add notebook container

.gitignore

@@ -0,0 +1,2 @@
+
+.idea

Dockerfile

@@ -28,7 +28,7 @@ ENV PYTHON_VER=python3.11
 RUN apt update && \
 	apt-get install -y software-properties-common && \
 	add-apt-repository ppa:deadsnakes/ppa && \
-	apt install -y $PYTHON_VER python3-pip && \
+	apt install -y $PYTHON_VER python3-pip $PYTHON_VER-dev && \
 	apt install -y r-base r-base-dev && \
 	rm -rf /var/lib/apt/lists/*
 
@@ -46,6 +46,10 @@ COPY --from=build /opt/$SPARK_VER/ /opt/spark/
 # this doesn't seem to actually work
 RUN echo "spark.pyspark.python /usr/bin/$PYTHON_VER" > /opt/spark/conf/spark-defaults.conf
 
+RUN /usr/bin/$PYTHON_VER -m pip install --upgrade pip && \
+    /usr/bin/$PYTHON_VER -m pip install --force-reinstall psutil==5.9.8 && \
+    /usr/bin/$PYTHON_VER -m pip install jupyterlab==4.2.0 pyspark==3.5.1 findspark==2.0.1
+
 COPY entrypoint.sh /opt/
 RUN chmod a+x /opt/entrypoint.sh
 

README.md

@@ -5,6 +5,7 @@ Rancher.
 
 ## Submit example
 
+### Submitting a job in Rancher
 Open a shell in a container in rancher with spark installed in `/opt/spark` (as of this writing,
 the `ci-core-cdm-prototype-spark-bash` container is appropriate and has environment variables set up
 for the necessary host and port configurations).
@@ -24,6 +25,58 @@ root@f058c872158d:/opt/spark# bin/spark-submit --master spark://10.58.1.104:7077
 Pi is roughly 3.144080
 ```
 
+### Submitting a Spark Job via Jupyter Notebook
+
+After launching the Jupyter Notebook, establish a Spark context or session with the master set to the environment variable 
+SPARK_MASTER_URL and proceed to submit your job. Once the job is submitted, you can monitor 
+the job status and logs in the Spark UI.
+
+Sample code to calculate Pi using `SparkContext`:
+```python
+import findspark
+findspark.init()
+from pyspark import SparkConf, SparkContext
+import random
+import os
+
+spark_master_url = os.environ['SPARK_MASTER_URL']
+
+conf = SparkConf().setMaster(spark_master_url).setAppName("Pi")
+sc = SparkContext(conf=conf)
+
+num_samples = 100000000
+def inside(p):     
+  x, y = random.random(), random.random()
+  return x*x + y*y < 1
+count = sc.parallelize(range(0, num_samples)).filter(inside).count()
+pi = 4 * count / num_samples
+print(pi)
+sc.stop()
+```
+
+Sample code to create a DataFrame and display the contents with `SparkSession`:
+```python
+import findspark
+findspark.init()
+from pyspark.sql import SparkSession
+import os
+
+spark_master_url = os.environ['SPARK_MASTER_URL']
+
+spark = SparkSession.builder \
+    .master(spark_master_url) \
+    .appName("TestSparkJob") \
+    .getOrCreate()
+
+data = [("Alice", 34), ("Bob", 45), ("Charlie", 56)]
+df = spark.createDataFrame(data, ["Name", "Age"])
+
+df.show()
+
+spark.stop()
+```
+
+
 ## Notes
 
 * When we switch to Rancher 2 should probably switch from the standalone scheduler to the k8s

docker-compose.yaml

@@ -15,7 +15,7 @@ services:
       - MODE=master
       - PORT=7077
       - WEBPORT=8080
-      
+
   spark-worker1:
     build:
       context: .
@@ -47,3 +47,17 @@ services:
       - CORES=2
       - MEM=1G
       - SPARK_MASTER_URL=spark://spark-master:7077
+
+  notebook:
+    build:
+      context: .
+      dockerfile: Dockerfile
+    ports:
+      - "4040:4040"
+      - "4041:4041"
+    depends_on:
+      - spark-master
+    environment:
+      - MODE=notebook
+      - PORT=4041
+      - SPARK_MASTER_URL=spark://spark-master:7077

entrypoint.sh

@@ -27,6 +27,18 @@ elif [ "$MODE" == "worker" ] ; then
     /opt/spark/sbin/start-worker.sh --cores "$CORES" --memory "$MEM" "$SPARK_MASTER_URL"
 elif [ "$MODE" == "bash" ] ; then
     bash
+elif [ "$MODE" == "notebook" ] ; then
+    echo "starting jupyter notebook"
+    WORKSPACE_DIR="/cdm_shared_workspace"
+    mkdir -p "$WORKSPACE_DIR"
+    cd "$WORKSPACE_DIR"
+    jupyter lab --ip=0.0.0.0 \
+                --port=$PORT \
+                --no-browser \
+                --allow-root \
+                --notebook-dir="$WORKSPACE_DIR" \
+                --ServerApp.token='' \
+                --ServerApp.password=''
 else
     echo "Unrecognized MODE env var: [$MODE]"
 fi