Kubernetes examples

Lab Minikube (based on official tutorial on https://kubernetes.io)

1. Launch GitHub Codespace
2. Run minikube start to start cluster
3. Run minikube dashboard --url to view dashboard in a new terminal
4. Hover over link and "follow link"
5. Create a deployment: kubectl create deployment hello-node --image=registry.k8s.io/e2e-test-images/agnhost:2.39 -- /agnhost netexec --http-port=8080
6. View deployment: kubectl get deployments
7. View pods: kubectl get pods
8. Create service and expose it: kubectl expose deployment hello-node --type=LoadBalancer --port=8080
9. View services: kubectl get services
10. Curl the url shown, for example: curl http://192.168.49.2:31839 or change to your URL.
11. Cleanup

kubectl delete service hello-node
kubectl delete deployment hello-node
minikube stop

Task 2: Deploy with Kubernetes FastAPI app

1. Push container to DockerHub (Optional): i.e. docker build -t <hub-user>/<repo-name>[:<tag>] and docker push <hub-user>/<repo-name>:<tag> Example of a pushed FastAPI container here: https://hub.docker.com/repository/docker/noahgift/fastapi-kube
2. minikube start
3. minikube dashboard --url
4. Hover over link and "follow link"
5. Create a deployment: kubectl create deployment hello-fastapi --image=registry.hub.docker.com/noahgift/fastapi-kube
6. View deployment: kubectl get deployments
7. Create service and expose it: kubectl expose deployment hello-fastapi --type=LoadBalancer --port=8080
8. View services: kubectl get service hello-fastapi
9. minikube service hello-fastapi --url
10. Curl web service: i.e. curl http://192.168.49.2:31224
11. Cleanup
12. Cleanup

kubectl delete service hello-fastapi
kubectl delete deployment hello-fastapi
minikube stop

Notes below

fastapi-from-zero

A repository to demonstrate FastAPI

/docs get to swagger

Docker

docker build . docker image ls #find image docker run -p 127.0.0.1:8080:8080 93fa55efa692

Cloud9 + ECR + App Runner

• Clone repo into Cloud9 (pick a machine with decent size CPU and RAM if possible, but students should use micro)
• Add ssh keys to GitHub
• resize to bigger disk
• Create virtualenv and add to bashrc and source python3 -m venv ~/.venv && echo 'source ~/.venv/bin/activate' >> ~/.bashrc && source ~/.bashrc
• cd into checkout and run make install
• Preview running FastAPI app after running: python main.py

• Create ECR repository by right-click in Cloud9

• Navigate to ECR repo created or whatever you named it and follow "push" instructions

  

• Navigate to AWS App Runner and Setup Continuous Delivery using ECR

  

• Setup AWS Code Build to push container after each build (which triggers auto-deploy)

  

See following buildspec.yml and Makefile

References

• FastAPI Docker docs
• Hello minikube
• Reference