18:59:02 <elbrus> #startmeeting
18:59:02 <MeetBot> Meeting started Wed Oct 26 18:59:02 2022 UTC.  The chair is elbrus. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:59:02 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
18:59:10 <elbrus> #topic Admin
18:59:27 <elbrus> #info Previous minutes: http://meetbot.debian.net/debian-release/2022/debian-release.2022-09-28-18.59.html
18:59:45 <elbrus> #info elbrus had an action to draft a new bits <- done
18:59:56 <elbrus> #info elbrus had an action to create a Wiki page with pros and cons about no-change NMU vs binNMU
19:00:08 <elbrus> https://wiki.debian.org/Teams/ReleaseTeam/binNMUvsNMU
19:00:24 <elbrus> #info elbrus had an action to check if he can patch udd to special case b-d-i in case of only -doc as arch:all
19:00:39 <elbrus> not done, but I'm also less convinced it's useful
19:00:49 <elbrus> or better phrased
19:01:18 <elbrus> the part of the bits about <!nodoc> may still be good enough for the purpose for now
19:01:55 <elbrus> I'll leave the action for now, but ...
19:02:19 <elbrus> #topic Transitions
19:02:37 <Sebastinas> gcc-12 is currently a problem for transitions.
19:03:06 <Sebastinas> It fails to build on mips64el (see mail from bunk on wb-team@)
19:03:28 <Sebastinas> But some packages get rather tight dependencies on gcc-12 on armel
19:03:30 <elbrus> and it interferes because it's always the latest gcc that builds the libraries?
19:03:52 <Sebastinas> I'm currently trying to work around that for Qt 5 which is stuck because of that.
19:04:18 <Sebastinas> (yay tpu)
19:04:35 <elbrus> for the future, would it help if the gcc libraries are build by the default gcc instead of the newest one?
19:04:40 <elbrus> is that even possible?
19:04:52 <Sebastinas> gcc-12 is the default
19:04:59 <Sebastinas> There is a problem with some symbols on armel
19:05:00 <elbrus> oops, missed that
19:05:44 <elbrus> I assume you meant https://lists.debian.org/debian-wb-team/2022/10/msg00009.html
19:05:54 <Sebastinas> Yes
19:06:49 <Sebastinas> From what I can tell for perl, we're waiting for autopkgtest results.
19:07:08 <elbrus> and gnumerics? or isn't that a problem?
19:07:15 * elbrus is working on the NMU
19:07:27 <elbrus> it's in the tracker, but not in the britney output
19:07:46 <elbrus> so it's a bit unclear if it will be a blocker once perl becomes candidate
19:08:24 * elbrus worked around the mariadb-10.6 issue, so the results should be in soon
19:08:37 <elbrus> and I have a fix in autopkgtest
19:08:52 <Sebastinas> Ah, gnumeric, yes. I nthe worst case it could be removed or we wait for your NMU.
19:09:04 <Sebastinas> *In the
19:09:07 <elbrus> I'll upload after the meeting
19:09:12 <elbrus> if you urgent it
19:09:16 <Sebastinas> Let's do that
19:09:20 <elbrus> no, then you urgent it
19:09:28 <elbrus> :)
19:09:59 <elbrus> any other transition news?
19:10:06 <Sebastinas> ffmpeg will soon be done.
19:10:11 <elbrus> nice
19:10:18 <Sebastinas> My openjfx NMU should hit tpu tomorrow
19:11:20 <Sebastinas> Otherwise, no big news. I guess we'll have a bunch of transitions to work through until the end of the year
19:11:45 <pochu> one step^W^Wmany transitions at a time
19:12:09 <Sebastinas> icu, python3.11, php8.2 are the big ones.
19:12:39 <elbrus> how much do they intersect?
19:13:21 <Sebastinas> icu and python 3.11 definitely do.
19:14:20 <elbrus> o well
19:14:24 <Sebastinas> but icu intersects with almost everything
19:14:48 <elbrus> any particular order in mind?
19:15:02 <Sebastinas> I'm thinking of starting icu directly after perl
19:15:07 <elbrus> python3.11 can just first add support in the shadow of other things
19:15:14 <Sebastinas> Right
19:15:31 <elbrus> php8.2 isn't ready yet anyways from the last reponses
19:15:38 <elbrus> so, sounds like a plan
19:16:18 <elbrus> once pyasn1 migrates to testing I (once again) hope we can remove pypy and once pam-python migrates I think we're down to one package for python2.7 removal
19:16:42 <Sebastinas> I talked to mitya57 about qtwebengine
19:16:57 <Sebastinas> But that fixed is currently blocked on the Qt transition ;)
19:17:30 <Sebastinas> s/fixed/fix
19:17:48 <elbrus> so there is a fix...
