[
https://issues.apache.org/jira/browse/SPARK-4986?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14291158#comment-14291158
]
Sean Owen commented on SPARK-4986:
----------------------------------
[~koudelka] This sounds pretty related to the issue I raised in SPARK-4545. I'm
not sure, but your changes might address the problem too -- graceful shutdown
takes a long time if no batch succeeds. If you have a moment would you mind
taking a look? If it's really a subset of the problem you're solving we can
close the other one.
> Graceful shutdown for Spark Streaming does not work in Standalone cluster mode
> ------------------------------------------------------------------------------
>
> Key: SPARK-4986
> URL: https://issues.apache.org/jira/browse/SPARK-4986
> Project: Spark
> Issue Type: Bug
> Components: Streaming
> Affects Versions: 1.2.0
> Reporter: Jesper Lundgren
> Fix For: 1.3.0, 1.2.1
>
>
> When using the graceful stop API of Spark Streaming in Spark Standalone
> cluster the stop signal never reaches the receivers. I have tested this with
> Spark 1.2 and Kafka receivers.
> ReceiverTracker will send StopReceiver message to ReceiverSupervisorImpl.
> In local mode ReceiverSupervisorImpl receives this message but in Standalone
> cluster mode the message seems to be lost.
> (I have modified the code to send my own string message as a stop signal from
> ReceiverTracker to ReceiverSupervisorImpl and it works as a workaround.)
--
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
