HPE Proliant ML30 - the Host OS run back when the Processor Core Disable = 1

[Anh_Cop]

Dear All,

my Server is HPE Proliant ML30 Gen 9.

Host OS is ESXI 6.5.0 

RAM 64G, in each 16G in 4 slot 

CPU xeon E3-1240v5 

it turn to PSOD after run in 2 years.

-------------------------------------------------

check IML:

Uncorrectable Machine Check Exception (Board 0, Processor 0, APIC ID 0x00000006, Bank 0x00000000, Status 0xBD800000'00000006, Address 0x00000000'00000000, Misc 0x000007FE'8A97D000)

And when i change core diable in BIOS from [0] to [1]

the host run back.

Question:

- CPU broken or not, do i need to replace cpu?

- can i run the host as normal with this situation?

- why i change the processor core diable = 1 per socket the host OS work again as CPU not failed?

-  what is the meaning of the processor core diable?

Thanks all for your time

[Kamalji]

To answer the query on Processor core Disable ,

Please Refer
UEFI System Utilities User Guide for HPE ProLiant Gen9 Servers and HPE Synergy
https://support.hpe.com/hpsc/doc/public/display?docId=c04398276

Page 46

Processor Core Disable
Use this option to specify the number of cores to enable per processor socket using Intel's Core MultiProcessing (CMP) Technology. Unused cores are disabled. Setting this option can:
• Reduce processor power usage
• Improve overall performance for applications that benefit from higher performance cores rather than more
processing cores
• Solve issues with software that is licensed on a per-core basis

Troubleshooting suggested 

You may like to check the last change made to see the begining of UMCE error .This will guide you to reverse the change and check if this is due to the changes .
Keep a check with VMware support ,if this is not due to a bug or curruption .
If UMCE error started during the operation without any changes ,in that case it is suggested to swap processor0 with known working and check if error moved with the Processor or stay back .