Certified Kubernetes Application Developer (CKAD) Study Guide

Chapter 5. Observability

Applications running in containers do not operate under the premise of “fire and forget.” Once the container has been started, you’ll want to know if the application is ready for consumption and is still working as expected in an hour, a week, or a month. The Observability section of the exam addresses the concern.

In this chapter, we’ll discuss container health probes—more specifically, startup, readiness, and liveness probes. You’ll learn about the different health verification methods and how to define them for the proper use cases. Moreover, the exam expects you to be familiar with strategies for debugging a misconfigured or misbehaving Kubernetes object. We’ll round out the chapter by ttalking about monitoring cluster nodes and Pods.

At a high level, this chapter covers the following concepts:

• Readiness probe

• Liveness probe

• Startup probe

• Troubleshooting Kubernetes objects

• Monitoring

Understanding Health Probing

Even with the best automated test coverage, it’s nearly impossible to find all bugs before deploying software to a production environment. That’s especially true for failure situations that only occur after operating the software for an extended period of time. It’s not uncommon to see memory leaks, deadlocks, infinite loops, and similar conditions crop up once the application has been put under load by end users.

Proper monitoring can help with identifying those issues; however, you still need to take an action to mitigate the ...