/justudior

Files to generate/run the Docker image/container containing the R interpreter, Jupyter Notebook and the RStudio IDE.

Primary LanguageDockerfileMIT LicenseMIT

justudior

This repository holds the file to generate the Docker image (and run the Docker container) containing the following tools:

  • the R interpreter;
  • Jupyter Notebook and associated tools (JupyterLabs, Voilà, etc.);
  • the RStudio IDE (server version), running inside the container.

We also include the necessary packages and utilities to glue R and Jupyter together.

Notice the name of the repo (and the tag of the Docker image) is justudior, a contraction of "Jupyter", "Studio" and "R", recalling the tools contained in the image.

Prerequisites

The only prerequisites is to install Docker on a Linux machine.

One can use instructions provided on Docker website: https://docs.docker.com/engine/install/

Building the image and running the container

Simply run the script located at the root of the repository:

./buildrun.sh

The image takes some time to build.

After the image is built, the script automatically runs the container that starts RStudio and a tmux session with Jupyter Notebook, allowing you to interact with a shell while Jupyter is running.

Connecting to RStudio server

Once the container is started, RStudio starts listening on port 8787 (and the port is exposed to the host).

On a browser on your host, you can connect to http://127.0.0.1:8787/ to access the RStudio server. Use the Linux user/password credentials pre-configured in the Dockerfile (modify it according to your needs) to connect through the WebUI.

Connecting to Jupyter Notebook

As the container starts, the tmux session loads Jupyter Notebook and provides a URL with a token. You can use the URL with the localhost IP address (127.0.0.1) to connect to the WebUI of Jupyter Notebook, as the port 8888 is exposed to the host.