endpoint now gets broken when that endpoint becomes non-ready
      (rather than continuing until the endpoint is fully deleted).
  
    - Traffic to a service IP now starts getting rejected (as opposed to
      merely dropped) as soon as there are no longer any *usable*
      endpoints, rather than waiting until all of the terminating

saikai.nagasaki.jp
saiki.oita.jp
saitama.jp
saitama.saitama.jp
saito.miyazaki.jp
saka.hiroshima.jp
sakado.saitama.jp
sakae.chiba.jp
sakae.nagano.jp
sakahogi.gifu.jp
sakai.fukui.jp
sakai.ibaraki.jp
sakai.osaka.jp
sakaiminato.tottori.jp
sakaki.nagano.jp
sakata.yamagata.jp
sakawa.kochi.jp
sakegawa.yamagata.jp
saku.nagano.jp
sakuho.nagano.jp
sakura
sakura.chiba.jp

* kubeadm: use kubeconfig flag instead of kubeconfig-dir on init phase bootstrap-token ([#71803](https://github.com/kubernetes/kubernetes/pull/71803), [@yagonobre](https://github.com/yagonobre))
* On GCI, NPD starts to monitor kubelet, docker, containerd crashlooping, read-only filesystem and corrupt docker overlay2 issues. ([#71522](https://github.com/kubernetes/kubernetes/pull/71522), [@wangzhen127](https://github.com/wangzhen127))

        the layer's dtype. If you had a model that used `float64`, it will
        probably silently use `float32` in TensorFlow2, and a warning will be
        issued that starts with Layer "layer-name" is casting an input tensor
        from dtype float64 to the layer's dtype of float32. To fix, either set
        the default dtype to float64 with

- Fixed a bug where containers would not start on cgroupv2 systems where swap is disabled. ([#120924](https://github.com/kubernetes/kubernetes/pull/120924), [@klueska](https://github.com/klueska)) [SIG Node]
- Fixed a regression in kube-proxy where it might refuse to start if given
  single-stack IPv6 configuration options on a node that has both IPv4 and

- Reduced GCE log rotation check from 1 hour to every 5 minutes.  Rotation policy is unchanged (new day starts, log file size > 100MB). ([#76352](https://github.com/kubernetes/kubernetes/pull/76352), [@jpbetz](https://github.com/jpbetz))

  or aggregated API servers. The garbage collector controller refreshes periodically.
  Therefore, expect a latency of about 30 seconds between when an API is added and when
  the garbage collector starts to manage it.


#### Monitoring/Prometheus

,GAAOvS,GAk9IxDsiB,GAAQliB,eAx1IR,SAAwBjD,GACpB,IAAIrE,EAASpC,KAAKusB,gBAAgB9lB,GAC9B+lB,EAAcxsB,KAAKusB,gBAAgB9lB,EAAIgmB,eAE3C,OAAIrqB,IAAWoqB,EACJpqB,GAGXpC,KAAKusB,gBAAgB9lB,GAAO+lB,EACvBrjB,MAAMd,GACN1G,IAAI,SAAU+qB,GACX,MACY,SAARA,GACQ,OAARA,GACQ,OAARA,GACQ,SAARA,EAEOA,EAAI/lB,MAAM,GAEd+lB,IAEV9lB,KAAK,IAEH5G,KAAKusB,gBAAgB9lB,KAk0IhCmlB,GAAQpiB,YA7zIR,WACI,OAAOxJ,KAAK2sB,cA6zIhBf,GAAQhjB,QAvzIR,SAAiBjB,GACb,OAAO3H,KAAK4sB,SAASxjB,QAAQ,KAAMzB,IAuzIvCikB,GAAQ9N,SAAW6I,GACnBiF,GAAQnE,WAAad,G...

- 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))

- Fixed a bug where init containers may fail to start due to a temporary container runtime failure. ([#127213](https://github.com/kubernetes/kubernetes/pull/127213), [@SergeyKanzhelev](https://github.com/SergeyKanzhelev)) [SIG Node]