Skip to content

Containerize a Python application

You are part of a remote and distributed team who is developing a Python web application. Lately, your peers have been complaining about Python version differences causing the app to behave differently or not work at all. The problems are not consistent across all team members and you see "Works on my machine!" in Slack a lot more now.

The first "production" launch of the app is a low traffic private beta that is scheduled for next week. To support the launch, you need to build a production ready setup. Since the application is still in heavy development you don't want to make the version differences worse by adding yet another environment. You therefore decide to use Docker to build a setup that can be used for development and production.

The Task

Using Docker and docker-compose, containerize a Python application fronted by an nginx reverse proxy and show us your understanding about how to deploy an application securely while still allowing for low friction development.

Specifically we will be looking for the following in your submission:

  • Secure and performant settings in your nginx configuration, including SSL/TLS specifics
  • Usage of the principle of least privilege/surprise
  • Docker best practices including security, configuration, and image size

Part One

Complete the provided nginx/nginx.conf by writing a server directive(s) that proxies to the upstream application.

Requirements:

  • Nginx should accept requests on ports 80 and 443
  • All HTTP requests should permanently redirect to their HTTPS equivalent
  • Use the provided SSL keypair found in nginx/files/localhost.crt and nginx/files/localhost.key for your SSL configuration
  • The SSL configuration should use modern and secure protocols and ciphers
  • Nginx should proxy requests to the application using an upstream directive
  • Pass headers X-Forwarded-For, X-Real-IP, and X-Forwarded-Proto to the upstream application with appropriate values

Part Two

Complete app/Dockerfile, nginx/Dockerfile and docker-compose.yml to produce a production ready image that can be also be used for development with Compose.

Requirements:

  • The app found in ./app/src is securely built, installed, and configured into a container.
  • When run by itself, the app container should start the app, serving traffic with a production quality server, on port 8000 without any extra configuration.
  • Running docker-compose up should:

    • Start the app container in development mode listening on port 8000
    • Allow local edits of the app source to be reflected in the running app container without restart (eg: hot code reload)
    • Start an nginx container configured with the files from Part One
    • The app service should not reachable directly from the host and can only be accessed through the nginx service.

Tying it all together

After running docker-compose up a command such as curl -k https://localhost/ should return output similar to:

It's easier to ask forgiveness than it is to get permission.
X-Forwarded-For: 172.20.0.1
X-Real-IP: 172.20.0.1
X-Forwarded-Proto: https

You may also run the included validation tool to further test your work. This script will stop and remove running and built containers, so please be careful. You can run the tool like so:

# Stop, remove, rebuild containers, and run tests (see -h for help)
./validate.sh

Tips & Guidance

Tips

  • Windows 10 Home users will need to use Docker toolbox or will need to do the exercise in a Linux VM / cloud instance since Docker Desktop for Windows requires Hyper-V, a Pro only feature.
  • The Python app is a Flask app and can be configured with environment variables
  • You can use gunicorn for the production server service. Since we're not testing a full production config in this exercise, you don't need to tune it. Simply show that the app image boots and serves traffic using a real server. If you are more familiar with a different standalone WSGI server, feel free to substitute it and discuss the change in your COMMENTS.md file.

Dos

  • Do add notes on running your solution and why you choose your particular solution in a COMMENTS.md file. Remember, you are working with a remote team. Written communication is important!
  • Do feel free to offer suggestions or feedback on this exercise

Do Nots

  • Do not worry about data persistence, scaling, or OCSP stapling
  • Do not alter validate.sh or the SSL key pair in nginx/files.
  • Do not modify the app and nginx service names in docker-compose.yml.
  • Do not create any additional Dockerfile or docker-compose files.

Included files

Here are the files we'll give to you to get you started. Download them when you are ready to begin.

Ready to submit?

Don't include anything in your files that could identify you. We assign submissions a random number when they are received so our team does not know whose homework they are evaluating. Multiple team members will review your submission before a decision is made.

Submit homework