So, the cluster administrator either provisions some persistent volumes or, alternatively, the cluster supports dynamic provisioning. We can now claim some storage for our workload by creating a persistent volume claim. But, first, it's important to understand the difference between ephemeral and persistent storage. We'll create an ephemeral file in a pod, restart the pod, and check that the file vanished. Then, we'll do the same thing again, but, this time, write the file to the persistent storage and check that the file still exists once the pod is restarted.
Before we start, let me share some convenient shell functions and aliases that I created in order to quickly launch an interactive session in specific pods. ...
