held by a prior request.
 *  Fix: Ensure network interceptors always have access to the underlying
    connection.
 *  Fix: Use `X509TrustManagerExtensions` on Android 17+.
 *  Fix: Unblock waiting dispatchers on MockWebServer shutdown.


## Version 3.2.0

_2016-02-25_

 *  Fix: Change the certificate pinner to always build full chains. This

    assertThat(stream1.writeBytesMaximum)
      .isEqualTo(Settings.DEFAULT_INITIAL_WINDOW_SIZE.toLong())

    // receiving a window update on the connection will unblock new streams.
    connection.readerRunnable.windowUpdate(0, 3)
    assertThat(connection.writeBytesTotal)
      .isEqualTo(Settings.DEFAULT_INITIAL_WINDOW_SIZE.toLong())
    assertThat(connection.writeBytesMaximum)

- ReplicaSets and Deployments should always count `.status.availableReplicas` at the correct time without a delay. This results in faster reconciliation of Deployment conditions and faster, unblocked Deployment rollouts. ([#132121](https://github.com/kubernetes/kubernetes/pull/132121), [@atiratree](https://github.com/atiratree)) [SIG Apps]

Guaranteed pods with integer CPU requests on nodes with static CPU & Memory policy configured is not allowed for the beta release of in-place resize. The feature gate `InPlacePodVerticalScalingExclusiveCPUs` defaults to `false`, but can be enabled to unblock development on ([#127262](https://github.com/kubernetes/kubernetes/issues/127262), [@tallclair](https://github.com/tallclair)) [SIG Node]. ([#128287](https://github.com/kubernetes/kubernetes/pull/128287), [@esotsal](https://github.com/esotsal)) [SIG...

* 👥 Update FastAPI People. PR [#4752](https://github.com/tiangolo/fastapi/pull/4752) by [@github-actions[bot]](https://github.com/apps/github-actions).
* ➖ Temporarily remove typer-cli from dependencies and upgrade Black to unblock Pydantic CI. PR [#4754](https://github.com/tiangolo/fastapi/pull/4754) by [@tiangolo](https://github.com/tiangolo).

        3.  v1 optimizer
        4.  target tensors in compile If you are experiencing any issues because
            of this change, please inform us (file an issue) about your use case
            and you can unblock yourself by setting
            `experimental_run_tf_function=False` in compile meanwhile. We have
            seen couple of use cases where the model usage pattern is not as