Prerequisites

Install minikube on your local machine: ● https://minikube.sigs.k8s.io/docs/start/ ● Python3.8 ●

Minishift was not used for this implementation, due to its lack of support for Python 3.7, which is required for FastAPI. Minikube, however, is able to be used.

docker image build -t api-server .

This was ran on WSL 2 Ubuntu with Docker as the default driver.

To start the cluster, use

minikube start --driver=docker

or set Docker as your default driver with:

minikube config set driver docker

This was developed with Python3.8 and mkvirtualenv

Run the command

mkvirtualenv -p /usr/bin/python3.8

Also utilized the in-cluster Docker daemon (docker-env) for faster development/deployment.

To do this, run:

eval $(minikube docker-env)

To spin up the Kubernetes deployment and service:

kubectl apply -f src/api.yaml

To view resources, run this command to see the dashboard:

minikube dashboard

6400* minikube delete 6401* minikube start

To make queries against the cluster locally, run:

minikube tunnel

Then run

kubectl get svc api-server-svc

and use the external ip address when making HTTP requests via curl or a tool like Postman