Chapter 10. Image pipelines

 

This chapter covers

  • The goals of Docker image pipelines
  • Patterns for building images and using metadata to help consumers use your image
  • Common approaches for testing that images are configured correctly and secure
  • Patterns for tagging images so they can be identified and delivered to consumers
  • Patterns for publishing images to runtime environments and registries

In chapter 8, you learned how to build Docker images automatically by using Dockerfiles and the docker build command. However, building the image is merely one critical step in a longer process for delivering functioning and trustworthy images. Image publishers should perform tests to verify that the image works under the expected operating conditions. Confidence in the correctness of the image artifact grows as it passes those tests. Next, it can finally be tagged and published to a registry for consumption. Consumers can deploy these images with confidence, knowing that many important requirements have already been verified.

These steps—preparing image material, building an image, testing, and finally publishing images to registries—are together called an image build pipeline. Pipelines help software authors quickly publish updates and efficiently deliver new features and fixes to consumers.

10.1. Goals of an image build pipeline

10.2. Patterns for building images

10.3. Record metadata at image build time

10.4. Testing images in a build pipeline

10.5. Patterns for tagging images

Summary

sitemap