written as a defaulted config.' ([#124480](https://github.com/kubernetes/kubernetes/pull/124480), [@neolit123](https://github.com/neolit123))
- .status.terminating field now gets tracked faster when active Pods are deleted, specifically when Job is failed, gets suspended or has too many active pods. ([#125175](https://github.com/kubernetes/kubernetes/pull/125175), [@dejanzele](https://github.com/dejanzele)) [SIG Apps and Testing]

### Bug or Regression

- Fixes a race condition in the iptables mode of kube-proxy in 1.27 and later
  that could result in some updates getting lost (e.g., when a service gets a
  new endpoint, the rules for the new endpoint might not be added until
  much later). ([#122815](https://github.com/kubernetes/kubernetes/pull/122815), [@wedaly](https://github.com/wedaly)) [SIG Network]

beta. ([#45247](https://github.com/kubernetes/kubernetes/pull/45247), [@mbohlool](https://github.com/mbohlool))

* The Kubernetes API server now exits if it encounters a networking failure (e.g. the networking interface hosting its address goes away) to allow a process manager (systemd/kubelet/etc) to react to the problem. Previously the server would log the failure and try again to bind to its configured address:port. ([#42272](https://github.com/kubernetes/kubernetes/pull/42272), [@mar...

- Fix memory leak in kube-apiserver when underlying time goes forth and back. ([#96266](https://github.com/kubernetes/kubernetes/pull/96266), [@chenyw1990](https://github.com/chenyw1990)) [SIG API Machinery]

*   Added a new --etcd-upgrade keyword to kubeadm upgrade apply. When this keyword is specified, etcd's static pod gets upgraded to the etcd version officially recommended for a target kubernetes release. ([#55010](https://github.com/kubernetes/kubernetes/pull/55010),[ @sbezverk](https://github.com/sbezverk))

- Ensure only one LoadBalancer rule is created when HA mode is enabled ([#99825](https://github.com/kubernetes/kubernetes/pull/99825), [@feiskyer](https://github.com/feiskyer)) [SIG Cloud Provider]
- Ensure that client-go's EventBroadcaster is safe (non-racy) during shutdown. ([#95664](https://github.com/kubernetes/kubernetes/pull/95664), [@DirectXMan12](https://github.com/DirectXMan12)) [SIG API Machinery]

`kubeadm init --config /etc/kubeadm/kubeadm.yaml`

([#55479](https://github.com/kubernetes/kubernetes/pull/55479), [@ijc](https://github.com/ijc))

* scheduler: Fix issue were a new pod with affinity gets stuck at `creating` because the node had been deleted but the pod still exists. ([#56835](https://github.com/kubernetes/kubernetes/pull/56835), [@wenlxie](https://github.com/wenlxie))

- DRA: scheduling pods is up to 16x faster, depending on the scenario. Scheduling throughput depends a lot on cluster utilization. It is higher for lightly loaded clusters with free resources and gets lower when the cluster utilization increases. ([#127277](https://github.com/kubernetes/kubernetes/pull/127277), [@pohly](https://github.com/pohly)) [SIG API Machinery, Apps, Architecture, Auth, Etcd, Instrumentation, Node, Scheduling and Testing]

- Fix a bug on the endpoints controller that does not reconcile the Endpoint object after this is truncated (it gets more than 1000 endpoints addresses) ([#127417](https://github.com/kubernetes/kubernetes/pull/127417), [@aojea](https://github.com/aojea)) [SIG Apps, Network and Testing]