A quick update for everyone: Reviewboard is now available for use by anyone wanting to use it when submitting patches to Bigtop. INFRA-5594 is now closed.
Mark On Mon, Dec 3, 2012 at 11:22 AM, Mark Grover <grover.markgro...@gmail.com> wrote: > Thanks, everyone for your input. > > The general consensus seems to be the following: > 1. Regardless of whether a review tool is used or not, the patch should > always be uploaded on the JIRA. > 2. We don't force people to upload the patch on the review tool. For patches > that are non-trivial, uploaders would be encouraged to upload the patch on > the review tool. > > I didn't see any particular objection against review board and it seems > commonly used in other projects so I went ahead and created an INFRA JIRA > for setting up Bigtop as a project on Reviewboard > (https://reviews.apache.org/dashboard/): > https://issues.apache.org/jira/browse/INFRA-5594 > > Let me know if there is anything I missed. > > Thanks again! > Mark > > On Wed, Nov 28, 2012 at 1:02 PM, Konstantin Boudnik <c...@apache.org> wrote: >> >> Sounds like an awful amount of work for patches that usually like 15 lines >> long. >> >> On Thu, Nov 29, 2012 at 01:59AM, Harsh J wrote: >> > FWIW, Hive uses Phabricator for this that helps automate the >> > discussion cross-posting, and also uploads all revisions of patches. >> > >> > There's also a script Hadoop uses to keep the comments cross posted >> > from RB to JIRA, to not lose them in case of RB outage or extinction, >> > but I don't think it does revision patch uploads (or not yet anyway). >> > >> > Personally, I like the RB UI more, but maybe Phabricator would feel >> > easier to setup. >> > >> > On Thu, Nov 29, 2012 at 1:53 AM, Konstantin Boudnik <c...@apache.org> >> > wrote: >> > > On Wed, Nov 28, 2012 at 02:22PM, Brock Noland wrote: >> > >> Flume does this, that is patches must be posted on JIRA as well >> > >> even if >> > >> they are on RB. >> > > >> > > Yup, that would make sense. >> > > >> > > Cos >> > > >> > >> On Wed, Nov 28, 2012 at 2:11 PM, Konstantin Boudnik >> > >> <c...@apache.org> >> > >> wrote: >> > >> >> > >> To chime in with Bruno earlier - having a review board doesn't >> > >> mean that >> > >> patches shouldn't posted on JIRAs as we do right now. Otherwise, >> > >> this >> > >> would be >> > >> an enforcement of the tooling, which I oppose big time. >> > >> >> > >> Cos >> > >> On Wed, Nov 28, 2012 at 12:02PM, Jonathan Hsieh wrote: >> > >> > I'm not that active here, but if you want bigtop setup on >> > >> > reviews.apache.org all it just takes filing a apache INFRA >> > >> jira and >> > >> a >> > >> > little patience to get it done. Once it is up, you post >> > >> patches >> > >> there. >> > >> > I'm from HBase-land and I generally ask for review board >> > >> only on >> > >> patches >> > >> > that are larger than a 1-2 screen-fulls. >> > >> > >> > >> > Jon. >> > >> > On Tue, Nov 27, 2012 at 5:08 PM, Mark Grover >> > >> <grover.markgro...@gmail.com> >> > >> > wrote: >> > >> > >> > >> > Hi all, >> > >> > When reviewing Bigtop patches, I often find myself >> > >> downloading >> > >> the >> > >> > patch, applying it to my local repo and using a diff tool >> > >> on my >> > >> computer >> > >> > to review the patch (and have some context around it). I >> > >> know for >> > >> a fact >> > >> > that some of the other Apache projects are very good >> > >> about asking >> > >> people >> > >> > to post reviews on reviewboard (or something similar) >> > >> when >> > >> uploading a >> > >> > new patch. That makes the process of reviewing the diff >> > >> and >> > >> commenting >> > >> > on patches much easier. >> > >> > Would it make sense for us to start leveraging something >> > >> > like https://reviews.apache.org/dashboard/ for this? >> > >> > Thoughts? >> > >> > Mark >> > >> > >> > >> > -- >> > >> > // Jonathan Hsieh (shay) >> > >> > // Software Engineer, Cloudera >> > >> > // j...@cloudera.com >> > >> > >> > >> >> > >> -- >> > >> Apache MRUnit - Unit testing MapReduce - >> > >> http://incubator.apache.org/mrunit/ >> > >> > >> > >> > -- >> > Harsh J > >
