[
https://issues.apache.org/jira/browse/HADOOP-11933?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14543941#comment-14543941
]
Sean Busbey commented on HADOOP-11933:
--------------------------------------
{quote}
bq. If the improvements aren't critical, I'd say push them to a follow on.
One of them I had in mind was removing the requirement of the patchprocess
directory to be relative to the git checkout. So you might consider that one
critical to the nifi use case.
{quote}
I'd like that, but I can just have the nifi tester not use docker for now,
right? the docker stuff is more needed for HBase ATM (HBASE-12497), but even in
that case it just enables new things I want rather then current functionality
in the project's forked precommit tester.
I think HADOOP-11929 is the only blocker for Nifi officially starting to use
things.
> run test-patch.sh in a docker container under Jenkins
> -----------------------------------------------------
>
> Key: HADOOP-11933
> URL: https://issues.apache.org/jira/browse/HADOOP-11933
> Project: Hadoop Common
> Issue Type: Test
> Reporter: Allen Wittenauer
> Assignee: Allen Wittenauer
> Attachments: HADOOP-11933.00.patch, HADOOP-11933.01.patch,
> HADOOP-11933.02.patch, HADOOP-11933.03.patch
>
>
> because of how hard it is to control the content of the Jenkins environment,
> it would be beneficial to run it in a docker container so that we can have
> tight control of the environment
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
