[FLINK-35722] CoordinatorEventsToStreamOperatorRecipientExactlyOnceITCase.testCheckpoint fails because of missed operator event

• Type: Bug
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Affects Version/s: 2.0.0, 1.20.0
• Component/s: Runtime / Coordination
• Labels: test-stability
• Assignee: Unassigned
• Reporter: Matthias Pohl
• Created: 28/Jun/24 10:34
• Updated: 28/Jun/24 10:40

Description

A test instability in CoordinatorEventsToStreamOperatorRecipientExactlyOnceITCase.testCheckpoint was observed where an expected OperatorEvent was missed:

Test org.apache.flink.streaming.runtime.tasks.CoordinatorEventsToStreamOperatorRecipientExactlyOnceITCase.testCheckpoint[testCheckpoint()] failed with:
java.lang.AssertionError:
Expecting actual:
  [0,
    1,
    3,
    4,
[...]
    98,
    99]
to contain exactly (and in same order):
  [0,
    1,
    2,
    3,
    4,
[...]
but could not find the following elements:
  [2]        at org.apache.flink.runtime.operators.coordination.CoordinatorEventsExactlyOnceITCase.checkListContainsSequence(CoordinatorEventsExactlyOnceITCase.java:175)
        at org.apache.flink.streaming.runtime.tasks.CoordinatorEventsToStreamOperatorRecipientExactlyOnceITCase.executeAndVerifyResults(CoordinatorEventsToStreamOperatorRecipientExactlyOnceITCase.java:178)
        at org.apache.flink.streaming.runtime.tasks.CoordinatorEventsToStreamOperatorRecipientExactlyOnceITCase.testCheckpoint(CoordinatorEventsToStreamOperatorRecipientExactlyOnceITCase.java:124)
        at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) 

The build failure happened on commit 2e853ce39a for FLINK-35552.

I attached the logs for further relevant build artifact and the extract logs for this specific test failure.

Issue Links

• Discovered while testing: FLINK-35552 Move CheckpointStatsTracker out of ExecutionGraph into Scheduler