Github user vanzin commented on a diff in the pull request:
https://github.com/apache/spark/pull/4747#discussion_r25289178
--- Diff:
core/src/main/scala/org/apache/spark/deploy/worker/ExecutorRunner.scala ---
@@ -135,7 +135,7 @@ private[spark] class ExecutorRunner(
logInfo("Launch command: " + command.mkString("\"", "\" \"", "\""))
builder.directory(executorDir)
- builder.environment.put("SPARK_LOCAL_DIRS",
appLocalDirs.mkString(","))
+ builder.environment.put("SPARK_EXECUTOR_DIRS",
appLocalDirs.mkString(File.pathSeparator))
--- End diff --
Without this, `Utils.scala` would always create a subdirectory for the
executor under the directory already created by the Worker. So you'd always
have "spark-xxxxx/spark-yyyyy" for every executor. It would be always like that
(no extra nesting), but, since I'm changing this, sounded like a good
enhancement.
And dirs should be joined with a path separator because that's the right
way to do it.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
