Docker cluster with Apache Ranger and a minimal Hadoop ecosystem to perform some basic experiments. This branch is forked from kadensungbincho/apache-ranger-docker-poc and I'm only reorganized and updated the docker-compose file, trying to make the start process in one step. This cluster provides:
- Apache Ranger (with ElasticSearch to store and retrieve audit logs);
- HDFS;
- Yarn;
- Hive.
All credits go to @kadensungbincho. It's highly recommended read his article on Medium platform.
- Run the command:
docker-composes up
- The process will starts (the first execution requires some time since it needs to download all images);
- Waits until the Ranger Web UI is ready in
localhost:6080
(when is started, a log in ranger-admin component likeStarting Apache Ranger Admin Service
will appears)
- Starting all components can take some time, and in my machine, all processes used approximately 4GB RAM.