[jira] [Commented] (YETUS-105) reject patches that hit too many modules

["Sean Busbey (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/YETUS-105?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14964447#comment-14964447 ] 

Sean Busbey commented on YETUS-105:
-----------------------------------

Would the general solution just be a list of modules and a threshold count? or are we talking something fancier like a history of execution times and an allowed total estimated runtime?

> reject patches that hit too many modules
> ----------------------------------------
>
>                 Key: YETUS-105
>                 URL: https://issues.apache.org/jira/browse/YETUS-105
>             Project: Yetus
>          Issue Type: New Feature
>          Components: Test Patch
>            Reporter: Allen Wittenauer
>
> I *think* this can be a general solution, but...
> In Hadoop, we should reject a patch that hits any three of hadoop-hdfs, hadoop-yarn-resourcemanager, hadoop-common, and hadoop-mapreduce-client because the unit tests are too expensive to run.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)