[
https://issues.apache.org/jira/browse/SPARK-16148?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15349920#comment-15349920
]
Tom Magrino edited comment on SPARK-16148 at 6/26/16 1:41 AM:
--------------------------------------------------------------
For posterity (in case others look here and don't find the other discussion on
github), I'm adding Tejas Patil's response to this question from my first PR:
"Since there is a pluggable cluster manager (CM) support in Spark, it is
possible to use CM apart from YARN or Mesos.... which can follow its own naming
convention for executors. Spark should not be depending on the naming format of
executors."
was (Author: tmagrino):
For posterity, I'm adding Tejas Patil's response to this question from my first
PR:
"Since there is a pluggable cluster manager (CM) support in Spark, it is
possible to use CM apart from YARN or Mesos.... which can follow its own naming
convention for executors. Spark should not be depending on the naming format of
executors."
> TaskLocation does not allow for Executor ID's with underscores
> --------------------------------------------------------------
>
> Key: SPARK-16148
> URL: https://issues.apache.org/jira/browse/SPARK-16148
> Project: Spark
> Issue Type: Bug
> Components: Scheduler
> Affects Versions: 1.6.1
> Reporter: Tom Magrino
> Priority: Minor
>
> Currently, the logic in TaskLocation does not allow for Executor IDs which
> contain underscores, leading to an IllegalArgumentException being thrown from
> core/src/scala/org/apache/spark/scheduler/TaskLocation.scala in the apply
> method.
--
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
