[
https://issues.apache.org/jira/browse/SPARK-8643?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Yin Huai updated SPARK-8643:
----------------------------
Attachment: HiveSparkSubmitSuite (SPARK-8368).txt
> local-cluster may not shutdown SparkContext gracefully
> ------------------------------------------------------
>
> Key: SPARK-8643
> URL: https://issues.apache.org/jira/browse/SPARK-8643
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
> Reporter: Yin Huai
> Attachments: HiveSparkSubmitSuite (SPARK-8368).txt
>
>
> When I was debugging SPARK-8567, I found that when I was using local-cluster,
> at the end of an application, executors were first killed and then launched
> again. From the log (attached), seems the master/driver side does not know
> it's in the shutdown process. So, it detected executor loss and then called
> the worker to launch new executors.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
