Hi, So far we had 4 duplicate reports of https://hibernate.atlassian.net/browse/HHH-12332 (so not counting the users having seen the bug and not opening a new one). I think this issue qualifies as "we should really get a hotfix release soon".

I let the duplicated issues open for now as I saw a couple of reproducers and we should check that all the reported cases are fixed once we have a proper fix. Apart from fixing the NPE, the OP has a point when he says there's something wrong here: https://github.com/hibernate/hibernate-orm/blame/5.2/hibernate-core/src/m... . I suspect c1 and c2 should be reassigned somehow. Christian, I saw you planned to fix it, what's your plan about it? Thanks. -- Guillaume _______________________________________________ hibernate-dev mailing list hibernate-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/hibernate-dev

On 03/05/2018 08:26 AM, Sanne Grinovero wrote: > On 5 March 2018 at 12:27, Guillaume Smet <guillaume.smet(a)gmail.com&gt; > wrote: >> Hi, So far we had 4 duplicate reports of >> https://hibernate.atlassian.net/browse/HHH-12332 (so not counting the >> users having seen the bug and not opening a new one). I think this >> issue qualifies as "we should really get a hotfix release soon". > +1 that would be nice, and the most effective way to prevent further > noise and pain. I also would agree. _______________________________________________ hibernate-dev mailing list hibernate-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/hibernate-dev

I agree that in general it is ok to cut a release with a partial solution in a case like this On Mon, Mar 5, 2018 at 8:40 AM andrea boriero <andrea(a)hibernate.org&gt; wrote: > If there are no objections I can manage a release as soon as a fix is > ready. > > On 5 March 2018 at 14:20, Chris Cranford <crancran(a)gmail.com&gt; wrote: > > > On 03/05/2018 08:26 AM, Sanne Grinovero wrote: > > > On 5 March 2018 at 12:27, Guillaume Smet <guillaume.smet(a)gmail.com&gt; > > > wrote: > > >> Hi, So far we had 4 duplicate reports of > > >> https://hibernate.atlassian.net/browse/HHH-12332 (so not counting the > > >> users having seen the bug and not opening a new one). I think this > > >> issue qualifies as "we should really get a hotfix release soon". > > > +1 that would be nice, and the most effective way to prevent further > > > noise and pain. > > I also would agree. > > _______________________________________________ > > hibernate-dev mailing list > > hibernate-dev(a)lists.jboss.org > > https://lists.jboss.org/mailman/listinfo/hibernate-dev > > > _______________________________________________ > hibernate-dev mailing list > hibernate-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/hibernate-dev > _______________________________________________ hibernate-dev mailing list hibernate-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/hibernate-dev

Just so that nobody misses it, Christian pushed a PR here: https://github.com/hibernate/hibernate-orm/pull/2180 . Will require someone more familiar with ORM than me to review it :). Thanks. On Mon, Mar 5, 2018 at 7:54 PM, Gail Badner <gbadner(a)redhat.com&gt; wrote: > +1 to a quick release, with a partial solution if necessary. > > On Mon, Mar 5, 2018 at 7:28 AM, Steve Ebersole <steve(a)hibernate.org&gt; > wrote: > > > I agree that in general it is ok to cut a release with a partial > solution > > in a case like this > > > > On Mon, Mar 5, 2018 at 8:40 AM andrea boriero <andrea(a)hibernate.org&gt; > > wrote: > > > > > If there are no objections I can manage a release as soon as a fix is > > > ready. > > > > > > On 5 March 2018 at 14:20, Chris Cranford <crancran(a)gmail.com&gt; wrote: > > > > > > > On 03/05/2018 08:26 AM, Sanne Grinovero wrote: > > > > > On 5 March 2018 at 12:27, Guillaume Smet < > guillaume.smet(a)gmail.com&gt; > > > > > wrote: > > > > >> Hi, So far we had 4 duplicate reports of > > > > >> https://hibernate.atlassian.net/browse/HHH-12332 (so not > counting > > the > > > > >> users having seen the bug and not opening a new one). I think > this > > > > >> issue qualifies as "we should really get a hotfix release soon". > > > > > +1 that would be nice, and the most effective way to prevent > further > > > > > noise and pain. > > > > I also would agree. > > > > _______________________________________________ > > > > hibernate-dev mailing list > > > > hibernate-dev(a)lists.jboss.org > > > > https://lists.jboss.org/mailman/listinfo/hibernate-dev > > > > > > > _______________________________________________ > > > hibernate-dev mailing list > > > hibernate-dev(a)lists.jboss.org > > > https://lists.jboss.org/mailman/listinfo/hibernate-dev > > > > > _______________________________________________ > > hibernate-dev mailing list > > hibernate-dev(a)lists.jboss.org > > https://lists.jboss.org/mailman/listinfo/hibernate-dev > > > _______________________________________________ > hibernate-dev mailing list > hibernate-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/hibernate-dev >

