- Sped up kubelet start time by executing an immediate runtime and node status update when the Kubelet sees that it has a CIDR. ([#67031](https://github.com/kubernetes/kubernetes/pull/67031), [@krzysztof-jastrzebski](https://github.com/krzysztof-jastrzebski))

om/kubernetes/kubernetes/pull/59052), [@pospispa](https://github.com/pospispa))

* iSCSI sessions managed by kubernetes will now explicitly set startup.mode to 'manual' to  prevent automatic login after node failure recovery. This is the default open-iscsi mode, so this change will only impact users who have changed their startup.mode to be 'automatic' in /etc/iscsi/iscsid.conf. ([#57475](https://github.com/kubernetes/kubernetes/pull/57475), [@stmcginnis](https://github.com/stmcginnis))...

- Force deleted pods may fail to terminate until the kubelet is restarted when the container runtime returns an error during termination. We have strengthened testing for runtime failures and now perform a more rigorous reconciliation to ensure static pods (especially those that use fixed UIDs) are restarted.  As a side effect of these changes static pods will be restarted with lower latency than before (2s vs 4s, on average) and rapid updates to pod configuration should...

- Return zero time (midnight on Jan. 1, 1970) instead of negative number when reporting startedAt and finishedAt of the not started or a running Pod when using `dockershim` as a runtime. ([#99585](https://github.com/kubernetes/kubernetes/pull/99585), [@Iceber](https://github.com/Iceber))

* Kubelet re-binds /var/lib/kubelet directory with rshared mount propagation during startup if it is not shared yet. ([#45724](https://github.com/kubernetes/kubernetes/pull/45724), [@jsafrane](https://github.com/jsafrane))

* kubeadm now detects the Docker cgroup driver and starts the kubelet with the matching driver. This eliminates a common error experienced by new users in when the Docker cgroup driver is not the same as the one set for the kubelet due to different Linux distributions setting different cgroup drivers for Docker, making it hard to start the kubelet properly.

      - `kubelet_pod_worker_latency_microseconds` -> `kubelet_pod_worker_duration_seconds`
      - `kubelet_pod_start_latency_microseconds` -> `kubelet_pod_start_duration_seconds`
      - `kubelet_cgroup_manager_latency_microseconds` -> `kubelet_cgroup_manager_duration_seconds`
      - `kubelet_pod_worker_start_latency_microseconds` -> `kubelet_pod_worker_start_duration_seconds`
      - `kubelet_pleg_relist_latency_microseconds` -> `kubelet_pleg_relist_duration_seconds`

* The following deprecated flags were removed from `kubelet`:

  * `api-servers` - add apiserver addresses to the kubeconfig file instead.

  Don't use these flags. Using deprecated flags causes the kubelet to print a warning. Using a removed flag causes the kubelet to abort the startup.

- CSI Migration: Fixes issue where all volumes with the same inline volume inner spec name were staged in the same path. Migrated inline volumes are now staged at a unique path per unique volume. ([#84754](https://github.com/kubernetes/kubernetes/pull/84754), [@davidz627](https://github.com/davidz627))

  
  This fixes a bug where clients with requests using hand size x max seats greater than the nominal concurrency limit can starve other requests in the same priority level. ([#118601](https://github.com/kubernetes/kubernetes/pull/118601), [@andrewsykim](https://github.com/andrewsykim))