Production Kubernetes: Building Successful Application Platforms

This book is pretty thorough on the interworkings of Kubernetes and has helped me a lot at work.

almost as good as Kinetics for Dummies. not nearly enough drawings.

Extremely well-structured, guiding readers through advanced concepts with clarity and depth. While it assumes a solid grasp of Kubernetes fundamentals, those with a foundational knowledge will find valuable insights that can be adapted to their own use cases.The book does not provide easy-to-follow examples, which might pose a challenge for some readers. Nonetheless, its comprehensive approach allows for customization and extension to various production scenarios, making it a worthwhile read for anyone serious about mastering Kubernetes.

This book has a lot of useful k8s Production/Design experience, best practices, tips and advices inside. A long book ~500 pages total with 16 Chapters. A lot of easy to understand and colorful diagrams to help us understand some internal designs/flow, e.g. end to end flow of a PVC storage request, how Calico using BGP, how Cilium using eBPF, how IPAM daemon working in AWS CNI.A lot of useful tips running/maintaining Production k8s, e.g. dedicated etcd cluster upgrade should be done subtractively because "Adding a fourth member to a three-node etcd cluster, for example, will require an update to all etcd nodes’ member list, which will require a restart."(p.58) The other control planes and worker nodes can be replaced additively (p.59)Useful Production tip: Why unexpected request errors during application rolling updates ? (p.145) because of race conditions between kubelet doing Pod Shutdown while Endpoints controller removing Pod IP addresses at the SAME time. If Endpoints controller/kubeproxy removed the deleted Pod IP too late, then request will fail. So, good suggestions to SIGTERM handlers and sleep pre-stop hooks to solve this !Another tip: use "DNS cache add-on called NodeLocal DNSCache" to help DNS service performance. How to tune DNS resolver config with ndots and search. There is an example of of the impact of ndots config on Pods to resolve external names. It can result in multiple DNS queries. (p.149)It shows users a lot of good design factors to consider, e.g. migrating strategy. When to use Cluster replacement, node replacement and in-place upgrades ? Many overview of different choices and examples, e.g. Which container runtimes or CRI ? Which Storage classes or CSI ? Which service meshes ? Which CNI ? Which Ingress controller ? Which Secrets provider ? ...etc.A lot of valuable discussions and tips in Admission control (OPA), Observability(audit, logging, events, metrics, showbacks, tracings), Identity (e.g.OIDC, kiam, SAT, PSAT), Multitenancy (Namespace Boundary, RBAC, Resource Quotas, Admission Webhooks), Autoscaling (HPA, VPA, Cluster Proportional Autoscaler, Cluster Autoscaling),Application Considerations (State Probes, termination, logs), Software Supply Chain (images, registries, Vulnerability Scanning, Image Signing, CD pipelines, Push-Based Deployments), Platform Abstractions (Command-Line Tooling, Abstraction Through Templating).I am a CKA and still learn a lot important and useful insights, experience, best practices and tips in this very practical field book! Strongly recommend!

I've been heavily involved with Kubernetes for the past 5-years, and run several large (EKS and bare metal) clusters in production. I learned a ton from this book, and it definitely took my K8S knowledge to 11. While I enjoyed the entire book, I especially liked the "application considerations," "admission control," and "secret management" chapters. Picked up a number of useful tidbits from these chapters. Can't recommend this book enough! Kudos to all of the authors.

This book covers the theory. Lacks any practical use. Overall my recommendation is do not buy, any other book will be more useful.

This book is amazing. As a solo developer, I am NOT really the target audience, but I still think it's a fantastic book.It's really written for folks who are operating enterprise-scale Kubernetes clusters for large organizations. The authors explain almost everything in relation to teams which are running a very large k8s clusters (or _multiple_ clusters) as a platform – whose clients in turn are others development teams at their company writing containerized apps for that internal k8s-based platform.Even tho none of those concerns apply to me, I still learned just tons from this book. The authors do an outstanding job of methodically going through all the major components of k8s (The API, essential controllers, core types, common patterns, etc.) that comprise the core K8s platform itself, explaining in granular detail how they all work, how to best configure/use them, common gotchas, etc.This is one of those books that is so dense with value and information that I often found myself having to re-read a paragraph (or sometimes a whole section) multiple times before I felt like I got it. It's that rich in information.My only warning would be that this book is not one with many examples, and it contains little in the way of tutorials or step-by-step examples, sample code, etc. It's really all about sharing operational information and know-how based on the authors' knowledge and deep experience. That said, you can find plenty of tutorials and sample code online – This is a great read for anyone who is *serious* about wanting to make the most of Kubernetes, imho.