You are viewing a plain text version of this content. The canonical link for it is here.
Posted to mapreduce-dev@hadoop.apache.org by Andrew Wang <an...@cloudera.com> on 2017/07/28 22:29:33 UTC

3.0.0-beta1 release plan and branch cuts

Hi all,

Here's a long overdue update on Hadoop 3. We've made great progress through
the alpha releases, thanks to the hard work of numerous contributors. With
alpha4 out the door, it's time to look toward beta1!

I updated the wiki page with a proposed release date for beta1: September
15th.

https://cwiki.apache.org/confluence/display/HADOOP/Hadoop+3.0.0+release

I went through the JIRA dashboard and pinged the blockers that looked
stalled. I also wrote a status update, and will be resuming more regular
updates as we close on beta1.

https://cwiki.apache.org/confluence/display/HADOOP/Hadoop+3+release+status+updates

It also sounds like there are some pending branch merges. I'm keeping my
eye on them, with an eye toward not breaking existing functionality and
compatibility. Please don't be offended if I ask that we delay your feature
until after 3.0 GA. We've been baking the current 3.0 feature set with the
alphas, and I don't want to possibly destabilize things when we're so close
to beta/GA.

Finally, a note on branching. I've been cutting release branches directly
off of trunk, to reduce the number of cherry-picks for backporting.

Since beta1 represents a point of stability, I plan to cut branch-3 shortly
before the release, and then cut the 3.0.0-beta1 release branch from
branch-3. trunk would move to 4.0.0-SNAPSHOT.

For GA, I plan to cut branch-3.0 off of branch-3, with branch-3 becoming
3.1.0-SNAPSHOT. At this point, the scheme will be the same as the branch-2s.

If needed, we can do these branch cuts sooner to unblock branch merges.

As always, please reach out with any comments/questions.

Cheers,
Andrew