(SOLVED) mce: [Hardware Error]: Machine check events logged

[DivideBy0]

Can anyone please help me decipher this mce error   This is a brand new build so is already acting funny  

 

 

241831.858564] mce: [Hardware Error]: Machine check events logged

[241831.858570] mce: [Hardware Error]: Machine check events logged

 

 

root@NAS-UNRAID:~# mcelog

Hardware event. This is not a software error.

MCE 0

CPU 0 BANK 5

ADDR 22f43abc0

TIME 1591113315 Tue Jun  2 10:55:15 2020

MCG status:

MCi status:

Corrected error

Error enabled

MCi_ADDR register valid

MCA: MEMORY CONTROLLER RD_CHANNEL1_ERR

Transaction: Memory read error

STATUS 9400004000910091 MCGSTATUS 0

MCGCAP 806 APICID 0 SOCKETID 0

MICROCODE 12d

CPUID Vendor Intel Family 6 Model 77

Hardware event. This is not a software error.

MCE 1

CPU 1 BANK 5

ADDR 22f43abc0

TIME 1591113315 Tue Jun  2 10:55:15 2020

MCG status:

MCi status:

Corrected error

Error enabled

MCi_ADDR register valid

MCA: MEMORY CONTROLLER RD_CHANNEL1_ERR

Transaction: Memory read error

STATUS 9400004000910091 MCGSTATUS 0

MCGCAP 806 APICID 2 SOCKETID 0

MICROCODE 12d

CPUID Vendor Intel Family 6 Model 77

root@NAS-UNRAID:~#

 

 

nas-unraid-diagnostics-20200602-2313.zip

Edited June 14, 2020 by johnwhicker

[JorgeB]

That looks like an ECC RAM corrected error, there might be more information on the board's SEL (system event log)

[DivideBy0]

Nope that's pretty much 2 lines in the syslog so I run the mcelog to get more info.  This was buried between some USB/UPS issues I am having. 

 

240021.315170] usb 1-1.1: USB disconnect, device number 10

[240021.487384] usb 1-1.1: new full-speed USB device number 11 using ehci-pci

[240021.572741] hid-generic 0003:0764:0501.0009: hiddev96,hidraw0: USB HID v1.10 Device [CPS CST135XLU] on usb-0000:00:16.0-1.1/input0

 

[241831.858564] mce: [Hardware Error]: Machine check events logged

[241831.858570] mce: [Hardware Error]: Machine check events logged

 

[243974.212849] usb 1-1.1: USB disconnect, device number 11

[243974.386787] usb 1-1.1: new full-speed USB device number 12 using ehci-pci

[243974.471684] hid-generic 0003:0764:0501.000A: hiddev96,hidraw0: USB HID v1.10 Device [CPS CST135XLU] on usb-0000:00:16.0-1.1/input0

[JorgeB]

I said the board's system event log, usually accessible in the BIOS or over IPMI.

[DivideBy0]

On 6/3/2020 at 8:57 AM, johnnie.black said:

I said the board's system event log, usually accessible in the BIOS or over IPMI.

Thanks much Sir. I didn't see anything in the BIOS or IMPI logs.  I even send the IPMI logs to syslog and nothing on this MCE error.

 

That being said I did run an extensive memtest86 Pro test for 24 hours straight and no errors on memory so I guess it was just ECC doing its job during this heavy data set copy?  perhaps ECC corrected some corrupted data?

[JorgeB]

Memtest won't show errors with ECC RAM, if you can't find more info on the affected DIMM, just remove one at a time and test for a few days, or disable ECC in the BIOS (if that's an option) and run memtest again.

[DivideBy0]

On 6/8/2020 at 3:24 AM, johnnie.black said:

Memtest won't show errors with ECC RAM, if you can't find more info on the affected DIMM, just remove one at a time and test for a few days, or disable ECC in the BIOS (if that's an option) and run memtest again.

Thanks partner. I run memtest pro version for 2 days and nothing.  I think is ok as I haven't seen that error anymore.  It was just during a heavy copy and mdsum check from drive to drive, about 8TG of data.