Please see http://www.apache.org/dev/release-publishing.html <http://www.apache.org/dev/release-publishing.html>.
Ralph > On Oct 3, 2016, at 9:20 AM, Mikael Ståldal <mikael.stal...@magine.com> wrote: > > Why do we need to vote on the exact artifact version to release? Isn't it > enough that we vote on the RC and then trust the release manager to not mess > it up when changing the version number and rebuilding? > > On Mon, Oct 3, 2016 at 6:04 PM, Remko Popma <remko.po...@gmail.com > <mailto:remko.po...@gmail.com>> wrote: > I used 2.7 and 2.7-rc1 in my previous message as an example. The point stands > regardless of the actual number. We need to vote on the version we want to > release. > > On Tue, Oct 4, 2016 at 1:01 AM, Remko Popma <remko.po...@gmail.com > <mailto:remko.po...@gmail.com>> wrote: > How would you solve the checksum problem? > > On Tue, Oct 4, 2016 at 12:50 AM, Mikael Ståldal <mikael.stal...@magine.com > <mailto:mikael.stal...@magine.com>> wrote: > I don't think we should change this now, for the 2.7 release. But consider > for the next release. > > On Mon, Oct 3, 2016 at 5:48 PM, Remko Popma <remko.po...@gmail.com > <mailto:remko.po...@gmail.com>> wrote: > If we were to modify the version from 2.7-rc1 to 2.7 after the vote passed, > the checksums would not match. We need to vote on the release version (2.7 in > this case) until the vote passes. > > If the release fails after the 2.7 release candidate we are voting on was > installed in a local repository you would have to manually remove the failed > candidate from that local repository. > > > > On Mon, Oct 3, 2016 at 11:52 PM, Matt Sicker <boa...@gmail.com > <mailto:boa...@gmail.com>> wrote: > My work projects use gradle, and I like to test release candidates with work > projects before voting. There's one use case that Mikael is alluding to. > > On 3 October 2016 at 09:42, Ralph Goers <ralph.go...@dslextreme.com > <mailto:ralph.go...@dslextreme.com>> wrote: > Why would you be using those while you are testing the release? > > To be clear, the release process requires that we build the artifacts to be > released and as part of the release process they are deployed to the Apache > Nexus staging repository (which is in the link in the email). If the release > passes I click a button in Nexus and the artifacts are pushed to Maven > central, so they have to have the real release version in them. The > distribution artifacts (the source and binaries) are extracted from what is > in the staging repository and pushed to the ASF distribution directory. I > cannot rebuild them again and change the version as that would require a new > vote. > > Ralph > >> On Oct 3, 2016, at 7:23 AM, Mikael Ståldal <mikael.stal...@magine.com >> <mailto:mikael.stal...@magine.com>> wrote: >> >> But some build tools, such as Gradle, cache stuff in other places. >> >> On Mon, Oct 3, 2016 at 4:08 PM, Ralph Goers <ralph.go...@dslextreme.com >> <mailto:ralph.go...@dslextreme.com>> wrote: >> They have the same version because if the vote passes these are the >> artifacts that will be released to Maven central. What we are doing is the >> correct process. When you test the release build locally you should delete >> it from your local repo if the build fails. >> >> Ralph >> >>> On Oct 3, 2016, at 5:45 AM, Mikael Ståldal <mikael.stal...@magine.com >>> <mailto:mikael.stal...@magine.com>> wrote: >>> >>> Currently, the Release Candidate artifacts have the same version as the >>> release will have ("2.7" currently). This can cause issues since Maven (and >>> other similar build tools) cache non-snapshot artifacts quite aggressively, >>> they are supposed to be immutable. >>> >>> Would it be possible to use another version, such as "2.7-rc1" or >>> "2.7-SNAPSHOT"? >>> >>> We don't have to change this now, but for the next release after 2.7 >>> >>> -- >>> >>> >>> Mikael Ståldal >>> Senior software developer >>> >>> Magine TV >>> mikael.stal...@magine.com <mailto:mikael.stal...@magine.com> >>> Grev Turegatan 3 | 114 46 Stockholm, Sweden | www.magine.com >>> <http://www.magine.com/> >>> >>> Privileged and/or Confidential Information may be contained in this >>> message. If you are not the addressee indicated in this message >>> (or responsible for delivery of the message to such a person), you may not >>> copy or deliver this message to anyone. In such case, >>> you should destroy this message and kindly notify the sender by reply >>> email. >> >> >> >> >> -- >> >> >> Mikael Ståldal >> Senior software developer >> >> Magine TV >> mikael.stal...@magine.com <mailto:mikael.stal...@magine.com> >> Grev Turegatan 3 | 114 46 Stockholm, Sweden | www.magine.com >> <http://www.magine.com/> >> >> Privileged and/or Confidential Information may be contained in this message. >> If you are not the addressee indicated in this message >> (or responsible for delivery of the message to such a person), you may not >> copy or deliver this message to anyone. In such case, >> you should destroy this message and kindly notify the sender by reply email. >> > > > > > -- > Matt Sicker <boa...@gmail.com <mailto:boa...@gmail.com>> > > > > > -- > > > Mikael Ståldal > Senior software developer > > Magine TV > mikael.stal...@magine.com <mailto:mikael.stal...@magine.com> > Grev Turegatan 3 | 114 46 Stockholm, Sweden | www.magine.com > <http://www.magine.com/> > > Privileged and/or Confidential Information may be contained in this message. > If you are not the addressee indicated in this message > (or responsible for delivery of the message to such a person), you may not > copy or deliver this message to anyone. In such case, > you should destroy this message and kindly notify the sender by reply email. > > > > > > > -- > > > Mikael Ståldal > Senior software developer > > Magine TV > mikael.stal...@magine.com <mailto:mikael.stal...@magine.com> > Grev Turegatan 3 | 114 46 Stockholm, Sweden | www.magine.com > <http://www.magine.com/> > > Privileged and/or Confidential Information may be contained in this message. > If you are not the addressee indicated in this message > (or responsible for delivery of the message to such a person), you may not > copy or deliver this message to anyone. In such case, > you should destroy this message and kindly notify the sender by reply email. >
