Release 1.2.0 – Changelog

Changelog

The 1.2.0 release resolved 650 JIRA issues in total.

Sub-task

  • [FLINK-3202] - Make Timer/Trigger Service Scoped to Key/Namespace
  • [FLINK-3580] - Reintroduce Date/Time and implement scalar functions for it
  • [FLINK-3660] - Measure latency of elements and expose it as a metric
  • [FLINK-3761] - Refactor State Backends/Make Keyed State Key-Group Aware
  • [FLINK-3950] - Add Meter Metric Type
  • [FLINK-4174] - Enhance Window Evictor
  • [FLINK-4221] - Show metrics in WebFrontend
  • [FLINK-4273] - Refactor JobClientActor to watch already submitted jobs
  • [FLINK-4312] - Remove Serializabiliy of ExecutionGraph
  • [FLINK-4317] - Restructure documentation layout
  • [FLINK-4320] - Fix misleading ScheduleMode names
  • [FLINK-4324] - Enable Akka SSL
  • [FLINK-4325] - Implement Web UI HTTPS
  • [FLINK-4380] - Introduce KeyGroupAssigner and Max-Parallelism Parameter
  • [FLINK-4381] - Refactor State to Prepare For Key-Group State Backends
  • [FLINK-4384] - Add a "scheduleRunAsync()" feature to the RpcEndpoint
  • [FLINK-4386] - Add as way to assert that code runs in the RpcEndpoint's Main Thread
  • [FLINK-4389] - Expose metrics to Webfrontend
  • [FLINK-4392] - Make RPC Service Thread Safe
  • [FLINK-4403] - RPC proxy classloading should use Flink class' classloader
  • [FLINK-4404] - Implement Data Transfer SSL
  • [FLINK-4405] - Implement Blob Server SSL
  • [FLINK-4434] - Add a testing RPC service
  • [FLINK-4490] - Decouple Slot and Instance
  • [FLINK-4507] - Deprecate savepoint backend config
  • [FLINK-4509] - Specify savepoint directory per savepoint
  • [FLINK-4510] - Always create CheckpointCoordinator
  • [FLINK-4512] - Add option for persistent checkpoints
  • [FLINK-4670] - Add watch mechanism on current RPC framework
  • [FLINK-4685] - Gather operator checkpoint durations data sizes from the runtime
  • [FLINK-4691] - Add group-windows for streaming tables
  • [FLINK-4697] - Gather more detailed checkpoint stats in CheckpointStatsTracker
  • [FLINK-4698] - Visualize additional checkpoint information
  • [FLINK-4748] - Fix shutdown of automatic watermark context
  • [FLINK-4749] - Remove redundant processing time timers and futures in the window operator
  • [FLINK-4750] - Ensure that active processing time triggers complete before closing operators
  • [FLINK-4768] - Migrate High Availability configuration options
  • [FLINK-4826] - Add keytab based kerberos support for Mesos environment
  • [FLINK-4877] - Refactorings around FLINK-3674 (User Function Timers)
  • [FLINK-4884] - Eagerly Store MergingWindowSet in State in WindowOperator
  • [FLINK-4899] - Implement DCOS package
  • [FLINK-4900] - Implement Docker image support
  • [FLINK-4901] - Build DCOS Docker image
  • [FLINK-4918] - Add SSL support to Mesos artifact server
  • [FLINK-4921] - Upgrade to Mesos 1.0.1
  • [FLINK-4922] - Write documentation
  • [FLINK-4937] - Add incremental group window aggregation for streaming Table API
  • [FLINK-4959] - Write Documentation for ProcessFunction
  • [FLINK-4984] - Add Cancellation Barriers to BarrierTracker and BarrierBuffer
  • [FLINK-4985] - Report Declined/Canceled Checkpoints to Checkpoint Coordinator
  • [FLINK-4987] - Harden slot pool logic
  • [FLINK-4993] - Don't Allow Trigger.onMerge() to return TriggerResult
  • [FLINK-4994] - Don't Clear Trigger State and Merging Window Set When Purging
  • [FLINK-5042] - Convert old savepoints to new savepoints
  • [FLINK-5043] - Converting keyed state from Flink 1.1 backend implementations to their new counterparts in 1.2
  • [FLINK-5044] - Converting operator and function state from Flink 1.1 for all changed operators in 1.2
  • [FLINK-5091] - Formalize the AppMaster environment for docker compability
  • [FLINK-5290] - Ensure backwards compatibility of the hashes used to generate JobVertexIds
  • [FLINK-5293] - Make the Kafka consumer backwards compatible.
  • [FLINK-5294] - Make the WindowOperator backwards compatible.
  • [FLINK-5295] - Migrate the AlignedWindowOperators to the WindowOperator and make it backwards compatible.
  • [FLINK-5317] - Make the continuous file processing code backwards compatible,
  • [FLINK-5318] - Make the Rolling sink backwards compatible.
  • [FLINK-5386] - Refactoring Window Clause
  • [FLINK-5454] - Add Documentation about how to tune Checkpointing for large state
  • [FLINK-5455] - Create documentation how to upgrade jobs and Flink framework versions
  • [FLINK-5456] - Add docs about new state and checkpointing interfaces
  • [FLINK-5457] - Create documentation for Asynchronous I/O
  • [FLINK-5458] - Add documentation how to migrate from Flink 1.1. to Flink 1.2
  • [FLINK-5459] - Add documentation how to debug classloading issues
  • [FLINK-5460] - Add documentation how to use Flink with Docker
  • [FLINK-5474] - Extend DC/OS documentation
  • [FLINK-5494] - Improve Mesos documentation
  • [FLINK-5502] - Add documentation about migrating functions from 1.1 to 1.2
  • [FLINK-5529] - Improve / extends windowing documentation
  • [FLINK-5555] - Add documentation about debugging watermarks

Bug

  • [FLINK-2662] - CompilerException: "Bug: Plan generation for Unions picked a ship strategy between binary plan operators."
  • [FLINK-2821] - Change Akka configuration to allow accessing actors from different URLs
  • [FLINK-2839] - Failing test: OperatorStatsAccumulatorTest.testAccumulatorAllStatistics
  • [FLINK-3138] - Method References are not supported as lambda expressions
  • [FLINK-3680] - Remove or improve (not set) text in the Job Plan UI
  • [FLINK-3706] - YARNSessionCapacitySchedulerITCase.testNonexistingQueue unstable
  • [FLINK-3813] - YARNSessionFIFOITCase.testDetachedMode failed on Travis
  • [FLINK-3888] - Custom Aggregator with Convergence can't be registered directly with DeltaIteration
  • [FLINK-3918] - Not configuring any time characteristic leads to a ClassCastException
  • [FLINK-4021] - Problem of setting autoread for netty channel when more tasks sharing the same Tcp connection
  • [FLINK-4035] - Add Apache Kafka 0.10 connector
  • [FLINK-4073] - YARNSessionCapacitySchedulerITCase.testTaskManagerFailure failed on Travis
  • [FLINK-4081] - FieldParsers should support empty strings
  • [FLINK-4100] - RocksDBStateBackend#close() can throw NPE
  • [FLINK-4108] - NPE in Row.productArity
  • [FLINK-4140] - CheckpointCoordinator fails to discard completed checkpoint
  • [FLINK-4148] - incorrect calculation distance in QuadTree
  • [FLINK-4155] - Get Kafka producer partition info in open method instead of constructor
  • [FLINK-4177] - CassandraConnectorTest.testCassandraCommitter causing unstable builds
  • [FLINK-4219] - Quote PDSH opts in start-cluster.sh
  • [FLINK-4226] - Typo: Define Keys using Field Expressions example should use window and not reduce
  • [FLINK-4236] - Flink Dashboard stops showing list of uploaded jars if main method cannot be looked up
  • [FLINK-4252] - Table program cannot be compiled
  • [FLINK-4255] - Unstable test WebRuntimeMonitorITCase.testNoEscape
  • [FLINK-4257] - Handle delegating algorithm change of class
  • [FLINK-4259] - Unclosed FSDataOutputStream in FileCache#copy()
  • [FLINK-4263] - SQL's VALUES does not work properly
  • [FLINK-4270] - 'as' in front of join does not work
  • [FLINK-4271] - There is no way to set parallelism of operators produced by CoGroupedStreams
  • [FLINK-4276] - TextInputFormatTest.testNestedFileRead fails on Windows OS
  • [FLINK-4285] - Non-existing example in Flink quickstart setup documentation
  • [FLINK-4289] - Source files have executable flag set
  • [FLINK-4290] - CassandraConnectorTest deadlocks
  • [FLINK-4291] - No log entry for unscheduled reporters
  • [FLINK-4292] - HCatalog project incorrectly set up
  • [FLINK-4293] - Malformatted Apache Headers
  • [FLINK-4296] - Scheduler accepts more tasks than it has task slots available
  • [FLINK-4297] - Yarn client can't determine fat jar location if path contains spaces
  • [FLINK-4298] - Clean up Storm Compatibility Dependencies
  • [FLINK-4304] - Jar names that contain whitespace cause problems in web client
  • [FLINK-4307] - Broken user-facing API for ListState
  • [FLINK-4309] - Potential null pointer dereference in DelegatingConfiguration#keySet()
  • [FLINK-4311] - TableInputFormat fails when reused on next split
  • [FLINK-4313] - Inconsistent code for Key/Value in the CheckpointCoordinator
  • [FLINK-4314] - Test instability in JobManagerHAJobGraphRecoveryITCase.testJobPersistencyWhenJobManagerShutdown
  • [FLINK-4316] - Make flink-core independent of Hadoop
  • [FLINK-4323] - Checkpoint Coordinator Removes HA Checkpoints in Shutdown
  • [FLINK-4329] - Fix Streaming File Source Timestamps/Watermarks Handling
  • [FLINK-4332] - Savepoint Serializer mixed read()/readFully()
  • [FLINK-4333] - Name mixup in Savepoint versions
  • [FLINK-4337] - Remove unnecessary Scala suffix from Hadoop1 artifact
  • [FLINK-4341] - Kinesis connector does not emit maximum watermark properly
  • [FLINK-4342] - Fix dependencies of flink-connector-filesystem
  • [FLINK-4385] - Union on Timestamp fields does not work
  • [FLINK-4387] - Instability in KvStateClientTest.testClientServerIntegration()
  • [FLINK-4388] - Race condition during initialization of MemorySegmentFactory
  • [FLINK-4394] - RMQSource: The QueueName is not accessible to subclasses
  • [FLINK-4398] - Unstable test KvStateServerHandlerTest.testSimpleQuery
  • [FLINK-4409] - class conflict between jsr305-1.3.9.jar and flink-shaded-hadoop2-1.1.1.jar
  • [FLINK-4411] - [py] Chained dual input children are not properly propagated
  • [FLINK-4412] - [py] Chaining does not properly handle broadcast variables
  • [FLINK-4417] - Checkpoints should be subsumed by CheckpointID not, by timestamp
  • [FLINK-4418] - ClusterClient/ConnectionUtils#findConnectingAddress fails immediately if InetAddress.getLocalHost throws exception
  • [FLINK-4425] - "Out Of Memory" during savepoint deserialization
  • [FLINK-4436] - Unclosed DataOutputBuffer in Utils#setTokensFor()
  • [FLINK-4437] - Lock evasion around lastTriggeredCheckpoint may lead to lost updates to related fields
  • [FLINK-4453] - Scala code example in Window documentation shows Java
  • [FLINK-4454] - Lookups for JobManager address in config
  • [FLINK-4482] - numUnsuccessfulCheckpointsTriggers is accessed without holding triggerLock
  • [FLINK-4485] - Finished jobs in yarn session fill /tmp filesystem
  • [FLINK-4486] - JobManager not fully running when yarn-session.sh finishes
  • [FLINK-4488] - Prevent cluster shutdown after job execution for non-detached jobs
  • [FLINK-4506] - CsvOutputFormat defaults allowNullValues to false, even though doc and declaration says true
  • [FLINK-4514] - ExpiredIteratorException in Kinesis Consumer on long catch-ups to head of stream
  • [FLINK-4522] - Gelly link broken in homepage
  • [FLINK-4526] - ApplicationClient: remove redundant proxy messages
  • [FLINK-4544] - TaskManager metrics are vulnerable to custom JMX bean installation
  • [FLINK-4556] - Make Queryable State Key-Group Aware
  • [FLINK-4559] - Kinesis Producer not setting credentials provider properly when AWS_CREDENTIALS_PROVIDER is "AUTO"
  • [FLINK-4563] - [metrics] scope caching not adjusted for multiple reporters
