Playing with Docker and Jekyll

Almost 2 years since the last post so let’s be proactive and have a look at Docker with Jekyll!

While giving this blog a fresh new look (with a modified version of the jekyll template Strata), I used Docker for the development phase. I have been using docker at work within a vagrant machine, running 5 to 6 docker containers but I never really played with it so I decided to install jekyll and serve this blog with Docker.

1 - Install Docker

Right, so the first thing to do is installing docker, which works on Mac, Linux and Windows!

2 - Create a Dockerfile

The Dockerfile is to docker what wheels are to a car: you won’t go far without it. The Dockerfile defines what your Docker image will contain and do.

Here is the one I came up with, and I’ll explain what it does.

FROM ruby:2.3

COPY . /jekyll-blog
WORKDIR /jekyll-blog

EXPOSE 4000

RUN bundler install
ENTRYPOINT jekyll serve -w -H 0.0.0.0 -P 4000

• Line 1: FROM {image} specifies that our image will be built from another image (here ruby, version 2.3) so we don’t really start from scratch.
• Line 3: COPY {source} {destination} will copy the content of our current folder into the docker image.
• Line 4: WORKDIR {path} to set the default working directory for docker next commands/processes.
• Line 6: EXPOSE {port} will forward ports to open within your docker container - here jekyll needs the port 4000.
• Line 8: RUN {command} to run commands (you don’t say?) - I want to install Jekyll dependencies for the Gemfile.
• Line 9: ENTRYPOINT {process} is the processus that will run by default - I am serving jekyll with a couple of parameters.

4 - Create your Docker Image

So you now have your Dockerfile, let’s build the image from it now using the command build [parameters] [dockerfile_source]:

docker build -t jekyll-blog .

I use -t to name my image jekyll-blog so it is easier to find it when running docker images.

Building the Jekyll docker image will take a couple of minutes since it will download the ruby image needed, run your commands from the dockerfile and finally create the image.

5 - Run your Docker container

The image should have been created successfully and appear in your list of local images with docker images so now, let’s use it to run the jekyll-blog container.

docker run -itd -p 1234:4000 -v "$(pwd):/jekyll-blog" --name blog jekyll-blog

I used a bunch of parameters which are:

• -i will keep STDIN open even if not attached (interactive).
• -t will allocate a pseudo-TTY.
• -d to run container in background and print container ID (detaching).
• -p to deal with ports.
• -v to mount volumes which is usefull while developing.
• --name will give your new container a nice name.

jekyll-blog is the name of the image we previously created.

Profit

If your container is running correctly, you should be able to access your jekyll blog using localhost:1234 (or whatever port you set while running your docker container).

You can get some logs from docker with the docker logs command:

docker logs -f blog

You can also bash into your container if you need more information using the docker exec command:

docker exec -it blog bash

Stopping and Removing all traces

While playing with Docker, I found it helpful to cleanup all my test containers and images.

The proper way to remove a container would be to stop it and then remove it.

docker stop [container_name_or_id]
docker rm [container_name_or_id]

# I can be lazy so I also used
docker rm -f [container_name_or_id]

To remove an image, you will have to stop any container which is currently using it.

docker rmi [image_id]

# or just force delete it
docker rmi -f [image_id]

At some point I just had too many containers/images to remove so I simply used:

docker rm $(docker ps -qa)
docker rmi $(docker images -q)

which will remove all existing container and images. Run the above command only if you absolutely want to clean up everything!

If you want to play more, have a look at the list of existing docker images and have fun!