setup jupyterhub config

Dockerfile

@@ -13,7 +13,19 @@ RUN groupadd -r spark && useradd -r -g spark spark_user
 
 RUN apt-get update && apt-get install -y \
     # GCC required to resolve error during JupyterLab installation: psutil could not be installed from sources because gcc is not installed.
-    gcc curl git npm nodejs graphviz graphviz-dev libgdal-dev build-essential python3-dev\
+    gcc \
+    curl \
+    git \
+    wget \
+    vim \
+    npm \
+    nodejs \
+    graphviz \
+    graphviz-dev \
+    libgdal-dev \
+    build-essential \
+    python3-dev \
+    sudo \
     && rm -rf /var/lib/apt/lists/*
 
 ENV HADOOP_AWS_VER=3.3.4
@@ -42,22 +54,26 @@ RUN pipenv sync --system
 
 RUN chown -R spark_user:spark /opt/bitnami
 
-# Set up Jupyter Lab directories
+# Set up JupyterLab directories
 ENV JUPYTER_CONFIG_DIR=/.jupyter
 ENV JUPYTER_RUNTIME_DIR=/.jupyter/runtime
 ENV JUPYTER_DATA_DIR=/.jupyter/data
 RUN mkdir -p ${JUPYTER_CONFIG_DIR} ${JUPYTER_RUNTIME_DIR} ${JUPYTER_DATA_DIR}
 RUN chown -R spark_user:spark /.jupyter
 
-# Set up Jupyter Hub directories
+# Set up JupyterHub directories
 ENV JUPYTERHUB_CONFIG_DIR=/srv/jupyterhub
 RUN mkdir -p ${JUPYTERHUB_CONFIG_DIR}
 COPY ./src/notebook_utils/startup.py ${JUPYTERHUB_CONFIG_DIR}/startup.py
+COPY ./config/jupyterhub_config.py ${JUPYTERHUB_CONFIG_DIR}/jupyterhub_config.py
+COPY ./scripts/spawn_notebook.sh ${JUPYTERHUB_CONFIG_DIR}/spawn_notebook.sh
+RUN chmod +x ${JUPYTERHUB_CONFIG_DIR}/spawn_notebook.sh
 RUN chown -R spark_user:spark ${JUPYTERHUB_CONFIG_DIR}
 
 # Jupyter Hub user home directory
-RUN mkdir -p /jupyterhub/users_home
-RUN chown -R spark_user:spark /jupyterhub/users_home
+ENV JUPYTERHUB_USER_HOME=/jupyterhub/users_home
+RUN mkdir -p $JUPYTERHUB_USER_HOME
+RUN chown -R spark_user:spark $JUPYTERHUB_USER_HOME
 
 RUN npm install -g configurable-http-proxy
 
@@ -82,6 +98,10 @@ ENV CDM_SHARED_DIR=/cdm_shared_workspace
 RUN mkdir -p ${CDM_SHARED_DIR} && chmod -R 777 ${CDM_SHARED_DIR}
 RUN chown -R spark_user:spark $CDM_SHARED_DIR
 
+# Allow spark_user to use sudo without a password
+# TODO: use `sudospawner` in JupyterHub to avoid this (https://jupyterhub.readthedocs.io/en/stable/howto/configuration/config-sudo.html)
+RUN echo "spark_user ALL=(ALL) NOPASSWD: ALL" >> /etc/sudoers
+
 # Switch back to non-root user
 USER spark_user
 

