeded.

At that point, HP Server Agents log an error indicating the CPU should be replaced.

I have run complete tests via HP offline diagnostics and the processor tests fine, so it seems less likely there is a hardware fault.

I'm wondering if others might have seen this before and perhaps it could be a software glitch or issue.

I would really appreciate any suggestions or ideas.

Thanks everyone.
9 REPLIES
marcus1234
Honored Contributor

Re: DL380G6 Corrected Machine Check error Win2008R2

updates firmware with cd 8.6

and drivers with psp softwarecdrom 8.3 if issue continues


http://support.microsoft.com/?kbid=912310

enjoy :)
Simon Peak
Occasional Advisor

Re: DL380G6 Corrected Machine Check error Win2008R2

Thanks very much for your reply Mark.

I did install the server using SmartStart 8.30 - sorry should have mentioned that in my original post.

Today I spent the day making sure drivers and firmware are up-to-date on the server.

That didn't help - the same warning is logged.

I also installed Windows 2003 X64 to check if the warning appears under that OS and it does appear exactly the same as a warning in the system event log.

So that leaves me with what looks like a hardware fault. I guess my next step is to call HP as I am not sure what else it could be.

I would appreciate any further advice or ideas...

Thank you
Simon


marcus1234
Honored Contributor

Re: DL380G6 Corrected Machine Check error Win2008R2

appears culprit is the cpu..
marcus1234
Honored Contributor

Re: DL380G6 Corrected Machine Check error Win2008R2

Simon Peak
Occasional Advisor

Re: DL380G6 Corrected Machine Check error Win2008R2

I had seen that page but hadn't applied the System Programmable Logic Device update.

So I applied that update to see if it would help with my issue.

Unfortunately, it didn't.

I'm still seeing the same errors logged in the event log, later followed by a message from hpilo2 stating:

Processor 1 has corrected an excessive number of internal errors. The system will continue to operate.

At this point the processor is flagged on the system management home page with an action: Replace CPU.

The strange thing is the system seems stable, even when running a CPU intensive stability test (Prime 95). In fact, I've noticed the machine check warning seems to not be logged when CPU usage is high; only when the processor is idle.

Thanks again for taking the time to help out Mark. I wish we could figure this out but it seems to be a tricky one.

I welcome any further suggestions or comments.

Thanks - Simon
marcus1234
Honored Contributor

Re: DL380G6 Corrected Machine Check error Win2008R2



as I mentioned

appears culprit is the cpu..


friendly hint:)



This is how we thank each other in the forum

https://community.hpe.com/t5/Community-FAQ/FAQ-Kudos/m-p/6838486/highlight/true#M7

Enjoy:)

Simon Peak
Occasional Advisor

Re: DL380G6 Corrected Machine Check error Win2008R2

I am updating this thread with the results of my call to HP.

I provided details about the processor warnings.

The HP tech asked me to change a setting in the BIOS, to see if it would help. He mentioned there have been similar calls to this one mentioning correctable memory and processor warnings.

I am documenting those steps in case this helps someone else.

The steps are:

1. Reboot the server and press F9 to enter RBSU.
2. Select "Power Management Options."
3. Select "Advanced Power Management Options."
4. Select "Minimum Processor Idle Power State."
5. Select "C1E" or "No C-States."

After changing this setting, I no longer receive any warnings about the processor.

So it appears there isn't a hardware fault but some kind of firmware/OS/power management issue perhaps.

I wonder if there are any disadvantages to changing the idle mode setting... I suspect slightly reduced power efficiency.

Anyone please feel free to chip in with any comments about what this change actually does.

Hope this helps someone else.
Simon

marcus1234
Honored Contributor

Re: DL380G6 Corrected Machine Check error Win2008R2

marcus1234
Honored Contributor

Re: DL380G6 Corrected Machine Check error Win2008R2

this server as you can see from various advisories has had quite a few issues and but appears they are actually getting sorted out as we go along.