JRuby and a working AvroStorage. Woohoo! Russell Jurney http://datasyndrome.com
On Apr 4, 2012, at 10:48 PM, Alan Gates <[email protected]> wrote: > +1. Let's roll 0.10. > > Alan. > > On Apr 4, 2012, at 9:34 PM, Jon Coveney wrote: > >> I think we got everything important in. Am excited >> >> On Apr 3, 2012, at 4:22 PM, Daniel Dai <[email protected]> wrote: >> >>> Is everything ready now? Can we start the release process? >>> >>> Daniel >>> >>> On Tue, Mar 27, 2012 at 11:54 AM, Alan Gates <[email protected]> wrote: >>>> Sure, that's fine. >>>> >>>> Alan. >>>> >>>> On Mar 27, 2012, at 11:46 AM, Jonathan Coveney wrote: >>>> >>>>> I agree, Alan, though I think we should make the deadline Monday, 4/2. >>>>> Most >>>>> testing is going to happen during the week anyway, and I'd like to give >>>>> true believers the weekend to fix up anything that is lacking. >>>>> >>>>> Sound ok? >>>>> >>>>> 2012/3/27 Alan Gates <[email protected]> >>>>> >>>>>> I propose we set a deadline and say patches need to be ready by that day, >>>>>> or they don't go in. Otherwise this time next week there will be another >>>>>> set of patches people would like to see. I'd say the deadline should be >>>>>> 3/30 (this Friday). That would mean we could address any final test >>>>>> failures we see and start rolling the RC next week. >>>>>> >>>>>> Alan. >>>>>> >>>>>> On Mar 27, 2012, at 12:00 AM, Jonathan Coveney wrote: >>>>>> >>>>>>> Bill, >>>>>>> >>>>>>> Thanks for going through those tickets. I marked them all as >>>>>>> 0.10/0.11... >>>>>>> 2616 seems like the only critical one, and the others are nice to haves >>>>>> (I >>>>>>> imagine 2587and 2574 won't be super critical until we build out the >>>>>> systems >>>>>>> that rely on them). All of them look pretty close so I'd have no problem >>>>>>> with that. >>>>>>> >>>>>>> Hopefully tomorrow others can weigh in on their priorities. Then we can >>>>>>> label things, work to finish them, and roll the RC. >>>>>>> >>>>>>> 2012/3/26 Daniel Dai <[email protected]> >>>>>>> >>>>>>>> Yes, but CI anyway only run unit tests, we will have to run e2e test >>>>>>>> by ourselves. >>>>>>>> >>>>>>>> On Mon, Mar 26, 2012 at 10:44 PM, Jonathan Coveney <[email protected]> >>>>>>>> wrote: >>>>>>>>> Perhaps a big initiative for 0.11 could be to get the builds to the >>>>>> point >>>>>>>>> where flaky tests don't essentially gut the usefulness of the CI >>>>>>>>> server >>>>>>>>> (Pig 0.9 hasn't passed in 6 months, etc). I'm not sure how difficult >>>>>> that >>>>>>>>> would be though, and what the benefit would be. But it's pretty >>>>>> annoying >>>>>>>> to >>>>>>>>> be able to wrap up a release when there are a bunch of "known useless" >>>>>>>>> tests. >>>>>>>>> >>>>>>>>> 2012/3/26 Daniel Dai <[email protected]> >>>>>>>>> >>>>>>>>>> Once PIG-2317 check in, only document patch (PIG-2601) and some of >>>>>>>>>> Russell's patches left. Before roll up RC, we need to make sure all >>>>>>>>>> unit tests and e2e tests pass, I can help with this. >>>>>>>>>> >>>>>>>>>> Daniel >>>>>>>>>> >>>>>>>>>> On Mon, Mar 26, 2012 at 10:29 PM, Jonathan Coveney < >>>>>> [email protected]> >>>>>>>>>> wrote: >>>>>>>>>>> It's time to get serious about rolling an RC. 0.10 seems pretty >>>>>>>> strong... >>>>>>>>>>> are there any pressing JIRAs that people want to try and get in? >>>>>>>> Looking >>>>>>>>>> at >>>>>>>>>>> the open JIRAs with a fix labeled of 0.10 (which is admittedly >>>>>>>> possibly >>>>>>>>>>> incomplete), it looks like there isn't anything terribly seriously >>>>>>>>>> left... >>>>>>>>>>> I'd love to get the ball rolling on an RC so we can test it and >>>>>>>> identify >>>>>>>>>>> any bugs. >>>>>>>>>>> >>>>>>>>>>> Thoughts? >>>>>>>>>> >>>>>>>> >>>>>> >>>>>> >>>> >