  • [FLINK-4564] - [metrics] Delimiter should be configured per reporter
  • [FLINK-4566] - ProducerFailedException does not properly preserve Exception causes
  • [FLINK-4567] - Enhance SerializedThrowable to properly preserver cause chains
  • [FLINK-4570] - Scalastyle Maven plugin fails undeterministically
  • [FLINK-4572] - Convert to negative in LongValueToIntValue
  • [FLINK-4573] - Potential resource leak due to unclosed RandomAccessFile in TaskManagerLogHandler
  • [FLINK-4581] - Table API throws "No suitable driver found for jdbc:calcite"
  • [FLINK-4586] - NumberSequenceIterator and Accumulator threading issue
  • [FLINK-4588] - Fix Merging of Covering Window in MergingWindowSet
  • [FLINK-4589] - Fix Merging of Covering Window in MergingWindowSet
  • [FLINK-4590] - Some Table API tests are failing when debug lvl is set to DEBUG
  • [FLINK-4592] - Fix flaky test ScalarFunctionsTest.testCurrentTimePoint
  • [FLINK-4594] - Validate lower bound in MathUtils.checkedDownCast
  • [FLINK-4601] - Check for empty string properly
  • [FLINK-4603] - KeyedStateBackend cannot restore user code classes
  • [FLINK-4609] - Remove redundant check for null in CrossOperator
  • [FLINK-4612] - Close FileWriter using try with resources
  • [FLINK-4614] - Kafka connector documentation refers to Flink 1.2-SNAPSHOT
  • [FLINK-4618] - FlinkKafkaConsumer09 should start from the next record on startup from offsets in Kafka
  • [FLINK-4619] - JobManager does not answer to client when restore from savepoint fails
  • [FLINK-4622] - CLI help message should include 'savepoint' action
  • [FLINK-4624] - Gelly's summarization algorithm cannot deal with null vertex group values
  • [FLINK-4626] - Missing break in MetricStore#add()
  • [FLINK-4628] - User class loader unavailable during input split assignment
  • [FLINK-4631] - NullPointerException during stream task cleanup
  • [FLINK-4640] - Serialization of the initialValue of a Fold on WindowedStream fails
  • [FLINK-4645] - Hard to register Kryo Serializers due to generics
  • [FLINK-4651] - Re-register processing time timers at the WindowOperator upon recovery.
  • [FLINK-4652] - Don't pass credentials explicitly to AmazonClient - use credentials provider instead
  • [FLINK-4659] - Potential resource leak due to unclosed InputStream in SecurityContext#populateSystemSecurityProperties()
  • [FLINK-4663] - Flink JDBCOutputFormat logs wrong WARN message
  • [FLINK-4665] - Remove boxing/unboxing to parse a primitive
  • [FLINK-4666] - Make constants to be final in ParameterTool
  • [FLINK-4667] - Yarn Session CLI not listening on correct ZK namespace when HA is enabled to use ZooKeeper backend
  • [FLINK-4671] - Table API can not be built
  • [FLINK-4672] - TaskManager accidentally decorates Kill messages
  • [FLINK-4677] - Jars with no job executions produces NullPointerException in ClusterClient
  • [FLINK-4696] - Limit the number of Akka Dispatcher Threads in LocalMiniCluster
  • [FLINK-4700] - Harden the TimeProvider test
  • [FLINK-4702] - Kafka consumer must commit offsets asynchronously
  • [FLINK-4708] - Scope Mini Kerberos Cluster dependencies as test dependencies
  • [FLINK-4709] - InputStreamFSInputWrapper does not close wrapped stream
  • [FLINK-4710] - Remove transitive Guice dependency from Hadoop
  • [FLINK-4711] - TaskManager can crash due to failing onPartitionStateUpdate call
  • [FLINK-4718] - Confusing label in Parallel Streams Diagram
  • [FLINK-4727] - Kafka 0.9 Consumer should also checkpoint auto retrieved offsets even when no data is read
  • [FLINK-4730] - Introduce CheckpointMetaData
  • [FLINK-4731] - HeapKeyedStateBackend restoring broken for scale-in
  • [FLINK-4732] - Maven junction plugin security threat
  • [FLINK-4741] - WebRuntimeMonitor does not shut down all of it's threads (EventLoopGroups) on exit.
  • [FLINK-4743] - The sqrt/power function not accept the real data types.
  • [FLINK-4744] - Introduce usercode class loader to deserialize partitionable operator state
  • [FLINK-4747] - Instability due to pending processing timers on operator close
  • [FLINK-4753] - Kafka 0.8 connector's Periodic Offset Committer should synchronize on checkpoint lock
  • [FLINK-4774] - QueryScopeInfo scope concatenation broken
  • [FLINK-4777] - ContinuousFileMonitoringFunction may throw IOException when files are moved
  • [FLINK-4784] - MetricQueryService actor name collision
  • [FLINK-4786] - BarrierBufferTest test instability
  • [FLINK-4788] - State backend class cannot be loaded, because fully qualified name converted to lower-case
  • [FLINK-4791] - Fix issues caused by expression reduction
  • [FLINK-4793] - Using a local method with :: notation in Java 8 causes index out of bounds
  • [FLINK-4794] - partition_by_hash() crashes if no parameter is provided
  • [FLINK-4795] - CsvStringify crashes in case of tuple in tuple, t.e. ("a", True, (1,5))
  • [FLINK-4798] - CEPITCase.testSimpleKeyedPatternCEP test failure
  • [FLINK-4801] - Input type inference is faulty with custom Tuples and RichFunctions
  • [FLINK-4804] - Grouping.first() function usage fails
  • [FLINK-4805] - Stringify() crashes with Python3 (run with pyflink3)
  • [FLINK-4824] - CliFrontend shows misleading error message when main() method returns before env.execute()
  • [FLINK-4827] - The scala example of SQL on Streaming Tables with wrong variable name in flink document
  • [FLINK-4829] - Accumulators are not thread safe
  • [FLINK-4833] - Unstable test OperatorStatsAccumulatorTest.testAccumulatorHeavyHitterCountMinSketch
  • [FLINK-4838] - remove STREAM keyword in StreamSQLExample
  • [FLINK-4845] - Fix Job Exceptions page
  • [FLINK-4852] - ClassCastException when assigning Watermarks with TimeCharacteristic.ProcessingTime
  • [FLINK-4862] - NPE on EventTimeSessionWindows with ContinuousEventTimeTrigger
  • [FLINK-4866] - Make Trigger.clear() Abstract to Enforce Implementation
  • [FLINK-4870] - ContinuousFileMonitoringFunction does not properly handle absolut Windows paths
  • [FLINK-4872] - Type erasure problem exclusively on cluster execution
  • [FLINK-4875] - operator name not correctly inferred
  • [FLINK-4879] - class KafkaTableSource should be public just like KafkaTableSink
