You're reading from Mastering Elastic Kubernetes Service on AWS Deploy and manage EKS clusters to support cloud-native applications in AWS

Product type Paperback

Published in Jul 2023

Publisher Packt

ISBN-13 9781803231211

Length 448 pages

Edition 1st Edition

Languages

Python

Tools

AWS

Concepts

Cloud Computing

Authors (2):

Yang-Xin Cao

Malcolm Orr

View More author details

Table of Contents (28) Chapters

Preface

1. Part 1: Getting Started with Amazon EKS

2. Chapter 1: The Fundamentals of Kubernetes and Containers FREE CHAPTER

3. Chapter 2: Introducing Amazon EKS

4. Chapter 3: Building Your First EKS Cluster

5. Chapter 4: Running Your First Application on EKS

6. Chapter 5: Using Helm to Manage a Kubernetes Application

7. Part 2: Deep Dive into EKS

8. Chapter 6: Securing and Accessing Clusters on EKS

9. Chapter 7: Networking in EKS

10. Chapter 8: Managing Worker Nodes on EKS

11. Chapter 9: Advanced Networking with EKS

12. Chapter 10: Upgrading EKS Clusters

13. Part 3: Deploying an Application on EKS

14. Chapter 11: Building Applications and Pushing Them to Amazon ECR

15. Chapter 12: Deploying Pods with Amazon Storage

16. Chapter 13: Using IAM for Granting Access to Applications

17. Chapter 14: Setting Load Balancing for Applications on EKS

18. Chapter 15: Working with AWS Fargate

19. Chapter 16: Working with a Service Mesh

20. Part 4: Advanced EKS Service Mesh and Scaling

21. Chapter 17: EKS Observability

22. Chapter 18: Scaling Your EKS Cluster

23. Chapter 19: Developing on EKS

24. Part 5: Overcoming Common EKS Challenges

25. Chapter 20: Troubleshooting Common Issues

26. Index

Why subscribe?

27. Other Books You May Enjoy

Summary

In this chapter, we explored why you will need to upgrade your cluster: either you want to use a new Kubernetes feature or a bug fix or AWS is deprecating the version you are using. We identified you will need to perform three actions for each cluster: upgrade the control plane, upgrade the worker nodes, and upgrade the core components such as kube-proxy and coreDNS.

We also discussed how the control plane upgrade is pretty straightforward as it’s a managed service, but node group and component upgrades can be more challenging. Using managed node groups and add-ons simplifies this, but you could also use a second cluster and move the workload between them, upgrading the non-active cluster. This approach—sometimes referred to as blue/green cluster deployments—will add cost and complexity, so it is not recommended, but it can minimize application outages due to upgrades.

In the next chapter, we will look at how you can use AWS Elastic Container Repository...