Challenge: stabilize a Pod with resources and probes

The Pods section of the book devotes whole subsections to probes, requests, limits and QoS classes. This challenge is its practical translation: two Pods whose problem is not in the application but in the contract they declare with the cluster.

The first one asks, through its requests, for more CPU and memory than any node has, so the scheduler leaves it in Pending indefinitely. The second one has a livenessProbe pointing at a path that does not exist, so the kubelet kills, every few seconds, a container that works perfectly.

The rule of the challenge mirrors the real world's: deleting the requests or the probe does not count. You have to fix them.

Passing criterion: the Pod procesador is Running while keeping reasonable requests, and the Pod estado is Ready and without new restarts while keeping its liveness probe. Remember from the first lesson that a Pod with neither requests nor limits gets the BestEffort QoS class; after this challenge, procesador will be Burstable.