Run the application:

git clone https://github.com/pssudo/kubernetes-fastapi.git
cd kubernetes-fastapi

Deploy using Helm

helm install products ./products_helm

Deploy using kubectl

kubectl apply -f products_k8s_manifest.yaml

Access the application

minikube service fastapi-svc
minikube addons enable metrics-server

Explaination & Steps to follow

• minikube service fastapi-svc will expose the service running inside the minikude to localhost
• port details will be displayed on the terminal console
• The REST endpoint will be accessible in the browser
• Go to ENDPOINT/docs to get the swagger format of the rest API
• Use the "Try it out" button to test the application
• Use POST and GET methods as per the swagger documentation
• Once new records are created using POST method run the below command to test the load

  for i in {1..1000}; do curl -s http://127.0.0.1:63538/products/1234 > /dev/null; done

:::note

• Please change the port number from 63538 to the one which is running in your local
• Change the product id from 1234 to the value you have passed as p_id :::

Auto Scaling

• While the above load test is running, you can observe the scaling policy (Up & Down are configured on CPU Utilization) using

kubectl describe hpa fastapi-hpa

• Alternatively you can enable the dashboard addon and observe the events

minikube addons enable dashboard
minikube dashboard