Using Docker

Chapter 7. Image Distribution

Once you’ve created your images, you’ll want to make them available, be it to coworkers, continuous integration servers, or end users. There are several ways to distribute images: you can rebuild them from Dockerfiles, pull from a registry, or use the docker load command to install from an archive file.

In this chapter, we’ll take a deeper look at the differences between these methods and explore the best ways to handle image distribution both internally in a team and externally to users. We’ll see how we can tag and upload our identidock image so that it can be used in other parts of our workflow and downloaded by others.

Tip

The code for this chapter is available at this book’s GitHub. The tag v0 is the code as it was at the end of the last chapter, with later tags representing the progression of the code through this chapter. To get this version of the code:

$ git clone -b v0 \
https://github.com/using-docker/image-dist/
...

Alternatively, you can download the code for any tag from the Releases page on the GitHub project.

Image and Repository Naming

We saw in “Working with Registries” how to tag images appropriately and upload them to remote repositories. When distributing images, it’s very important to use descriptive and accurate names and tags. To recap, image names and tags are set when building the image or by using the docker tag command:

$ cd identidock
$ docker build -t "identidock:0.1" . 
$ docker tag "identidock:0.1" "amouat/identidock:0.1" ...