Conio nano

This repository is a fork of the big-data-europe/docker-hadoop but configured to run and test Conio shim easily.

About quick start, configurations and other details visit docker-hadoop's README.md.

Changes between

Conio nano:

• Exposed ResourceManager, NodeManager and JobHistory Server UI
• Uses the Linux Container Executor (with configurable container-executor.cfg) for the NodeManager
• Mounts the Docker socket to the NodeManager's container thus creating a dind-variant container
• Is configured to schedule Docker containers
• Is added a Conio make target for this project which builds a container where the Conio shim client can run

Get started

1. Copy the yaml file and the fat conio shim jar (with dependencies) to the /conio folder.

2. Type the following command: make conio

3. Start the dockerized Hadoop service according to the docker-hadoop docs.

4. Start the Conio shim client in a Docker container with some options like:

docker run -it --env-file hadoop.env --network docker-hadoop_default -v $(PWD)/conio:/conio conio/base:master -- sudo -u conio java -jar /conio/conio-1.0-SNAPSHOT-jar-with-dependencies.jar -yaml /conio/pod.yaml -queue default