* fix race condition when attach/delete azure disk in same time ([#84917](https://github.com/kubernetes/kubernetes/pull/84917), [@andyzhangx](https://github.com/andyzhangx))
* fix azure load balancer update dns label issue ([#85329](https://github.com/kubernetes/kubernetes/pull/85329), [@nilo19](https://github.com/nilo19))

- Kubeadm: during 'upgrade apply', if the kube-proxy ConfigMap is missing, assume that kube-proxy should not be upgraded. Same applies to a missing kube-dns/coredns ConfigMap for the DNS server addon. Note that this is a temporary workaround until 'upgrade apply' supports phases. Once phases are supported the kube-proxy/dns upgrade should be skipped manually. ([#89593](https://github.com/kubernetes/kubernetes/pull/89593), [@neolit123](https://github.com/neolit123)) [SIG Cluster Lifecycle]...

- Fixes a regression introduced in 1.29 where conntrack entries for UDP connections
  to deleted pods did not get cleaned up correctly, which could (among other things)
  cause DNS problems when DNS pods were restarted. ([#127806](https://github.com/kubernetes/kubernetes/pull/127806), [@danwinship](https://github.com/danwinship)) [SIG Network]

- github.com/mattn/go-colorable: [v0.0.9](https://github.com/mattn/go-colorable/tree/v0.0.9)
- github.com/mattn/go-isatty: [v0.0.3](https://github.com/mattn/go-isatty/tree/v0.0.3)
- github.com/miekg/dns: [v1.0.14](https://github.com/miekg/dns/tree/v1.0.14)
- github.com/mitchellh/cli: [v1.0.0](https://github.com/mitchellh/cli/tree/v1.0.0)
- github.com/mitchellh/go-homedir: [v1.1.0](https://github.com/mitchellh/go-homedir/tree/v1.1.0)

          call = call,
          fastFallback = fastFallback,
        )
      routeSelector = newRouteSelector
    }

    // List available IP addresses for the current proxy. This may block in Dns.lookup().
    if (!newRouteSelector.hasNext()) throw IOException("exhausted all routes")
    val newRouteSelection = newRouteSelector.next()
    routeSelection = newRouteSelection

### Feature

- Added `k8s-short-name` and `k8s-long-name` format validation tags to enforce DNS label and DNS subdomain compliance. ([#133894](https://github.com/kubernetes/kubernetes/pull/133894), [@lalitc375](https://github.com/lalitc375))

- Kubelet: fixed reloading of kubelet server certificate files when they are changed on disk, and kubelet is dialed by IP address instead of DNS/hostname ([#133654](https://github.com/kubernetes/kubernetes/pull/133654), [@kwohlfahrt](https://github.com/kwohlfahrt)) [SIG API Machinery, Auth, Node and Testing]

- github.com/mattn/go-colorable: [v0.0.9](https://github.com/mattn/go-colorable/tree/v0.0.9)
- github.com/mattn/go-isatty: [v0.0.3](https://github.com/mattn/go-isatty/tree/v0.0.3)
- github.com/miekg/dns: [v1.0.14](https://github.com/miekg/dns/tree/v1.0.14)
- github.com/mitchellh/cli: [v1.0.0](https://github.com/mitchellh/cli/tree/v1.0.0)
- github.com/mitchellh/go-homedir: [v1.0.0](https://github.com/mitchellh/go-homedir/tree/v1.0.0)

- Fixes a regression introduced in 1.29 where conntrack entries for UDP connections
  to deleted pods did not get cleaned up correctly, which could (among other things)
  cause DNS problems when DNS pods were restarted. ([#127808](https://github.com/kubernetes/kubernetes/pull/127808), [@danwinship](https://github.com/danwinship)) [SIG Network]

- `client-go`, using log level 9, traces the following events of a HTTP request:
      - DNS lookup
      - TCP dialing
      - TLS handshake
      - Time to get a connection from the pool