I can't remember exactly but i think it's since i upgraded to 6.8.x.

When i reboot my unraid server all is working as expected.

After a few days the first symptom is: cpu load on dashboard stays at 0%

Then i have a 502 error when trying to open web terminal.

 To return to normal i have to restart nginx or the server.

2020/03/12 10:16:02 [error] 30369#30369: *663562 connect() to unix:/var/run/ttyd.sock failed (111: Connection refused) while connecting to upstream, client: 192.168.0.121, server: , request: "GET /webterminal/ HTTP/1.1", upstream: "http://unix:/var/run/ttyd.sock:/", host: "192.168.0.6:3080", referrer: "http://192.168.0.6:3080/Docker"

 I hope someone get an idea. Since 6.8.x my best uptime is around 2 weeks.

I didn't mention it but many (but not all) docker gui throw the same error, as some vm.

godzilla-diagnostics-20200312-1024.zip

Try running is safe mode for a week or two.

It will not be convenient as there are some plugins i need (nvidia, nerdpack, userscript, community app).

I know the nvidia plugin is not the culprit as i was having the issue before installing it.

Perhaps  i will try with a minimal number of plugins. For example i could remove cosmetic plugins (dynamix stats, system temp,...) 

 

And now i can't update or remove plugin within the gui.

It's been 2 weeks since i uninstalled several plugins and i have no more trouble except vnc for vm.

The list of plugins i uninstalled:

- dynamix active streams

- dynamix system info

- dynamix system stats

- dynamix system temp

- dynamix apcupsd

- fix common problems

Edited April 1, 20206 yr by caplam