High CPU Use After Maintenance and Restart
I have php-fpm processes that are consuming 100% of the available CPU since my Linode was restarted last night after planned maintenance. Everything was working fine before the server was shut down and restarted. This is what I'm seeing in my nginx error log:
2019/10/17 10:33:17 [error] 12792#12792: *976 connect() to unix:/run/php/php7.2-fpm.sock failed (11: Resource temporarily unavailable) while connecting to upstream, client: X.X.X.X, server: mysite.org, request: "GET /forum/index.php?PHPSESSID=hflc1dr0nsfvjdvp60alinagc0&type=rss;action=.xml HTTP/1.1", upstream: "fastcgi://unix:/run/php/php7.2-fpm.sock:", host: "www.mysite.org"
htop shows a number of php-fpm processes as the CP use culprits. Here's what's in the PHP fpm log:
[17-Oct-2019 10:00:07] NOTICE: fpm is running, pid 12276
[17-Oct-2019 10:00:07] NOTICE: ready to handle connections
[17-Oct-2019 10:00:07] NOTICE: systemd monitor interval set to 10000ms
[17-Oct-2019 10:05:06] WARNING: [pool 428] server reached pm.max_children setting (40), consider raising it
I haven't tried changing pm.max_children since the current value was working just fine until now. Could this be a DDoS attack?
