A repo to learn the basics of kubernetes
Minikube used to set up local kubernetes cluster.
- The docker images needed for the project needs to set up inside the minikuve env.
eval $(minikube docker-env) # unix shell
- Once the env has been changed to minikube docker, we can set up the images.
docker pull mysql:8.0
docker build . -t flask-sample:latest
- Run kubectl
kubectl apply -f secret.yml
kubectl apply -f volume.yml
kubectl apply -f deployment.yml
kubectl apply -f flask-deployment.yml
build.sh executes the above commands. (Before executing build.sh, apply the volume kubectl apply -f volume.yml)
teardown.sh is to delete the deployments and services.
In case of errors while setting up, logs and describe commands of kubectl can be used.
Logs:
kubectl logs <pod-name>
kubectl logs mysql-65bfd8d9bc-lvrrz
Describe:
kubectl describe <kind> <name>
kubectl describe pods mysql-65bfd8d9bc-lvrrz
kubectl exec -it <pod-name> -- <command>
kubectl exec -it mysql-65bfd8d9bc-lvrrz -- mysql -h localhost -u root -p
kubectl exec -it flaskapi-deployment-5995dc67d5-5m9mm -- /bin/bash
Scaling of the pods based on conditions resource limit (CPU, memory), http_request count, etc.
Here the hpa.yml has the autoscaling configuration, it has been configured to scale the flaskapi-development pod based on its memory consumption.
kubectl apply -f hpa.yml