-include Makefile.overrides.mk

# Set the environment variable BUILD_WITH_CONTAINER to use a container
# to build the repo. The only dependencies in this mode are to have make and
# docker. If you'd rather build with a local tool chain instead, you'll need to
# figure out all the tools you need in your environment to make that work.
export BUILD_WITH_CONTAINER ?= 0

ifeq ($(BUILD_WITH_CONTAINER),1)

to use the certificates written to disk, assuming the downtime is less than certificate expiration. This is why
the certificates are persisted to disk, rather than kept in memory like in the standard Kubernetes deployment.

## Certificate Rotation

          taskRunner.lock.withLock {
            activeThreads--
            startNextTask()
          }
        }
      }
    }
  }

  /**
   * This blocking queue hooks into a fake clock rather than using regular JVM timing for functions
   * like [poll]. It is only usable within task faker tasks.
   */
  private inner class TaskFakerBlockingQueue<T>(
    val delegate: BlockingQueue<T>,

    some applications.
 *  New: Override `equals()` and `hashCode()` in our new cookie class. This
    class now defines equality by value rather than by reference.
 *  New: Handle 408 responses by retrying the request. This allows servers to
    direct clients to retry rather than failing permanently.
 *  New: Expose the framed protocol in `Connection`. Previously this would

#### Webhooks

Istio contains both Validation and Mutating webhook configurations. These need a `caBundle` specified in order to provision the TLS trust. Because Istiod's CA certificate is somewhat dynamic, this is patched at runtime (rather than part of the install). The webhook controllers handle this patching.

    /**
     * Appends a quoted-string to a StringBuilder.
     *
     * RFC 2388 is rather vague about how one should escape special characters in form-data
     * parameters, and as it turns out Firefox and Chrome actually do rather different things, and
     * both say in their comments that they're not really sure what the right approach is. We go

  way of predicting and reporting when network-related failures will take place.
  It means warning users that since the network is missing, certain features and
  operations will be unavailable, rather than simply waiting for those
  operations to fail, then trying to help users decipher the error messages they
  get as a result.

* Implications for Resolution

** Dependency Resolution

    pragmatic software license, and a desire to create high quality software that leads the way in its field.
    We consider ourselves not simply a group of projects sharing a server, but rather a community of developers
    and users.
  </description>
  <licenses>
    <license>
      <name>The Apache Software License, Version 2.0</name>
      <url>http://www.apache.org/licenses/LICENSE-2.0.txt</url>

*   Add a `tf.distribute.cluster_resolver.TPUClusterResolver.connect` API to
    simplify TPU initialization.
*   Add `tf.distribute.Strategy.gather` and
    `tf.distribute.ReplicaContext.all_gather` methods to gather and concatenate
    `tf.distribute.DistributedValues` across workers and devices.

### `tf.keras`:

*   Introduces experimental preprocessing layers API

### Other notable changes

* Connections from Pods to Services with 0 endpoints will now ICMP reject immediately, rather than blackhole and timeout. ([#72534](https://github.com/kubernetes/kubernetes/pull/72534), [@thockin](https://github.com/thockin))
* Services of type=LoadBalancer which have no endpoints will now immediately ICMP reject connections, rather than time out. ([#74394](https://github.com/kubernetes/kubernetes/pull/74394), [@thockin](https://github.com/thockin))