unicorn_api

This is our API Proxy code for use by trusted Unicorn Rentals partners, and less trusted DevOps teams. Its purpose is to proxy API calls to our backend. Every 4 minutes it uses a different auth secret to auth to the backend, and to validate the result with caller. Currently the auth is updated by a gitpush every 4 minutes.

Architecture

Below was/is the proposed architecture for deploying this application. We have partially fulfilled the design in our MVP deployment.

Getting Started

If you want to locally test etc:

• Have python 2.7

This will deploy the python dependencies, and run a local web server.

pip install -r requirements.txt
python app.py

Tests

If deploying with CICD tooling, run tests via CodeBuild or any other CI tool.

python tests_app.py -v

Or you can manually test with cURL, httpie or a browser.

Deployment

Configure AWS Parameter Store BACKEND_API value.

Then deploy how suits you, but included is a Dockerfile if you want to use Docker, and an appspec.yml for CodeDeploy.

docker build -t unicorn_api .
docker run unicorn_api -p 5000:5000

Optionally have AWS xray deployed (executable not in this repo):

curl https://s3.dualstack.us-east-1.amazonaws.com/aws-xray-assets.us-east-1/xray-daemon/aws-xray-daemon-linux-2.x.zip -o xray.zip
unzip xray.zip
chmod +x xray 
nohup ./xray &

And have a default AWS region set, either via environment variable, or aws configure.

Built With

Python
Flask
flask_restful
Requests
watchtower