Flask-Gunicorn
Flask-Gunicorn lets you simply run your Flask Application using the gunicorn application server easily from the command line.
Unfortunately Gunicorn doesn't work on Windows machines, so you may want to look into alternative solutions for serving your application.
Installation
Install the extension with pip:
$ pip install flask-gunicornUsage
Once installed, Flask-Gunicorn automatically overides the run
command in the flask command line tool.
To use, simply let flask know where your application is by setting
an environment variable:
export FLASK_APP=myapp.pyThen run the server:
flask runBy default Flask-Gunicorn will make a sensible guess at how many
workers to allocate to the application server based the number of
CPU cores on your machine, but this can be specified using the
--workers x argument or WORKERS environment variable.
the flask run command also takes serveral optional arguments to
help you customize your gunicorn server.
| Argument | Description |
|---|---|
--workers |
How many workers should Gunicorn spawn |
--worker_class |
Should Gunicorn use a specific class of worker? E.g. gevent |
--debugger |
Run the server with the interactive debugger |
--no-debugger |
Turn off the debugger mode |
--host |
What address should the server bind to (e.g. 127.0.0.1) |
--port |
What port should be used (e.g. 5000) |
--reload |
Turn on the reloader (gunicorn will notice changes to code and restart if noticed) This is on by default in debug mode. |
--noreload |
Turn off the reloader |
Contributing
Contributions will be gleefully received! Check out the current issues, or feel free to crate new issues for any problems you've encountered and we'll push this little project forwards.
Thanks to Flask-Common and the Gunicorn Docs on Custom Applications for hints on getting this all working.
