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

Posted to issues@spark.apache.org by "Devesh Agrawal (Jira)" <ji...@apache.org> on 2020/08/18 00:30:00 UTC

[jira] [Created] (SPARK-32643) [Cleanup] Consolidate state kept in ExecutorDecommissionInfo with TaskSetManager.tidToExecutorKillTimeMapping

Devesh Agrawal created SPARK-32643:
--------------------------------------

             Summary: [Cleanup] Consolidate state kept in ExecutorDecommissionInfo with TaskSetManager.tidToExecutorKillTimeMapping
                 Key: SPARK-32643
                 URL: https://issues.apache.org/jira/browse/SPARK-32643
             Project: Spark
          Issue Type: Sub-task
          Components: Spark Core
    Affects Versions: 3.1.0
            Reporter: Devesh Agrawal


The decommissioning state is a bit fragment across two places in the TaskSchedulerImpl:
 * [https://github.com/apache/spark/pull/29014/|https://github.com/apache/spark/pull/29014/files] stored the incoming decommission info messages in _TaskSchedulerImpl.executorsPendingDecommission._

 * While [https://github.com/apache/spark/pull/28619/|https://github.com/apache/spark/pull/28619/files] was storing just the executor end time in the map _TaskSetManager.tidToExecutorKillTimeMapping_ (which in turn is contained in TaskSchedulerImpl).

While the two states are not really overlapping, its a bit of a code hygiene concern to save this state in two places. 

With [https://github.com/apache/spark/pull/29422], TaskSchedulerImpl is emerging as the place where all decommissioning book keeping is kept within the driver. So consolidate the information in _tidToExecutorKillTimeMapping_ into _ExecutorDecommissionInfo._

 

 



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

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