Microserver Gen10 Plus crashing - "Uncorrectable Machine Check Exception"

[wrongecho]

Hello,

I have a HPE Microserver Gen10 Plus at home running Server 2019 Hyper-V edition that keeps unexpectedly rebooting/crashing.

I have applied all OS and HPE BIOS updates, the latest SPP and even adjusted the power profile / C-states as per this post, but the crashes continue. I do not think this is a thermal issue as I have additional fans for airflow where the server is located.

There is no error in Windows other than the shutdown being unexpected - no BSOD.

The error message from AHS reads:

"Uncorrectable Machine Check Exception (Processor 1, APIC ID 0x00000000, Bank 0x00000006, Status 0xEE000000'0040110A, Address 0x00000000'FEF1E9C0, Misc 0x00000438'80000086). ACTION:Update the system firmware. If the issue persists, contact support."

I did contact support but the server is out of warranty and they are unable to assist without payment. As I am a homelab user, this is not feasible for me.

Does anyone have any other suggestions, please? I would be incredibly grateful. This is driving me crazy!

[Suman_1978]

Hi,

Have you tried updating other firmwares like iLO, SPS, etc.

Usually for this error the solution could be to swap or replace the processor.
https://support.hpe.com/hpesc/public/docDisplay?docId=a00029456en_us

Thank You!
I work with HPE but opinions expressed here are mine.

[wrongecho]

Hi @Suman_1978 

Thanks for your reply.

The server is running the latest Service Pack 2024.04.00.00. I've applied all relevant updates the Service Pack updater recommended.

Is there a specific firmware I'm missing that solves this problem? I'd prefer to avoid swapping the processor.

Thanks!

[BPSingh]

Bank 0x00000006 UMCE typically indicates an issue with a PCI device. I recommend logging a support ticket to identify the device and check its firmware and driver.

[wrongecho]

Hi BPSingh
Appreciate your reply.

My expansion ports are all empty, is there anything else that could cause this?

Again, as per the original post, I cannot raise this with support as the server is out of warranty.

Thanks in advance.

[BPSingh]

What is the workload profile set in BIOS?

Please ensure the workload profile is set to HPC (High Performance Compute) for non-virtualized host OS, or to Virtualization Max Performance for virtualized host OS.

Refer to below web-link:

https://support.hpe.com/hpesc/public/docDisplay?docId=sd00001068en_us&docLocale=en_US&page=GUID-E7671EDC-DC0F-4D82-B4E8-2B5FD6E76E54.html

[wrongecho]

Hi @BPSingh 

The Workload profile is set to "Virtualisationn - Max Performance". This option also disabled C-States.

Appreciate the suggestion however this was changed some time ago when I found a blog here, and has not helped with the crashes sadly.

[BPSingh]

Then I think the logs should be analysed to find the cause of the issue.