Verifying application health
In this section, we are going to review how application Pods are considered healthy. Pods always start in the Pending state and continue to the Running state once the main container is considered healthy. If the Pod executes a Service process, it will stay in this Running state. If the Pod is associated with a Job resource, it may end successfully (the Succeeded state) or fail (the Failed state).
If we remove a Pod resource, it will go to Terminating until it is completely removed from Kubernetes.
Important note
If Kubernetes cannot retrieve the Pod’s status, its state will be Unknown. This is usually due to communication issues between the hosts’ kubelet and the API server.
Kubernetes reviews the state of the containers to set the Pod’s state, and containers can be either Waiting, Running, or Terminated. We can use kubectl describe pod <POD_NAME> to review the details of these phases. Let’s quickly review these...