  • [FLINK-4887] - Replace ActorGateway by TaskManagerGateway interface
  • [FLINK-4890] - FileInputFormatTest#testExcludeFiles fails on Windows OS
  • [FLINK-4907] - Add Test for Timers/State Provided by AbstractStreamOperator
  • [FLINK-4932] - Don't let ExecutionGraph fail when in state Restarting
  • [FLINK-4933] - ExecutionGraph.scheduleOrUpdateConsumers can fail the ExecutionGraph
  • [FLINK-4941] - Show ship strategy in web interface
  • [FLINK-4945] - KafkaConsumer logs wrong warning about confirmation for unknown checkpoint
  • [FLINK-4967] - RockDB state backend fails on Windows
  • [FLINK-4973] - Flakey Yarn tests due to recently added latency marker
  • [FLINK-4977] - Enum serialization does not work in all cases
  • [FLINK-4991] - TestTask hangs in testWatchDogInterruptsTask
  • [FLINK-4998] - ResourceManager fails when num task slots > Yarn vcores
  • [FLINK-5002] - Lack of synchronization in LocalBufferPool#getNumberOfUsedBuffers
  • [FLINK-5007] - Retain externalized checkpoint on suspension
  • [FLINK-5011] - TraversableSerializer does not perform a deep copy of the elements it is traversing
  • [FLINK-5014] - RocksDBStateBackend misses good toString
  • [FLINK-5016] - EventTimeWindowCheckpointingITCase testTumblingTimeWindowWithKVStateMaxMaxParallelism with RocksDB hangs
  • [FLINK-5019] - Proper isRestored result for tasks that did not write state
  • [FLINK-5027] - FileSource finishes successfully with a wrong path
  • [FLINK-5028] - Stream Tasks must not go through clean shutdown logic on cancellation
  • [FLINK-5032] - CsvOutputFormatTest fails on Windows OS
  • [FLINK-5033] - CEP operators don't properly advance time
  • [FLINK-5037] - Instability in AbstractUdfStreamOperatorLifecycleTest
  • [FLINK-5038] - Errors in the "cancelTask" method prevent closeables from being closed early
  • [FLINK-5039] - Avro GenericRecord support is broken
  • [FLINK-5040] - Set correct input channel types with eager scheduling
  • [FLINK-5048] - Kafka Consumer (0.9/0.10) threading model leads problematic cancellation behavior
  • [FLINK-5049] - Instability in QueryableStateITCase.testQueryableStateWithTaskManagerFailure
  • [FLINK-5050] - JSON.org license is CatX
  • [FLINK-5055] - Security feature crashes JM for certain Hadoop versions even though using no Kerberos
  • [FLINK-5056] - BucketingSink deletes valid data when checkpoint notification is slow.
  • [FLINK-5057] - Cancellation timeouts are picked from wrong config
  • [FLINK-5058] - taskManagerMemory attribute set wrong value in FlinkShell
  • [FLINK-5063] - State handles are not properly cleaned up for declined or expired checkpoints
  • [FLINK-5071] - YARN: yarn.containers.vcores config not respected when checking for vcores
  • [FLINK-5073] - ZooKeeperCompleteCheckpointStore executes blocking delete operation in ZooKeeper client thread
  • [FLINK-5075] - Kinesis consumer incorrectly determines shards as newly discovered when tested against Kinesalite
  • [FLINK-5082] - Pull ExecutionService lifecycle management out of the JobManager
  • [FLINK-5085] - Execute CheckpointCoodinator's state discard calls asynchronously
  • [FLINK-5097] - The TypeExtractor is missing input type information in some Graph methods
  • [FLINK-5102] - Connection establishment does not react to interrupt
  • [FLINK-5109] - Invalid Content-Encoding Header in REST API responses
  • [FLINK-5111] - Change the assignToKeyGroup() method to not use Object.hashCode()
  • [FLINK-5112] - Remove unused accumulator code from ArchivedExecutionJobVertex
  • [FLINK-5114] - PartitionState update with finished execution fails
  • [FLINK-5118] - Inconsistent records sent/received metrics
  • [FLINK-5119] - Last taskmanager heartbeat not showing in web frontend
  • [FLINK-5143] - Add EXISTS to list of supported operators
  • [FLINK-5144] - Error while applying rule AggregateJoinTransposeRule
  • [FLINK-5147] - StreamingOperatorsITCase.testGroupedFoldOperation failed on Travis
  • [FLINK-5149] - ContinuousEventTimeTrigger doesn't fire at the end of the window
  • [FLINK-5150] - WebUI metric-related resource leak
  • [FLINK-5154] - Duplicate TypeSerializer when writing RocksDB Snapshot
  • [FLINK-5158] - Handle ZooKeeperCompletedCheckpointStore exceptions in CheckpointCoordinator
  • [FLINK-5160] - SecurityContextTest#testCreateInsecureHadoopCtx fails on windows
  • [FLINK-5162] - BlobRecoveryItCase#testBlobServerRecovery fails on Windows
  • [FLINK-5164] - Hadoop-compat IOFormat tests fail on Windows
  • [FLINK-5173] - Upgrade RocksDB dependency
  • [FLINK-5179] - MetricRegistry life-cycle issues with HA
  • [FLINK-5184] - Error result of compareSerialized in RowComparator class
  • [FLINK-5193] - Recovering all jobs fails completely if a single recovery fails
  • [FLINK-5197] - Late JobStatusChanged messages can interfere with running jobs
  • [FLINK-5206] - Flakey PythonPlanBinderTest
  • [FLINK-5209] - Fix TaskManager metrics
  • [FLINK-5214] - Clean up checkpoint files when failing checkpoint operation on TM
  • [FLINK-5216] - CheckpointCoordinator's 'minPauseBetweenCheckpoints' refers to checkpoint start rather then checkpoint completion
  • [FLINK-5218] - Eagerly close checkpoint streams on cancellation
  • [FLINK-5221] - Checkpointed workless in Window Operator
  • [FLINK-5228] - LocalInputChannel re-trigger request and release deadlock
  • [FLINK-5229] - Cleanup StreamTaskStates if a checkpoint operation of a subsequent operator fails
  • [FLINK-5247] - Fix incorrect check in allowedLateness() method. Make it a no-op for non-event time windows.
  • [FLINK-5248] - SavepointITCase doesn't catch savepoint restore failure
  • [FLINK-5261] - ScheduledDropwizardReporter does not properly clean up metrics
  • [FLINK-5267] - TaskManager logs not scrollable
  • [FLINK-5274] - LocalInputChannel throws NPE if partition reader is released
  • [FLINK-5275] - InputChanelDeploymentDescriptors throws misleading Exception if producer failed/cancelled
  • [FLINK-5285] - CancelCheckpointMarker flood when using at least once mode
  • [FLINK-5289] - NPE when using value state on non-keyed stream
  • [FLINK-5292] - Make the operators for 1.1->1.2 backwards compatible.
