This is an etcd question-- you need to understand whether your cluster is using etcd2 format or etcd3 format. In your case, the cluster is probably using etcd3 format. To make etcdctl work, you'll have to set an env var.
Try running `ETCDCTL_API=3 etcdctl --help`. You will find that the available commands have changed (
`ls` is gone).
Additionally, depending on the flags passed to apiserver, you may find that your objects are being stored as protobufs instead of json, in which case there is actually no good way to view them (we are working on providing a tool for this, which may directly read the boltdb database etcd writes, for use in emergencies).
TPRs have no proto representation and aren't affected, if that's all you want to view you'll be OK.