While the worker server has more routine issues and higher load to manage, the web front-end recently went down.

Leaving myself or others a paper trail:

Steps to fix on the webclient container:

1. Restart PHP-FPM: /etc/init.d/php7.0-fpm restart

2. Restart nginx: /etc/init.d/nginx restart

Comments

nerdstein created an issue. See original summary.

nerdstein credited daggerhart.

nerdstein’s picture

nerdstein
he/him
Primary language English
Location United States
commented

PHP-FPM crashed again, signifying a larger issue.

Error logs showed that it was continually hitting the threshold for pm.max_children.

The following code was added to vi /etc/php/7.0/fpm/pool.d/www.conf to resolve:

pm = dynamic
pm.max_children = 100
pm.start_servers = 20
pm.min_spare_servers = 10
pm.max_spare_servers = 20
pm.max_requests = 1000

PHP-FPM restarted. Thanks to daggerhart for the configs.

Status: Fixed » Closed (fixed)

Automatically closed - issue fixed for 2 weeks with no activity.