We recently came across the situation where we had a lot of old ReplicaSets inside one of the Kubernetes clusters of one of our clients which did not any DESIRED, CURRENT or READY instances.
We created a single-line command to clean this mess up which we want to share with you: