I'll help Kelvin out with some answers here: > 1. I was looking at the commits going in the release and found that we are missing mention of certain changes in the changelog such as https://github.com/apache/tinkerpop/commit/e92d53fa0767fafa4e74f12ca84542d77d7d2976 , https://github.com/apache/tinkerpop/commit/81f884e2782939ddaf338ec87b38ad67690bc0a8 and https://github.com/apache/tinkerpop/commit/ae720befe507cd3385d32e7bd91229bd51758ea8. Please correct me if I am wrong but I would assume that all relevant changes (even if they are part of RC version of go client) should be added to the changelog? If you agree, I can do an exhaustive run though the commits on Monday and fix the change log.
Most of those links for commits are for gremlin-go. I think that was fine because we hadn't released gremlin-go yet so CHANGELOG entries on something that wasn't official (beyond the release candidate) didn't seem necessary. As for the one Java commit example there, yes, we probably should have included that. Of course, I think the big error is that we didn't include gremlin-go at all in the CHANGELOG for 3.5.4. It is however in the Upgrade Documentation. My personal opinion here is that the CHANGELOG omissions should not block release. Changelog should have been caught during code freeze, though the official code freeze reminder wasn't sent so maybe this release caught folks a bit off guard. otoh, i'm not sure how many folks actually do any sort of reviews during that week aside from the actual release manager so it wouldn't have been caught anyway. What goes into CHANGELOG has always been highly subjective but generally speaking there should be one or more entries per JIRA and zero or more for a CTR commit. We tend to avoid entries that deal with tests or documentation and focus more on user content and refactoring or behavioral shifts. When I used to keep my eye on TinkerPop on a more daily basis, I'd try to ensure the CHANGELOG stayed perfect but without someone actively doing that I imagine it will lose a bit of quality. We've had this kind of thing happen before and I usually just make note to come back and fix the CHANGELOG post release once the branches have reopened. I would definitely add gremlin-go to 3.5.4 to denote its official release and tuck in an entry about that "backport" commit you found. Please feel free to do that when the branches reopen. > 2. I performed some sanity checks by downloading the binary and connecting the server via the console. I observed some warnings while starting the console but I think those are not release blocking. That's a Java 11 warning given Groovy 2.x - it stinks but can't be avoided I'm afraid. We've been releasing with that since 3.5.x I think > 3. Since Kelvin, you are performing the release and signing it, would we require you to add your public keys Kelvin is still working through some setup issues with Apache and build environment and indeed, his keys should be in that list. In an effort to not let the release slip further, we signed with my keys. > 4. I ran the tests for the 3.5.4 branch and ensured that all tests passed. I did not verify that the number of tests running for 3.5.4 >= number of tests running for 3.5.3. I'm not sure If i'm crediting it right, but Lyndon/Yang setup codecov for gremlin-go. I thought the idea was to try to get that running across the whole project. It would be neat to have those metrics since codecov seems to be working out for us. > 5. I found a broken link in the upgrade documentation. At https://tinkerpop.apache.org/docs/3.5.4/upgrade/#_gremlin_go, when we click on the "here" hyperlink on the line "For full documentation on the available API to give Gremlin Go a try, click here <https://pkg.go.dev/github.com/apache/tinkerpop/gremlin-go/[email protected]/driver>." it takes us to a page which says "Oops! We couldn't find “ github.com/apache/tinkerpop/gremlin-go/v3/[email protected]”. That is expected and you will find others like it. Until the VOTE is