Zeppelin for Spark + Hadoop (SIGNAL 2019)

Signal Docker Container (spark,hdfs,etc)

Docker

1. Install Docker Desktop (https://www.docker.com/products/docker-desktop) Additional Docker Resources:

• https://docs.docker.com/get-started/
• https://hub.docker.com/

Docker Runtime Recommendations

1. 2 or more cpu cores.
2. 8gb/ram or higher.

Getting Started

1. Build the Docker Container ./build.sh
2. Start the Environment ./start.sh
3. If all goes well then http://localhost:8080/ should have Zeppelin up and running. (park Primer: http://localhost:8080/#/notebook/2EHFGMBH5)

Build

The Dockerfile is used to build Zeppelin which has dependency on the rest of other docker containers Hadoop and Spark

Build the Zeppelin Docker Container

Note: Building the initial docker container can take anywhere from 5-20 minutes. Please be patient!

./build.sh

Start up the Environment

To start the entire image with spark + hadoop + zeppelin:

./start.sh

Stopping the Environment

./stop.sh

Docker Info

List all running docker processes

docker ps

Log into a Docker Container VM

You just need to launch a bash session.

docker exec -it CONTAINER_NAME command

docker exec -it zeppelin /bin/bash