iPython Kernel in Docker

Basic example of connecting to an iPython kernel running within a Docker container.

Prerequisites

• jq - needed to parse the connection file

10 Second Tutorial

Build the docker image. The docker container will launch a kernel running Python3 and listen on the ports defined in the connection file.

make build

Make Jupyter aware of our kernel

make install

Launch a Jupyter Console, which will create a connection file and a container, then attach to the kernel running inside of the container.

make run

Debugging and Testing

To start the kernel yourself and attach using the --existing flag, use:

make test

And then in a separate terminal, from the project directory, run

jupyter console --existing $(pwd)/sample_connection_file.json