SUSE warm reset generated uncorrectable error on server

[Bios123]

I am running memory test on SUSE, Memory test is just writing some patterns into a memory and reading it back in all CPU threads. After a while, when memory test is running in the background , I am trying to reset the system by echo b> /proc/sysrq-trigger command, Upon SUSE warm reset , I am seeing some UNCORRECTABLE MEMORY ERROR in model specific register (MSR : 0x421),

I am seeing this behavior on one of my server which has huge amount of memory.

,Have you ever seen this UCECC error when SUSE generates warm reset when memory test in the background ? Could anyone please let me know!

[gogalthorp]

Memory test is not meant to be run with other process. It is changing memory and thus can corrupt other running threads.

[Bios123]

Thanks. During force shutdown in suse, CPU reset is occured before the memory controller finish its operation. This might also corrupt the memory . Is that correct ?

[gogalthorp]

If you want to test the memory boot from the install media and select memory test. Run over night. any errors indicate memory problems.

[Bios123]

Thank you!