CODEWITHSUNDEEP
ansiblesupervisord
vabada
vabada

Reputation: 1814

Ansible supervisorctl, difference between state 'restarted' and 'present'

I am deploying a Django app to both production and staging environments (which happen to be in the same server). In order to do that, I run supervisord with two different process for both environments.

Thus in my ansible-playbook, I template my uwsgi config (according to the environment variables) and copy it to the /etc/supervisor/conf.d folder.

Say I want to deploy the first time to a new staging environment. Since there are not any present config files, I would need to supervisorctl reread and supervisorctl update.

As I read from the docs:

When state = present, the module will call supervisorctl reread then supervisorctl add if the program/group does not exist.

When state = restarted, the module will call supervisorctl update then call supervisorctl restart.

But does it mean I need to call both of them if I want to support both new staging instances and my current ones? Or is it enough to just use the state=restarted? Or am I doing something wrong?

Would this code be ok? Or is it duplicating some information?

    - name: Add uwsgi app (reread + add)
      become: yes
      supervisorctl: name={{uwsgi_app}} state=present

    - name: Start uwsgi app (update + restart)
      become: yes
      supervisorctl: name={{uwsgi_app}} state=restarted

Upvotes: 1

Views: 2227

Answers (1)

Aida Paul
Aida Paul

Reputation: 2722

You will need to either call both or set in your applications config files (the one you set in /etc/supervisor/conf.d) for the application to start automatically by adding a line:

autostart: true

While it is true by default, I prefer to be explicit.

Upvotes: 2

Related Questions