Kubernetes has become an essential part of the daily work for most system, network, and cluster administrators today. But to work effectively together on a production-scale Kubernetes system, they must be able to speak the same language. This book provides a clear guide to the layers of complexity and abstraction that come with running a Kubernetes network.
Authors James Strong and Vallery Lancey bring you up to speed on the intricacies that Kubernetes has to offer for large container deployments. If you're to be effective in troubleshooting and maintaining a production cluster, you need to be well versed in the abstraction provided at each layer. This practical book shows you how.
Learn the Kubernetes networking modelChoose the best interface for your clusters from the CNCF Container Network Interface projectExplore the networking and Linux primitives that power KubernetesQuickly troubleshoot networking issues and prevent downtimeExamine cloud networking and Kubernetes using the three major Amazon Web Services, Google Cloud, and Microsoft AzureLearn the pros and cons of various network tools--and how to select the best ones for your stack
James Strong began his career in Networking, first attending Cisco Networking Academy in High School. He then went on to be a Network Engineer at the University of Dayton and GE Appliances. While attending GE’s Information Technology Leadership program, James was able to see many of the problems that face system administrators and developers in an Enterprise environment. As the Cloud Native Director at Contino, James leads many large-scale enterprises and financial institutions through their Cloud and DevOps journeys. He is deeply involved in his local cloud-native community, running local meetups, both AWS User Group and Cloud-Native Louisville. He holds a Master of Science in Computer Science from the University of Louisville, six AWS Certifications, including the Certified Advanced Networking Specialty, along with the CNCF’s CKA.
Any book on the subject is an ambitious endeavor but this book does a fine job explaining the platform.
Kubernetes is a demanding subject because in order to understand it properly you also need to have a working knowledge of Containers and, to a lesser degree, Namespaces. That's however the easy part because once you have done it another challenge begins: understanding Networking. Not that it is overly difficult if you have an solid understanding of TCP/IP networking but requiring it all (Containers plus Networking) raises the bar and fewer people meet it. The author tackles de subject by starting from the ground up with enough depth to make the book quite comprehensive.
More into the actual subject of Kubernetes the book covers the principal set of networking abstractions in the platform and presents plenty of the operational sessions showing what happens under the hood. Last but not least it shows the different adaptations supported in each of the major public cloud providers.
It is by no means a substitute of the necessary "learning by doing" stage anyone has to go through to master Kubernetes Networking but it is a very good starting point. It will not all solve the issues that will arise in your Kubernetes clusters, it will clear plenty of your doubts but it will also make you raise some other questions.
It is not the definitive reference on the matter (after all the subject evolves) but it is a fine workbook which I will happily recommend. You will learn plenty from it.
Pretty good book. I didn't understand everything in Chapter 5 and I felt it was the worst chapter "apply this, now look at this output" etc.. hence the rating for the delivery of content sometimes. But I'm sure eventually, like many other things, the more you know the easier it is to learn things that were previously unclear.
However, I definitely learned a lot in this book -- even from the refresher on namespaces, veth pairs etc, so author did a good job of making sure the prerequisites were clear.
Worth a read, for sure -- it's a good 3 stars. Don't be put off by that.
Book cover linux and k8s networking aspects in good way. Only issue I have with Section 6 Kubernetes and Cloud Networking. This last section looks as very unbalanced between content on providers. - I would like add one more provider as Oracle OCI (OKE K8S). - Make structure more systematic how describe features in every cloud provider and how work with them. - Use using infrastructure as code in Terraform or CLI. In AWS part you had CLI examples, in Azure were almost screens only. In GCP ware missing examples at all.
Summary and overview was useful for all clouds but make more in aspects rest of book will be great. Some parts was just describing obvious (for example Azure AZ that can be move in foot note and not spend 2 pages on that).
This book has usefull material, but i must say I did not like the way concepts were explained. It misses a more didactical approach where you test things out as things are explained.
Concepts like IPTables are explained with a mix of low level command line but with no real possibility for us to follow along on our side through a sample project. As a result, there are too much details that I feel will not be remembered
as a result, I have been skipping several sections with too much details and not enough concept explanations.
Not a bad book though. But I feel it could have been much better with more structured examples
I work in this field and wanted a conceptual run through - so this book was an overkill for me. If time/situations permit, I would definitely love to run all the commands and examples the authors have labouriously provided (I had to skim through them).Super thankful for the intro part of the book where the authors start from the absolute basics of networking, packet transmission and protocols - amongst the best I've read on the topic. The K8S world moves fast enough that things like Cilium and eBPF based load-balancing aren't here. Even service meshes didn't get that much coverage.