// when the connection is no longer used. Closing the channel can cause subtle race conditions
	// with push. According to the spec: "It's only necessary to close a channel when it is important
	// to tell the receiving goroutines that all data have been sent."

	// Block until either a request is received or a push is triggered.
	// We need 2 go routines because 'read' blocks in Recv().
	go s.receiveDelta(con, ids)

	// housekeeping routine. This test registers the pod, removes the unwanted pod, then calls into
	// HandlePodCleanups a few more times. We should only see one Destroy() event. podKiller runs
	// within a goroutine so a two second delay should be enough time to
	// mark the pod as killed (within this test case).

	kubelet.HandlePodCleanups(ctx)

	// assert that unwanted pods were killed

	// notified of config changes, updates, periodic reconciliation, container runtime updates, and
	// evictions of all desired pods and will invoke reconciliation methods per pod in separate
	// goroutines. The podWorkers are authoritative in the kubelet for what pods are actually being
	// run and their current state:
	//
	// * syncing: pod should be running (syncPod)