big-data-europe/docker-hadoop ? reverse-engineered prompt

Build me a Docker based Hadoop demo cluster that I can start with one command. I want the setup to spin up the main Hadoop services, let me open the web dashboards in my browser, and include a simple example job I can run to prove everything is working, like a word count demo.

Make it easy to change Hadoop settings through environment variables or a single env file, so I can tweak things without editing a bunch of config files. If it makes sense, include support for running it with normal Docker Compose and also in swarm mode. I’d like the containers to come up in the right order and wait for each other so the cluster starts reliably.

If you need to check current Hadoop or Docker docs to get the ports and startup details right, please do that.