spark-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Mateusz Pieniak (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (SPARK-24334) Race condition in ArrowPythonRunner causes unclean shutdown of Arrow memory allocator
Date Wed, 23 May 2018 07:13:02 GMT

    [ https://issues.apache.org/jira/browse/SPARK-24334?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16486832#comment-16486832
] 

Mateusz Pieniak commented on SPARK-24334:
-----------------------------------------

[~icexelloss] I didn't notice exception in my UDF when I tried to collect 1000 rows to driver
and display dataframe. However when I tried to dump it to Parquet, I noticed some executors
failed because of the exception in my UDF and others because of exception related to this
issue. 

I made sure that there is no exception thrown in my UDF. Right now my executors fail because
of different issue. I don't know whether it's related.
{code:java}
ExecutorLostFailure (executor 1 exited caused by one of the running tasks) Reason: Executor
heartbeat timed out after 390732 ms{code}

> Race condition in ArrowPythonRunner causes unclean shutdown of Arrow memory allocator
> -------------------------------------------------------------------------------------
>
>                 Key: SPARK-24334
>                 URL: https://issues.apache.org/jira/browse/SPARK-24334
>             Project: Spark
>          Issue Type: Sub-task
>          Components: PySpark
>    Affects Versions: 2.3.0
>            Reporter: Li Jin
>            Priority: Major
>
> Currently, ArrowPythonRunner has two thread that frees the Arrow vector schema root and
allocator - The main writer thread and task completion listener thread. 
> Having both thread doing the clean up leads to weird case (e.g., negative ref cnt, NPE,
and memory leak exception) when an exceptions are thrown from the user function.
>  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

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


Mime
View raw message