* Fixed missing error checking that could cause kubelet to crash in a race condition. ([#60962](https://github.com/kubernetes/kubernetes/pull/60962), [@technicianted](https://github.com/technicianted))
* Bump Heapster to v1.5.2 ([#61396](https://github.com/kubernetes/kubernetes/pull/61396), [@kawych](https://github.com/kawych))

   * create a directory, not a regular file. A common pitfall is to call {@code createTempFile},
   * delete the file and create a directory in its place, but this leads a race condition which can
   * be exploited to create security vulnerabilities, especially when executable files are to be
   * written into the directory.
   *

* Fix a race condition in the scheduler preemption logic that could cause nominatedNodeName of a pod not to be considered in one or more scheduling cycles. ([#72504](https://github.com/kubernetes/kubernetes/pull/72504), [@bsalamat](https://github.com/bsalamat))

* Increase Azure default maximumLoadBalancerRuleCount to 250. ([#72621](https://github.com/kubernetes/kubernetes/pull/72621), [@feiskyer](https://github.com/feiskyer))
* fix race condition issue for smb mount on windows ([#75371](https://github.com/kubernetes/kubernetes/pull/75371), [@andyzhangx](https://github.com/andyzhangx))

   * create a directory, not a regular file. A common pitfall is to call {@code createTempFile},
   * delete the file and create a directory in its place, but this leads a race condition which can
   * be exploited to create security vulnerabilities, especially when executable files are to be
   * written into the directory.
   *

- EndpointSlice controller now handles terminating pods correctly and is better at preventing race conditions. ([#89118](https://github.com/kubernetes/kubernetes/pull/89118), [@robscott](https://github.com/robscott)) [SIG Apps and Network]

  // introduce a very specific kind of data-race. And given the other operations performed by these
  // methods that involve volatile read/write operations, in practice there is no issue. Also, the
  // way in such a visibility issue would surface is most likely as a failure of cancel() to
  // propagate to the input. Cancellation propagation is fundamentally racy so this is fine.
  //

* Fix race causing apiserver crashes during etcd healthchecking ([#60069](https://github.com/kubernetes/kubernetes/pull/60069), [@wojtek-t](https://github.com/wojtek-t))
* Increase allowed lag for ssh key sync loop in tunneler to allow for one failure ([#60068](https://github.com/kubernetes/kubernetes/pull/60068), [@wojtek-t](https://github.com/wojtek-t))

- Fixed a race condition in kube-controller-manager and the scheduler, caused by
  a bug in the transforming informer during the Resync operation, 
  by making the transforming function idempotent. ([#124352](https://github.com/kubernetes/kubernetes/pull/124352), [@wojtek-t](https://github.com/wojtek-t)) [SIG API Machinery and Scheduling]

   * need to be signalled because it is not known which if any of them have waiters (and hasWaiters
   * can't be used reliably because of a check-then-act race). With our Monitor guards, we only
   * signal the first active guard that is satisfied. But the corresponding thread may have already
   * been interrupted and is waiting to reacquire the lock while still registered in activeGuards,