"Your server has detected hardware errors. You should install mcelog via the NerdPack plugin, post your diagnostics and ask for assistance on the unRaid forums. The output of mcelog (if installed) has been logged."

 

I had one impromptu shut down, and on my trial install, on a different USB flash drive I had a variety of problems that reading up seemed to indicate was due to problems with the flash drive. 

 

I'm not sure what all the diagnostics will tell people who are accustomed to reading them, but "Fix Common Problems" seems pretty sure that I should be freaking out here. Happy to answer any questions for anyone interested in suggesting what's going on.

Thanks!

 

tower-diagnostics-20180208-0003.zip

Edited February 8, 20188 yr by hominid

Oh, I'm preclearing my last drive, so I'm pretty sure I can't update to 6.4.1 yet.

It crashed totally again at 5:45 or so. unresponsive to mouse, ssh. Here's the diagnostics from immediately after rebooting.

tower-diagnostics-20180208-2315.zip

I ran memtest; no errors.

This power supply, a Corsair - CXM 750W 80+ Bronze Certified Semi-Modular ATX Power Supply, has served faithfully for an Intel motherboard for 4 years before powering the current ASRock - X370 Taichi ATX AM4 Motherboard with a Ryzen 7 1700. In the few minutes between posting that I'd run memtest and while I looked up the power supply details, my system did a quick little reboot without having been asked to do anything after booting. 

 

Now, I'm also seeing the word "undefined" peppered along the top left portion of the LimeTech banner:

 

tower-diagnostics-20180209-0050.zip

Are you using the Ryzen C-states workaround? See here:

 

 

 

You should also upgrade to 6.4.1. There isn't much point in troubleshooting an older release.

 

(Edit) sorry I just saw your note about being in the middle of a preclear

Edited February 9, 20188 yr by ljm42

Thanks so much for your suggestions. I'll make the indicated changes next time I can safely shut down, or the next time it shuts down by itself.

On 1/16/2018 at 10:38 PM, ljm42 said:

If you are running Ryzen:

•  Edit your \\tower\flash\config\go script (using a good editor like Notepad++ (not Notepad)) and add the "zenstates" command right before "emhttp", like this:

  
  /usr/local/sbin/zenstates --c6-disable
  /usr/local/sbin/emhttp &

   Also, go into your BIOS and disable "Global C-state control"

 

Those two changes (or potentially one of them, since I made them together) seem to have done it. Not only have the alarms stopped going off, but it's been up for a personal record of 2.5 days!

 

Thanks All!