Sets up a cronjob to defrag etcd pods running in cluster. Optionally backs up pre-defrag to an existing static or dynamically created pvc.
The job runs on a control plane node and ensures cluster health after each node is defragmented and targets the leader last.