Hello everyone! Thank you for discussing this.
There were two bugs during the release of 8.10.0. I considered the first bug wasn't a blocker to respin 8.10.0 RC1 and it made into 8.10.0 RC2. However, Mayya discovered that the second bug had a severe impact on search after requests with sort, and we didn't fully understand its severity until 8.10.0 was out. Although the bug has been in the previous versions, I am +1 to release 8.10.1 to reduce the impact. Best Regards, Nhat On Wed, Oct 6, 2021 at 2:36 PM Timothy Potter <[email protected]> wrote: > I agree with Mike on this one as well. In addition, I'm surprised > nobody asked to halt the RC1 and make RC2 with Nhat's fix while I was > doing 8.10. Nhat made it sound like it was not a big deal at the time, > but now there's some urgency in releasing it? > > Tim > > On Wed, Oct 6, 2021 at 11:15 AM Mike Drob <[email protected]> wrote: > > > > It feels weird to say that I’m against releases, but generally I feel > like bug fix releases should be scoped either for a regression discovered > in that release or for rapid security fixes. Otherwise, what’s the harm in > waiting for the next release train? > > > > Obviously any committee is free to create a release candidate on any > commit, and if there are three PMC members in support then a release can > happen, but I don’t want to be putting pressure on ourselves where we are > constantly in the middle of a release cycle. > > > > Or waiting a month and doing 8.11 seems fine too? > > > > On Wed, Oct 6, 2021 at 8:17 AM Adrien Grand <[email protected]> wrote: > >> > >> Mike, it's unclear to me if you are suggesting waiting before doing a > 8.10.1 release? On my end I'm good with doing a 8.10.1 release now, we > could still do a 8.10.2 release later in case we find new bugs? > >> > >> On Tue, Oct 5, 2021 at 10:41 PM Mayya Sharipova < > [email protected]> wrote: > >>> > >>> No, the bug is not new and was present in the previous versions as > well, but was discovered quite recently. > >>> > >>> On Tue, Oct 5, 2021 at 3:54 PM Mike Drob <[email protected]> wrote: > >>>> > >>>> Is the bug new in 8.10? If it affects older versions as well then I > feel like 8.10.1 might be less urgent. > >>>> > >>>> Mike > >>>> > >>>> On Tue, Oct 5, 2021 at 2:14 PM Adrien Grand <[email protected]> > wrote: > >>>>> > >>>>> +1 to a 8.10.1 patch release > >>>>> > >>>>> On Tue, Oct 5, 2021 at 2:03 AM Mayya Sharipova < > [email protected]> wrote: > >>>>>> > >>>>>> Thanks for the update, Robert. Would be nice to have these bug > fixes as well. > >>>>>> > >>>>>> On Mon, Oct 4, 2021 at 7:56 PM Robert Muir <[email protected]> > wrote: > >>>>>>> > >>>>>>> FYI Looks like there are already six items currently listed under > >>>>>>> "Bugfixes" for 8.11.0, so those could be candidates for the patch > >>>>>>> release. > >>>>>>> > >>>>>>> Bug Fixes > >>>>>>> --------------------- > >>>>>>> * LUCENE-10110: MultiCollector now handles single leaf collector > that > >>>>>>> wants to skip low-scoring hits > >>>>>>> but the combined score mode doesn't allow it. (Jim Ferenczi) > >>>>>>> * LUCENE-10111: Missing calculating the bytes used of > DocsWithFieldSet > >>>>>>> in NormValuesWriter. > >>>>>>> (Lu Xugang) > >>>>>>> * LUCENE-10116: Missing calculating the bytes used of > DocsWithFieldSet > >>>>>>> and currentValues in SortedSetDocValuesWriter. > >>>>>>> (Lu Xugang) > >>>>>>> * LUCENE-10070 Skip deleted docs when accumulating facet counts for > >>>>>>> all docs. (Ankur Goel, Greg Miller) > >>>>>>> * LUCENE-10126: Sort optimization with a chunked bulk scorer > >>>>>>> can wrongly skip documents (Nhat Nguyen, Mayya Sharipova) > >>>>>>> * LUCENE-10134: ConcurrentSortedSetDocValuesFacetCounts shouldn't > >>>>>>> share liveDocs Bits across threads. > >>>>>>> (Ankur Goel) > >>>>>>> > >>>>>>> On Mon, Oct 4, 2021 at 7:46 PM Mayya Sharipova <[email protected]> > wrote: > >>>>>>> > > >>>>>>> > Hello everyone! > >>>>>>> > Thank you, Timothy, for the recent 8.10 release. > >>>>>>> > > >>>>>>> > I wonder if we are ok to do a 8.10.1 patch release and do it > fairly soon? this week? > >>>>>>> > Nhat fixed a bad bug where "sort with after" on a numeric field > can incorrectly miss documents. This bug only manifests when sort > optimization on numeric fields is explicitly enabled. > >>>>>>> > > >>>>>>> > Given that we are not sure when the next minor release will be, > it would be useful to have a patch release. If nobody is opposed to it, I > can volunteer to be the Release Manager. > >>>>>>> > > >>>>>>> > Thanks. > >>>>>>> > >>>>>>> > --------------------------------------------------------------------- > >>>>>>> To unsubscribe, e-mail: [email protected] > >>>>>>> For additional commands, e-mail: [email protected] > >>>>>>> > >>>>> > >>>>> > >>>>> -- > >>>>> Adrien > >> > >> > >> > >> -- > >> Adrien > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > >
