* Fixed mutation warning in Attach/Detach controller ([#35273](https://github.com/kubernetes/kubernetes/pull/35273), [@jsafrane](https://github.com/jsafrane))
* Don't count failed pods as "not-ready" ([#35404](https://github.com/kubernetes/kubernetes/pull/35404), [@brendandburns](https://github.com/brendandburns))

- kube-apiserver now reloads the `--authorization-config` file when it changes. Reloads increment the `apiserver_authorization_config_controller_automatic_reload_last_timestamp_seconds` timestamp metric, with `status="success"` for successful reloads and `status="failed"` for failed reloads. Failed reloads keep using the previously loaded authorization configuration.

- Specific error message on failed rolling update issued by older kubectl against 1.4 master ([#32751](https://github.com/kubernetes/kubernetes/issues/32751))
- bump master cidr range from /30 to /29 ([#32886](https://github.com/kubernetes/kubernetes/issues/32886))
- non-hostNetwork daemonsets will almost always have a pod that fails to schedule ([#32900](https://github.com/kubernetes/kubernetes/issues/32900))