ragecryx / k8s-op-djs

K8s Operator that is reading another service API in the cluster and dispatches jobs based on the entries returned.

k8s-op-djs

How to run

Ensure you have kubectl and minikube installed. It's configured to run with docker driver.

1. make minikube-start to create and start the cluster.
2. kubectl apply -f ./djs/crd.yaml to register the custom resource.
3. make docker-image to build docker image.
4. minikube image load djs-demo:1.0 to load the docker image from your local docker into minikube.
5. In mocker directory run make service-on-minikube. This will build & load docker image of mock service and start it in the cluster.
6. kubectl apply -f ./example-rbac.yaml so Operator has access to the k8s API.
7. kubectl create -f ./example-djs.yaml to create a k8s object of DynamicJobScheduler kind.
8. kubectl get jobs to observe jobs being created, their naming will be djs-job-<person name>-<uuidv4>.
9. Use kubectl logs -f <the name of the djs-operator pod> to observe what the Operator is processing.

To run step 5 you need to have a Python VirtualEnv created and packages installed with pip install -r requirements.txt.