Okay, I reran the smoke tests using releaseWizard after following this
suggestion from
https://cwiki.apache.org/confluence/display/LUCENE/ReleaseTodo.
To speed up tests while building the RC, you can increase parallelism by
adding the following settings to your *~/lucene.build.properties*
tests.jvms=8
tests.jvms.override=8
That sped up the smoke test process significantly and I could get a
successful run:
verify that each binary artifact has a deployed POM...
verify Maven artifacts' md5/sha1 digests...
check for javadoc and sources artifacts...
verify deployed POMs' coordinates...
.
unpack lucene-9.12.3.tgz...
verify that Maven artifacts are same as in the binary distribution...
verify JAR metadata/identity/no javax.* or java.* classes...
SUCCESS! [0:28:33.753611]
Sending out the vote request shortly.
Thanks
Ankit
On Thu, Sep 18, 2025 at 8:16 AM Michael Sokolov <[email protected]> wrote:
> I ran that command ... and it finished SUCCESS! [0:07:40.669238]
> without ever running any tests at all?
>
> On Thu, Sep 18, 2025 at 3:49 AM Ankit Jain <[email protected]> wrote:
> >
> > I was planning to send a VOTE email today, but was unable to get smoke
> tests to pass. During smoke test run from the release wizard, only
> TestTieredMergePolicy was failing which seems flaky -
> https://github.com/apache/lucene/issues/13667#issuecomment-3305466611
> > Hence, I staged the artifacts and ran the smoke tests from another
> machine. But now, a bunch of codes and backward-codecs tests also failed.
> >
> > I am wondering if this has something to do with my machine or issue in
> general. Is someone able to run the smoke tests and confirm one way or the
> other?
> >
> > python3 -u dev-tools/scripts/smokeTestRelease.py \
> >
> >
> https://dist.apache.org/repos/dist/dev/lucene/lucene-9.12.3-RC1-rev-d2cf5568e3965cc642cae3b695fcbeff5bb5c2ed
> >
> >
> >
> >
> > Thanks
> > Ankit
> >
> > On Wed, Sep 17, 2025 at 1:15 AM Ankit Jain <[email protected]>
> wrote:
> >>
> >> Thanks Mike for the reminder. I created a signing gpg key and uploaded
> the same to https://dist.apache.org/repos/dist/release/lucene/KEYS.
> Hopefully that should avoid any key related issues.
> >> Will also get my key signed by one of the committers to prevent tests
> from complaining. I am assuming that the gpg public key block does not
> change after getting it signed.
> >>
> >> On Tue, Sep 16, 2025 at 7:27 PM Michael Sokolov <[email protected]>
> wrote:
> >>>
> >>> Have you gotten a code signing gpg key? That usually seems to cause
> some delays since the key must be committed to a file in order for the
> release to go through
> >>>
> >>> On Tue, Sep 16, 2025, 3:17 PM Ankit Jain <[email protected]>
> wrote:
> >>>>
> >>>> Thanks Benjamin and Chris. Will kick off the release process later
> today or tomorrow!
> >>>>
> >>>> I can see that all the proposed changes above have already been
> backported into branch_9_12. Additionally, I see another change
> >>>>
> >>>> Ensure vector queries handle advanceShallow correctly (#14858)
> >>>>
> >>>> that will get released as part of 9.12.3.
> >>>>
> >>>>
> >>>> Thanks
> >>>> Ankit
> >>>>
> >>>> On Mon, Sep 15, 2025 at 4:41 AM Chris Hegarty
> <[email protected]> wrote:
> >>>>>
> >>>>> Thanks Ankit,
> >>>>>
> >>>>> I backported the missing DOAP changes for release 10.2.2 and 10.3.0
> [*]. I hope the release goes smoothly! :fingers_crossed:
> >>>>>
> >>>>> -Chris.
> >>>>>
> >>>>> [*]
> https://github.com/apache/lucene/commit/d2cf5568e3965cc642cae3b695fcbeff5bb5c2ed
> >>>>>
> >>>>>
> >>>>> > On 15 Sep 2025, at 11:59, Benjamin Trent <[email protected]>
> wrote:
> >>>>> >
> >>>>> > Ankit,
> >>>>> >
> >>>>> > You got it :). No push back and folks seem ready for a release.
> You can start the process as soon as today/tomorrow
> >>>>> >
> >>>>> > Good luck!
