[00:03] -queuebot:#ubuntu-release- New binary: peony [amd64] (noble-proposed/universe) [4.0.0.0-1.1] (ubuntukylin)
[00:24] <vorlon> well, someone made python-cryptography build on riscv64 in the main archive, which means it has to be built again
=== chris14_ is now known as chris14
[02:52] <vorlon> jbicha: why does glib2.0 build-depend on dh-sequence-gnome at all? I can't see that it does anything useful or relevant
[02:53] <vorlon> it has a whole lot of gir and python dependencies, and all it's used for here is to mangle .la files (which shouldn't be shipped at all) and export some substvars that are unused in this package
[03:41] <vorlon> jbicha: I failed to include it in the latest upload, but you might want to pick up this fix to the nogir profile: https://launchpadlibrarian.net/718074260/gobject-introspection_1.79.1-1ubuntu4_1.79.1-1ubuntu5~pre1.diff.gz
[03:46] <vorlon> starting no-change rebuilds for libglib2.0-0t64.  (I really only need policykit-1 at the moment, but these are all going to be relevant fairly soon)
[03:47] <vorlon> i.e. gobject-introspection now has a clean build https://launchpad.net/~canonical-foundations/+archive/ubuntu/archive-bootstrap/+build/27883336 and glib2.0 is only missing python and shared-mime-info bits
[04:17] <mwhudson> vorlon: exciting
[05:09] <vorlon> hah trying to bootstrap pygobject, and that's not going to work either without source changes, because once again, dh-sequence-gnome
[05:09] <vorlon> which is even more useless for pygobject because python and therefore no .la files
[05:10] <vorlon> ah no I was looking at a wrong version of the source package
[05:52] -queuebot:#ubuntu-release- New: accepted litehtml [amd64] (noble-proposed) [0.6-1.2ubuntu1]
[05:52] -queuebot:#ubuntu-release- New: accepted litehtml [ppc64el] (noble-proposed) [0.6-1.2ubuntu1]
[05:52] -queuebot:#ubuntu-release- New: accepted litehtml [s390x] (noble-proposed) [0.6-1.2ubuntu1]
[05:52] -queuebot:#ubuntu-release- New: accepted litehtml [arm64] (noble-proposed) [0.6-1.2ubuntu1]
[05:53] -queuebot:#ubuntu-release- New: accepted litehtml [riscv64] (noble-proposed) [0.6-1.2ubuntu1]
[06:00] -queuebot:#ubuntu-release- New: accepted peony [amd64] (noble-proposed) [4.0.0.0-1.1]
[06:00] -queuebot:#ubuntu-release- New: accepted peony [arm64] (noble-proposed) [4.0.0.0-1.1]
[06:12] -queuebot:#ubuntu-release- New sync: charliecloud (noble-proposed/primary) [0.37-1]
[06:15] <vorlon> oh, well of *course* nvidia-cuda-toolkit depends on libglib2.0-0 and has to be uploaded
[06:17] -queuebot:#ubuntu-release- New: accepted charliecloud [sync] (noble-proposed) [0.37-1]
[06:24] -queuebot:#ubuntu-release- New binary: charliecloud [amd64] (noble-proposed/none) [0.37-1] (no packageset)
[06:24] <vorlon> very rude of launchpad to have published pygobject on amd64 but not armhf
[06:31] -queuebot:#ubuntu-release- New: accepted charliecloud [amd64] (noble-proposed) [0.37-1]
[07:28] <doko> vorlon: did you make progress on chafa/harfbuzz?