19:17:53 <elbrus> that's good news
19:18:06 <Sebastinas> There's a patch as far as I understand, but it might need some work
19:19:30 <elbrus> anything else on transitions
19:19:45 <Sebastinas> I think that's it
19:19:53 <elbrus> #topic current status of bookworm
19:20:27 <elbrus> I'm hoping for bug squashing parties...
19:20:40 <elbrus> there's quite some bugs with patches or simple solutions
19:21:00 <elbrus> which cloud the list a bit
19:21:24 <elbrus> are we already aware of blockers?
19:21:51 <elbrus> otherwise it looks pretty OK
19:22:37 <Sebastinas> I'm not currently aware of any
19:22:51 <Sebastinas> But I think the vast majority of upgrade bugs will come later.
19:23:29 <elbrus> we should mention that in the next bits
19:23:44 <elbrus> that people should start trying upgrades....
19:24:00 <elbrus> luckily piuparts catches a lot
19:24:25 <Sebastinas> At least regarding usrmerge there were less bugs than I feared.
19:24:34 <elbrus> #action mention upgrade in the next bits (to get those upgrade bugs filed)
19:24:47 <elbrus> ack
19:24:57 <Sebastinas> And if I'm not mistaken, most (all?) of them have been addressed.
19:25:30 <elbrus> I've seen good turn around on the bugs I've seen
19:25:40 <elbrus> but I don't claim to have the overview
19:27:00 <elbrus> #topic current status of d-i
19:27:10 * elbrus wonders if kibi is around
19:27:51 <elbrus> I've seen a plan around non-free-firmware
19:28:09 <elbrus> I guess it will be quite some work, but it's ongoing
19:28:29 <elbrus> https://lists.debian.org/debian-boot/2022/10/msg00044.html
19:29:12 <elbrus> anyways, not much *I* can say about the progress
19:29:25 <elbrus> #topic AOB
19:30:15 <elbrus> not much comes to my mind at the moment
19:30:52 <elbrus> hmm, except maybe
19:30:57 <elbrus> Enabling branch protection on amd64 and arm64
19:31:03 <elbrus> what do we think of that?
19:31:27 <Sebastinas> It's late.
19:31:41 <elbrus> yes
19:32:09 <elbrus> either we plan to rebuild everything... or it's for trixie?
19:32:28 <Sebastinas> With the current buildd situation, I'm not fond of a rebuild of the archive
19:32:30 <elbrus> like jmm_ said in his reply 15 minutes ago
19:32:51 <elbrus> ppc64el you mean?
19:33:02 <elbrus> or mips*
19:33:08 <Sebastinas> Both
19:33:14 <elbrus> ack
19:33:42 <elbrus> maybe we should reply that to the thread
19:33:47 <elbrus> do you want to do that?
19:33:51 <Sebastinas> I can, yes.
19:34:10 <elbrus> #action Sebastinas replies to the branch protection thread
19:34:13 <elbrus> thanks
19:34:26 <elbrus> anything else for AOB?
19:34:46 <Sebastinas> Nothing from my side
19:35:05 <elbrus> I noticed you replied on #d-devel about i386
19:35:16 <elbrus> was that with or without RT hat on ;)
19:35:54 <Sebastinas> Without
19:36:02 <Sebastinas> But we should have this discussion after the release
19:36:30 <elbrus> that's what I said just before the bullseye release, but I agree
19:36:40 <Sebastinas> :)
19:37:11 <elbrus> see https://lists.debian.org/debian-devel-announce/2021/01/msg00002.html under "bullseye architectures"
19:37:33 <elbrus> I think that whole section is still valid
19:38:00 <elbrus> except we were earlier with our call for porters
19:38:32 <elbrus> anyways, I propose to end the meeting... agreed?
19:38:37 <Sebastinas> ACK
19:38:43 <elbrus> #topic Next meeting
19:38:53 <elbrus> #info Next meeting is 23 November at 19:00 UTC (import into your calendar via https://release.debian.org/release-calendar.ics)
19:39:01 <elbrus> #endmeeting