[FLINK-24972] FlinkKafkaProducerITCase testScaleUpAfterScalingDown Timeout due to failed to complete checkpoint

• Type: Bug
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Affects Version/s: 1.14.0
• Component/s: Connectors / Kafka
• Labels: stale-assigned test-stability
• Assignee: Fabian Paul
• Reporter: Yun Gao
• Created: 21/Nov/21 11:35
• Updated: 01/Jan/22 10:39

Description

Nov 20 23:38:33 [ERROR] Tests run: 15, Failures: 10, Errors: 1, Skipped: 0, Time elapsed: 270.661 s <<< FAILURE! - in org.apache.flink.streaming.connectors.kafka.FlinkKafkaProducerITCase
Nov 20 23:38:33 [ERROR] testScaleUpAfterScalingDown  Time elapsed: 64.049 s  <<< ERROR!
Nov 20 23:38:33 org.apache.flink.runtime.checkpoint.CheckpointException: Could not complete snapshot 1 for operator MockTask (2/4)#0. Failure reason: Checkpoint was declined.
Nov 20 23:38:33 	at org.apache.flink.streaming.api.operators.StreamOperatorStateHandler.snapshotState(StreamOperatorStateHandler.java:265) 

After this failure the other tests failed with 

Nov 20 23:38:33 
Nov 20 23:38:33 [ERROR] testFailAndRecoverSameCheckpointTwice  Time elapsed: 5.697 s  <<< FAILURE!
Nov 20 23:38:33 java.lang.AssertionError: Detected producer leak. Thread name: kafka-producer-network-thread | producer-MockTask-000000000000002a000000000000002c-5
Nov 20 23:38:33 	at org.junit.Assert.fail(Assert.java:89) 

Issue Links

• relates to: FLINK-21214 FlinkKafkaProducerITCase.testScaleDownBeforeFirstCheckpoint Failed
• relates to: FLINK-18807 FlinkKafkaProducerITCase.testScaleUpAfterScalingDown failed with "Timeout expired after 60000milliseconds while awaiting EndTxn(COMMIT)"