hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Allen Wittenauer ...@effectivemachines.com>
Subject Re: Same jenkins build running on 2 patches.
Date Sun, 03 Dec 2017 05:19:59 GMT

> On Dec 1, 2017, at 1:36 PM, Jason Lowe <jlowe@oath.com> wrote:
>  I thought the admin precommit build was kicking off the project-specific precommit build
with an attachment ID argument so the project precommit can be consistent with the admin precommit
build on what triggered the precommit process.

	 precommit-admin generates a list of JIRA issue ids and passes those to the build jobs. 
The JIRA issue ids are generated from a simple JQL statement.  precommit build is the first
time the attachment id’s are determined.

> If the patch is tracked by attachment ID then I think the build would remain consistent
even when users attach new patches in the middle of the precommit process.

	Implementation of YETUS-504 will almost certainly eliminate the double download in Dockerized
situations since the credentials will only be available outside of it.  In the process, certain
testing scenarios also get dropped, but security is more important.

	That said, fixing users attempting to prematurely optimize test runs is pretty low on the
priority list.  For my own project list, some way to inform a user that only 20-30% of the
HDFS unit tests in branch-2 are getting executed is way higher.  [trunk comes in at around
80% post-surefire 2.20.1.]  precommit’s output doesn’t reflect tests that don’t run
because surefire doesn’t report them either. See also SUREFIRE-1447.  

To unsubscribe, e-mail: common-dev-unsubscribe@hadoop.apache.org
For additional commands, e-mail: common-dev-help@hadoop.apache.org

View raw message