Hi, So, AFAICS, Andrea found another failing case for the new PR. I would vote for reverting the original patch [ https://hibernate.atlassian.net/browse/HHH-11544] for now and let Christian the time to think of a proper fix. I'm a bit worried we will miss something if we try to find a fix in a hurry. AFAICS, the original issue was a bit of a corner case so better get the other more common cases working. Anyone against reverting the original patch? Christian?

On 6 March 2018 at 11:21, Christian Beikov <christian.beikov(a)gmail.com&gt; wrote: > Hey, > > I'm fine with reverting the patch as well, though I just fixed the issue > Andrea found and am confident about the solution now. Hi Christian, that's great! Let's go with your fix then. Thanks, Sanne > > > Mit freundlichen Grüßen, > ------------------------------------------------------------------------ > *Christian Beikov* > Am 06.03.2018 um 11:57 schrieb Guillaume Smet: >> Hi, >> >> So, AFAICS, Andrea found another failing case for the new PR. >> >> I would vote for reverting the original patch [ >> https://hibernate.atlassian.net/browse/HHH-11544] for now and let Christian >> the time to think of a proper fix. I'm a bit worried we will miss something >> if we try to find a fix in a hurry. >> >> AFAICS, the original issue was a bit of a corner case so better get the >> other more common cases working. >> >> Anyone against reverting the original patch? Christian? >> > > _______________________________________________ > hibernate-dev mailing list > hibernate-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/hibernate-dev _______________________________________________ hibernate-dev mailing list hibernate-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/hibernate-dev

When are you planning to release? I'd like to take a look at the original fix and PR as well. On Tue, Mar 6, 2018 at 5:44 AM, Steve Ebersole <steve(a)hibernate.org&gt; wrote: > +1 > > If you are confident, that's good for me > > > On Tue, Mar 6, 2018, 7:24 AM Sanne Grinovero <sanne(a)hibernate.org&gt; wrote: > >> On 6 March 2018 at 11:21, Christian Beikov <christian.beikov(a)gmail.com&gt; >> wrote: >>> Hey, >>> >>> I'm fine with reverting the patch as well, though I just fixed the > issue >>> Andrea found and am confident about the solution now. >> Hi Christian, that's great! Let's go with your fix then. >> >> Thanks, >> Sanne >> >> >>> >>> Mit freundlichen Grüßen, >>> ------------------------------------------------------------ > ------------ >>> *Christian Beikov* >>> Am 06.03.2018 um 11:57 schrieb Guillaume Smet: >>>> Hi, >>>> >>>> So, AFAICS, Andrea found another failing case for the new PR. >>>> >>>> I would vote for reverting the original patch [ >>>> https://hibernate.atlassian.net/browse/HHH-11544] for now and let >> Christian >>>> the time to think of a proper fix. I'm a bit worried we will miss >> something >>>> if we try to find a fix in a hurry. >>>> >>>> AFAICS, the original issue was a bit of a corner case so better get > the >>>> other more common cases working. >>>> >>>> Anyone against reverting the original patch? Christian? >>>> >>> _______________________________________________ >>> hibernate-dev mailing list >>> hibernate-dev(a)lists.jboss.org >>> https://lists.jboss.org/mailman/listinfo/hibernate-dev >> _______________________________________________ >> hibernate-dev mailing list >> hibernate-dev(a)lists.jboss.org >> https://lists.jboss.org/mailman/listinfo/hibernate-dev > _______________________________________________ > hibernate-dev mailing list > hibernate-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/hibernate-dev > _______________________________________________ hibernate-dev mailing list hibernate-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/hibernate-dev

