Strange stack when job encounter back pressure

[aitozi <gj...@gmail.com>]

Hi,

I notice the job encounter a strange case: the upstream operator is
underpressure, all task in back pressure sample shows HIGH, but when i
jstack the downstream task, I only see the stack below:


   java.lang.Thread.State: WAITING (on object monitor)
        at java.lang.Object.wait(Native Method)
        at java.lang.Object.wait(Object.java:502)
        at
org.apache.flink.runtime.io.network.partition.consumer.SingleInputGate.getNextBufferOrEvent(SingleInputGate.java:745)
        - locked <0x00000000b35221a8> (a java.util.ArrayDeque)
        at
org.apache.flink.runtime.io.network.partition.consumer.SingleInputGate.getNextBufferOrEvent(SingleInputGate.java:704)
        at
org.apache.flink.streaming.runtime.io.BarrierBuffer.getNext(BarrierBuffer.java:170)
        at
org.apache.flink.streaming.runtime.io.BarrierBuffer.getNextNonBlocked(BarrierBuffer.java:162)
        at
org.apache.flink.streaming.runtime.io.StreamInputProcessor.processInput(StreamInputProcessor.java:269)
        at
org.apache.flink.streaming.runtime.tasks.OneInputStreamTask.run(OneInputStreamTask.java:113)
        at
org.apache.flink.streaming.runtime.tasks.StreamTask.invoke(StreamTask.java:398)
        at org.apache.flink.runtime.taskmanager.Task.run(Task.java:751)
        at java.lang.Thread.run(Thread.java:882)

<http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/file/t1056/20200722102524.jpg> 

It likes that : the downstream say "i am waiting to data" but the upstream
say "I am blocked by the downstream".

Our code is based on 1.5.x

Is there any one also encounter this case or know the reason?





--
Sent from: http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/