Your server has detected hardware errors

[Scyllx]

Hello,

 

Since 2 weeks i got these notifications errors : Your server has detected hardware errors. You should install mcelog via the NerdPack plugin, post your diagnostics and ask for assistance on the unRaid forums. The output of mcelog (if installed) has been logged

 

The only error I see in the logs in the following : 

Dec 15 08:52:41 Tower kernel: mce: [Hardware Error]: Machine check events logged
Dec 15 08:52:41 Tower kernel: mce: [Hardware Error]: CPU 15: Machine Check: 0 Bank 5: bea0000000000108
Dec 15 08:52:41 Tower kernel: mce: [Hardware Error]: TSC 0 ADDR 1f80645cace7e MISC d012000100000000 SYND 4d000000 IPID 500b000000000 
Dec 15 08:52:41 Tower kernel: mce: [Hardware Error]: PROCESSOR 2:870f10 TIME 1639554740 SOCKET 0 APIC f microcode 8701021

 

And later :

Dec 15 08:52:45 Tower ntpd[1947]: kernel reports TIME_ERROR: 0x41: Clock Unsynchronized

 

Dec 15 09:03:06 Tower root: mcelog: ERROR: AMD Processor family 23: mcelog does not support this processor.  Please use the edac_mce_amd module instead.
Dec 15 09:03:06 Tower root: CPU is unsupported

 

Config is the following : 

3700X 

32gb ram

B550I Aorus AX

3*14Tb WD White

Crucial P2 1Tb for cache

 

I have run a memtest for like 8h with no errors reporting

 

Is this a dying CPU issue ? I have seen some random reboots, and parity check launch on restart

 

Thanks in advance

tower-diagnostics-20211215-1027.zip

[Squid]

Safe to ignore.  Happened during processor initialization and is fairly common for some hardware combinations at that point in boot up

[Scyllx]

But can it be linked to the random reboots I encountered ?

I notice it happened since I enabled the XMP for my RAM, but Memtest show no errors

[JonathanM]

14 hours ago, Scyllx said:

I notice it happened since I enabled the XMP for my RAM

Overclocking is not recommended for server stability.

[Scyllx]

Disabling the XMP profile seem to remove the mcelog events, weird