[
https://issues.apache.org/jira/browse/SPARK-1770?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13993917#comment-13993917
]
Aaron Davidson commented on SPARK-1770:
---------------------------------------
Sorry, you're right, it was somehow committed when that PR was merged
(https://github.com/apache/spark/commit/06b15baab25951d124bbe6b64906f4139e037deb)
though the change doesn't actually show up in the PR itself.
> repartition and coalesce(shuffle=true) put objects with the same key in the
> same bucket
> ---------------------------------------------------------------------------------------
>
> Key: SPARK-1770
> URL: https://issues.apache.org/jira/browse/SPARK-1770
> Project: Spark
> Issue Type: Bug
> Affects Versions: 0.9.0, 1.0.0, 0.9.1
> Reporter: Matei Zaharia
> Priority: Blocker
> Labels: Starter
> Fix For: 1.0.0
>
>
> This is bad when you have many identical objects. We should assign each one a
> random key.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
