Github user tgravescs commented on the issue: https://github.com/apache/spark/pull/17238 Sorry if I'm missing something here but I don't see why this is a problem? If you have YARN misconfigured or not configured everything is going to default to DEFAULT_RACK. If you want them to be on different racks then fix the yarn configuration. The topology mapping in hadoop is a plugin so someone could write a plugin that just returned DEFAULT_RACK for anything, if you don't want that behavior you can write a plugin that does something different. @morenn520 I'm assuming in this case the tasks had some host locality level before looking at the rack level? I don't believe tez/MR do anything special with respect to the DEFAULT_RACK but I'll double check
--- 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 infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org