[05:34] <jibel> bdmurray, no I didn't
[08:23] <LocutusOfBorg> thanks kanashiro cpaelzer for doing ti!
[08:23] <LocutusOfBorg> *it
[08:49] <cpaelzer> yw LocutusOfBorg
[12:21] <kanashiro> LocutusOfBorg: o/ you're welcome
[13:28] <LocutusOfBorg> mitya57, https://launchpad.net/ubuntu/+source/pyside2/5.15.0-2 please? :) (arm64 fails)
[13:30] <LocutusOfBorg> can you please have a look?
[13:50] <realtime-neil-> How does one build the ubuntu installer? https://wiki.ubuntu.com/Installer/Development talks about checking out source, but not how to build it. Compare/contrast  https://wiki.debian.org/DebianInstaller/Build
[13:58] <juliank> realtime-neil-: which installer, I guess is the first question
[13:59] <juliank> We have the old d-i / classic server, subiquity (the server installer), curtin (the backend behind subiquity amongst others), and ubiquity (the desktop installer)
[13:59] <juliank> I guess there might be one or two more
[13:59] <juliank> e.g. image based things like rpi get configured on first boot, and don't really have an installer
[14:00] <Unit193> calamares too!
[14:00] <juliank> well, calamares is not an ubuntu installer
[14:00] <juliank> i guess a flavour uses it?
[14:00] <Unit193> I thought Lubuntu did.
[14:00] <juliank> i don't know, I'm just talking about insallers in main :)
[14:01] <Unit193> And as a non-coredev, I don't care about main. :P
[14:01] <Unit193> ...Except for when popcon has been broken since bionic, that's....fun.
[14:35] <mitya57> LocutusOfBorg: argh
[14:35] <mitya57> I don't have Ubuntu arm64 hardware so I can't even check what's the difference from Debian
[14:37] <realtime-neil-> juliank, the old d-i one
[14:52] <mitya57> LocutusOfBorg: ideally we need some glibc expert to fix this properly (what I did is just a workaround).
[15:12] <LocutusOfBorg> blacklist again?
[15:13] <mitya57> LocutusOfBorg: maybe, but that will only fix build, and it will be still impossible to use PySide2.QtWebEngine on arm64.
[15:14] <mitya57> The real bug cannot be fixed in pyside2 itself: it also happens with pyqt5 and even with pure Python+ctypes code.
[15:15] <mitya57> I already contacted glibc and mesa maintainers (in Debian) but got no reply… Maybe I should ask on debian-arm.
[16:23] <LocutusOfBorg> jamespage, I uploaded the openstack-pkg-tools merge without the --subunit drop. Do you think its still needed?
[16:24] <LocutusOfBorg> I'm upstreaming the Ubuntu delta to debian, and they don't see any reason to not use subunit
[17:47] <LocutusOfBorg> coreycb, I'm syncing python-amqp, used for new kombu
[17:50] <coreycb> LocutusOfBorg: sounds good, thank you
[20:06] <bdmurray> xnox: Do you have any hints on how to recreate bug 1874953? I tried an upgrade from E to F and got a conffile prompt but my pager worked for me
[20:28] <xnox> bdmurray:  one has to install with split-usr, so install xenial upgrade to focal.
[20:29] <xnox> bdmurray:  and the debconf prompt should be quite early, after less is upgraded, and in the same transaction.
[20:29] <bdmurray> xnox: alright, that'll be a fun slog
[20:29] <xnox> yes, sorry
[20:29] <xnox> bdmurray:  usr-merged systems are unaffected.
[20:55] <kanashiro> mwhudson: I was checking golang-github-miekg-dns and I was wondering if you do not want to submit the changes you proposed here in a PR: https://github.com/miekg/dns/issues/1129#issuecomment-648560142
[20:55] <kanashiro> it might be easier to collect feedback from upstream
[21:17] <mwhudson> kanashiro: i was hoping to get some feedback but yeah, maybe making that into a PR makes sense
[21:23] <mwhudson> kanashiro: https://github.com/miekg/dns/pull/1130
[21:24] <kanashiro> mwhudson: many thanks :)