3002704 - Linux: Hardware memory corruption logged on OS level

• OS logs (/var/log/messages, /var/log/warn, dmesg) contain entries referencing "hardware memory corruption":

Example 1:

2020-12-08T12:21:03.858622+01:00 server3 kernel: [1423822.815742] MCE: Killing <process:PID> due to hardware memory corruption fault at 7fe9bc16fb00

Example 2:

Dec 3 04:44:10 server1 kernel: [5105857.955669] [Hardware Error]: Machine check events logged
Dec 3 04:44:10 server1 kernel: [5105857.955693] Uncorrected hardware memory error in user-access at 48d3259dc0
Dec 3 04:44:10 server1 kernel: [5105857.956108] MCE 0x48d3259: huge page recovery: Delayed
Dec 3 04:44:10 server1 kernel: [5105857.956110] MCE 0x48d3259: huge page still referenced by 1 users
Dec 3 04:44:10 server1 kernel: [5105857.956112] Memory error not recovered
Dec 3 04:44:10 server1 kernel: [5105858.220111] MCE: Killing <process:PID> due to hardware memory corruption fault at 10ab458064

Example 3:

Jan 20 09:39:41 server2 kernel: [106912.808085] Disabling lock debugging due to kernel taint
Jan 20 09:39:41 server2 kernel: [106912.808181] [Hardware Error]: Machine check events logged
Jan 20 09:39:41 server2 kernel: [106912.808264] Uncorrected hardware memory error in user-access at f63ef7d300
Jan 20 09:39:41 server2 kernel: [106912.809617] MCE 0xf63ef7d: Killing <process:PID> due to hardware memory corruption
Jan 20 09:39:41 server2 kernel: [106912.809798] MCE 0xf63ef7d: dirty LRU page recovery: Recovered
Jan 20 09:39:41 server2 kernel: [106912.810001] MCE: Killing JobWrk0222:125751 due to hardware memory corruption fault at 7f4e5eb23330
Jan 20 09:39:53 server2 hpasmlited[3365]: CRITICAL: Uncorrectable Memory Error (Board 7, Memory Module 8)
Jan 20 09:39:53 server2 kernel: [106924.806296] [Hardware Error]: Machine check events logged
Jan 20 09:39:53 server2 kernel: [106924.806313] Uncorrected hardware memory error in user-access at d0e6d11f00
Jan 20 09:39:53 server2 kernel: [106924.806325] MCE 0xd0e6d11: corrupted page was clean: dropped without side effects

• SAP or any other process has crashed at the same time the logs are written

Linux (any distribution, any version)

Any SAP software

hardware memory corruption, MCE, memory corruption, Linux , KBA , BC-OP-LNX , Linux , Problem