Using Systemd to Make Sure Gunicorn Starts on Boot. As Ubuntu has switched to systemd as its service framework starting in 15.04 for all flavors, this is a migration of the Gunicorn upstart job that appears in "Test Driven Development with Django" (http://chimera.labs.oreilly.com/books/1234000000754/ch08.html#_using_upstart_to_make_sure_gunicorn…

gunicorn-SITENAME-staging.example.com.service

# Gunicorn Site systemd service file

[Unit]
Description=Gunicorn server for SITENAME-staging.example.com
After=network.target
After=syslog.target

Environment=sitedir=/Development/sites/SITENAME-staging.example.com
ExecStart=$(sitedir)/virtualenv/bin/gunicorn --chdir $(sitedir)/source workouts.wsgi:application --bind unix:/tmp/SITENAME-staging.example.com.socket
Restart=on-failure
RuntimeDirectory=gunicorn-stagingd
RuntimeDirectoryMode=755


#sudo systemctl start gunicorn-SITENAME-staging.example.com.service

Thank you so much. I just spent like 2 hours trying to figure this out. Hopefully it will work in Ubuntu 16.04...

The Restart=on-failure option doesn't work for me. I added it, then reloaded the config via systemctl daemon-reload, restarting my application and then killed its process, but it didn't restart on failure. Any idea?

@mkmoisen, if you did "kill -9 pid" it should restart, but if you did "killall gunicorn" will not restart because killall is sending a SIGTERM signal.

If set to on-failure, the service will be restarted when the process exits with a non-zero exit code, is terminated by a signal (including on core dump, but excluding the SIGHUP, SIGINT, SIGTERM or SIGPIPE signals). Because these four mentioned signals means clean exit.

New version described at https://simpleit.rocks/set-up-ubuntu-to-serve-a-django-website-step-by-step/.

Thank you. that --chdir saved me