Running the 6.5.1rc5 and I have received a few errors similar to this:

Apr 17 22:10:21 Tower kernel: mce: [Hardware Error]: Machine check events logged
Apr 17 22:10:21 Tower kernel: mce: [Hardware Error]: CPU 2: Machine Check: 0 Bank 0: 9000004000010005
Apr 17 22:10:21 Tower kernel: mce: [Hardware Error]: TSC 21546d59d59c6 
Apr 17 22:10:21 Tower kernel: mce: [Hardware Error]: PROCESSOR 0:306a9 TIME 1524017421 SOCKET 0 APIC 4 microcode 1f

Not very frequent but enough that I am concerned but not sure how to test/remedy. Enclosed are my diagnostics from today - would appreciate any guidance.

tower-diagnostics-20180419-1010.zip

Start by running memtest for a few hours, ideally 24.

Thanks johnnie.black - will do this over the next few days ...

Memtest didn't show any issues - although it should run for a few days but can't cripple the server that long - overnight test showed no errors.

 

 I have actually now added another 16GB of RAM as well (4x8GB) ...

 

Another error listed on the 24th but nothing since then:

Apr 24 10:53:06 Tower kernel: mce: [Hardware Error]: Machine check events logged
Apr 24 10:53:06 Tower kernel: mce: [Hardware Error]: CPU 2: Machine Check: 0 Bank 0: 9000004000010005
Apr 24 10:53:06 Tower kernel: mce: [Hardware Error]: TSC cabbb7762305 
Apr 24 10:53:06 Tower kernel: mce: [Hardware Error]: PROCESSOR 0:306a9 TIME 1524581586 SOCKET 0 APIC 4 microcode 1f

Same - only different!?

 

I am going to 'redo' my stock Intel CPU heatsink with a Noctua very shortly - waiting for parts - perhaps that will help ...

Is your server functional while receiving these errors?  I'm actually getting a ton of them after changing hardware.  And I can no longer get to the gui in addition to that.  Just wondering if we could compare notes on symptoms and possible fixes?

The server DOES seem functional after this happens - it ran along just fine for a while and I never noticed anything wrong at all

But now you mention it my local GUI was 'frozen' on this last error - I hardly ever boot to a local GUIbut I had for this instance.

I remember wanting to reboot from the server but had to do it from a remote connection - the remote GUI was fine and let me restart the server.

I am questioning my stock Intel cooler though and not really 'trusting' it to be efficient.

I updated from rc5 to 6.5.1 last night as well with no issues so fingers crossed ...

1 hour ago, dstanley said:

 I have actually now added another 16GB of RAM as well (4x8GB) ...

 

It actually try running on a single DIMM for a while instead of four, and try it in different sockets.

Since this is first thread that pops up in google search when Googling this specific error (Machine check events logged Bank 0: 9000004000010005 unraid). I figured I'd leave my two cents. I got exactly same error as OP. Same error was filling my unraid syslogs and the server was unstable, randomly crashing and rebooting. I managed to solve this by changing my motherboard's bios settings.

 

I set my CPU CORE RATIO from AUTO to SYNC ALL CORES.

Everything else related to CPU is set to AUTO. 

 

Motherboard: Asus z170-A

CPU: Intel i7-6700k