[
https://issues.apache.org/jira/browse/FLINK-31909?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Dong Lin updated FLINK-31909:
-----------------------------
Fix Version/s: (was: ml-2.3.0)
> Using BroadcastUtils#withBroadcast in iteration perround mode got stuck
> -----------------------------------------------------------------------
>
> Key: FLINK-31909
> URL: https://issues.apache.org/jira/browse/FLINK-31909
> Project: Flink
> Issue Type: Bug
> Components: Library / Machine Learning
> Reporter: Zhipeng Zhang
> Priority: Major
>
> Using BroadcastUtils#withBroadcastStream in iterations in per round mode
> could possibly lead to stuck.
>
> It seems that the there is a task waiting for the mail from the mailbox.
>
> {code:java}
> 793 "tail-map-head-Parallel Collection Source (1/1)#0" #200 prio=5
> os_prio=31 tid=0x00007faabb571800 nid=0x18c03 waiting on condition
> [0x0000700013aae000]
> 793 java.lang.Thread.State: TIMED_WAITING (parking)
> 794 at sun.misc.Unsafe.park(Native Method)
> 795 - parking to wait for <0x0000000747805568> (a
> java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject)
> 796 at
> java.util.concurrent.locks.LockSupport.parkNanos(LockSupport.java:215)
> 797 at
> java.util.concurrent.locks.AbstractQueuedSynchronizer$ConditionObject.await(AbstractQueuedSynchronizer.java:2163)
> 798 at
> org.apache.flink.streaming.runtime.tasks.mailbox.TaskMailboxImpl.take(TaskMailboxImpl.java:149)
> 799 at
> org.apache.flink.streaming.runtime.tasks.mailbox.MailboxProcessor.processMailsWhenDefaultActionUnavailable(MailboxProcessor.java:335)
> 800 at
> org.apache.flink.streaming.runtime.tasks.mailbox.MailboxProcessor.processMail(MailboxProcessor.java:324)
> 801 at
> org.apache.flink.streaming.runtime.tasks.mailbox.MailboxProcessor.runMailboxLoop(MailboxProcessor.java:201)
> 802 at
> org.apache.flink.streaming.runtime.tasks.StreamTask.runMailboxLoop(StreamTask.java:804)
> 803 at
> org.apache.flink.streaming.runtime.tasks.StreamTask.invoke(StreamTask.java:753)
> 804 at
> org.apache.flink.runtime.taskmanager.Task$$Lambda$1430/1226027100.run(Unknown
> Source)
> 805 at
> org.apache.flink.runtime.taskmanager.Task.runWithSystemExitMonitoring(Task.java:948)
> 806 at
> org.apache.flink.runtime.taskmanager.Task.restoreAndInvoke(Task.java:927)
> 807 at org.apache.flink.runtime.taskmanager.Task.doRun(Task.java:741)
> 808 at org.apache.flink.runtime.taskmanager.Task.run(Task.java:563)
> 809 at java.lang.Thread.run(Thread.java:748) {code}
--
This message was sent by Atlassian Jira
(v8.20.10#820010)