[ 
https://issues.apache.org/jira/browse/SPARK-26713?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Dongjoon Hyun updated SPARK-26713:
----------------------------------
    Fix Version/s: 2.4.5

> PipedRDD may holds stdin writer and stdout read threads even if the task is 
> finished
> ------------------------------------------------------------------------------------
>
>                 Key: SPARK-26713
>                 URL: https://issues.apache.org/jira/browse/SPARK-26713
>             Project: Spark
>          Issue Type: Bug
>          Components: Spark Core
>    Affects Versions: 2.1.3, 2.2.3, 2.3.0, 2.3.1, 2.3.2, 2.4.0
>            Reporter: Xianjin YE
>            Assignee: Xianjin YE
>            Priority: Major
>             Fix For: 2.4.5, 3.0.0
>
>
> During an investigation of OOM of one internal production job, I found that 
> PipedRDD leaks memory. After some digging, the problem lies down to the fact 
> that PipedRDD doesn't release stdin writer and stdout threads even if the 
> task is finished.
>  
> PipedRDD creates two threads: stdin writer and stdout reader. If we are lucky 
> and the task is finished normally, these two threads exit normally. If the 
> subprocess(pipe command) is failed, the task will be marked failed, however 
> the stdin writer will be still running until it consumes its parent RDD's 
> iterator. There is even a race condition with ShuffledRDD + PipedRDD: the 
> ShuffleBlockFetchIterator is cleaned up at task completion and hangs stdin 
> writer thread, which leaks memory. 



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org

Reply via email to