I would like to know how to generate an ECC correctable error or an FB-DIMM configuration write error (Non-Fatal Recoverable) that will be displayed to the dmesg or messages logfiles, like in the following examples:
7840 08/09/11 15:10:47 MIN BMC Memory #08 Correctable ECC, P1_DIMMF1 6f [20 ff 50] Edac MC0:UE row 0, channel -a = 0 labels "-" NON-Fatal recoverable (Branch = 0 DRAM - Bank = 0 Buffer ID = 0 RDWR= Read RAS = 0 Cas= 0 NON-Fatal recoverable Err=0x2000 (FB-DIMM configuration write error on first attempt)
I am using a CentOS 7 server inside a VirtualBox having a Windows 8.1 PC as host.
Except the aforementioned errors, any way on how to generate other RAM error types, would be so valuable. I have searched a little bit and tried to use edac in order to inject some errors, but without success (I cannot find any functions inside the path
/sys/devices/system/edac/mc/ to inject errors).
If you could guide me on how to produce such RAM errors in my system that would be displayable in dmesg and messages logfiles, I would much appreciate it. The goal is to gather several logfiles with hardware errors, that will be my dataset for research purposes.