Helm Charts to deploy Jupyter in Kubernetes Cluster for Machine Learning

• Prerequisite

  • kubernetes Setup kubectl
  • kubernetes cluster
  • Helm setup

• Installing the Docker image with jupyter/all-spark-notebook

  See here for more Docker images details.

  $ cd helm/jupyter/
  $ helm install jupyter . -f values.yaml
  

• Upgrade Jupyter

  $ helm upgrade jupyter . -f values.yaml
  

• Access jupyter with locally

  $ export POD_NAME=$(kubectl get pods --namespace default -l "app.kubernetes.io/name=jupyter,app.kubernetes.io/instance=jupyter" -o jsonpath="{.items[0].metadata.name}")
  
  $ export CONTAINER_PORT=$(kubectl get pod --namespace default $POD_NAME -o jsonpath="{.spec.containers[0].ports[0].containerPort}")
  
  $ kubectl --namespace default port-forward $POD_NAME 8080:$CONTAINER_PORT
  
  

• Find the existing token from Jupyter instance

  $ export POD_NAME=$(kubectl get pods --namespace default -l "app.kubernetes.io/name=jupyter,app.kubernetes.io/instance=jupyter" -o jsonpath="{.items[0].metadata.name}")
  
  $ kubectl exec -it $POD_NAME /bin/sh
  
  

• Print token from existing notebook after ssh into pod

  $ jupyter notebook list
  

• Open jupyter locally

  http://localhost:8080/

• Delete deployment

  $ helm delete jupyter