bluedragonx/wsgi repository overview

⁠Docker WSGI Image

A Docker image for hosting WSGI apps.

⁠What's In It

The image uses Nginx and uWSGI to host WSGI apps. Both run under the app user. All logs are piped through syslog-ng so there's nothing to collect. And it exposes ports 80 and 443. The app should be a volume mounted at /app and contain a uwsgi.yaml and nginx.conf file to configure uWSGI and the nginx server section(s) for the app. There may be multiple nginx config files loaded as long as they each match the pattern nginx*.conf.

⁠Building

We build it this way:

git clone https://github.com/BlueDragonX/docker-wsgi.git
docker build --rm -t bluedragonx/wsgi docker-wsgi

⁠Running

The /app directory contains an nginx.conf to listen on port 80 and a uwsgi.yml file to host the Python demo WSGI app. So to run the sampe app you could:

docker run -P bluedragonx/wsgi

And to run your own app you might:

docker run -P -v /path/to/my/app:/app bluedragonx/wsgi

By default Nginx starts with 2 workers and so does uWSGI. You can change that with environment variables as well:

docker run -P -e NGINX_WORKERS=4 -e UWSGI_WORKERS=16 bluedragonx/wsgi

⁠Inheriting

The image supports backing in the app by an inheriting container. This is as easy as loading your own app contents with an ADD. However, you may want to load your own system-wide Nginx or uWSGI configs. To do this you should set the DISABLE_APP_CONFIG to a non-empty value in your Dockerfile. This will prevent the image from loading configuration from the /app directory.

⁠License

Copyright (c) 2014 Ryan Bourgeois. Licensed under BSD-Modified. See the LICENSE file for a copy of the license.