> >>>>> >
> >>>>> >
> >>>>> > Ben
> >>>>> >
> >>>>> > On Sun, Sep 14, 2025 at 10:22 AM Michael McCandless <
> [email protected]> wrote:
> >>>>> > Thank you everyone! +1 for a rare N-2 bugfix release.
> >>>>> >
> >>>>> > Mike McCandless
> >>>>> >
> >>>>> > http://blog.mikemccandless.com
> >>>>> >
> >>>>> >
> >>>>> > On Sun, Sep 14, 2025 at 8:38 AM Uwe Schindler <[email protected]>
> wrote:
> >>>>> > Hi,
> >>>>> > all fine. It was already backported by Chris. I haven't seen the
> message, sorry!
> >>>>> > Uwe
> >>>>> > Am 14.09.2025 um 15:33 schrieb Uwe Schindler:
> >>>>> >> Hi,
> >>>>> >> I'd like to include the bugfix for the Spinloop bug in the shared
> Arena grouping. I'll backport that in a moment.
> >>>>> >> Uwe
> >>>>> >> Am 14.09.2025 um 07:41 schrieb Ankit Jain:
> >>>>> >>> Hi Chris,
> >>>>> >>>
> >>>>> >>> I would like to volunteer and be RM for the first time. Benjamin
> made a really good point about using bug fix release as stepping stone to
> minor version release.
> >>>>> >>>
> >>>>> >>>
> >>>>> >>> Thanks
> >>>>> >>> Ankit
> >>>>> >>>
> >>>>> >>> On Sat, Sep 13, 2025 at 7:28 AM Benjamin Trent <
> [email protected]> wrote:
> >>>>> >>> Hey Chris,
> >>>>> >>>
> >>>>> >>> Thank you for this proposal. I agree, these are pretty critical
> bug fixes.
> >>>>> >>>
> >>>>> >>> I can be released manager for this one :). Need to shake my
> "release manager dust" off.
> >>>>> >>>
> >>>>> >>> Unless there is someone who REALLY wants the opportunity to do
> their first time as RM. A big fix release is usually an easier option for
> your first release.
> >>>>> >>>
> >>>>> >>> On Sat, Sep 13, 2025, 10:13 AM Chris Hegarty
> <[email protected]> wrote:
> >>>>> >>> Hi,
> >>>>> >>>
> >>>>> >>> Lucene 10.3 is not yet out the door, but I’m back proposing yet
> another release! This time a bugfix release of 9.12.
> >>>>> >>>
> >>>>> >>> My primary motivation for proposing this release is to resolve a
> number of issues related to sharing of arenas in our mmap implementation.
> Specifically:
> >>>>> >>>
> >>>>> >>> 1. Reduce sharedArenaMaxPermits from 1024 to 64 (#15078)
> >>>>> >>> 2. Use READONCE to read segment infos (#15074)
> >>>>> >>> 3. Fix rare spin-loop in RefCountedSharedArena (#15112)
> >>>>> >>>
> >>>>> >>> Any other potential bug fixes for 9.12.3? I will try to find
> someone to be release manager or volunteer to do it myself if not. Let’s
> try to get an RC build out towards the end of next week, if possible.
> >>>>> >>>
> >>>>> >>> -Chris.
> >>>>> >>>
> >>>>> >>> P.S. I see that the following bug fix is already in branch_9_12
> - Ensure vector queries handle advanceShallow correctly #14858
> >>>>> >>>
> >>>>> >>>
> ---------------------------------------------------------------------
> >>>>> >>> To unsubscribe, e-mail: [email protected]
> >>>>> >>> For additional commands, e-mail: [email protected]
> >>>>> >>>
> >>>>> >> --
> >>>>> >> Uwe Schindler
> >>>>> >> Achterdiek 19, D-28357 Bremen
> >>>>> >> https://www.thetaphi.de
> >>>>> >> eMail: [email protected]
> >>>>> > --
> >>>>> > Uwe Schindler
> >>>>> > Achterdiek 19, D-28357 Bremen
> >>>>> > https://www.thetaphi.de
> >>>>> > eMail: [email protected]
> >>>>>
> >>>>>
> >>>>> ---------------------------------------------------------------------
> >>>>> To unsubscribe, e-mail: [email protected]
> >>>>> For additional commands, e-mail: [email protected]
> >>>>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]
>
>
