Skip to content

Releases: linkedin/brooklin

5.4.3

31 Oct 02:19
b31e41a
Compare
Choose a tag to compare

What's Changed

  • Fixing Deadlock Bug Between ZK Callback & Event Thread On Acquiring Coordinator Object by @shrinandthakkar in (#964)
  • Fixing the CPU Intensive RemoveAll with Lists in Sticky & Load Based Partition Assignment Strategies by @shrinandthakkar in (#965)
  • Optimizing By Using Precomputed Set in Partition Assignment Routine by @shrinandthakkar in (#960)
  • Standardize log messages for datastream operations by @amybubu in (#961)
  • Upgrade Pre Assignment Clean Up To Remove Dependency Tasks Across Hosts by @shrinandthakkar in (#952)
  • Quick Optimization For Partition Assignors by @shrinandthakkar in (#956)

Full Changelog: 5.4.0...5.4.3

5.4.0

15 Aug 22:51
8b4d916
Compare
Choose a tag to compare

What's Changed

  • BMM Restart Improvements Part 2. Followers claim their tokens after handling stop assignments by @jzakaryan in #921
  • BMM Restart Improvements Part 3. Leader polls for unclaimed tokens before declaring a datastream STOPPED by @jzakaryan in #922
  • BMM Restart Improvements Part 4. Edge cases, cleanup and general improvements by @jzakaryan in #924
  • Moved the datastream stop check poll to a separate thread to avoid blocking the event queue by @jzakaryan in #925
  • BMM Restart Improvements: Bug fixes and logging improvements by @jzakaryan in #926
  • Bump Kafka version by @vmaheshw in #929
  • Handling Metrics and SLA Reporting for Throughput Violating Topics via Datastream Update API Part 1 by @shrinandthakkar in #928
  • Added Support and Handling For Generic Headers by @shrinandthakkar in #931
  • Minor improvements w.r.t #928 & Releasing a newer version by @shrinandthakkar in #932
  • Exposed the Kafka Connector's task interrupt timeout as config by @jzakaryan in #933
  • Separated Metrics Handling for Throughput Violating Topics by @shrinandthakkar in #930
  • Released New Version 5.3.2 by @shrinandthakkar in #935
  • Increased the size of token claims thread pool by @jzakaryan in #938
  • Audit START/END log events for CoordinatorEvent operations by @ehoner in #936
  • Add metrics for CoordinatorEvent queue by @ehoner in #937
  • Minor improvements for assignment tokens feature by @jzakaryan in #941
  • Fix Metric Registration Issue for Throughput Violations Reporting by @shrinandthakkar in #942
  • Fix tests and reduce the overall test execution time by @vmaheshw in #939
  • Fixes for NoOpTransportProvider by @jzakaryan in #940
  • Add stop method to TransportProviderAdmin by @vmaheshw in #944
  • Log existing datastream for update and delete call by @vmaheshw in #948
  • EventQueue metrics are not exported by @ehoner in #945
  • Added log messages to debug assignPartitions by @jzakaryan in #951
  • Replacing Coordinator Queue With Deque & Fixing Usage Of toMap Util by @shrinandthakkar in #950
  • Fixing Flaky Test testLeaderDoAssignmentForNewlyElectedLeaderFailurePathVariation by @shrinandthakkar in #953
  • Minor version release - 5.4.0 by @ehoner in #954

Full Changelog: 5.1.0...5.4.0

5.1.0

20 Jan 19:54
bf47190
Compare
Choose a tag to compare

The zookeeper dependency update in version 5 includes includes breaking changes. Some of the features below were included in untagged point releases.

What's Changed

  • Add support for post datastream create,update,delete and state change action by @hshukla in #915
  • Update zookeeper dependency to import from LinkedIn published version by @surajkn in #917
  • Set datastream status to deleting for delete call by @hshukla in #918
  • Bump version after release by @hshukla in #920
  • BMM Restart Improvements Part 1. Leader Coordinator Issuing Assignment Tokens by @jzakaryan in #919

