- master — default target for all changes (new features + bug fixes).  
  If the change should also go to maven-4.0.x, request a backport in the PR.

- maven-3.10.x — target for 3.x changes (new features + bug fixes).  
  If the change should also go to maven-3.9.x, request a backport in the PR.
  maven-3.9.x should receive only serious bug fixes, as it is approaching EOL.

Backports are typically handled by maintainers; 

### Bug or Regression

- Fix a bug in the kube-apiserver where a malformed Service without name can cause high CPU usage. The bug is present on the new Cluster IP allocators enabled with the feature MultiCIDRServiceAllocator (enabled by default since 1.33) ([#135499](https://github.com/kubernetes/kubernetes/pull/135499), [@aojea](https://github.com/aojea)) [SIG Testing]

- Fixed a bug where init containers may fail to start due to a temporary container runtime failure. ([#126543](https://github.com/kubernetes/kubernetes/pull/126543), [@gjkim42](https://github.com/gjkim42)) [SIG Node]
- Fixed a bug which the scheduler didn't correctly tell plugins Node deletion.

* `bug`: Fixes
    * Something that was supported didn't work, and this fixes it. There are many PRs that claim to be bug fixes because the user is doing something in an unexpected way that is not supported, but they considered it what should be supported by default. Many of these are actually features or refactors. But in some cases there's an actual bug.
* `refactor`: Refactors

* you should: você deveria
* async context manager: gerenciador de contexto assíncrono
* autocomplete: autocompletar
* autocompletion: preenchimento automático
* auto-completion: preenchimento automático
* bug: bug
* context manager: gerenciador de contexto
* cross domain: cross domain (do not translate to "domínio cruzado")
* cross origin: cross origin (do not translate to "origem cruzada")

 *  Fix: Handle multiple 1xx responses.

 *  Fix: Address a performance bug in our internal task runner. We had a race condition that could
    result in it OkHttp starting a thread for each queued task, even when a single thread could run
    all of them.

 *  Fix: Address a performance bug in `MultipartReader`. We were scanning the entire input stream

    - [Node Binaries](#node-binaries)
    - [Container Images](#container-images)
  - [Changelog since v1.33.9](#changelog-since-v1339)
  - [Changes by Kind](#changes-by-kind)
    - [Bug or Regression](#bug-or-regression)
  - [Dependencies](#dependencies)
    - [Added](#added)
    - [Changed](#changed)
    - [Removed](#removed)
- [v1.33.9](#v1339)
  - [Downloads for v1.33.9](#downloads-for-v1339)

- Fixed a bug in CEL's common.UnstructuredToVal where `==` evaluates to false for identical objects when a field is present but the value is null.  This bug does not impact the Kubernetes API. ([#131559](https://github.com/kubernetes/kubernetes/pull/131559), [@jpbetz](https://github.com/jpbetz)) [SIG API Machinery]

  - [Changes by Kind](#changes-by-kind)
    - [Deprecation](#deprecation)
    - [API Change](#api-change)
    - [Feature](#feature)
    - [Documentation](#documentation)
    - [Failing Test](#failing-test)
    - [Bug or Regression](#bug-or-regression)
    - [Other (Cleanup or Flake)](#other-cleanup-or-flake)
  - [Dependencies](#dependencies)
    - [Added](#added)
    - [Changed](#changed)
    - [Removed](#removed)

      with:
        repo-token: ${{ secrets.GITHUB_TOKEN }}