Nigel, How the storage class knows the corresponding storage block? I don’t see mentioned any storage name in the yaml file. Am I missing something?
Each volume has an association with a provisioner (csi plugin). If no provisioner is applied then ‘default’ become the provisioner. Default is likely the local instance storage (SSD, NSF, etc) You can think of a provisioner like a plugin. Nigal covered this during the introduction of this section. For a little more context this CSI link might help (https://github.com/container-storage-interface/spec/blob/master/spec.md).
As Nigel mentioned, you create an identifiable name , in your metadata PV (persistent volume), that identifies the type of storage would like to provisioned to the kubernetes cluster. It’s then up to the developer to query the cluster and identify the available storage they want to use for their pods/containers (in this context is pods). Once the volume has been provisioned to the cluster developers can dynamically assign any available storage class according to what’s been made available to use in the cluster (which and easily be a array of various types of storage). This is done by using the same metadata name provisioned to the cluster by a cloud administrator when they deployed the provisional volume.
Re-watch the video and pay close attention to the metadata names and labels. I hope this adds some clarity.