- Fixed bug in reflector that couldn't recover from `Too large resource version` errors with API servers before 1.17.0. ([#115093](https://github.com/kubernetes/kubernetes/pull/115093), [@xuzhenglun](https://github.com/xuzhenglun))

- Fixed the bug where images that were pinned by the container runtime could be garbage collected by `kubelet`. ([#119986](https://github.com/kubernetes/kubernetes/pull/119986), [@ruiwen-zhao](https://github.com/ruiwen-zhao))

### Breaking Changes

* PR [#2434](https://github.com/tiangolo/fastapi/pull/2434) includes several improvements that shouldn't affect normal use cases, but could affect in advanced scenarios:
    * If you are testing the generated OpenAPI (you shouldn't, FastAPI already tests it extensively for you): the order for `tags` in `include_router` and *path operations* was updated for consistency, but it's a simple order change.

- Fixed issue where kubectl-convert would fail when encountering resources that could not be converted to the specified api version. New behavior is to warn the user of the failed conversions...

  // - ScheduleAnyway tells the scheduler to schedule the pod in any location,
  //   but giving higher precedence to topologies that would help reduce the
  //   skew.
  // A constraint is considered "Unsatisfiable" for an incoming pod
  // if and only if every possible node assignment for that pod would violate
  // "MaxSkew" on some topology.
  // For example, in a 3-zone cluster, MaxSkew is set to 1, and pods with the same

  were unnecessarily using `--node-ip` with an external cloud provider in 1.23,
  and then running into problems with 1.24, could not fix the problem by just
  removing the unnecessary `--node-ip` from the kubelet arguments, because