[07:29] <vorlon> doko: fixed chafa now (I hate shared-mime-info)
[07:29] <vorlon> so I think once chafa builds, harfbuzz will also
[07:30] <doko> building mesa without LLVM now
[07:31] <vorlon> policykit-1 now copied to the bootstrap ppa
[07:31] <vorlon> IIRC that's the last bit needed for systemd
[07:32] <vorlon> which, once that's done, we can copy systemd to the main archive and there's a good chance that unblocks all the builds that fail when trying to upgrade apt
[07:55] <vorlon> doko: heh apparently we raced on copying libxml-parser-perl
[07:55] <vorlon> harfbuzz building
[09:52] <juliank> vorlon: I synced perl (my changes were just a subset, and there was no other delta), but it ftbfs on armhf due to libpam login dependency, I guess I should wait for other arches to build, binary copy to bootstrap, then have armhf built there and binary copy it back to the archive, but the concept is arguably confusing because there's specific bootstraps in progress in the PPA hence why I
[09:52] <juliank> did not interfere so far to not break those
[09:53] <vorlon> juliank: as soon as systemd can build and be copied out to the archive, my expectation is that the libpam stuff clears up
[09:53] <juliank> ack
[09:53] <vorlon> systemd is afaik only waiting for policykit, which is pending publication
[09:54] <vorlon> I don't expect to be awake to see it all the way through, but I've at least set a trigger to copy systemd in in ~1h
[09:54] <juliank> fwiw you can also copy binaries that are pending publication
[09:54] <juliank> :D
[09:55] <vorlon> not alway
[09:55] <vorlon> s
[09:55] <vorlon> but in this case I'm not trying to copy them
[09:55] <schopin> vorlon: no objection to my uploading openssl 3.0.13 for adrien ? (rebased on top of the t64 changes)
[09:55] <vorlon> juliank: they need to be published in the ppa, because systemd needs to build against them
[09:55] <vorlon> schopin: no objection
[09:56] <schopin> ACK, should hit the archive once my local build is finished.
[10:00] <juliank> vorlon: So to make sure I understand correctly, I see an older systemd in the bootstrap PPA (u4~pp3) that did build and a newer one (u4) in the archive that is in dep-wait, so that copies to bootstrap and then builds and back out?
[10:00] <vorlon> juliank: yes
[10:01] <vorlon> and it won't build until policykit is published
[10:01] <juliank> indeed
[10:09] <vorlon> juliank: the other thing will be copying https://launchpad.net/ubuntu/+source/libxaw/2:1.0.14-1build1 and https://launchpad.net/ubuntu/+source/libxmu/2:1.1.3-3build1 to the ppa once riscv64 binaries publish (or at least, are confirmed to be part of the binary copy).  these are on the critical path for building texlive-bin (in fact, I think the only blockers now)
[10:09] <juliank> ack
[10:12] -queuebot:#ubuntu-release- Unapproved: crun (jammy-proposed/universe) [0.17+dfsg-1.1 => 0.17+dfsg-1.1ubuntu0.1] (no packageset)
[10:12] <vorlon> https://launchpad.net/~canonical-foundations/+archive/ubuntu/archive-bootstrap/+build/27892321 got past build-deps, yay
[10:13] -queuebot:#ubuntu-release- New binary: perl [i386] (noble-proposed/main) [5.38.2-3.2] (core, i386-whitelist)
[10:13] <doko> vorlon: mesa now built. why did we need that?
[10:13] <vorlon> so, setting a trigger to copy systemd back to the main archive in 90 minutes if no one gets to it first
[10:14] <vorlon> doko: I don't remember now
[10:15] -queuebot:#ubuntu-release- New binary: perl [amd64] (noble-proposed/main) [5.38.2-3.2] (core, i386-whitelist)
[10:18] <vorlon> heh, we have openldap from experimental, which builds with -Werror=implicit-function-declaration; unstable does not
[10:24] <vorlon> hmm but I can't see why
[10:24] -queuebot:#ubuntu-release- New binary: perl [ppc64el] (noble-proposed/main) [5.38.2-3.2] (core, i386-whitelist)
[10:28] <vorlon> well, it's because in Ubuntu it's already being built without -Werror=implicit-function-declaration
[10:28] <vorlon> I guess I'll do a maintainer upload of that in the morning (since they've been delinquent in removing me from the uploaders field)
[10:31] <vorlon> libxmu needs w3m, sigh; copied it over
[10:31] <vorlon> and that's where I'm stopping for the night
[10:33] -queuebot:#ubuntu-release- New binary: perl [s390x] (noble-proposed/main) [5.38.2-3.2] (core, i386-whitelist)
=== HerbY_NL is now known as HerbY_NL2
[12:06] <jbicha> vorlon: Ubuntu's version of dh-sequence-gnome (gnome-pkg-tools) runs dh_translations for language packs. It also runs dh_scour but Ubuntu Desktop might turn that off but that feels more like a 24.10 idea at this point.
[12:08] -queuebot:#ubuntu-release- New binary: perl [riscv64] (noble-proposed/main) [5.38.2-3.2] (core, i386-whitelist)
[12:08] <jbicha> glib & a lot of the GNOME packages now override dh_gnome_clean so they aren't doing the .la thing any more. I agree the substvar thing is a problem since there are only a few packages that even use that feature but the warnings clutter the build log
[12:08] <doko> harfbuzz, systemd, nvi copied to the distro
[12:08] <jbicha> before 24.04 LTS, we used to use dh-sequence-gnome for everything GNOME, not it's not used for quite a few packages (like gobject-introspection)
[12:10] <jbicha> oh & it looks like you noticed that we recently dropped that from pygobject too :)
[12:11] <jbicha> I believe it's the dh_scour thing that makes dh-sequence-gnome especially annoying for bootstrapping glib, and Debian's -gnome doesn't do that
[12:13] -queuebot:#ubuntu-release- New binary: perl [arm64] (noble-proposed/main) [5.38.2-3.2] (core, i386-whitelist)
[12:14] -queuebot:#ubuntu-release- New: accepted perl [amd64] (noble-proposed) [5.38.2-3.2]
[12:14] -queuebot:#ubuntu-release- New: accepted perl [i386] (noble-proposed) [5.38.2-3.2]
[12:14] -queuebot:#ubuntu-release- New: accepted perl [riscv64] (noble-proposed) [5.38.2-3.2]
[12:15] -queuebot:#ubuntu-release- New: accepted perl [arm64] (noble-proposed) [5.38.2-3.2]
[12:15] -queuebot:#ubuntu-release- New: accepted perl [s390x] (noble-proposed) [5.38.2-3.2]
[12:15] -queuebot:#ubuntu-release- New: accepted perl [ppc64el] (noble-proposed) [5.38.2-3.2]
[12:27] <doko> libxmu built in the ppa
[12:32] <jbicha> juliank: https://launchpad.net/ubuntu/+source/policykit-1/124-1build1/+latestbuild/armhf :(
[12:32] <doko> postgresql-16 copied to the ppa
[12:32] <doko> err, postgresql-16 copied to the distro
[12:36] <doko> policykit-1 and openssh copied to the distro
[12:49] <doko> libxmu copied to the distro
[13:20] <doko> libxaw copied to the distro
[14:30] <doko> texlive-bin copied to the distro
[14:42] <LocutusOfBorg> doko, any idea for this? https://launchpadlibrarian.net/718236563/buildlog_ubuntu-noble-armhf.llvm-toolchain-17_1%3A17.0.6-5ubuntu1_BUILDING.txt.gz
[14:42] <LocutusOfBorg> your llvm build is strange!
[14:43] <LocutusOfBorg> maybe you left some files?
[14:47] <LocutusOfBorg> -[!armel !armhf] usr/lib/llvm-@LLVM_VERSION@/lib/clang/@LLVM_VERSION@*/include/xray/
[14:47] <LocutusOfBorg> -[!armel !armhf] usr/lib/llvm-@LLVM_VERSION@/lib/clang/@LLVM_VERSION@*/include/profile/
[14:47] <LocutusOfBorg> doko, ^^
[14:53] <doko> yes, because compiler-rt is enabled again. so if you would have fixed the compiler-rt build properly, I think we wouldn't have to deal with all the compiler-rt stuff ;p
[14:53] <LocutusOfBorg> you are completely right!
