So because my Supermicro server has rebooted itself for unknown reasons at least twice in the past two months, I decided to run memtest last night. I ran one pass with no errors, then rebooted and activated SMP and ran it again and went to bed. This morning I came down to find that it had crashed after only about 17 minutes. Does this bode well/bad or it is unimportant? See attached screen shot.

Crashing while running Memtest certainly "bodes bad". Not necessarily a memory fault but quite possibly a "something else" fault. I don't think you can consider it unimportant in any way. It might be best to strip the system right down to the bare minimum - motherboard, CPU and one stick of RAM and then try again before swapping things systematically. The easiest thing to try first is probably a different power supply.

Remember, it only crashed in SMP mode, I don't know if that is significant or not, in 'normal' mode, it ran for a full 5 hours and completed one pass without any errors. I am just wondering if the program itself doesn't run in SMP mode very well?

I suppose the question then is, do you plan to run unRAID in SMP mode?

 

You could try MemTest version 7 from here: http://www.memtest86.com/ It installs as a dual boot USB stick. BIOS boot runs the old version 4; UEFI boot runs version 7.

 

Well unRAID is SMP aware, SMP is an acronym for Symmetric Multi Processing, I just happened to notice that the version of memtest on the unRAID flash drive had an SMP mode, figured I would run one pass in it.

Yes, I did know that actually.    I was being slightly facetious, but my point is that if a simple system like MemTest causes your server to crash, what hope is there of running a more complex one like unRAID. Of course you'll want to run unRAID on all cores but I think you'll continue to have problems until you get to the bottom of the issue. That's why I suggested the newer version 7 of MemTest.