New Contributors

Full Changelog: 4.1.0...5.1.0

4.1.0

29 Sep 15:40
0d6a7d2
Compare
Choose a tag to compare

What's Changed

  • Refactored StickyPartitionAssignmentStrategy and implemented task estimation logic in LoadBasedPartitionAssignmentStrategy by @jzakaryan in #835
  • Fix flaky test testConsumeFromMultipleTopics by @vmaheshw in #838
  • Refactor Optional parameters in the constructor of strategy by @vmaheshw in #836
  • Implemented logic to prevent tasks from having more than specified number of partitions by @jzakaryan in #837
  • Publishing artifacts to JFrog by @shrinandthakkar in #839
  • Make the Throughput based assignment and task estimation based on partition assignment configurable. by @vmaheshw in #841
  • Implemented metrics for LoadBasedPartitionAssignmentStrategy by @jzakaryan in #840
  • Clear the CallbackStatus entry from the map in FlushlessEventProducerHandler by @vmaheshw in #843
  • Handle the new session after session expiry by @vmaheshw in #770
  • Fixed issue with config key by @jzakaryan in #847
  • Print count of orphaned tasks and orphaned task locks in the log message by @surajkn in #844
  • Datastream stop transition redesign by @shrinandthakkar in #842
  • Fix flaky stopping simultaneous datastream test by @shrinandthakkar in #849
  • Update the DatabaseChunkedReader to take the Connection as input rather than the DataSource by @somandal in #850
  • Handle leaking TP exception in handleAssignmentChange by @surajkn in #845
  • Migrating from AdminUtils with AdminClient by @surajkn in #848
  • Fix running task data structure logic in AbstractKafkaConnector for failing to stop task by @surajkn in #851
  • Removing partial Helix ZkClient dependency by @surajkn in #852
  • Stop the tasks in parallel in AbstractKafkaConnector by @vmaheshw in #853
  • Add additional log in LoadBasedPartitionAssignmentStrategy by @vmaheshw in #856
  • Fixing flaky test testCreateDatastreamHappyPathDefaultRetention by @surajkn in #854
  • Added toString() override in PartitionThroughputInfo by @jzakaryan in #858
  • Add Stats to DatastreamTaskImpl by @vmaheshw in #855
  • Make PartitionThroughputProvider metricsAware by @vmaheshw in #859
  • Added base strategy metric info in LoadBasedPartitionAssignmentStrategy by @jzakaryan in #857
  • Add alert metrics to identify that elastic task configurations require adjustment by @vmaheshw in #860
  • Fix restartDeadTask logic when the task thread has died by @surajkn in #861
  • Fix metric infos in PartitionThroughputProvider by @vmaheshw in #862
  • Fix the metrics deregistration in AbstractKafkaConnector when multiple stop are called by @vmaheshw in #865
  • Fix logging in LoadBasedPartitionAssignmentStrategy by @vmaheshw in #866
  • Make Default byte in rate and Msg in rate configurable by @vmaheshw in #864
  • Metrics are getting emitted in LoadBasedPartitionAssignmentStrategy only when it needs adjustment by @vmaheshw in #867
  • Use topic level throughput information when partition level information is unavailable. by @vmaheshw in #871
  • Fix compilation error by @vmaheshw in #874
  • Loadbased Partition assigner not using topic level metrics to recognize partitions by @vmaheshw in #876
  • Flushless producer supporting both comparable and non comparable offsets by @shrinandthakkar in #873
  • LiveInstanceProvider subscription should be done only by the leader coordinator by @vmaheshw in #879
  • Fixed issue with missing exception message during task initialization by @jzakaryan in #882
  • Kafka upgrade by @srinagaraj in #881
  • skipping onPartitionsRevoked during consumer.close() call by @srinagaraj in #886
  • Scala 2.12 upgrade by @srinagaraj in #895
  • Upgrade avro and move jackson from codehaus to fasterxml by @vmaheshw in #894
  • Fix topic deletion when multiple duplicate streams expire at the same time by @surajkn in #897
  • Use 2.4.1.57 kafka version by @shrinandthakkar in #901
  • tests for min/max partitions per task metrics and minor code quality improvements by @ryannedolan in #887
  • fix rebalancing-tasks bug and added tests by @shrinandthakkar in #900
  • Refactor Stopping Tasks On Assignment Change of Tasks by @shrinandthakkar in #868
  • Change python dependency in commit-msg git hook by @surajkn in #904
  • Remove Scala Dependencies by @shrinandthakkar in #905
  • Introduce broadcast API to TransportProvider by @surajkn in #903
  • Dedupe tasks on LeaderDoAssignment by @jogrogan in #906
  • Fix Stopping Logic and Maintain Stopping Latch Counter by @shrinandthakkar in #877
  • Fixing test OnAssignmentChangeMultipleReassignments by @shrinandthakkar in #908
  • Update kafka version by @jogrogan in #910
  • Replace 101tec ZkClient with Helix ZkClient by @surajkn in #909
  • Add retry to query retention time for destination topic by @surajkn in #863
  • Upgrade Zookeeper version to 3.6.3 by @jogrogan in #913

New Contributors

Full Changelog: 1.1.0...4.1.0

1.1.0

16 Jun 09:34
90703bd
Compare
Choose a tag to compare
Added configs for fetching throughput info (#834)

1.0.2

01 Oct 21:16
6d5337b
Compare
Choose a tag to compare

1.0.1

16 Sep 17:52
2993725
Compare
Choose a tag to compare

1.0.0

14 Jul 12:13
5062a7f
Compare
Choose a tag to compare

First open-source release