Support a Workflow for Fixing Config Errors from this Role about ansible-role-nginx-config HOT 3 CLOSED

Hm that is some interesting behaviour. By all means, in my testing, if I run a playbook with an incorrect variable and then rerun it with a correct variable, the validation task will succeed on the second run. The Register NGINX Config task is run after a new template is deployed so it should not be failing if the config has been fixed. I'll try to do some further testing with the collection and see if I can replicate the issue.

from ansible-role-nginx-config.

So, after some more digging, I think I might now what the issue is. The check config handler only runs at the end of the playbook after the config is changed, so the issue is not with the nginx_config role.

However, you might be encountering an issue that'll happen if you run both the nginx and nginx_config roles within the same playbook if you set nginx_debug_output: true in the nginx role variables. The nginx role will try to verify the config and fail, since the nginx role would be running before the nginx_config role runs and fixes the config. If you set nginx_debug_output: false, you should be able to avoid running into the issue you are describing.

from ansible-role-nginx-config.

There's been no updates in a while so I'm going to go ahead and close this. I'm pretty sure the issue is related to what I described above. Do note that I would recommend only running the nginx role only once in your workflows, and then run the nginx_config role as many times as necessary.

from ansible-role-nginx-config.