You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@spark.apache.org by "sid (JIRA)" <ji...@apache.org> on 2015/06/06 00:45:00 UTC

[jira] [Created] (SPARK-8134) Lifecycle of RDDs in Cluster envs

sid created SPARK-8134:
--------------------------

             Summary: Lifecycle of RDDs in Cluster envs
                 Key: SPARK-8134
                 URL: https://issues.apache.org/jira/browse/SPARK-8134
             Project: Spark
          Issue Type: New Feature
          Components: Spark Core
    Affects Versions: 1.3.1
            Reporter: sid


Is there a way to implement simple driver lifecycle like this.
init() - connect to JMS
RDD - read a file and pump out messages
close() - close JMS 

Currently we are hacking RDDs to create singleton connections (worked node agnostic); but then unable to close them.
also when i call SparkContext.close(); it immediately kills jobs in worker nodes.
There is NO WAY of knowing when to close the context in driver
I looked into SparkListener and that doesnt help solve our problem.

I assumed driver would stay alive until jobs are done and cleanup. Driver simply spawns jobs and ends



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

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