Hey All,

 

I've seen similar topics on the forum so sorry if I'm not understanding things properly. Within ~10 minutes of my server starting & the array starting it dies with a Machine Check Exception:

 

I'm not able to type anything at this point as the kernel has panic'd. Similarly, I see no reasonable information in a syslog. Lastly, /var/log/mcelog doesn't exist at next reset reset, I assume because it's not persisted across resets. I have no idea how to actually run this through mcelog.

 

I have captured a diagnostics as well as a syslog from a minute or so before the system went down.

tower-diagnostics-20181120-1819.zip 

syslog (7)

 

I've also run memtest for ~18 hours with zero errors. On top of that, I've removed one of my two memory dimms at a time to isolate a potential memory issue. Each time I did this I still had a panic.

 

I'm lost where to go at this point. Any advice would be incredibly appreciated.

syslog (7)

 

Edit:

 

It's also worth noting that I swapped hardware recently from a working windows gaming machine. The hard drives & PSU are the same and everything else is new.

Edited November 21, 20187 yr by yousillygoose

try this during the brief period you can

mcelog --syslog --daemon
tail -f /var/log/syslog > /boot/syslog.txt

then wait for the system to crash..  After that, upload syslog.txt in the root of the flashdrive

Alrighty, I just did it.

 

When I ran `mcelog --syslog --daemon` it produced the below in the logs. 

'Nov 21 07:00:57 Tower mcelog: Cannot open `/dev/mcelog': No such file or directory'

syslog.txt

 

The server also went down at roughly 07:10 (there are no recent logs around that time). Let me know what more I need to do.

Edited November 21, 20187 yr by yousillygoose

Since my last post, I also swapped the PSU but it's still panicking. My MOBO Bios is really old so I think I'll try to flash it. After that, I'm entirely at a loss. Hopefully someone can assist in getting a useable log which points to the issue.

I've updated the BIOS (z68 board) and it has been stable for a bit over a day. I'm still using a different PSU and only 1 DIMM so I'll start adding back in original components to make sure it remains stable.

 

For a good forum archive of how to solve this, I'd still love for someone to advise on how to read a machine check exception log in the unraid world.