From dev-return-56445-archive-asf-public=cust-asf.ponee.io@phoenix.apache.org Mon May 13 14:53:13 2019 Return-Path: X-Original-To: archive-asf-public@cust-asf.ponee.io Delivered-To: archive-asf-public@cust-asf.ponee.io Received: from mail.apache.org (hermes.apache.org [207.244.88.153]) by mx-eu-01.ponee.io (Postfix) with SMTP id BF657180671 for ; Mon, 13 May 2019 16:53:12 +0200 (CEST) Received: (qmail 7343 invoked by uid 500); 13 May 2019 14:53:12 -0000 Mailing-List: contact dev-help@phoenix.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: dev@phoenix.apache.org Delivered-To: mailing list dev@phoenix.apache.org Received: (qmail 7332 invoked by uid 99); 13 May 2019 14:53:11 -0000 Received: from Unknown (HELO mailrelay1-lw-us.apache.org) (10.10.3.159) by apache.org (qpsmtpd/0.29) with ESMTP; Mon, 13 May 2019 14:53:11 +0000 Received: from hw13390.local (pool-173-69-177-211.bltmmd.fios.verizon.net [173.69.177.211]) by mailrelay1-lw-us.apache.org (ASF Mail Server at mailrelay1-lw-us.apache.org) with ESMTPSA id BA0891BC8 for ; Mon, 13 May 2019 14:53:11 +0000 (UTC) Subject: Re: [DISCUSS] Next Phoenix 5.x release (was "Board report due in ~1 week") To: dev@phoenix.apache.org References: <0f6efc7a-66f2-35be-6db4-d6abf706d59f@apache.org> From: Josh Elser Message-ID: <2add8426-b8c3-b144-3910-183d4e827b8e@apache.org> Date: Mon, 13 May 2019 10:53:11 -0400 User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.14; rv:67.0) Gecko/20100101 Thunderbird/67.0 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8; format=flowed Content-Language: en-US Content-Transfer-Encoding: 8bit I'd say we update to HBase-2.0.3 as that is the latest, known-good version. Of course, I thought the same of that for 2.0.5 up until a week or two ago :) Let's get a branch made and check it out. On 5/13/19 9:00 AM, Jaanai Zhang wrote: > So we can keep the current HBase version in 5.0.1? > > ---------------------------------------- > Jaanai Zhang > Best regards! > > > > Jaanai Zhang 于2019年5月13日周一 下午8:22写道： > >> +1 >> >> ---------------------------------------- >> Jaanai Zhang >> Best regards! >> >> >> >> Thomas D'Silva 于2019年5月10日周五 下午1:25写道： >> >>> +1 to this approach. >>> >>> On Thu, May 9, 2019 at 9:22 AM Josh Elser wrote: >>> >>>> After working on trying to make Phoenix compatible with >=HBase 2.0.4, >>>> I'm wondering if it would just be good to get 5.0.1 out the door and try >>>> to fix HBase compat in a 5.1.0, acknowledging that we don't work with >>>> the newer 2.0.x HBase versions (really, anything that contains >>>> HBASE-21401[1]). >>>> >>>> I feel like it's much less risky to just acknowledge that we're limited >>>> in target HBase version for a bug-fix release. >>>> >>>> What do folks think? That would lessen the burden on you, Jaanai. >>>> >>>> [1] https://issues.apache.org/jira/browse/HBASE-21401 >>>> >>>> On 5/1/19 5:35 PM, Josh Elser wrote: >>>>> I think it would be better to figure out if there is anything >>> currently >>>>> on master that _shouldn't_ be included in a 5.0.1. My guess would be >>>>> "no". It feels like branching for the sake of branching to keep 5.0.1 >>>>> and 5.1.0 distinct. >>>>> >>>>> On 5/1/19 12:38 AM, Thomas D'Silva wrote:> Should we use 2.0.5 ( the >>>>> latest released HBase version )? If PHOENIX-5250 >>>>> > is a blocker >>>> we >>>>> will >>>>> > need to fix it before we can release 5.0.1 (or 5.1). >>>>> > >>>>> > These are the list of bugs that have a fix version of 5.1 : >>>>> > >>>>> >>>> >>> https://issues.apache.org/jira/issues/?jql=project%20%3D%20PHOENIX%20AND%20fixVersion%20%3D%205.1.0%20AND%20issuetype%20%20%3D%20Bug >>>>> >>>>> > We should figure out which of these we want to cherrypick for the >>>> 5.0.1 >>>>> > release. >>>>> >>>>> On 4/30/19 10:39 AM, Josh Elser wrote: >>>>>> Let's leave the other thread for the board report, please. I've >>>>>> changed the subject as such. >>>>>> >>>>>> Thanks for volunteering to be RM. >>>>>> >>>>>> I'm of the opinion that we should just update to the latest HBase >>>>>> 2.0.x line. The compatibility assertions from HBase should make this >>> a >>>>>> no-op for us to change versions. There are arguments, however, in >>> both >>>>>> ways that we should use the earliest, non-breaking version of HBase. >>>>>> >>>>>> As the RM, it is your prerogative to tell everyone else how you are >>>>>> tracking it :). Figure out what fixVersion you're using on Jira, and >>>>>> then either manage the list of open issues in there yourself, or push >>>>>> out issues and have folks put issues back in which they feel must be >>>>>> included. >>>>>> >>>>>> Since this is your first time, I would prioritize an rc0 ASAP so that >>>>>> you can get comfortable with the process (even if we know that there >>>>>> are other fixes which must be made before the final release). >>>>>> >>>>>> On 4/30/19 2:34 AM, Jaanai Zhang wrote: >>>>>>> I will try doing release 5.0.1. Tow things need to confirm: >>>>>>> >>>>>>> 1. According to the thread[1], a new Phoenix 5.0.1 will support >>> HBase >>>>>>> 2.0.2, right? As far as I know, some improvements of HBase 2.0.2 >>> will >>>>>>> cause >>>>>>> some critical issues, for example, >>>>>>> https://issues.apache.org/jira/browse/PHOENIX-5250 (This issue from >>>> our >>>>>>> production environments, we had used HBase 2.0.2 in Phoenix 5.0). >>>>>>> >>>>>>> 2. We should have a clear JIRA list for a 5.0.1, I am not sure what >>>>>>> JIRAs >>>>>>> should go into it, only critical and blocked JIRAs apply for a >>> 5.0.1? >>>>>>> If I >>>>>>> know what priority need to pick, I will collect them. >>>>>>> >>>>>>> >>>>>>> ---------------------------------------- >>>>>>> Jaanai Zhang >>>>>>> Best regards! >>>>>>> >>>>>>> >>>>>>> >>>>>>> Thomas D'Silva 于2019年4月30日周二 >>>>>>> 上 午10:06写道： >>>>>>> >>>>>>>> Jaanai, >>>>>>>> >>>>>>>> We are waiting for a few omid bug fixes to do the 4.15/5.1 release >>>> that >>>>>>>> will have splittable system catalog and the omid integration. >>>>>>>> Are you interested in doing a 5.0.1 release that has the HBase >>> 2.0.x >>>>>>>> compatibility fixes that were discussed in a previous thread[1]? >>>>>>>> >>>>>>>> The steps to create a RC are straightforward and documented here: >>>>>>>> https://phoenix.apache.org/release.html. >>>>>>>> The main thing you need to do is to add your code signing key to >>>>>>>> https://dist.apache.org/repos/dist/dev/phoenix/KEYS (follow the >>>>>>>> steps at >>>>>>>> the start of that file) >>>>>>>> and then commit using svn. Then you can follow the rest of the >>> steps >>>>>>>> listed >>>>>>>> in "How to do a release" >>>>>>>> >>>>>>>> Thanks, >>>>>>>> Thomas >>>>>>>> >>>>>>>> [1] >>>>>>>> >>>>>>>> >>>> >>> https://lists.apache.org/thread.html/99fcc737d7a8f82ddffb1b34a64f7099f7909900b8bea36dd6afca16@%3Cdev.phoenix.apache.org%3E >>>>>>>> >>>>>>>> >>>>>>>> On Mon, Apr 29, 2019 at 6:33 PM Jaanai Zhang < >>> cloud.poster@gmail.com> >>>>>>>> wrote: >>>>>>>> >>>>>>>>> I would like to volunteer for a new 5.x release if someone can >>>>>>>>> guide me >>>>>>>>> release process. Thanks. >>>>>>>>> >>>>>>>>> ---------------------------------------- >>>>>>>>> Jaanai Zhang >>>>>>>>> Best regards! >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> Josh Elser 于2019年4月30日周二 上午12:39写道： >>>>>>>>> >>>>>>>>>> Hiya folks, >>>>>>>>>> >>>>>>>>>> It's about that time for another board report. Please reply here >>>> with >>>>>>>>>> anything of merit that you think the board might find >>>>>>>>>> interesting/useful. As a reminder, they board is typically more >>>>>>>>>> concerned with high-level project/community details than the >>>>>>>>>> nuts-and-bolts of the code changes for the project. >>>>>>>>>> >>>>>>>>>> On my radar already is... >>>>>>>>>> >>>>>>>>>> * Multiple new committers and PMC'ers (thanks so much to the >>> folks >>>>>>>>>> who >>>>>>>>>> have been driving votes!) >>>>>>>>>> * NoSQL day in May >>>>>>>>>> * 4.14.2 in vote >>>>>>>>>> * Need for a new 5.x.y release (if there are no volunteers, I may >>>>>>>>>> have >>>>>>>>>> to find the time to make this happen. It's been too long) >>>>>>>>>> >>>>>>>>>> Thanks! >>>>>>>>>> >>>>>>>>>> - Josh >>>>>>>>>> >>>>>>>>> >>>>>>>> >>>>>>> >>>> >>> >> >