Centos – How to find which memory has CE error

centoshardwarememory

I think I have a memory in my server which has errors and I am wondering how I can find which one it is.

Server model: Supermicro 6072R-EN3RFT

RAM: 128 GB

CentOS 7 with latest updates installed

The mcelog says the following:

:[  883.230897] EDAC sbridge MC0: HANDLING MCE MEMORY ERROR
:[  883.230904] EDAC sbridge MC0: CPU 0: Machine Check Event: 0 Bank 10: cc0001c7000800c1
:[  883.230906] EDAC sbridge MC0: TSC 0 
:[  883.230908] EDAC sbridge MC0: ADDR b71b18000 
:[  883.230909] EDAC sbridge MC0: MISC 908401000200e8c 
:[  883.504829] EDAC sbridge MC0: PROCESSOR 0:306e4 TIME 1469612575 SOCKET 0 APIC 0
:[  883.504841] mce: [Hardware Error]: Machine check events logged
:[  883.606151] EDAC MC0: 7 CE memory scrubbing error on CPU_SrcID#0_Ha#0_Chan#0_DIMM#0 (channel:0 slot:0 page:0xb71b18 offset:0x0 grain:32 syndrome:0x0 -  OVERFLOW area:DRAM err_code:0008:00c1 socket:0 ha:0 channel_mask:1 rank:1)
:[  899.306134] EDAC sbridge MC0: HANDLING MCE MEMORY ERROR
:[  899.306143] EDAC sbridge MC0: CPU 0: Machine Check Event: 0 Bank 10: cc000207000800c1
:[  899.306145] EDAC sbridge MC0: TSC 0 
:[  899.306148] EDAC sbridge MC0: ADDR c71b19000 
:[  899.306150] EDAC sbridge MC0: MISC 908410000200e8c 
:[  899.306153] EDAC sbridge MC0: PROCESSOR 0:306e4 TIME 1469612590 SOCKET 0 APIC 0
:[  899.306172] mce: [Hardware Error]: Machine check events logged
:[  899.644814] EDAC MC0: 8 CE memory scrubbing error on CPU_SrcID#0_Ha#0_Chan#0_DIMM#0 (channel:0 slot:0 page:0xc71b19 offset:0x0 grain:32 syndrome:0x0 -  OVERFLOW area:DRAM err_code:0008:00c1 socket:0 ha:0 channel_mask:1 rank:1)
:[  901.190512] {1}[Hardware Error]: Hardware error from APEI Generic Hardware Error Source: 1
:[  901.190528] {1}[Hardware Error]: It has been corrected by h/w and requires no further action
:[  901.190533] {1}[Hardware Error]: event severity: corrected
:[  901.190538] {1}[Hardware Error]:  Error 0, type: corrected
:[  901.190541] {1}[Hardware Error]:  fru_text: CorrectedErr
:[  901.190546] {1}[Hardware Error]:   section_type: memory error
:[  901.190549] [Firmware Warn]: error section length is too small
:[ 4916.540282] EDAC sbridge MC0: HANDLING MCE MEMORY ERROR
:[ 4916.540290] EDAC sbridge MC0: CPU 0: Machine Check Event: 0 Bank 10: cc000287000800c1
:[ 4916.540292] EDAC sbridge MC0: TSC 0 
:[ 4916.540294] EDAC sbridge MC0: ADDR b743ff000 
:[ 4916.540296] EDAC sbridge MC0: MISC 908400800240e8c 
:[ 4916.540298] EDAC sbridge MC0: PROCESSOR 0:306e4 TIME 1469616606 SOCKET 0 APIC 0
:[ 4916.540313] mce: [Hardware Error]: Machine check events logged
:[ 4916.540340] EDAC MC0: 10 CE memory scrubbing error on CPU_SrcID#0_Ha#0_Chan#0_DIMM#0 (channel:0 slot:0 page:0xb743ff offset:0x0 grain:32 syndrome:0x0 -  OVERFLOW area:DRAM err_code:0008:00c1 socket:0 ha:0 channel_mask:1 rank:1)

I tried the following:

grep "[0-9]" /sys/devices/system/edac/mc/mc*/csrow*/ch*_ce_count
/sys/devices/system/edac/mc/mc0/csrow0/ch0_ce_count:669
/sys/devices/system/edac/mc/mc0/csrow0/ch1_ce_count:0
/sys/devices/system/edac/mc/mc0/csrow0/ch2_ce_count:0
/sys/devices/system/edac/mc/mc0/csrow0/ch3_ce_count:0
/sys/devices/system/edac/mc/mc1/csrow0/ch0_ce_count:0
/sys/devices/system/edac/mc/mc1/csrow0/ch1_ce_count:0
/sys/devices/system/edac/mc/mc1/csrow0/ch2_ce_count:0
/sys/devices/system/edac/mc/mc1/csrow0/ch3_ce_count:0

Does this mean, I have 8 slots with 16 GB in each and the first slot contains the memory with error?

Any ideas which one is the memory module with errors? I am not a system administrator so I don't really know how to proceed…

Kind regards

Best Answer

I would expect your DIMM slots to perhaps be labelled BANK A DIMM 0, BANK A DIMM 1, etc. up to BANK B DIMM 3. You could make the assumption that BANK A DIMM 0 is the problem one, and so try swapping it with one of the other 7 assuming they're all equal and repeat your tests until it generates an error again. If a different /sys/devices/system/edac/mc/mc?/csrow0/ch?_ce_count counter is incremented then you can be reasonably sure you've found the problem DIMM.

Related Topic