sorry but I had an horrible night and not feeling great right now so I'll do the release late this afternoon or tomorrow morning. On 6 March 2018 at 19:32, Chris Cranford <chris(a)hibernate.org&gt; wrote: > Gail - > > I asked Andrea earlier and he hoped Wednesday March 7th; however he did > say it depends on the status of Christian's fix. > > On 03/06/2018 02:13 PM, Gail Badner wrote: > > When are you planning to release? I'd like to take a look at the original > > fix and PR as well. > > > > On Tue, Mar 6, 2018 at 5:44 AM, Steve Ebersole <steve(a)hibernate.org&gt; > wrote: > > > >> +1 > >> > >> If you are confident, that's good for me > >> > >> > >> On Tue, Mar 6, 2018, 7:24 AM Sanne Grinovero <sanne(a)hibernate.org&gt; > wrote: > >> > >>> On 6 March 2018 at 11:21, Christian Beikov < christian.beikov(a)gmail.com > > > >>> wrote: > >>>> Hey, > >>>> > >>>> I'm fine with reverting the patch as well, though I just fixed the > >> issue > >>>> Andrea found and am confident about the solution now. > >>> Hi Christian, that's great! Let's go with your fix then. > >>> > >>> Thanks, > >>> Sanne > >>> > >>> > >>>> > >>>> Mit freundlichen Grüßen, > >>>> ------------------------------------------------------------ > >> ------------ > >>>> *Christian Beikov* > >>>> Am 06.03.2018 um 11:57 schrieb Guillaume Smet: > >>>>> Hi, > >>>>> > >>>>> So, AFAICS, Andrea found another failing case for the new PR. > >>>>> > >>>>> I would vote for reverting the original patch [ > >>>>> https://hibernate.atlassian.net/browse/HHH-11544] for now and let > >>> Christian > >>>>> the time to think of a proper fix. I'm a bit worried we will miss > >>> something > >>>>> if we try to find a fix in a hurry. > >>>>> > >>>>> AFAICS, the original issue was a bit of a corner case so better get > >> the > >>>>> other more common cases working. > >>>>> > >>>>> Anyone against reverting the original patch? Christian? > >>>>> > >>>> _______________________________________________ > >>>> hibernate-dev mailing list > >>>> hibernate-dev(a)lists.jboss.org > >>>> https://lists.jboss.org/mailman/listinfo/hibernate-dev > >>> _______________________________________________ > >>> hibernate-dev mailing list > >>> hibernate-dev(a)lists.jboss.org > >>> https://lists.jboss.org/mailman/listinfo/hibernate-dev > >> _______________________________________________ > >> hibernate-dev mailing list > >> hibernate-dev(a)lists.jboss.org > >> https://lists.jboss.org/mailman/listinfo/hibernate-dev > >> > > _______________________________________________ > > hibernate-dev mailing list > > hibernate-dev(a)lists.jboss.org > > https://lists.jboss.org/mailman/listinfo/hibernate-dev > > _______________________________________________ > hibernate-dev mailing list > hibernate-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/hibernate-dev > _______________________________________________ hibernate-dev mailing list hibernate-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/hibernate-dev