  • [FLINK-5296] - Expose the old AlignedWindowOperators to the user through explicit commands.
  • [FLINK-5298] - TaskManager crashes when TM log not existant
  • [FLINK-5307] - Log configuration for every reporter
  • [FLINK-5320] - Fix result TypeInformation in WindowedStream.fold(ACC, FoldFunction, WindowFunction)
  • [FLINK-5321] - FlinkMiniCluster does not start Jobmanager MetricQueryService
  • [FLINK-5323] - CheckpointNotifier should be removed from docs
  • [FLINK-5326] - IllegalStateException: Bug in Netty consumer logic: reader queue got notified by partition about available data, but none was available
  • [FLINK-5329] - Metric list is being cut off in the WebFrontend
  • [FLINK-5330] - Harden KafkaConsumer08Test
  • [FLINK-5332] - Non-thread safe FileSystem::initOutPathLocalFS() can cause lost files/directories in local execution
  • [FLINK-5344] - docs don't build in dockerized jekyll; -p option is broken
  • [FLINK-5345] - IOManager failed to properly clean up temp file directory
  • [FLINK-5349] - Fix code sample for Twitter connector
  • [FLINK-5350] - Don't overwrite existing Jaas config property
  • [FLINK-5356] - librocksdbjni*.so is not deleted from /tmp directory
  • [FLINK-5357] - WordCountTable fails
  • [FLINK-5359] - Job Exceptions view doesn't scroll
  • [FLINK-5361] - Flink shouldn't require Kerberos credentials
  • [FLINK-5364] - Rework JAAS configuration to support user-supplied entries
  • [FLINK-5365] - Mesos AppMaster/TaskManager should obey sigterm
  • [FLINK-5369] - Rework jsr305 and logging dependencies
  • [FLINK-5375] - Fix watermark documentation
  • [FLINK-5379] - Flink CliFrontend does not return when not logged in with kerberos
  • [FLINK-5380] - Number of outgoing records not reported in web interface
  • [FLINK-5381] - Scrolling in some web interface pages doesn't work (taskmanager details, jobmanager config)
  • [FLINK-5382] - Taskmanager log download button causes 404
  • [FLINK-5392] - flink-dist build failed when change scala version to 2.11
  • [FLINK-5394] - the estimateRowCount method of DataSetCalc didn't work
  • [FLINK-5397] - Fail to deserialize savepoints in v1.1 when there exist missing fields in class serialization descriptors
  • [FLINK-5407] - Savepoint for iterative Task fails.
  • [FLINK-5408] - RocksDB initialization can fail with an UnsatisfiedLinkError in the presence of multiple classloaders
  • [FLINK-5417] - Fix the wrong config file name
  • [FLINK-5418] - Estimated row size does not support nested types
  • [FLINK-5419] - Taskmanager metrics not accessible via REST
  • [FLINK-5427] - Typo in the event_timestamps_watermarks doc
  • [FLINK-5432] - ContinuousFileMonitoringFunction is not monitoring nested files
  • [FLINK-5434] - Remove unsupported project() transformation from Scala DataStream docs
  • [FLINK-5443] - Create a path to migrate from the Rolling to the BucketingSink.
  • [FLINK-5444] - Flink UI uses absolute URLs.
  • [FLINK-5450] - WindowOperator logs about "re-registering state from an older Flink version" even though its not a restored window
  • [FLINK-5464] - MetricQueryService throws NullPointerException on JobManager
  • [FLINK-5467] - Stateless chained tasks set legacy operator state
  • [FLINK-5468] - Restoring from a semi async rocksdb statebackend (1.1) to 1.2 fails with ClassNotFoundException
  • [FLINK-5470] - Requesting non-existing log/stdout file from TM crashes the TM
  • [FLINK-5473] - setMaxParallelism() higher than 1 is possible on non-parallel operators
  • [FLINK-5482] - QueryableStateClient does not recover from a failed lookup due to a non-running job
  • [FLINK-5483] - Link to "linking modules not contained in binary distribution" broken in all connector documentations
  • [FLINK-5484] - Kryo serialization changed between 1.1 and 1.2
  • [FLINK-5492] - BootstrapTools log wrong address of started ActorSystem
  • [FLINK-5493] - FlinkDistributionOverlay does not properly display missing environment variables
  • [FLINK-5495] - ZooKeeperMesosWorkerStore cannot be instantiated
  • [FLINK-5496] - ClassCastException when using Mesos HA mode
  • [FLINK-5504] - mesos-appmaster.sh logs to wrong directory
  • [FLINK-5518] - HadoopInputFormat throws NPE when close() is called before open()
  • [FLINK-5520] - Disable outer joins with non-equality predicates
  • [FLINK-5530] - race condition in AbstractRocksDBState#getSerializedValue
  • [FLINK-5531] - SSl code block formatting is broken
  • [FLINK-5532] - Make the marker WindowAssigners for the fast aligned windows non-extendable.
  • [FLINK-5549] - TypeExtractor fails with RuntimeException, but should use GenericTypeInfo
  • [FLINK-5556] - BarrierBuffer resets bytes written on spiller roll over
  • [FLINK-5560] - Header in checkpoint stats summary misaligned
  • [FLINK-5561] - DataInputDeserializer#available returns one too few
  • [FLINK-5562] - Driver fixes
  • [FLINK-5578] - Each time application is submitted to yarn, application id increases by two
  • [FLINK-5580] - Kerberos keytabs not working for YARN deployment mode
  • [FLINK-5585] - NullPointer Exception in JobManager.updateAccumulators
  • [FLINK-5602] - Migration with RocksDB job led to NPE for next checkpoint
  • [FLINK-5612] - GlobPathFilter not-serializable exception
  • [FLINK-5613] - QueryableState: requesting a non-existing key in RocksDBStateBackend is not consistent with the MemoryStateBackend and FsStateBackend
  • [FLINK-5617] - Check new public APIs in 1.2 release
  • [FLINK-5626] - Improve resource release in RocksDBKeyedStateBackend
  • [FLINK-5637] - Default Flink configuration contains whitespace characters, causing parser WARNings
  • [FLINK-5638] - Deadlock when closing two chained async I/O operators
  • [FLINK-5643] - StateUtil.discardStateFuture fails when state future contains null value
  • [FLINK-5660] - Not properly cleaning PendingCheckpoints up
  • [FLINK-5663] - Checkpoint fails because of closed registry
  • [FLINK-5666] - Blob files are not cleaned up from ZK storage directory
  • [FLINK-5667] - Possible state data loss when task fails while checkpointing
  • [FLINK-5670] - Local RocksDB directories not cleaned up