over and we official release, we can't expose any convenience binaries for those versions, so [email protected] won't exist yet until after the vote and the post-release tasks are completed: https://tinkerpop.apache.org/docs/current/dev/developer/#_post_release_tasks If the problem remains after the post-release tasks for some reason, I typically just manually go edit the HTML and correct it (as well as the asciidoc separately). > 6. https://issues.apache.org/jira/browse/TINKERPOP-2708 (added to 3.5.4 changelog <https://tinkerpop.apache.org/docs/3.5.4/upgrade/#_deprecated_connectonstartup_option_in_gremlin_javascript>) is a breaking change for users of gremlin-javascript. Are we sure we want to introduce a breaking change in a patch version release 3.5.4? well spotted. the good news is that i think it's not as bad as it seems. the actual breaking change occurs in 3.7.0 and not 3.5.4. In 3.5.4 the user will just get a deprecation warning if they use that feature and normal usage should remain unchanged. We probably needed a JIRA for 3.5.4 change and one for 3.7.0 change, but they were squished into one so it's confusing. My opinion would be to not block release over that. If we were really bugged by it we could just manually edit the 3.5.4 Upgrade Docs to remove (or replace) the JIRA reference. Not sure it's necessary though unless someone feels really strongly about it. As an aside, Divij, I know you'd wanted to look in on the release process while it was happening. Dave was interested too, but Kelvin and I had a hard time connecting to get these releases out and scheduling was such that we ended up working on it little 20 to 45 minute increments whenever we could each find some ad-hoc moments to meet and they were usually deep in our afternoons (my evenings) which wouldn't have fit your time zone well. If you choose to pick up the release manager hat next time around, I'd be happy to schedule in your timezone so you can see how it works. Well, I hope I covered everything. Thanks for all your diligence on this release review! On Sat, Jul 23, 2022 at 6:22 AM Divij Vaidya <[email protected]> wrote: > 5. I found a broken link in the upgrade documentation. At > https://tinkerpop.apache.org/docs/3.5.4/upgrade/#_gremlin_go, when we > click on the "here" hyperlink on the line "For full documentation on the > available API to give Gremlin Go a try, click here > <https://pkg.go.dev/github.com/apache/tinkerpop/gremlin-go/[email protected]/driver>." > it takes us to a page which says "Oops! We couldn't find “ > github.com/apache/tinkerpop/gremlin-go/v3/[email protected]”. > > 6. https://issues.apache.org/jira/browse/TINKERPOP-2708 (added to 3.5.4 > changelog > <https://tinkerpop.apache.org/docs/3.5.4/upgrade/#_deprecated_connectonstartup_option_in_gremlin_javascript>) > is a breaking change for users of gremlin-javascript. Are we sure we want > to introduce a breaking change in a patch version release 3.5.4? > > Divij Vaidya > > > > On Sat, Jul 23, 2022 at 12:11 PM Divij Vaidya <[email protected]> > wrote: > >> Fixing a typo -> "I am completely *not* sure about this but perhaps..." >> >> -- >> Divij Vaidya >> >> >> >> On Sat, Jul 23, 2022 at 12:04 PM Divij Vaidya <[email protected]> >> wrote: >> >>> Hey Kelvin >>> >>> Thank you for picking up the responsibility for the release process. I >>> have a couple of questions before I can vote to go ahead with the release. >>> >>> 1. I was looking at the commits going in the release and found that we >>> are missing mention of certain changes in the changelog such as >>> https://github.com/apache/tinkerpop/commit/e92d53fa0767fafa4e74f12ca84542d77d7d2976, >>> >>> https://github.com/apache/tinkerpop/commit/81f884e2782939ddaf338ec87b38ad67690bc0a8 >>> and >>> https://github.com/apache/tinkerpop/commit/ae720befe507cd3385d32e7bd91229bd51758ea8. >>> Please correct me if I am wrong but I would assume that all relevant >>> changes (even if they are part of RC version of go client) should be added >>> to the changelog? If you agree, I can do an exhaustive run though the >>> commits on Monday and fix the change log. >>> >>> 2. I performed some sanity checks by downloading the binary and >>> connecting the server via the console. I observed some warnings while >>> starting the console but I think those are not release blocking. >>> >>> dev-dsk-diviv-1a-190badd5 % ./bin/gremlin.sh >>> WARNING: An illegal reflective access operation has occurred >>> WARNING: Illegal reflective access by >>> org.codehaus.groovy.reflection.CachedClass >>> (file:/local/home/diviv/apache-tinkerpop-gremlin-console-3.5.4/lib/groovy-2.5.14-indy.jar) >>> to method java.lang.Object.finalize() >>> WARNING: Please consider reporting this to the maintainers of >>> org.codehaus.groovy.reflection.CachedClass >>> WARNING: Use --illegal-access=warn to enable warnings of further illegal >>> reflective access operations >>> WARNING: All illegal access operations will be denied in a future release >>> >>> 3. Since Kelvin, you are performing the release and signing it, would we >>> require you to add your public keys to >>> https://dist.apache.org/repos/dist/dev/tinkerpop/KEYS? I am completely >>> sure about this but perhaps @Stephen Mallette <[email protected]> might >>> be able to answer this one. >>> >>> dev-dsk-diviv-1a-190badd5 % gpg --verify >>> apache-tinkerpop-gremlin-console-3.5.4-bin.zip.asc >>> apache-tinkerpop-gremlin-console-3.5.4-bin.zip >>> gpg: Signature made Tue 19 Jul 2022 09:27:18 PM UTC using RSA key ID >>> 01B1989A >>> gpg: Good signature from "Stephen Mallette <[email protected]>" >>> gpg: WARNING: This key is not certified with a trusted signature! >>> gpg: There is no indication that the signature belongs to the >>> owner. >>> Primary key fingerprint: 8D8C 72F9 4E0D E6A6 7E1B 87A7 6F38 BF38 01B1 >>> 989A >>> >>> 4. I ran the tests for the 3.5.4 branch and ensured that all tests >>> passed. I did not verify that the number of tests running for 3.5.4 >= >>> number of tests running for 3.5.3. >>> >>> -- >>> Divij Vaidya >>> >>> >>> >>> On Fri, Jul 22, 2022 at 11:52 PM Florian Hockmann < >>> [email protected]> wrote: >>> >>>> I checked the links and everything looks good. Thanks for taking care >>>> of the releases. >>>> >>>> VOTE +1 >>>> >>>> ________________________________ >>>> Von: Kelvin Lawrence <[email protected]> >>>> Gesendet: Mittwoch, 20. Juli 2022 23:05 >>>> An: [email protected] >>>> Betreff: Subject: [VOTE] TinkerPop 3.5.4 Release >>>> >>>> Hello, >>>> >>>> We are happy to announce that TinkerPop 3.5.4 is ready for release. >>>> >>>> The release artifacts can be found at this location: >>>> https://dist.apache.org/repos/dist/dev/tinkerpop/3.5.4/ >>>> >>>> The source distribution is provided by: >>>> apache-tinkerpop-3.5.4-src.zip >>>> >>>> Two binary distributions are provided for user convenience: >>>> apache-tinkerpop-gremlin-console-3.5.4-bin.zip >>>> apache-tinkerpop-gremlin-server-3.5.4-bin.zip >>>> >>>> The GPG key used to sign the release artifacts is available at: >>>> https://dist.apache.org/repos/dist/dev/tinkerpop/KEYS >>>> >>>> The online docs can be found here: >>>> https://tinkerpop.apache.org/docs/3.5.4/ (user docs) >>>> https://tinkerpop.apache.org/docs/3.5.4/upgrade/ (upgrade >>>> docs) >>>> https://tinkerpop.apache.org/javadocs/3.5.4/core/ (core >>>> javadoc) >>>> https://tinkerpop.apache.org/javadocs/3.5.4/full/ (full >>>> javadoc) >>>> https://tinkerpop.apache.org/dotnetdocs/3.5.4/ (.NET API docs) >>>> https://tinkerpop.apache.org/jsdocs/3.5.4/ (Javascript API >>>> docs) >>>> >>>> The Git tag can be found here: >>>> https://github.com/apache/tinkerpop/tree/3.5.4 >>>> >>>> The release notes are available here: >>>> >>>> https://github.com/apache/tinkerpop/blob/3.5.4/CHANGELOG.asciidoc >>>> >>>> The [VOTE] will be open for the next 72 hours --- closing Saturday >>>> (July >>>> 23rd 2022) at 21:00 GMT (16:00 US CDT) >>>> >>>> My vote is +1. >>>> >>>> Thank you very much, >>>> Kelvin Lawrence >>>> >>>