[14:53] <doko> committed to the branches, starting new ppa builds
[14:53] <LocutusOfBorg> I did *lot* of useless work because I failed to do a 3 lines patch
[14:54] <LocutusOfBorg> but I'm pretty sure when I read the compiler-rt solaris code it was written "don't do that, we need time_t to be exactly 32 bit because otherwise you will break compatibility with solaris tools"
[14:55] <LocutusOfBorg> so, I had *reasons* to do it, but if your solution works, I'm really happy, and pretty sure nobody will ever need compiler-rt on solaris 32bit arch :)
[14:55] <LocutusOfBorg> doko, please hold on
[14:55] <LocutusOfBorg> also remove the two patches
[14:55] <LocutusOfBorg> -fix-unwind-detection-logic.patch
[14:55] <LocutusOfBorg> -runtimes-Don-t-link-against-compiler-rt-explicitly.patch
[14:56] <juliank> jbicha: seems fine now
[14:57] <LocutusOfBorg> please doko check my additional push
[14:57] <LocutusOfBorg> if it works, I'm really happier
[14:58] <doko> afk for 30min
[14:59] <LocutusOfBorg> please ack the changes, I can't backport the two patches on llvm-14, so I really need the two patches for compiler-rt to disappear
[15:54] -queuebot:#ubuntu-release- New binary: fastdds [amd64] (noble-proposed/universe) [2.11.2+ds-6.1] (no packageset)
[15:56] -queuebot:#ubuntu-release- New binary: fastdds [ppc64el] (noble-proposed/universe) [2.11.2+ds-6.1] (no packageset)
[15:58] -queuebot:#ubuntu-release- New binary: fastdds [s390x] (noble-proposed/universe) [2.11.2+ds-6.1] (no packageset)
[15:59] <doko> texlive-bin, perl, mesa, lua5.[1-4] copied to the distro
[15:59] <doko> vorlon: ^^^
[15:59] <doko> now I get "new" errors like "libssl3t64 : Breaks: libssl3 (< 3.0.10-1ubuntu5) but 3.0.10-1ubuntu2 is to be installed"
[16:00] <doko> llvm-toolchain-17 copied to the archive. will follow-up with a fresh upload directly to unstable. the armhf build will keep installable
[16:01] <doko> LocutusOfBorg: running out of time. please test your changes in a PPA later. I don't think anything has to be reverted, and you just don't backport the backports to earlier versions
[16:16] -queuebot:#ubuntu-release- New binary: dart [amd64] (noble-proposed/universe) [6.12.1+dfsg4-13.1] (no packageset)
[16:28] -queuebot:#ubuntu-release- New binary: dart [ppc64el] (noble-proposed/universe) [6.12.1+dfsg4-13.1] (no packageset)
[16:33] -queuebot:#ubuntu-release- New binary: fastdds [arm64] (noble-proposed/universe) [2.11.2+ds-6.1] (no packageset)
[16:37] -queuebot:#ubuntu-release- New binary: sssd [amd64] (noble-proposed/main) [2.9.4-1.1ubuntu1] (i386-whitelist, ubuntu-desktop, ubuntu-server)
[16:37] -queuebot:#ubuntu-release- New binary: sssd [i386] (noble-proposed/main) [2.9.4-1.1ubuntu1] (i386-whitelist, ubuntu-desktop, ubuntu-server)
[16:38] <doko> great, on Monday I asked for fast merges, now that we were almost ready, people upload new t64 stuff. not amused about the timing
