I think it could make sense to EOL of 1.0 if we have a consensus on it. I'd be in favor of reducing workload. I think we should make one more release that rolls up the fixes and improvements we've already committed to branch-1.0, though.
> On Nov 4, 2015, at 11:09 AM, Enis Söztutar <[email protected]> wrote: > > Related, I was waiting on 1.2.0 RC to start the discussion, but we can do > it now. > > How do you guys feel about retiring 1.0 branch and releases? With the > guarantees we have upgrading to 1.1 is a relatively painless job. With 1.1 > and 1.2 active releases, we will be encouraging the users to upgrade to > them via rolling upgrades. > > It is basically a tradeoff between lessening the burden on committers and > PMC to keep backporting patches and creating and testing RCs, and the users > need to upgrade to have important bug fixes. > > Enis > >> On Wed, Nov 4, 2015 at 9:15 AM, Andrew Purtell <[email protected]> wrote: >> >> Discussion on HBASE-14712 talks about 1.2. Has this been observed with 1.1? >> Should this hold up a minor 1.1 release? If present in 1.1 this bug is >> already shipping in three (?) prior releases. Assuming we can keep up a >> regular release cadence, if there's a fix for this problem applicable to >> 1.1 we can get it out in the next timely minor release. >> >> Also, I don't mean to step on any toes. I don't think I have, but please >> accept my apologies if so. (smile) I'm just volunteering to drive release >> candidates so, hopefully if the PMC accepts them, the many queued up bug >> fixes get out into the hands of users and overall we reduce the size of the >> deltas between minor releases. I'd like to see monthly releases of all >> active code lines - 0.98, 1.0, 1.1. Happy to help with making that possible >> when others are busy with work or life. Likewise, if I get busy and can't >> get out a 0.98 release on the monthly mark, I would not mind at all someone >> stepping up to do it. >> >> I also had someone volunteer off list to help me with spinning the bits, so >> thank you. You know who you are. (smile) >> >> I guess at this point we should settle on who is doing the release >> candidates for November. Then we can do it again for December, January, >> etc. Would a google doc spreadsheet help? Or I can just start a discussion >> thread at the start of every month. Let me know. >> >> For November, >> >> 0.98.16 - Andrew >> 1.0.3 - Andrew (thanks, Enis) >> 1.1.3 - ? >> >> >> >>> On Tue, Nov 3, 2015 at 11:21 PM, Nick Dimiduk <[email protected]> wrote: >>> >>> Apologies for my absence as of late; making some transitions on this >>> end. I circled back on 1.1 this week, it's well overdue. Looks like >>> HBASE-14712 is holding up that show. I haven't looked closely at the >>> conversation there (it's too late tonight, maybe tomorrow evening?). >>> >>>> On Tue, Nov 3, 2015 at 7:42 PM, Enis Söztutar <[email protected]> wrote: >>>> Sounds good. Thanks Andrew for doing the work. >>>> >>>> Enis >>>> >>>>> On Tue, Nov 3, 2015 at 6:02 PM, Andrew Purtell <[email protected]> >>>> wrote: >>>> >>>>> 1.0.2 was released on Mon Aug 31 2015, 64 days ago. There are 71 >> issues >>>>> marked as complete with a fix version of 1.0.3. (There are also 18 >>>>> unresolved issues targeting this version.) >>>>> >>>>> 1.1.2 was released on Tue Sep 01 2015, 63 days ago. There are 83 >> issues >>>>> marked as complete with a fix version of 1.1.3. (There are also 3 >>>>> unresolved issues targeting this version.) >>>>> >>>>> Unless objection, I'll RM 1.0.3 and 1.1.3 candidates along with >> 0.98.16 >>>>> when it's ready to go. There are two backport issues pending review >> for >>>>> 0.98 and 5 other unresolved issues I hope to get to. Probably I will >>> end up >>>>> spinning all of the RC bits over the weekend ready for Monday of next >>> week. >>>>> >>>>> -- >>>>> Best regards, >>>>> >>>>> - Andy >>>>> >>>>> Problems worthy of attack prove their worth by hitting back. - Piet >> Hein >>>>> (via Tom White) >> >> >> >> -- >> Best regards, >> >> - Andy >> >> Problems worthy of attack prove their worth by hitting back. - Piet Hein >> (via Tom White) >>
