Thanks Yongjun and Allen for the feedback. I agree that option 2 could be a safer option if any concern on option 1. Will defer this change to 2.6.5.
Thanks, Junping ________________________________________ From: Yongjun Zhang <yzh...@cloudera.com> Sent: Wednesday, February 10, 2016 7:11 PM To: hdfs-dev@hadoop.apache.org Cc: Hadoop Common; mapreduce-...@hadoop.apache.org; yarn-...@hadoop.apache.org Subject: Re: [VOTE] Release Apache Hadoop 2.6.4 RC0 Thanks Junping and Allen. It'd be nice to have HDFS-9629 but I'm ok with option 2, given the fact that the issue is not critical (and will be addressed in all future releases), and the concern Allen raised. Best, --Yongjun On Wed, Feb 10, 2016 at 8:37 AM, Allen Wittenauer <a...@altiscale.com> wrote: > > > On Feb 9, 2016, at 6:27 PM, Junping Du <j...@hortonworks.com> wrote: > > > > Thanks Yongjun for identifying and proposing this change to 2.6.4. I > think this is the right thing to do and check for following releases. For > 2.6.4, it seems unnecessary to create another release candidate for this > issue as we only kicking off a new RC build when last RC has serious > problem in functionality. The vote progress is quite smoothly so far, so it > seems unlikely that we will create a new RC. However, I think there are > still two options here: > > Option 1: in final build, adopt change of HDFS-9629 that only updates > the footer of Web UI to show year 2016. > > Option 2: skip HDFS-9629 for 2.6.4 and adopt it later for 2.6.5. > > I prefer Option 1 as this is a very low risky change without affecting > any functionality, and we allow non-functional changes (like release date, > etc.) happen on final build after RC passed. I would like to hear the > voices in community here before acting for the next step. Thoughts? > > > > I’d think having PMC votes apply to what is not actually the final > artifact is against the ASF rules. > > >