[16:38] -queuebot:#ubuntu-release- New binary: sssd [ppc64el] (noble-proposed/main) [2.9.4-1.1ubuntu1] (i386-whitelist, ubuntu-desktop, ubuntu-server)
[16:42] -queuebot:#ubuntu-release- New: accepted dart [amd64] (noble-proposed) [6.12.1+dfsg4-13.1]
[16:42] -queuebot:#ubuntu-release- New: accepted fastdds [arm64] (noble-proposed) [2.11.2+ds-6.1]
[16:42] -queuebot:#ubuntu-release- New: accepted fastdds [s390x] (noble-proposed) [2.11.2+ds-6.1]
[16:42] -queuebot:#ubuntu-release- New: accepted dart [ppc64el] (noble-proposed) [6.12.1+dfsg4-13.1]
[16:42] -queuebot:#ubuntu-release- New: accepted fastdds [ppc64el] (noble-proposed) [2.11.2+ds-6.1]
[16:43] -queuebot:#ubuntu-release- New: accepted fastdds [amd64] (noble-proposed) [2.11.2+ds-6.1]
[16:43] -queuebot:#ubuntu-release- New: accepted sssd [amd64] (noble-proposed) [2.9.4-1.1ubuntu1]
[16:43] -queuebot:#ubuntu-release- New: accepted sssd [ppc64el] (noble-proposed) [2.9.4-1.1ubuntu1]
[16:43] -queuebot:#ubuntu-release- New: accepted sssd [i386] (noble-proposed) [2.9.4-1.1ubuntu1]
[17:06] -queuebot:#ubuntu-release- New binary: dart [arm64] (noble-proposed/universe) [6.12.1+dfsg4-13.1] (no packageset)
[17:07] -queuebot:#ubuntu-release- New: accepted dart [arm64] (noble-proposed) [6.12.1+dfsg4-13.1]
[18:59] -queuebot:#ubuntu-release- New binary: pacemaker [i386] (noble-proposed/main) [2.1.6-4.1ubuntu1] (i386-whitelist, ubuntu-server)
[19:01] -queuebot:#ubuntu-release- New binary: pacemaker [ppc64el] (noble-proposed/main) [2.1.6-4.1ubuntu1] (i386-whitelist, ubuntu-server)
[19:01] -queuebot:#ubuntu-release- New binary: pacemaker [s390x] (noble-proposed/main) [2.1.6-4.1ubuntu1] (i386-whitelist, ubuntu-server)
[19:03] -queuebot:#ubuntu-release- New binary: pacemaker [arm64] (noble-proposed/main) [2.1.6-4.1ubuntu1] (i386-whitelist, ubuntu-server)
[19:04] -queuebot:#ubuntu-release- New binary: pacemaker [amd64] (noble-proposed/main) [2.1.6-4.1ubuntu1] (i386-whitelist, ubuntu-server)
[20:12] <vorlon> LocutusOfBorg: as soon as https://launchpad.net/ubuntu/noble/armhf/libcryptsetup12 is visible on the mirrors, buildd chroots should be un-broken on armhf in the main archive and we can start doing mass-givebacks of armhf packages if you want to help with that
[20:14] -queuebot:#ubuntu-release- New binary: pacemaker [riscv64] (noble-proposed/main) [2.1.6-4.1ubuntu1] (i386-whitelist, ubuntu-server)
[20:18] -queuebot:#ubuntu-release- Unapproved: update-notifier (jammy-proposed/main) [3.192.54.7 => 3.192.54.8] (ubuntu-desktop, ubuntu-server)
[20:25] <vorlon> jbicha: are there strings in any of these graphics libraries to be translated? and I thought I was looking at the Debian version before but missed the translations + scour, ok
[20:27] <vorlon> jbicha: we already have optipng as part of pkgbinarymangler. Would it make sense to move scour there instead?