I agree with the revert solution Guillaume proposed. On 7 March 2018 at 12:25, Guillaume Smet <guillaume.smet(a)gmail.com&gt; wrote: > Apparently, there is still an issue with the latest fix but a different > one: > https://hibernate.atlassian.net/browse/HHH-12332?page=com. > atlassian.jira.plugin.system.issuetabpanels%3Acomment- > tabpanel&focusedCommentId=101427#comment-101427 > > I posted a message to try to help the OP isolating the issue. > > I know it's not very satisfactory but I would really prefer that we revert > this all for now. It's too risky just to fix a corner case. > > -- > Guillaume > > On Wed, Mar 7, 2018 at 12:11 PM, andrea boriero <andrea(a)hibernate.org&gt; > wrote: > >> sorry but I had an horrible night and not feeling great right now so I'll >> do the release late this afternoon or tomorrow morning. >> >> On 6 March 2018 at 19:32, Chris Cranford <chris(a)hibernate.org&gt; wrote: >> >> > Gail - >> > >> > I asked Andrea earlier and he hoped Wednesday March 7th; however he did >> > say it depends on the status of Christian's fix. >> > >> > On 03/06/2018 02:13 PM, Gail Badner wrote: >> > > When are you planning to release? I'd like to take a look at the >> original >> > > fix and PR as well. >> > > >> > > On Tue, Mar 6, 2018 at 5:44 AM, Steve Ebersole <steve(a)hibernate.org&gt; >> > wrote: >> > > >> > >> +1 >> > >> >> > >> If you are confident, that's good for me >> > >> >> > >> >> > >> On Tue, Mar 6, 2018, 7:24 AM Sanne Grinovero <sanne(a)hibernate.org&gt; >> > wrote: >> > >> >> > >>> On 6 March 2018 at 11:21, Christian Beikov < >> christian.beikov(a)gmail.com >> > > >> > >>> wrote: >> > >>>> Hey, >> > >>>> >> > >>>> I'm fine with reverting the patch as well, though I just fixed the >> > >> issue >> > >>>> Andrea found and am confident about the solution now. >> > >>> Hi Christian, that's great! Let's go with your fix then. >> > >>> >> > >>> Thanks, >> > >>> Sanne >> > >>> >> > >>> >> > >>>> >> > >>>> Mit freundlichen Grüßen, >> > >>>> ------------------------------------------------------------ >> > >> ------------ >> > >>>> *Christian Beikov* >> > >>>> Am 06.03.2018 um 11:57 schrieb Guillaume Smet: >> > >>>>> Hi, >> > >>>>> >> > >>>>> So, AFAICS, Andrea found another failing case for the new PR. >> > >>>>> >> > >>>>> I would vote for reverting the original patch [ >> > >>>>> https://hibernate.atlassian.net/browse/HHH-11544] for now and let >> > >>> Christian >> > >>>>> the time to think of a proper fix. I'm a bit worried we will miss >> > >>> something >> > >>>>> if we try to find a fix in a hurry. >> > >>>>> >> > >>>>> AFAICS, the original issue was a bit of a corner case so better >> get >> > >> the >> > >>>>> other more common cases working. >> > >>>>> >> > >>>>> Anyone against reverting the original patch? Christian? >> > >>>>> >> > >>>> _______________________________________________ >> > >>>> hibernate-dev mailing list >> > >>>> hibernate-dev(a)lists.jboss.org >> > >>>> https://lists.jboss.org/mailman/listinfo/hibernate-dev >> > >>> _______________________________________________ >> > >>> hibernate-dev mailing list >> > >>> hibernate-dev(a)lists.jboss.org >> > >>> https://lists.jboss.org/mailman/listinfo/hibernate-dev >> > >> _______________________________________________ >> > >> hibernate-dev mailing list >> > >> hibernate-dev(a)lists.jboss.org >> > >> https://lists.jboss.org/mailman/listinfo/hibernate-dev >> > >> >> > > _______________________________________________ >> > > hibernate-dev mailing list >> > > hibernate-dev(a)lists.jboss.org >> > > https://lists.jboss.org/mailman/listinfo/hibernate-dev >> > >> > _______________________________________________ >> > hibernate-dev mailing list >> > hibernate-dev(a)lists.jboss.org >> > https://lists.jboss.org/mailman/listinfo/hibernate-dev >> > >> _______________________________________________ >> hibernate-dev mailing list >> hibernate-dev(a)lists.jboss.org >> https://lists.jboss.org/mailman/listinfo/hibernate-dev >> > > _______________________________________________ hibernate-dev mailing list hibernate-dev(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/hibernate-dev

On Wed, Mar 7, 2018 at 1:53 PM, Sanne Grinovero <sanne(a)hibernate.org&gt; wrote: > > Let's not rush things and see what Christian thinks: he should be > allowed to analyze this - if he has time and wants to - w/o too much > pressure. I see the last report is an NPE, that's typically not too > hard to fix so I'd rather try again. Well, not having too much pressure is exactly the reason why I'd like us to revert. We would then have time to iterate with the people having reported the issue without having a last release out there that does not work as expected. We apparently have at least one person willing to test the WIP PR so that's a good thing for that. > > The risk with a full revert is that it's never getting fixed: we're > missing some unit tests for "unknown" corner cases, so I propose we > put ourselves at least in the position to be able to collect enough > feedback to be able to (eventually) reproduce them and then proceed > with confidence. Reverting and releasing does not mean not pursuing the issues and making progress on this. That only means we can do it without pressure.

-- Guillaume