Crashing server :(

[Henry Thomas]

Crashes in about 4 hours. 6.4.1 most recent version. One time it crashed, Parity Disk 2 was 'not assigned' but it did reassign. Yesterday it crashed, the cache drive SSD was unassigned. It did reassign. I do see CRC errors and I've tightened the cables but to no effect on crashing. 

unraid-diagnostics-20180215-1853.zip

[JorgeB]

Are you disabling C-States? It's a known issue with Ryzen:

 

 

[Henry Thomas]

It was disabled because I was fiddling with VM. But I did hit 'default' button last night. I'll check that but I don't think that's going to be it. But I'll check / reboot right now.

[JorgeB]

C-States have nothing to do with VMs, also make sure besides disabling it in the bios you're using this, from the thread above:

 

On 17/01/2018 at 4:38 AM, 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"

 

 

 

[Henry Thomas]

Yeah, c-states are off, but I skipped over that script edit. I have no idea how to get to that file. I installed notepad++ but that file isn't on the flash drive at least plugged into a windows laptop. Can I get to this file from a windows laptop from the server? I always get stuck here.

[JorgeB]

3 minutes ago, Henry Thomas said:

that file isn't on the flash drive at least plugged into a windows laptop

It's in the config folder, filename is just "go"

[Henry Thomas]

DUH. I got it. It looks like an override in case the BIOS gets flipped but I'm up for anything. Booting ....

 

It keeps kicking my cache drive off as "unmounted" so I've just physically disconnected it.

[Henry Thomas]

Fix Common Problems says "Your server has issued one more more call traces. This could be caused by a Kernel Issue, Bad Memory, etc. You should post your diagnostics" ok

[Henry Thomas]

Anyone see a reason why I can't plug my SSD back in? Diagnostics in first post

unRaid_Syslog

[JorgeB]

13 minutes ago, Henry Thomas said:

Anyone see a reason why I can't plug my SSD back in?

What do you mean by plug back in? There's no cache device assigned on the diags.

 

Feb 15 18:50:05 unRaid emhttpd: import 30 cache device: no device

 

[Henry Thomas]

that's because I unplugged it because after the crashes it was always 'unmounted' so I suspected it as being a culprit (crashed once since I unplugged it.)

[JorgeB]

We can't see the problem like that, try to mount it again, if the webGUI is unresponsive try grabbing the diags by typing diagnostics on the console.

[Henry Thomas]

It's been stable overnight. I am ripping a bluray now and I swear it's faster than with the SSD unplugged. I'm theorizing I had two issues - one the edit to the "go" file. Lessons learned - read the release notes carefully. I always do but I think when I processed that I translated it into "optional" or something because I've never had to edit a file like that on an incremental upgrade - ever. Second issue - flaky SSD. 

 

Is the cache drive supposed to be THAT much faster than a non-cache drive? I'm not using a stopwatch but I'm not noticing much or any difference. Perhaps that's due to the drive being flaky - OR is it because Ryzen has faster throughput to the array? Any thoughts?