- Kube-apiserver: errors decoding objects in etcd are now recorded in an `apiserver_storage_decode_errors_total` counter metric ([#114376](https://github.com/kubernetes/kubernetes/pull/114376), [@baomingwang](https://github.com/baomingwang)) [SIG API Machinery and Instrumentation]

- Evictions due to pods breaching their ephemeral storage limits are now recorded by the `kubelet_evictions` metric and can be alerted on. ([#87906](https://github.com/kubernetes/kubernetes/pull/87906), [@smarterclayton](https://github.com/smarterclayton)) [SIG Node]

- Fixed two scheduler metrics (pending_pods and schedule_attempts_total) not being recorded ([#87692](https://github.com/kubernetes/kubernetes/pull/87692), [@everpeace](https://github.com/everpeace)) [SIG Scheduling]
- Kubelet metrics have been changed to buckets. 

- kubelet
  - OS and Arch information is now recorded in `kubernetes.io/os` and `kubernetes.io/arch` labels on Node objects. The previous labels (`beta.kubernetes.io/os` and `beta.kubernetes.io/arch`) are still recorded, but are deprecated and targeted for removal in v1.18. ([#73333](https://github.com/kubernetes/kubernetes/pull/73333), [@yujuhong](https://github.com/yujuhong))

  with `scheme: HTTPS` that encounter errors indicating the endpoint is actually
  using HTTP fall back to making the request over HTTP for compatibility with
  previous releases. When this happens, a `LifecycleHTTPFallback` event is recorded
  in the namespace of the pod and a `kubelet_lifecycle_handler_http_fallbacks_total`
  metric in the kubelet is incremented. Cluster administrators can opt out of the
  expanded lifecycle handler capabilities by setting

- UpdateContainerResources is no longer recorded as a `container_status` operation. It now uses the label `update_container`. ([#75278](https://github.com/kubernetes/kubernetes/pull/75278), [@Nessex](https://github.com/Nessex))

- Resolves non-deterministic behavior of the garbage collection controller when ownerReferences with incorrect data are encountered. Events with a reason of `OwnerRefInvalidNamespace` are recorded when namespace mismatches between child and owner objects are detected. The [kubectl-check-ownerreferences](https://github.com/kubernetes-sigs/kubectl-check-ownerreferences) tool can be run prior to upgrading to locate existing objects with invalid...

- Kube-apiserver: DELETECOLLECTION API requests are now recorded in metrics with the correct verb. ([#113133](https://github.com/kubernetes/kubernetes/pull/113133), [@sxllwx](https://github.com/sxllwx)) [SIG API Machinery]

    * as being in state "Unknown", along with a longer description in `kubectl describe` output.
* The hostname of the node (as autodetected by the kubelet, specified via --hostname-override, or determined by the cloudprovider) is now recorded as an address of type "Hostname" in the status of the Node API object. The hostname is expected to be resolveable from the apiserver. ([#25532](https://github.com/kubernetes/kubernetes/pull/25532), [@mkulke](https://github.com/mkulke))

- Adds profile label to kube-scheduler metric schedule_attempts_total
  - Adds result and profile label to e2e_scheduling_duration_seconds. Times for unschedulable and error attempts are now recorded. ([#92202](https://github.com/kubernetes/kubernetes/pull/92202), [@alculquicondor](https://github.com/alculquicondor)) [SIG Instrumentation and Scheduling]