Improvement

  • [FLINK-2090] - toString of CollectionInputFormat takes long time when the collection is huge
  • [FLINK-2765] - Upgrade hbase version for hadoop-2 to 1.2 release
  • [FLINK-3042] - Define a way to let types create their own TypeInformation
  • [FLINK-3347] - TaskManager (or its ActorSystem) need to restart in case they notice quarantine
  • [FLINK-3615] - Add support for non-native SQL types
  • [FLINK-3670] - Kerberos: Improving long-running streaming jobs
  • [FLINK-3677] - FileInputFormat: Allow to specify include/exclude file name patterns
  • [FLINK-3702] - DataStream API PojoFieldAccessor doesn't support nested POJOs
  • [FLINK-3719] - WebInterface: Moving the barrier between graph and stats
  • [FLINK-3779] - Add support for queryable state
  • [FLINK-3787] - Yarn client does not report unfulfillable container constraints
  • [FLINK-3866] - StringArraySerializer claims type is immutable; shouldn't
  • [FLINK-3869] - WindowedStream.apply with FoldFunction is too restrictive
  • [FLINK-3899] - Document window processing with Reduce/FoldFunction + WindowFunction
  • [FLINK-3904] - GlobalConfiguration doesn't ensure config has been loaded
  • [FLINK-3921] - StringParser not specifying encoding to use
  • [FLINK-4023] - Move Kafka consumer partition discovery from constructor to open()
  • [FLINK-4037] - Introduce ArchivedExecutionGraph without any user classes
  • [FLINK-4068] - Move constant computations out of code-generated `flatMap` functions.
  • [FLINK-4104] - Restructure Gelly docs
  • [FLINK-4129] - Remove the example HITSAlgorithm
  • [FLINK-4161] - Quickstarts can exclude more flink-dist dependencies
  • [FLINK-4179] - Update TPCHQuery3Table example
  • [FLINK-4189] - Introduce symbols for internal use
  • [FLINK-4190] - Generalise RollingSink to work with arbitrary buckets
  • [FLINK-4203] - Improve Table API documentation
  • [FLINK-4204] - Clean up gelly-examples
  • [FLINK-4218] - Sporadic "java.lang.RuntimeException: Error triggering a checkpoint..." causes task restarting
  • [FLINK-4222] - Allow Kinesis configuration to get credentials from AWS Metadata
  • [FLINK-4241] - Cryptic expression parser exceptions
  • [FLINK-4242] - Improve validation exception messages
  • [FLINK-4245] - Metric naming improvements
  • [FLINK-4247] - CsvTableSource.getDataSet() expects Java ExecutionEnvironment
  • [FLINK-4248] - CsvTableSource does not support reading SqlTimeTypeInfo types
  • [FLINK-4251] - Add possiblity for the RMQ Streaming Sink to customize the queue
  • [FLINK-4253] - Rename "recovery.mode" config key to "high-availability"
  • [FLINK-4260] - Allow SQL's LIKE ESCAPE
  • [FLINK-4268] - Add a parsers for BigDecimal/BigInteger
  • [FLINK-4281] - Wrap all Calcite Exceptions in Flink Exceptions
  • [FLINK-4282] - Add Offset Parameter to WindowAssigners
  • [FLINK-4299] - Show loss of job manager in Client
  • [FLINK-4302] - Add JavaDocs to MetricConfig
  • [FLINK-4306] - Fix Flink and Storm dependencies in flink-storm and flink-storm-examples
  • [FLINK-4308] - Allow uploaded jar directory to be configurable
  • [FLINK-4310] - Move BinaryCompatibility Check plugin to relevant projects
  • [FLINK-4318] - Make master docs build target version-specific
  • [FLINK-4322] - Unify CheckpointCoordinator and SavepointCoordinator
  • [FLINK-4340] - Remove RocksDB Semi-Async Checkpoint Mode
  • [FLINK-4366] - Enforce parallelism=1 For AllWindowedStream
  • [FLINK-4369] - EvictingWindowOperator Must Actually Evict Elements
  • [FLINK-4378] - Enable RollingSink to custom HDFS client configuration
  • [FLINK-4396] - GraphiteReporter class not found at startup of jobmanager
  • [FLINK-4410] - Report more information about operator checkpoints
  • [FLINK-4431] - Introduce a "VisibleForTesting" annotation
  • [FLINK-4435] - Replace Guava's VisibleForTesting annotation with Flink's annotation
  • [FLINK-4439] - Error message KafkaConsumer08 when all 'bootstrap.servers' are invalid
  • [FLINK-4445] - Ignore unmatched state when restoring from savepoint
  • [FLINK-4447] - Include NettyConfig options on Configurations page
  • [FLINK-4457] - Make the ExecutionGraph independent of Akka
  • [FLINK-4459] - Introduce SlotProvider for Scheduler
  • [FLINK-4478] - Implement heartbeat logic
  • [FLINK-4495] - Running multiple jobs on yarn (without yarn-session)
  • [FLINK-4525] - Drop the "eager split pre-assignment" code paths
  • [FLINK-4527] - Drop the "flinkAccumulators" from the Execution
  • [FLINK-4539] - Duplicate/inconsistent logic for physical memory size in classes "Hardware" and "EnvironmentInformation"
  • [FLINK-4541] - Support for SQL NOT IN operator
  • [FLINK-4543] - Race Deadlock in SpilledSubpartitionViewTest
  • [FLINK-4549] - Test and document implicitly supported SQL functions
  • [FLINK-4550] - Clearly define SQL operator table
  • [FLINK-4552] - Refactor WindowOperator/Trigger Tests
  • [FLINK-4560] - enforcer java version as 1.7
  • [FLINK-4579] - Add StateBackendFactory for RocksDB Backend
  • [FLINK-4599] - Add 'explain()' also to StreamTableEnvironment
  • [FLINK-4611] - Make "AUTO" credential provider as default for Kinesis Connector
  • [FLINK-4623] - Create Physical Execution Plan of a DataStream
  • [FLINK-4625] - Guard Flink processes against blocking shutdown hooks
  • [FLINK-4638] - Fix exception message for MemorySegment
  • [FLINK-4654] - clean up docs
  • [FLINK-4662] - Bump Calcite version up to 1.9
  • [FLINK-4668] - Fix positive random int generation
  • [FLINK-4669] - scala api createLocalEnvironment() function add default Configuration parameter
  • [FLINK-4684] - Remove obsolete classloader from CheckpointCoordinator
  • [FLINK-4704] - Move Table API to org.apache.flink.table
  • [FLINK-4715] - TaskManager should commit suicide after cancellation failure
  • [FLINK-4720] - Implement an archived version of the execution graph
  • [FLINK-4723] - Unify behaviour of committed offsets to Kafka / ZK for Kafka 0.8 and 0.9 consumer
  • [FLINK-4728] - Replace reference equality with object equality
  • [FLINK-4729] - Use optional VertexCentric CombineFunction
  • [FLINK-4733] - Port WebFrontend to new metric system
  • [FLINK-4734] - Remove use of Tuple setField for fixed position
  • [FLINK-4736] - Don't duplicate fields in Ordering
  • [FLINK-4737] - Add more compression algorithms to FileInputFormat
  • [FLINK-4739] - Adding packaging details for the Elasticsearch connector
  • [FLINK-4745] - Convert KafkaTableSource test to unit tests
  • [FLINK-4751] - Extend Flink's futures to support combining two futures
  • [FLINK-4752] - Improve session window documentation
  • [FLINK-4762] - Use plural in time interval units
  • [FLINK-4764] - Introduce config options
  • [FLINK-4771] - Compression for AvroOutputFormat
  • [FLINK-4772] - Store metrics in MetricStore as strings
  • [FLINK-4773] - Introduce an OperatorIOMetricGroup
  • [FLINK-4775] - Simplify access to MetricStore
  • [FLINK-4776] - Move ExecutionGraph initialization into a dedicated class
  • [FLINK-4780] - Ability to use UDP protocol in metrics-graphite
  • [FLINK-4787] - Add REST API call for cancel-with-savepoints
  • [FLINK-4792] - Update documentation - FlinkML/QuickStart Guide
  • [FLINK-4825] - Implement a RexExecutor that uses Flink's code generation
  • [FLINK-4832] - Count/Sum 0 elements
  • [FLINK-4876] - Allow web interface to be bound to a specific ip/interface/inetHost
  • [FLINK-4881] - Docker: Remove dependency on shared volumes
  • [FLINK-4892] - Snapshot TimerService using Key-Grouped State
  • [FLINK-4894] - Don't block on buffer request after broadcastEvent
  • [FLINK-4906] - Use constants for the name of system-defined metrics
