* adjusted http/2 buffer sizes for apiservers to prevent starvation issues between concurrent streams ([#67902](https://github.com/kubernetes/kubernetes/pull/67902), [@liggitt](https://github.com/liggitt))

* fluentd config for GKE clusters updated: detect exceptions in container log streams and forward them as one log entry. ([#39656](https://github.com/kubernetes/kubernetes/pull/39656), [@thomasschickinger](https://github.com/thomasschickinger))
* Make fluentd pods critical ([#39146](https://github.com/kubernetes/kubernetes/pull/39146), [@crassirostris](https://github.com/crassirostris))

                                    StreamUtil.stream(groupsAndRoles.getFirst()).of(stream1 -> stream1.forEach(groupList::add));
                                    StreamUtil.stream(groupsAndRoles.getSecond()).of(stream2 -> stream2.forEach(roleList::add));
                                }
                            }
                        }

   *
   * <p>For example:
   *
   * {@snippet :
   * Stream.of("foo", "quux", "banana", "elephant")
   *     .collect(least(2, comparingInt(String::length)))
   * // returns {"foo", "quux"}
   * }
   *
   * <p>This {@code Collector} uses O(k) memory and takes expected time O(n) (worst-case O(n log
   * k)), as opposed to e.g. {@code Stream.sorted(comparator).limit(k)}, which currently takes O(n

* adjusted http/2 buffer sizes for apiservers to prevent starvation issues between concurrent streams ([#67902](https://github.com/kubernetes/kubernetes/pull/67902), [@liggitt](https://github.com/liggitt))

- SSA treats certain structs as atomic ([#100684](https://github.com/kubernetes/kubernetes/pull/100684), [@Jefftree](https://github.com/Jefftree)) [SIG API Machinery, Auth, Node and Storage]

* 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))
* Introduced `--http2-max-streams-per-connection` command line flag on api-servers and set default to 1000 for aggregated API servers. ([#60054](https://github.com/kubernetes/kubernetes/pull/60054), [@MikeSpreitzer](https://github.com/MikeSpreitzer))

         * Executes the provided {@link Consumer} with a {@link Stream} obtained from the supplier.
         * The stream is automatically closed after the consumer is executed.
         *
         * @param stream the {@link Consumer} to process the {@link Stream}
         */
        public void of(final Consumer<Stream<T>> stream) {
            try (Stream<T> s = supplier.get()) {
                stream.accept(s);
            }
        }

        }
        return list.stream().map(p -> {
            final String key = p.getFirst();
            if (StringUtil.isEmpty(key) || userBean
                    .map(user -> stream(user.getRoles()).get(stream -> stream.anyMatch(s -> (ROLE_VALUE_PREFIX + s).equals(key)))
                            || stream(user.getGroups()).get(stream -> stream.anyMatch(s -> (GROUP_VALUE_PREFIX + s).equals(key))))

      Map<K, V> map, ObjectOutputStream stream) throws IOException {
    stream.writeInt(map.size());
    for (Map.Entry<K, V> entry : map.entrySet()) {
      stream.writeObject(entry.getKey());
      stream.writeObject(entry.getValue());
    }
  }

  /**
   * Populates a map by reading an input stream, as part of deserialization. See {@link #writeMap}
   * for the data format.
   */