You can CTR the commit to master now so that it's there for when I go through the binary release process tomorrow after the 3.4.0 vote closes.
On Mon, Jan 7, 2019 at 7:38 AM Florian Hockmann <[email protected]> wrote: > Oh no, completely forgot about that one -.- > Sorry about that! > > I don't think that it really matters whether it's changed after tagging > 3.4.0 or not as it only affects the tags in hub.docker.com. It would only > be good if we could change that before 3.4.0 gets deployed. Should I make a > CTR commit to master or would you prefer another approach since we're > currently in the release process? > > Alternatively, you could also add the tag manually this one time and then > we'll resolve the ticket later for the next release. These commands should > do the trick (assuming that you built the 3.4.0 images already, e.g., with > 'mvn clean install -pl gremlin-server,gremlin-console -DdockerImages'): > docker tag tinkerpop/gremlin-console:3.4.0 tinkerpop/gremlin-console:latest > docker push tinkerpop/gremlin-console:latest > docker tag tinkerpop/gremlin-server:3.4.0 tinkerpop/gremlin-server:latest > docker push tinkerpop/gremlin-server:latest > > This works as long as 3.3.5 is already deployed as that would otherwise > use the latest tag again and therefore overwrite the tagging of latest for > the 3.4.0 images > > -----Ursprüngliche Nachricht----- > Von: Stephen Mallette <[email protected]> > Gesendet: Montag, 7. Januar 2019 13:07 > An: [email protected] > Betreff: Re: [DISCUSS] code freeze 3.2.11/3.3.5/3.4.0 > > Florian - i suppose this one needs to be done? > > https://issues.apache.org/jira/browse/TINKERPOP-1980 > > should it have been done before we tagged 3.4.0 and built all the > artifacts? does it matter? > > On Fri, Jan 4, 2019 at 4:52 PM Stephen Mallette <[email protected]> > wrote: > > > well - that was absolute insanity right down the > > valid-distiribution.sh on 3.4.0. Please make sure you pull master > > before testing 3.4.0 as kuppitz had to drop in a patch.......... > > > > On Wed, Jan 2, 2019 at 4:50 PM Stephen Mallette <[email protected]> > > wrote: > > > >> so, 3.2.11 and 3.3.5 releases are prepared - vote threads were just > sent. > >> i would have had 3.4.0 done too but i flubbed the 2018/2019 date on > >> 3.3.5 and had to redo the whole thing basically - so that was rotten. > >> I'll have > >> 3.4.0 out tomorrow. > >> > >> On Wed, Jan 2, 2019 at 7:34 AM Stephen Mallette > >> <[email protected]> > >> wrote: > >> > >>> yes. i can also use a specific commit, but it invariably leaves out > >>> commits for a release. i guess those commits left out will just be > >>> release manager commits typically, but i dunno, it just bugged me > >>> that it wasn't the actual 3.2.10..3.2.11 > >>> > >>> On Wed, Jan 2, 2019 at 7:15 AM Robert Dale <[email protected]> wrote: > >>> > >>>> Can't you just use HEAD? > >>>> > >>>> git shortlog -sn 3.2.10..HEAD > >>>> > >>>> Robert Dale > >>>> > >>>> > >>>> On Wed, Jan 2, 2019 at 7:11 AM Stephen Mallette > >>>> <[email protected]> > >>>> wrote: > >>>> > >>>> > All PRs appear merged....I've starting the release process > >>>> > beginning > >>>> with > >>>> > 3.2.11. Should see VOTE threads coming soon. > >>>> > > >>>> > Separately, I already found a problem with using git shortlog to > >>>> generate > >>>> > the "contributor list" as there is no 3.2.11 tag to use in the > >>>> arguments > >>>> > given to it. It's a bit of a cart before the horse problem. I > >>>> > will > >>>> come up > >>>> > with something else to deal with that. Probably related to the > >>>> > website which can be updated out of band with the release. Just > >>>> > wanted to > >>>> point out > >>>> > the problem so that no one would be expecting to see that content > >>>> > in > >>>> the > >>>> > release notes when the VOTE threads go out. > >>>> > > >>>> > On Tue, Jan 1, 2019 at 8:31 AM Stephen Mallette > >>>> > <[email protected] > >>>> > > >>>> > wrote: > >>>> > > >>>> > > I don't know what is going on but yesterday i spent a few > >>>> > > minutes > >>>> messing > >>>> > > with the failing python build with radish and after a few > >>>> > > failures > >>>> it > >>>> > just > >>>> > > started working again locally. Then I re-ran failed travis jobs > >>>> > > and > >>>> they > >>>> > > started passing. I guess the problem was fixed externally - weird. > >>>> I > >>>> > > expect to start the release process soon - there are just a > >>>> > > couple > >>>> more > >>>> > > longstanding PRs to merge that delayed a little during the > >>>> > > holiday > >>>> > period. > >>>> > > > >>>> > > > >>>> > > > >>>> > > On Fri, Dec 28, 2018 at 6:10 PM Daniel Kuppitz > >>>> > > <[email protected]> > >>>> wrote: > >>>> > > > >>>> > >> A quick & dirty fix for the broken radish libs to at least get > >>>> > >> a > >>>> clean > >>>> > >> build in docker: > >>>> > >> > >>>> > >> diff --git a/docker/scripts/build.sh b/docker/scripts/build.sh > >>>> > >> index e172ddaaa9..8b56cf53c3 100755 > >>>> > >> --- a/docker/scripts/build.sh > >>>> > >> +++ b/docker/scripts/build.sh > >>>> > >> *@@ -78,7 +78,10 @@* if [ -r "settings.xml" ]; then > >>>> > >> cp settings.xml ~/.m2/ > >>>> > >> fi > >>>> > >> > >>>> > >> *-mvn clean install process-resources > >>>> > >> ${TINKERPOP_BUILD_OPTIONS} > >>>> || exit > >>>> > >> 1* > >>>> > >> *+mvn clean install -DskipTests* *+sed -i > >>>> > >> 's/background=background,/background=background/g' > >>>> > >> > >>>> > >> > >>>> > > >>>> gremlin-python/target/python2/env/local/lib/python2.7/site-packages > >>>> /radish/parser.py* > >>>> > >> *+* > >>>> > >> +mvn install process-resources ${TINKERPOP_BUILD_OPTIONS} || > >>>> > >> +exit 1 > >>>> > >> [ -z "${BUILD_JAVA_DOCS}" ] || mvn process-resources > >>>> > >> -Djavadoc || > >>>> exit > >>>> > 1 > >>>> > >> > >>>> > >> if [ ! -z "${BUILD_USER_DOCS}" ]; then > >>>> > >> > >>>> > >> > >>>> > >> > >>>> > >> Cheers, > >>>> > >> Daniel > >>>> > >> > >>>> > >> > >>>> > >> On Thu, Dec 27, 2018 at 6:18 AM Stephen Mallette < > >>>> [email protected]> > >>>> > >> wrote: > >>>> > >> > >>>> > >> > Hi all, just checking in during the holiday period. My > >>>> > >> > laptop > >>>> will > >>>> > >> return > >>>> > >> > to the off position shortly, but I wanted to point out that > >>>> > >> > we > >>>> seem to > >>>> > >> have > >>>> > >> > a problem with Python: > >>>> > >> > > >>>> > >> > https://travis-ci.org/apache/tinkerpop/jobs/472444735 > >>>> > >> > > >>>> > >> > every current PR seems to be in fail mode right now. We will > >>>> want to > >>>> > >> sort > >>>> > >> > that out before release.Looks like something in radish. If > >>>> anyone can > >>>> > >> have > >>>> > >> > a look this week that would be helpful as I won't have time > >>>> > >> > to > >>>> dig on > >>>> > it > >>>> > >> > too deeply until next week (which is when we are supposed to > >>>> > >> > be > >>>> > >> releasing). > >>>> > >> > > >>>> > >> > On Mon, Dec 17, 2018 at 7:09 AM Stephen Mallette < > >>>> > [email protected]> > >>>> > >> > wrote: > >>>> > >> > > >>>> > >> > > While I think we could go on for every adding things to > >>>> > >> > > 3.4.0 I > >>>> > think > >>>> > >> > it's > >>>> > >> > > time to cut it off and release. There's too many good > >>>> > >> > > things in > >>>> > there > >>>> > >> to > >>>> > >> > > hold for any longer. We are looking at releasing 3.2.11, > >>>> > >> > > 3.3.5 > >>>> and > >>>> > >> 3.4.0. > >>>> > >> > > > >>>> > >> > > I'd propose we polish up remaining items this week, set > >>>> > >> > > for > >>>> code > >>>> > >> freeze > >>>> > >> > > 12/22 and then build the release for VOTE the week of the > >>>> 31st. I > >>>> > >> assume > >>>> > >> > > that there are enough PMC members around the holiday > >>>> > >> > > period to > >>>> VOTE > >>>> > on > >>>> > >> > the > >>>> > >> > > release artifacts. If the VOTE has to stay open a bit > >>>> > >> > > longer > >>>> than is > >>>> > >> > > typical then that's ok. I'm happy to just do all three > >>>> > >> > > releases > >>>> > myself > >>>> > >> > this > >>>> > >> > > time as it might be hard to coordinate with others during > >>>> > >> > > the > >>>> > holiday > >>>> > >> > > period. > >>>> > >> > > > >>>> > >> > > We still have a number of important things to finish - > >>>> specifically: > >>>> > >> > > > >>>> > >> > > 1. code reviews on open PRs with ids > 1000 2. finish up > >>>> > >> > > the GraphBinary - jorge is adding two more > >>>> serializers > >>>> > >> and > >>>> > >> > > then i think we can call this a day and put it up for review. > >>>> > >> > > 3. documentation review > >>>> > >> > > 4. anything else? > >>>> > >> > > > >>>> > >> > > As usual, let's continue to use this thread for release > >>>> coordination > >>>> > >> > > heading into code freeze. > >>>> > >> > > > >>>> > >> > > > >>>> > >> > > > >>>> > >> > > >>>> > >> > >>>> > > > >>>> > > >>>> > >>> > >