[20:27] <vorlon> (but, maybe that wouldn't have helped any with the bootstrapping problem)
[20:27] -queuebot:#ubuntu-release- Unapproved: update-notifier (focal-proposed/main) [3.192.30.18 => 3.192.30.19] (ubuntu-desktop, ubuntu-server)
[20:33] <vorlon> jbicha: fwiw I just checked pango1.0 and it contains no .po files in the source, and the only .svg files are in docs that appear not to be shipped
[20:34] <vorlon> jbicha: so from my perspective, dropping the build-dep is a "bootstrap" fix that should be made permanent (cc: liushuyu )
[20:42] -queuebot:#ubuntu-release- New: accepted pacemaker [amd64] (noble-proposed) [2.1.6-4.1ubuntu1]
[20:42] -queuebot:#ubuntu-release- New: accepted pacemaker [i386] (noble-proposed) [2.1.6-4.1ubuntu1]
[20:42] -queuebot:#ubuntu-release- New: accepted pacemaker [riscv64] (noble-proposed) [2.1.6-4.1ubuntu1]
[20:42] -queuebot:#ubuntu-release- New: accepted pacemaker [arm64] (noble-proposed) [2.1.6-4.1ubuntu1]
[20:42] -queuebot:#ubuntu-release- New: accepted pacemaker [s390x] (noble-proposed) [2.1.6-4.1ubuntu1]
[20:42] -queuebot:#ubuntu-release- New: accepted pacemaker [ppc64el] (noble-proposed) [2.1.6-4.1ubuntu1]
[20:59] <arraybolt3> ubuntu-release: Gentle ping w.r.t. the following two FFes: https://bugs.launchpad.net/ubuntu/+source/calamares-settings-ubuntu/+bug/2055799 https://bugs.launchpad.net/ubuntu/+source/lubuntu-installer-prompt/+bug/2056603
[20:59] -ubottu:#ubuntu-release- Launchpad bug 2055799 in calamares-settings-ubuntu (Ubuntu) "FFe: Switch Ubuntu Unity to using Calamares" [Undecided, New]
[20:59] -ubottu:#ubuntu-release- Launchpad bug 2056603 in lubuntu-installer-prompt (Ubuntu) "FFe: Open WiFi network support" [Undecided, New]
[21:00] <arraybolt3> UI freeze and beta freeze are coming up quick and these are both critical feature goals (especially the first one) for the affected flavors.
[21:01] <arraybolt3> (The latter I just filed, the first one I filed several days ago.)
[21:01] <vorlon> ginggs: ^ did you have any time to review FFes today?
[21:01] -queuebot:#ubuntu-release- Unapproved: update-notifier (bionic-proposed/main) [3.192.1.20 => 3.192.1.21] (ubuntu-desktop, ubuntu-server)
[21:08] <jbicha> vorlon: I agree & am making that change now in Debian & will sync pango tomorrow
[21:09] <vorlon> jbicha: cheers; in the meantime I've uploaded to Ubuntu to move the bootstrap along, and have also just made the same change to cairo
[21:09] <jbicha> thanks
[21:09] <vorlon> (cairo had bootstrap builds done in the main archive because I was initially expecting this to be a quick in-archive thing; so the cairo upload rolls back those changes *except* for dh-sequence-gnome)
[21:11] <vorlon> and some further bootstrapping is still needed here, but that's ok
[21:11] <jbicha> I hadn't realized the time_t transition was going to be a bootstrap problem or we would have tried doing some cleanup sooner
[21:13] <vorlon> jbicha: honestly I hadn't thought about that being the case either
[21:13] <vorlon> or maybe I assumed things were in a more bootstrappable state generally :)
[21:13] <jbicha> we're making it easier for Debian to adopt new exotic architectures, lol
[21:14] -queuebot:#ubuntu-release- Unapproved: update-notifier (xenial-proposed/main) [3.168.21 => 3.168.22] (ubuntu-desktop, ubuntu-server)
[21:16] <vorlon> jbicha: well it also makes it easier for me to do the armel+armhf rebootstrap in Debian, so I thank you for that ;)
[21:16] <ginggs> vorlon: i did not :(
[21:30] <vorlon> happy happy in-archive builds https://launchpad.net/ubuntu/+source/netplan.io/1.0-1/+build/27853854
[23:39] <vorlon> ok, in anger I am now trying to make the ubuntu-build --batch option actually useful
[23:39] <vorlon> here come some build retries