[jira] [Created] (FLINK-31903) Caching records fails in BroadcastUtils#withBroadcastStream

["Zhipeng Zhang (Jira)" <ji...@apache.org>]

Zhipeng Zhang created FLINK-31903:
-------------------------------------

             Summary: Caching records fails in BroadcastUtils#withBroadcastStream
                 Key: FLINK-31903
                 URL: https://issues.apache.org/jira/browse/FLINK-31903
             Project: Flink
          Issue Type: Bug
          Components: Library / Machine Learning
    Affects Versions: ml-2.3.0
            Reporter: Zhipeng Zhang


When caching more than 1,000,000 records using BroadcastUtils#withBroadcast, it leads to exception as follows:
{code:java}
Caused by: org.apache.flink.util.FlinkRuntimeException: Exceeded checkpoint tolerable failure threshold.
    at org.apache.flink.runtime.checkpoint.CheckpointFailureManager.checkFailureAgainstCounter(CheckpointFailureManager.java:206)
    at org.apache.flink.runtime.checkpoint.CheckpointFailureManager.handleTaskLevelCheckpointException(CheckpointFailureManager.java:191)
    at org.apache.flink.runtime.checkpoint.CheckpointFailureManager.handleCheckpointException(CheckpointFailureManager.java:124)
    at org.apache.flink.runtime.checkpoint.CheckpointCoordinator.abortPendingCheckpoint(CheckpointCoordinator.java:2078)
    at org.apache.flink.runtime.checkpoint.CheckpointCoordinator.receiveDeclineMessage(CheckpointCoordinator.java:1038)
    at org.apache.flink.runtime.scheduler.ExecutionGraphHandler.lambda$declineCheckpoint$2(ExecutionGraphHandler.java:103)
    at org.apache.flink.runtime.scheduler.ExecutionGraphHandler.lambda$processCheckpointCoordinatorMessage$3(ExecutionGraphHandler.java:119)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
    at java.lang.Thread.run(Thread.java:748)
 {code}
It seems that the bug comes from caching too many records when calling 

AbstractBroadcastWrapperOperator#snapshot. 



--
This message was sent by Atlassian Jira
(v8.20.10#820010)