  • [FLINK-4913] - Per-job Yarn clusters: include user jar in system class loader
  • [FLINK-4917] - Deprecate "CheckpointedAsynchronously" interface
  • [FLINK-4923] - Expose input/output buffers and bufferPool usage as a metric for a Task
  • [FLINK-4924] - Simplify Operator Test Harness Constructors
  • [FLINK-4925] - Integrate meters into IOMetricGroups
  • [FLINK-4935] - Submit job with savepoint via REST API
  • [FLINK-4936] - Operator names for Gelly inputs
  • [FLINK-4939] - GenericWriteAheadSink: Decouple the creating from the committing subtask for a pending checkpoint
  • [FLINK-4943] - flink-mesos/ConfigConstants: Typo: YYARN -> YARN
  • [FLINK-4946] - Load jar files from subdirectories of lib
  • [FLINK-4963] - Tabulate edge direction for directed VertexMetrics
  • [FLINK-4970] - Parameterize vertex value for SSSP
  • [FLINK-4972] - CoordinatorShutdownTest relies on race condition for success
  • [FLINK-4975] - Add a limit for how much data may be buffered during checkpoint alignment
  • [FLINK-4983] - Web UI: Add favicon
  • [FLINK-4996] - Make CrossHint @Public
  • [FLINK-5000] - Rename Methods in ManagedInitializationContext
  • [FLINK-5003] - Provide Access to State Stores in Operator Snapshot Context
  • [FLINK-5004] - Add task manager option to disable queryable state server
  • [FLINK-5008] - Update quickstart documentation
  • [FLINK-5010] - Decouple the death watch parameters from the Akka ask timeout
  • [FLINK-5012] - Provide Timestamp in TimelyFlatMapFunction
  • [FLINK-5020] - Make the GenericWriteAheadSink rescalable.
  • [FLINK-5021] - Makes the ContinuousFileReaderOperator rescalable.
  • [FLINK-5022] - Suppress RejectedExecutionException when the Executor is shut down
  • [FLINK-5026] - Rename TimelyFlatMap to Process
  • [FLINK-5034] - Don't Write StateDescriptor to RocksDB Snapshot
  • [FLINK-5035] - Don't Write TypeSerializer to Heap State Snapshot
  • [FLINK-5046] - Avoid redundant serialization when creating the TaskDeploymentDescriptor
  • [FLINK-5054] - Make the BucketingSink rescalable.
  • [FLINK-5096] - Make the RollingSink rescalable.
  • [FLINK-5110] - Remove the AbstractAlignedProcessingTimeWindowOperator.
  • [FLINK-5113] - Make all Testing Functions implement CheckpointedFunction Interface.
  • [FLINK-5123] - Add description how to do proper shading to Flink docs.
  • [FLINK-5124] - Support more temporal arithmetic
  • [FLINK-5128] - Get Kafka partitions in FlinkKafkaProducer only if a partitioner is set
  • [FLINK-5145] - WebInterface to aggressive in pulling metrics
  • [FLINK-5146] - Improved resource cleanup in RocksDB keyed state backend
  • [FLINK-5155] - Deprecate ValueStateDescriptor constructors with default value
  • [FLINK-5159] - Improve perfomance of inner joins with a single row input
  • [FLINK-5163] - Make the production functions rescalable (apart from the Rolling/Bucketing Sinks)
  • [FLINK-5168] - Scaladoc annotation link use [[]] instead of {@link}
  • [FLINK-5169] - Make consumption of input channels fair
  • [FLINK-5181] - Add Tests in StateBackendTestBase that verify Default-Value Behaviour
  • [FLINK-5185] - Decouple BatchTableSourceScan with TableSourceTable
  • [FLINK-5186] - Move Row and RowTypeInfo into Flink core
  • [FLINK-5192] - Provide better log config templates
  • [FLINK-5194] - Log heartbeats on TRACE level
  • [FLINK-5196] - Don't log InputChannelDescriptor
  • [FLINK-5198] - Overwrite TaskState toString
  • [FLINK-5199] - Improve logging of submitted job graph actions in HA case
  • [FLINK-5201] - Promote loaded config properties to INFO
  • [FLINK-5207] - Decrease HadoopFileSystem logging
  • [FLINK-5211] - Include an example configuration for all reporters
  • [FLINK-5213] - Missing @Override in Task
  • [FLINK-5223] - Add documentation of UDTF in Table API & SQL
  • [FLINK-5224] - Improve UDTF: emit rows directly instead of buffering them
  • [FLINK-5226] - Eagerly project unused attributes
  • [FLINK-5237] - Consolidate and harmonize Window Translation Tests
  • [FLINK-5240] - Properly Close StateBackend in StreamTask when closing/canceling
  • [FLINK-5249] - description of datastream rescaling doesn't match the figure
  • [FLINK-5250] - Make AbstractUdfStreamOperator aware of WrappingFunction
  • [FLINK-5251] - Decouple StreamTableSourceScan with TableSourceTable
  • [FLINK-5255] - Improve single row check in DataSetSingleRowJoinRule
  • [FLINK-5257] - Display optimized logical plan when explaining table
  • [FLINK-5258] - reorganize the docs to improve navigation and reduce duplication
  • [FLINK-5259] - wrong execution environment in retry delays example
  • [FLINK-5266] - Eagerly project unused fields when selecting aggregation fields
  • [FLINK-5278] - Improve Task and checkpoint logging
  • [FLINK-5280] - Refactor TableSource
  • [FLINK-5304] - Change method name from crossApply to join in Table API
  • [FLINK-5306] - Display checkpointing configuration details in web UI "Configuration" tab
  • [FLINK-5310] - Harden the RocksDB JNI library loading
  • [FLINK-5311] - Write user documentation for BipartiteGraph
  • [FLINK-5327] - Remove IOException from StateObject::getStateSize
  • [FLINK-5335] - Allow ListCheckpointed user functions to return null
  • [FLINK-5343] - Add more option to CsvTableSink
  • [FLINK-5366] - Add end-to-end tests for Savepoint Backwards Compatibility
  • [FLINK-5367] - restore updates lost when merging recent doc refactoring
  • [FLINK-5368] - Let Kafka consumer show something when it fails to read one topic out of topic list
  • [FLINK-5371] - Add documentation for async I/O
  • [FLINK-5377] - Improve savepoint docs
  • [FLINK-5395] - support locally build distribution by script create_release_files.sh
  • [FLINK-5404] - Consolidate and update S3 documentation
  • [FLINK-5412] - Enable RocksDB tests on Windows
  • [FLINK-5421] - Explicit restore method in Snapshotable
  • [FLINK-5424] - Improve Restart Strategy Logging
  • [FLINK-5430] - Bring documentation up to speed for current feature set
  • [FLINK-5438] - Typo in JobGraph generator Exception
  • [FLINK-5442] - Add test to fix ordinals of serialized enum StateDescriptor.Type
  • [FLINK-5446] - System metrics reference incomplete
  • [FLINK-5447] - Sync documentation of built-in functions for Table API with SQL
  • [FLINK-5448] - Fix typo in StateAssignmentOperation Exception
  • [FLINK-5451] - Extend JMX metrics documentation section
  • [FLINK-5452] - Make table unit tests pass under cluster mode
  • [FLINK-5466] - Make production environment default in gulpfile
  • [FLINK-5480] - User-provided hashes for operators
  • [FLINK-5485] - Mark compiled web frontend files as binary when processed by git diff
  • [FLINK-5507] - remove queryable list state sink
  • [FLINK-5508] - Remove Mesos dynamic class loading
  • [FLINK-5512] - RabbitMQ documentation should inform that exactly-once holds for RMQSource only when parallelism is 1
  • [FLINK-5515] - fix unused kvState.getSerializedValue call in KvStateServerHandler
  • [FLINK-5521] - remove unused KvStateRequestSerializer#serializeList
  • [FLINK-5528] - tests: reduce the retry delay in QueryableStateITCase
  • [FLINK-5557] - Fix link in library methods
  • [FLINK-5559] - queryable state: KvStateRequestSerializer#deserializeKeyAndNamespace() throws an IOException without own failure message if deserialisation fails
  • [FLINK-5574] - Add checkpoint statistics docs
  • [FLINK-5576] - extend deserialization functions of KvStateRequestSerializer to detect unconsumed bytes
  • [FLINK-5609] - Add last update time to docs
  • [FLINK-5610] - Rename Installation and Setup to Project Setup
  • [FLINK-5615] - queryable state: execute the QueryableStateITCase for all three state back-ends

