nginx 504 errors

[BioHazardous]

Not sure what happened, but the UI just stopped responding. I can still FTP to the server, but Plex wasn't working, so was trying to restart the docker, but can't connect to the UI.

 

Checked the nginx logs and there are a few errors from today, mostly similar to this:
 

2018/04/09 15:33:37 [error] 3807#3807: *15222439 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 192.168.3.3, server: , request: "GET /Main HTTP/1.1", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "192.168.3.107", referrer: "http://192.168.3.107/Settings/Scheduler"
 

2018/04/09 15:35:38 [error] 3807#3807: *15222439 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 192.168.3.3, server: , request: "GET /favicon.ico HTTP/1.1", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "192.168.3.107", referrer: "http://192.168.3.107/Main"
 

Everything had been working fine for months. Haven't done anything new other than add media to Plex, and even then it was fine. I assume a restart will fix this, but still wanted to post it anyway.

[BioHazardous]

Went to syslog, and found something a little different from earlier today:
Apr  9 15:00:17 Tower php-fpm[3781]: [WARNING] [pool www] server reached max_children setting (20), consider raising it

 

Any other logs I should check or things I should look for?  I just don't want to try restarting it then lose access to all of this fun stuff if it doesn't come back up.

[BioHazardous]

Everything ended up freezing up on me. My Putty connection to the server was non-responsive after a few minutes, then FTP wouldn't let me connect anymore. So had to restart it manually. It's back up and running.. *Shrugs*  Just wish I knew why this issue occurred in the first place.