supervisord_vs_monit

supervisord_vs_monit.txt

supervisord [supervisord.org]
=============================

* Set up jobs via a config file (.ini style)

	Example:
	--------
	[program:forever]
	command=/usr/bin/python /root/supervisoreval/bin/forever.py &
	numprocs=1
	autorestart=unexpected
	exitcodes=0,2
	autostart=false

Fill list of options here: https://supervisord.org/configuration.html#program-x-section-settings

* Runs jobs as child processes of `supervisord` process
	* Pro: can restart processes almost instantaneously
	* Con: if supervisord process dies, so do its processes, or they become orphaned processes

* Requires system Python of at least 2.6
* Documentation notes it won't run under Python 3, though this might not really be a concern for now -- really depends on
  when RHEL moves to a Python 3 system version (probably not for a while)
* Installable via yum or pip
* Can view job statuses on command line using `supervisorctl`  
* Has an optional web interface that allows an admin to start/stop/restart supervised jobs
* Can be installed as root or as a less privileged user

monit [mmonit.com]
=================

* Set up jobs via a config file (each job requires a pid file)

	Example:
	--------
	check process forever with pidfile /var/run/forever.pid
		start program = "/etc/init.d/forever start"
		stop program = "/etc/init.d/forever stop"

Supports additional logic to determine what constitutes failure. Here's a more sophisticated example:

	check process apache with pidfile /usr/local/apache/logs/httpd.pid
    	start program = "/etc/init.d/httpd start" with timeout 60 seconds
    	stop program  = "/etc/init.d/httpd stop"
    	if cpu > 60% for 2 cycles then alert
    	if cpu > 80% for 5 cycles then restart
    	if totalmem > 200.0 MB for 5 cycles then restart
    	if children > 250 then restart
    	if loadavg(5min) greater than 10 for 8 cycles then stop
    	if failed host www.tildeslash.com port 80 protocol http
       		and request "/somefile.html"
       		then restart
    	if failed port 443 type tcpssl protocol http
       		with timeout 15 seconds
       		then restart

See https://mmonit.com/monit/documentation/monit.html#EXISTENCE-TESTING for more details.

* Jobs are run on their own, not as child processes of monit. 
* Polls jobs every two minutes by default so there will be some delay in services being brought back to life post-failure.
* Installable via yum
* Has a web interface that shows monitored jobs and allows start/stop/reload of services via the web (it's nice!)
* Should be installed by root