When managing PHP-FPM services, you might encounter this puzzling scenario:

$ service php5-fpm start
php5-fpm start/running, process 4516

$ service php5-fpm status  
php5-fpm stop/waiting

$ ps aux | grep php
www-data 3552 0.0 0.7 338108 14960 ? S 05:43 0:00 php-fpm: pool www
www-data 3553 0.0 1.3 338168 27156 ? S 05:43 0:00 php-fpm: pool www

The service appears to start successfully, yet status checks report "stop/waiting" while worker processes continue running. This indicates the master process may have terminated unexpectedly while leaving child processes orphaned.

Several factors could contribute to this behavior:

• Init script miscommunication: The init script fails to properly track the master process PID
• FastCGI process manager crash: The master process exits but workers remain
• PID file issues: Stale or incorrectly located PID files
• Upstart/SysV conflicts: Common on Ubuntu systems with both init systems

First, verify the actual process hierarchy:

$ pstree -p | grep php
|-php-fpm(1234)-+-php-fpm(2345)
               |-php-fpm(2346)
               -php-fpm(2347)

$ sudo netstat -tulpn | grep php
tcp 0 0 127.0.0.1:9000 0.0.0.0:* LISTEN 1234/php-fpm: maste

Check PID file location and contents:

$ cat /var/run/php5-fpm.pid
1234

$ ps -p 1234
PID TTY TIME CMD
1234 ? 00:00:01 php-fpm: master

For Ubuntu 14.04 systems, try these solutions:

# Clean up any existing processes
$ sudo killall php-fpm

# Reset the init script
$ sudo update-rc.d php5-fpm defaults
$ sudo update-rc.d php5-fpm enable

# Alternative restart sequence
$ sudo service php5-fpm stop
$ sudo rm -f /var/run/php5-fpm.pid
$ sudo service php5-fpm start

Modify /etc/php5/fpm/php-fpm.conf:

; Ensure proper PID file handling
pid = /var/run/php5-fpm.pid

; Add emergency restart settings
emergency_restart_threshold = 10
emergency_restart_interval = 1m
process_control_timeout = 10s

If upgraded to systemd:

$ sudo systemctl daemon-reload
$ sudo systemctl restart php5-fpm.service
$ sudo systemctl status php5-fpm.service

Create a watchdog script /usr/local/bin/php-fpm-monitor:

#!/bin/bash
MASTER_PID=$(cat /var/run/php5-fpm.pid 2>/dev/null)

if [ -z "$MASTER_PID" ] || ! ps -p $MASTER_PID > /dev/null; then
    echo "Master process not running - restarting"
    service php5-fpm restart
fi

exit 0

Add to cron:

*/5 * * * * root /usr/local/bin/php-fpm-monitor

---

html

When running service php5-fpm status after a restart, you might encounter contradictory output where the service reports stop/waiting while PHP pages continue to load normally. Process inspection reveals active worker processes:

ps -aux | grep php
www-data 3552  0.0  0.7 338108 14960 ?  S    05:43  0:00 php-fpm: pool www
www-data 3553  0.0  1.3 338168 27156 ?  S    05:43  0:00 php-fpm: pool www

The core issue typically stems from the init system (Upstart in Ubuntu 14.04) losing track of the master process PID. Check for the master process explicitly:

ps -ef | grep php-fpm | grep master
root     4516     1  0 05:43 ?        00:00:00 php-fpm: master process (/etc/php5/fpm/php-fpm.conf)

If this command returns empty, your master process has detached from the init system's control.

Examine these critical settings in /etc/php5/fpm/php-fpm.conf:

; Process manager settings
pm = dynamic
pm.max_children = 50
pm.start_servers = 5
pm.min_spare_servers = 2
pm.max_spare_servers = 8
pm.process_idle_timeout = 10s

Ubuntu 14.04's Upstart script (/etc/init/php5-fpm.conf) may need modification. Add these stanzas:

respawn
respawn limit unlimited
expect daemon

Create a systemd service override (for Ubuntu 16.04+) or modify Upstart behavior:

# For systemd systems
sudo systemctl edit php7.4-fpm
[Service]
PIDFile=/run/php/php7.4-fpm.pid
Restart=always

Enable verbose logging temporarily:

; /etc/php5/fpm/php-fpm.conf
log_level = debug

Then monitor logs in real-time:

tail -f /var/log/php5-fpm.log

For critical systems, consider process supervision:

# Using supervisor
[program:php-fpm]
command=/usr/sbin/php-fpm --nodaemonize
autostart=true
autorestart=true