Bulk Operation

FLINK-4256 Set jobmanager.execution.failover-strategy to region in default flink-conf.yaml

FLINK-4256 Add documentation for AdaptedRestartPipelinedRegionStrategyNG

FLINK-4256 IT test for fine-grained recovery (task executor failures)

FLINK-4256 Run HA dataset E2E test with new RestartPipelinedRegionStrategy

FLINK-4256 Leverage JM side partition state to improve region failover experience

FLINK-4256 IT test for fine-grained recovery (user code failures)

FLINK-4256 Add an adapter of region failover NG for legacy scheduler

FLINK-4256 Implement ExecutionGraph to FailoverTopology Adapter

FLINK-4256 Introduce PartitionConnectionException for unreachable producer

FLINK-4256 Implement a region failover strategy based on new FailoverStrategy interfaces

FLINK-4256 Resetting ExecutionVertex in region failover may cause inconsistency of IntermediateResult status

FLINK-4256 Backtrack failover regions if intermediate results are unavailable

FLINK-4256 Batch Job: InputSplit Fault tolerant for DataSourceTask

FLINK-4256 Introduce a TerminationFuture for Execution

FLINK-4256 Introduce the abstract PartitionException for downstream task failure

FLINK-4256 ExecutionGraph use FailoverCoordinator to manage the failover of execution vertexes

FLINK-4256 Implement a new RestartStrategy that works for the FailoverRegion.

FLINK-4256 The implementation of RestartPipelinedRegionStrategy

FLINK-4256 The implementation of FailoverRegion.

FLINK-4256 Initialize TaskExecutions directly with their starting state

FLINK-4256 Fix misleading ScheduleMode names

FLINK-4256 Remove Serializabiliy of ExecutionGraph