Hey, this MCE problem has popped up a couple of times in FCP. I installed mcelog so my diagnostics should have all the info.

Any ideas?

Dec 29 05:36:59 tobor-server kernel: mce: [Hardware Error]: Machine check events logged
Dec 29 05:36:59 tobor-server kernel: EDAC sbridge MC0: HANDLING MCE MEMORY ERROR
Dec 29 05:36:59 tobor-server kernel: EDAC sbridge MC0: CPU 0: Machine Check Event: 0 Bank 10: 8c000041000800c1
Dec 29 05:36:59 tobor-server kernel: EDAC sbridge MC0: TSC 1b4f8b5ad885a 
Dec 29 05:36:59 tobor-server kernel: EDAC sbridge MC0: ADDR 5ad232000 
Dec 29 05:36:59 tobor-server kernel: EDAC sbridge MC0: MISC 90850001000028c 
Dec 29 05:36:59 tobor-server kernel: EDAC sbridge MC0: PROCESSOR 0:306e4 TIME 1546022219 SOCKET 0 APIC 0
Dec 29 05:36:59 tobor-server kernel: EDAC MC0: 1 CE memory scrubbing error on CPU_SrcID#0_Ha#0_Chan#0_DIMM#0 (channel:0 slot:0 page:0x5ad232 offset:0x0 grain:32 syndrome:0x0 - area:DRAM err_code:0008:00c1 socket:0 ha:0 channel_mask:1 rank:1)

 

tobor-server-diagnostics-20181230-0005.zip

Looks like an ECC memory error, check the boards system event log or just event log, it might have more info on the affected DIMM slot.

The machine is offsite so will take a few days to get in front of. Are you suggesting it may be ram related? I ran a 24h memtest a few months ago with no errors however there is definitely something wrong in the memory department somehow.

1 hour ago, enmesh-parisian-latest said:

I ran a 24h memtest a few months ago with no errors

 

13 hours ago, johnnie.black said:

Looks like an ECC memory error

ECC memory errors aren't exposed by memtest unless they are particularly severe. That's the whole purpose of ECC, single bit errors are corrected silently and seamlessly.