uWSGI docker image

Usage

• You need a requirements.txt file at the root of your project folder - even if empty. You can also set an alternate path using the REQUIREMENTS_FILE environment variable.
• Mount your project folder as /code volume of this image.
• To specify where your WSGI application module lives, you should use the WSGI_MODULE environment variable. It defaults to app:app, which will work if you have a WSGI callable named app inside a app.py file in your /code folder.
• Finally, by default, uWSGI listens on a UNIX socket named wsgi.sock that will be placed in a newly-created run/ directory in your project folder (mounted to /run inside the image). You should point your production webserver to this socket in order to serve your application. You can also affect this behavior using the ADDRESS environment variable - set it to any other UNIX socket path (default /run/wsgi.sock) or even to a TCP host/port (eg. 127.0.0.1:5000).

Basically:

docker run \
    -v /path/to/your/code:/code \
    -e WSGI_MODULE=yourmodule:yourapp \
    -e REQUIREMENTS_FILE=/path/to/your/code/requirements.txt \
    cheaterman/uwsgi

See also: Docker Hub repository