Reduced the number of cloud API calls and service downtime caused by excessive re-configurations of cluster LBs with externalTrafficPolicy=Local when node readiness changes (https://github.com/kubernetes/kubernetes/issues/111539). The service controller (in cloud-controller-manager) will avoid resyncing nodes which are transitioning between `Ready` / `NotReady` (only for ETP=Local Services).  The LBs used for these services will solely rely on the health check probe defined by the `healthCheckNodePort`...

 - Kubeadm does not set the deprecated '--cgroup-driver' flag in /var/lib/kubelet/kubeadm-flags.env, it will be set in the kubelet config.yaml. If you have this flag in /var/lib/kubelet/kubeadm-flags.env or /etc/default/kubelet (/etc/sysconfig/kubelet for RPMs) please remove it and set the value using KubeletConfiguration ([#90513](https://github.com/kubernetes/kubernetes/pull/90513),...

* "kubeadm init" now writes a structured and versioned kubelet ComponentConfiguration file to `/var/lib/kubelet/config.yaml` and an environment file with runtime flags that you can source in the systemd kubelet dropin to `/var/lib/kubelet/kubeadm-flags.env`. ([#63887](https://github.com/kubernetes/kubernetes/pull/63887), [@luxas](https://github.com/luxas))

- Support for the Azure File plugin has been added to `csi-translation-lib` (CSIMigrationAzureFile). ([#78356](https://github.com/kubernetes/kubernetes/pull/78356), [@andyzhangx](https://github.com/andyzhangx))
- Added support for Azure Disk plugin to csi-translation-lib (CSIMigrationAzureDisk) ([#78330](https://github.com/kubernetes/kubernetes/pull/78330), [@andyzhangx](https://github.com/andyzhangx))

        * Make the kubectl from k8s release the default on GCI
* kubelet summary rootfs now refers to the filesystem that contains the Kubelet RootDirectory (var/lib/kubelet) instead of cadvisor's rootfs ( / ), since they may be different filesystems. ([#35136](https://github.com/kubernetes/kubernetes/pull/35136), [@dashpole](https://github.com/dashpole))

* Enable containerized mounter only for nfs and glusterfs types ([#37990](https://github.com/kubernetes/kubernetes/pull/37990), [@jingxu97](https://github.com/jingxu97))
* GCI: Remove /var/lib/docker/network ([#37593](https://github.com/kubernetes/kubernetes/pull/37593), [@yujuhong](https://github.com/yujuhong))

- Kubeadm: ignore EINVAL when unmounting /var/lib/kubelet peer mounts during reset ([#137494](https://github.com/kubernetes/kubernetes/pull/137494), [@fuweid](https://github.com/fuweid)) [SIG Cluster Lifecycle]

* kubelet `--cert-dir` now defaults to `/var/lib/kubelet/pki`, in order to ensure bootstrapped and rotated certificates persist beyond a reboot. resolves an issue in kubeadm with false-positive `/var/lib/kubelet is not empty` message during pre-flight checks ([#53317](https://github.com/kubernetes/kubernetes/pull/53317), [@liggitt](https://github.com/liggitt))

  - build: Update to k/repo-infra@v0.1.1 (supports go1.15.2)
  - build: Use go-runner:buster-v2.0.1 (built using go1.15.1)
  - bazel: Replace --features with Starlark build settings flag
  - hack/lib/util.sh: some bash cleanups
    
    - switched one spot to use kube::logging
    - make kube::util::find-binary return an error when it doesn't find
      anything so that hack scripts fail fast instead of with "binary not

    1. Generate a `/var/lib/kubelet/instance-config.yaml` file to customize the `containerRuntimeEndpoint` field in per-node kubelet configurations.
    2. Remove the `kubeadm.alpha.kubernetes.io/cri-socket` annotation from nodes during upgrade operations.