config/jupyterhub_config.py

@@ -0,0 +1,26 @@
+"""
+This is the JupyterHub configuration file. It is used to configure the JupyterHub server.
+Refer to the JupyterHub documentation for more information:
+https://jupyterhub.readthedocs.io/en/latest/tutorial/getting-started/config-basics.html
+"""
+import os
+
+from jupyterhub_config.custom_spawner import VirtualEnvSpawner
+
+c = get_config()
+
+# Set the authenticator class
+# TODO: Change the authenticator class to a secure one (e.g. GitHubOAuthenticator)
+c.JupyterHub.authenticator_class = 'jupyterhub.auth.DummyAuthenticator'
+c.Authenticator.allowed_users = {'spark_user', 'test_user1', 'test_user2'}
+c.DummyAuthenticator.password = os.environ['JUPYTERHUB_ADMIN_PASSWORD']
+
+c.Authenticator.admin_users = {'spark_user'}
+
+c.JupyterHub.spawner_class = VirtualEnvSpawner
+
+# Set the JupyterHub IP address and port
+c.JupyterHub.ip = '0.0.0.0'
+c.JupyterHub.port = int(os.getenv('NOTEBOOK_PORT'))
+
+c.JupyterHub.log_level = 'DEBUG'

docker-compose.yaml

@@ -196,7 +196,7 @@ services:
         - JUPYTER_MODE=jupyterhub
         - YARN_RESOURCE_MANAGER_URL=http://yarn-resourcemanager:8032
         - SPARK_MASTER_URL=spark://spark-master:7077
-        - SPARK_DRIVER_HOST=dev-jupterhub
+        - SPARK_DRIVER_HOST=dev-jupyterhub
         - MINIO_URL=http://minio:9002
         - MINIO_ACCESS_KEY=minio-readwrite
         - MINIO_SECRET_KEY=minio123
@@ -207,6 +207,7 @@ services:
         - POSTGRES_DB=hive
         - POSTGRES_URL=postgres:5432
         - USAGE_MODE=dev
+        - JUPYTERHUB_ADMIN_PASSWORD=testpassword123
       volumes:
         - ./cdr/cdm/jupyter:/cdm_shared_workspace
         - ./cdr/cdm/jupyter/jupyterhub/users_home:/jupyterhub/users_home
@@ -237,6 +238,7 @@ services:
       - POSTGRES_PASSWORD=hivepassword
       - POSTGRES_DB=hive
       - POSTGRES_URL=postgres:5432
+      - JUPYTERHUB_ADMIN_PASSWORD=testpassword123
     volumes:
       - ./cdr/cdm/jupyter/jupyterhub/users_home:/jupyterhub/users_home
 

scripts/notebook_entrypoint.sh

@@ -28,7 +28,7 @@ if [ "$JUPYTER_MODE" = "jupyterlab" ]; then
 elif [ "$JUPYTER_MODE" = "jupyterhub" ]; then
   echo "starting jupyterhub"
 
-  echo "TO BE IMPLEMENTED"
+  jupyterhub -f "$JUPYTERHUB_CONFIG_DIR"/jupyterhub_config.py
 else
   echo "ERROR: JUPYTER_MODE is not set to jupyterlab or jupyterhub. Please set JUPYTER_MODE to either jupyterlab or jupyterhub."
   exit 1

scripts/spawn_notebook.sh

@@ -0,0 +1,9 @@
+#!/bin/bash
+
+USERNAME=${JUPYTERHUB_USER}
+
+echo "Starting Jupyter Notebook for user: $USERNAME"
+cd $JUPYTERHUB_USER_HOME/$USERNAME
+
+# Start the notebook server with current user
+exec jupyterhub-singleuser "$@"

src/jupyterhub_config/custom_spawner.py

@@ -0,0 +1,17 @@
+from jupyterhub.spawner import SimpleLocalProcessSpawner
+
+
+class VirtualEnvSpawner(SimpleLocalProcessSpawner):
+    """
+    A custom JupyterHub spawner that creates and manages a virtual environment
+    for each user, configuring their workspace based on their admin status.
+    """
+
+    def start(self):
+        """
+        Start the JupyterHub server for the user. This method ensures that the
+        user's directory and virtual environment are set up, configures environment
+        variables, and sets the notebook directory before starting the server.
+        """
+
+        return super().start()

test/src/jupyterhub_config/custom_spawner_test.py

@@ -0,0 +1 @@
+from jupyterhub_config.custom_spawner import *