New Feature

  • [FLINK-1984] - Integrate Flink with Apache Mesos
  • [FLINK-3060] - Add possibility to integrate custom types into the TypeExtractor
  • [FLINK-3097] - Add support for custom functions in Table API
  • [FLINK-3239] - Support for Kerberos enabled Kafka 0.9.0.0
  • [FLINK-3674] - Add an interface for Time aware User Functions
  • [FLINK-3755] - Introduce key groups for key-value state to support dynamic scaling
  • [FLINK-3848] - Add ProjectableTableSource interface and translation rule
  • [FLINK-3874] - Add a Kafka TableSink with JSON serialization
  • [FLINK-3929] - Support for Kerberos Authentication with Keytab Credential
  • [FLINK-3931] - Implement Transport Encryption (SSL/TLS)
  • [FLINK-3932] - Implement State Backend Security
  • [FLINK-3940] - Add support for ORDER BY OFFSET FETCH
  • [FLINK-4180] - Create a batch SQL example
  • [FLINK-4181] - Add a basic streaming Table API example
  • [FLINK-4294] - Allow access of composite type fields
  • [FLINK-4359] - Add INTERVAL type
  • [FLINK-4379] - Add Rescalable Non-Partitioned State
  • [FLINK-4391] - Provide support for asynchronous operations over streams
  • [FLINK-4420] - Introduce star(*) to select all of the columns in the table
  • [FLINK-4469] - Add support for user defined table function in Table API & SQL
  • [FLINK-4546] - Remove STREAM keyword in Stream SQL
  • [FLINK-4554] - Add support for array types
  • [FLINK-4576] - Low Watermark Service in JobManager for Streaming Sources
  • [FLINK-4639] - Make Calcite features more pluggable
  • [FLINK-4643] - Average Clustering Coefficient
  • [FLINK-4664] - Add translator to NullValue
  • [FLINK-4717] - Naive version of atomic stop signal with savepoint
  • [FLINK-4844] - Partitionable Raw Keyed/Operator State
  • [FLINK-4861] - Package optional project artifacts
  • [FLINK-4934] - Triadic Census
  • [FLINK-4960] - Allow the AbstractStreamOperatorTestHarness to test scaling down
  • [FLINK-4976] - Add a way to abort in flight checkpoints
  • [FLINK-5265] - Introduce state handle replication mode for CheckpointCoordinator

Task

  • [FLINK-4315] - Deprecate Hadoop dependent methods in flink-java
  • [FLINK-4429] - Move Redis Sink from Flink to Bahir
  • [FLINK-4446] - Move Flume Sink from Flink to Bahir
  • [FLINK-4676] - Merge flink-batch-connectors and flink-streaming-connectors modules
  • [FLINK-4778] - Update program example in /docs/setup/cli.md due to the change in FLINK-2021
  • [FLINK-4891] - Remove flink-contrib/flink-operator-stats
  • [FLINK-4895] - Drop support for Hadoop 1

Test

  • [FLINK-4842] - Introduce test to enforce order of operator / udf lifecycles
  • [FLINK-4843] - Introduce Test for FsCheckpointStateOutputStream::getPos

Wish

  • [FLINK-4799] - Re-add build-target symlink to project root