open data hub manual install examples (ODH circa 0.5.1)

install-odh.md

oc apply -f deploy/crds/opendatahub_v1alpha1_opendatahub_crd.yaml
oc apply -f deploy/service_account.yaml
oc apply -f deploy/role.yaml
oc apply -f deploy/role_binding.yaml
oc apply -f deploy/operator.yaml
# customize odh manifest
oc apply -f my_environment_cr.yaml

my_environment_cr.yaml

apiVersion: opendatahub.io/v1alpha1
kind: OpenDataHub
metadata:
  name: example-opendatahub
spec:
  # JupyterHub deployment developed by Graham Dumpleton - https://github.com/aicoe/jupyterhub-ocp-oauth
  aicoe-jupyterhub:
    # Deploy the ODH aicoe-jupyterhub role if True
    odh_deploy: true
    notebook_cpu: 512Mi
    notebook_memory: 512Mi
    deploy_all_notebooks: False
    registry: 'quay.io'
    repository: 'odh-jupyterhub'
    storage_class: ''
    db_memory: 256Mi
    jupyterhub_memory: 256Mi
    notebook_image: 's2i-minimal-notebook:3.6'
    s3_endpoint_url: ''
    #GPU mode defines a method used to spawn GPU enabled containers. Allow values are: None/empty (default), selinux, privileged
    gpu_mode: ''

    # Name of the configmap that will be used when spawning a notebook for the single user
    spark_configmap_template: 'jupyterhub-spark-operator-configmap'
    # PYSPARK args to use in the notebook pod
    # These submit args should be customized for the values passed for spark_memory and spark_cpu. You'll need to account for the available memory on the spark work nodes
    spark_pyspark_submit_args: "--conf spark.cores.max=1 --conf spark.executor.instances=1 --conf spark.executor.memory=1G --conf spark.executor.cores=1 --conf spark.driver.memory=1G --packages com.amazonaws:aws-java-sdk:1.7.4,org.apache.hadoop:hadoop-aws:2.7.3 pyspark-shell"
    spark_pyspark_driver_python: "jupyter"
    spark_pyspark_driver_python_opts: "notebook"
    spark_home: "/opt/app-root/lib/python3.6/site-packages/pyspark/"
    spark_pythonpath: "$PYTHONPATH:/opt/app-root/lib/python3.6/site-packages/:/opt/app-root/lib/python3.6/site-packages/pyspark/python/:/opt/app-root/lib/python3.6/site-packages/pyspark/python/lib/py4j-0.8.2.1-src.zip"

    # Number of master and worker nodes for the spark cluster
    spark_worker_nodes: 1
    spark_master_nodes: 1

    # Amount of cpu & memory to allocate to the each node in the cluster.
    # This value will be applied to all worker and master nodes
    spark_memory: 1Gi
    spark_cpu: 1
    # Spark image to use in the cluster
    spark_image: "quay.io/opendatahub/spark-cluster-image:spark22python36"

  # Spark operator developed by radanalyticsio - https://github.com/radanalyticsio/spark-operator
  spark-operator:
    # Deploy the ODH spark-operator role if True
    odh_deploy: false
    master_node_count: 0
    master_memory: 1Gi
    master_cpu: 1
    worker_node_count: 0
    worker_memory: 1Gi
    worker_cpu: 1

  # Seldon Delployment
  seldon:
    odh_deploy: false

  # JupyterHub deployment developed by Graham Dumpleton - https://github.com/jupyter-on-openshift/jupyterhub-quickstart
  jupyter-on-openshift:
    # Deploy the ODH jupyter-on-openshift role if True
    odh_deploy: false
    notebook_memory: 2Gi
    # Add these whitelisted environment variables from JupyterHub to the user's notebook pod
    jupyterhub_config: |
      c.KubeSpawner.env_keep = ['S3_ENDPOINT_URL', 'S3_ACCESS_KEY', 'S3_SECRET_KEY']
    # Environment variables that will be set on the JupyterHub pod
    extra_env_vars:
      S3_ENDPOINT_URL: "http://s3.foo.com:8000"
      S3_ACCESS_KEY: "YOURS3ACCESSKEYHERE"
      S3_SECRET_KEY: "this1is2just3gibberish"

  kafka:
    odh_deploy: false
    kafka_cluster_name: odh-message-bus
    kafka_broker_replicas: 3
    kafka_zookeeper_replicas: 3

  # Deployment of Prometheus and Grafana for Monitoring of ODH
  monitoring:
    odh_deploy: false

  # Deployment of Two Sigma's BeakerX Jupyter notebook
  beakerx:
    odh_deploy: false

  # Deployment of AI Library models as SeldonDeployments.  Can only be done if Seldon is also deployed.
  ai-library:
    odh_deploy: false
 
  # Deployment of Argo 
  argo:
    odh_deploy: false

single-user-config-map.yaml

apiVersion: v1
kind: ConfigMap
metadata:
  name: eje-profiles
  labels:
    jupyterhub: singleuser-profiles
data:
  jupyterhub-singleuser-profiles.yaml: |
    profiles:
      - name: gh-repo-env-var
        env:
          JUPYTER_PRELOAD_REPOS: 'https://github.com/erikerlandson/anomaly-detection-workshop'
    sizes:
      - name: 2 gig single core
        resources:
          mem_limit: 2Gi
          cpu_limit: 1
      - name: 3 gig single core
        resources:
          mem_limit: 3Gi
          cpu_limit: 1
      - name: 4 gig single core
        resources:
          mem_limit: 4Gi
          cpu_limit: 1
      - name: 6 gig single core
        resources:
          mem_limit: 6Gi
          cpu_limit: 1