Github user zhzhan commented on the pull request:
https://github.com/apache/spark/pull/3409#issuecomment-65156197
@tgravescs Following is the code flow. The last call will have problem if
we don't have system property set correctly. In addition, other configuration
also have the similar problem, e.g., yarn-site.xml. mapred-site.xml is just for
illustration.
ExecutorRunnableUtil.prepareEnvironment->ClientBase.populateClasspath->
ClientBase.populateHadoopClasspath->ClientBase.getMRAppClasspath->conf.getStrings("mapreduce.application.classpathâ)
By the way, mapred-site.xml is introduced by spark-1064. Agree with you
here, spark should not relies on mapred-site.xml. But removing it may have
impact on existing user or causing regression issue.
---
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]
