This tutorial is about the Medium article available here: https://medium.com/p/c0efc5ed4ae4
To run you will need a Kubernetes engine running (I'm using Minikube). To launch the entire application you will need to:
- Compile the docker file in app/Dockerfile. Since I'm running minikube I launch it with the contained docker engine
docker build . -t "tutorialk8sjob" - Deploy the service account
kubectl apply -f service_account/jobrobot_sa.yaml - Deploy the job role
kubectl apply -f service_account/jobrobot_role.yaml - Deploy the role binding
kubectl apply -f service_account/jobrobot_rolebinding.yaml - Deploy your pod
kubectl apply -f app.yml
To watch your cluster instantiate pods you want to run kubectl get pods. With this you will be able to see the evolving tasks being done and been deleted.
You need to install the package 'kubernetes' on your Python environment. You can do that by running 'pip install kubernetes'.