How does the agent discover PHP-FPM pools?

The agent parses /etc/php-fpm.d/*.conf and discovers each pool's pm.status_path. It then polls the status page once a minute per pool. Multiple pools on the same host are tracked independently.

Does it detect max_children saturation?

Yes. php.fpm.max_children_reached and php.fpm.listen_queue are graphed, and an alert rule fires when either is non-zero for sustained intervals — the usual sign a pool is undersized.

Does it work with Nginx Unit or Apache?

PHP-FPM specifically. For Apache + mod_php, install a PHP-FPM sidecar and configure pools there; Amplify collects from the FPM pool regardless of upstream web server.

PHP-FPM Monitoring
Know which pool is saturated, right now.

Per-pool workers, listen queue, slow requests, and max-children-reached events. Auto-discovers every pool on the host. Alerts when any pool starts queueing.

Start free See demo

Per-pool metrics collected

php.fpm.active_processes

Workers currently handling requests.

php.fpm.idle_processes

Spare workers available for the next request.

php.fpm.listen_queue

Requests waiting for a free worker — non-zero means the pool is under-sized.

php.fpm.max_children_reached

How many times the pool hit pm.max_children. Any non-zero value is an outage risk.

php.fpm.slow_requests

Requests that exceeded request_slowlog_timeout.

php.fpm.accepted_conn

Cumulative requests served per pool.

Required pool config

# /etc/php-fpm.d/www.conf
pm.status_path = /status
ping.path      = /ping
request_slowlog_timeout = 5s
slowlog = /var/log/php-fpm/www-slow.log

The agent auto-discovers this and starts reporting within one minute of restart.

Right-size every PHP-FPM pool.

Start free View pricing

PHP-FPM MonitoringKnow which pool is saturated, right now.