[
https://issues.apache.org/jira/browse/SPARK-13519?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16646128#comment-16646128
]
Chao Fang commented on SPARK-13519:
-----------------------------------
We encounter this issue in our environment which used spark 1.5.2, so could I
just merge this pr code into 1.5.2 version without other change?
I am not sure whether there exists similar discussion. Also will this merge
operation impact the use of yarn deploy mode of spark? thx
> Driver should tell Executor to stop itself when cleaning executor's state
> -------------------------------------------------------------------------
>
> Key: SPARK-13519
> URL: https://issues.apache.org/jira/browse/SPARK-13519
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
> Affects Versions: 1.6.0
> Reporter: Shixiong Zhu
> Assignee: Shixiong Zhu
> Priority: Major
> Fix For: 2.0.0
>
>
> When the driver removes an executor's state, the connection between the
> driver and the executor may be still alive so that the executor cannot exit
> automatically (E.g., Master will send RemoveExecutor when a work is lost but
> the executor is still alive), so the driver try to tell the executor to stop
> itself. Otherwise, we will leak an executor.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
