Kubernetes is very good at telling you what is wrong. It is very bad at telling you where to look.

In two hours we will walk through the handful of failures that account for most of the time people lose to a cluster, and I will show you the exact command that ends each one.

A pod stuck in Pending. A container in CrashLoopBackOff with an empty log. A deploy that hangs at 1 of 3 forever. A service that resolves for one pod and not another. A process that dies with no exception, no stack trace, and no explanation, because the kernel killed it and nothing wrote that down where you would find it.

You will break each one on your own cluster, then find it. Not by reading a slide about it, by actually watching the events stream tell you what happened while you were looking somewhere else.

This is deliberately narrow. I am not going to teach you the whole platform in two hours and neither is anybody else. I am going to teach you the five things that will otherwise cost your team an afternoon each.