[05:15] morning all [05:18] Morning jibel === JanC is now known as Guest56745 === JanC_ is now known as JanC [07:12] cpaelzer, I pushed a fix for bug 1856574 . Should I resubscribe ubuntu-sponsors? [07:13] morning jibel and duflu [07:13] hi marcustomlinson [07:13] Error: Could not gather data from Launchpad for bug #1856574 (https://launchpad.net/bugs/1856574). The error has been logged [07:13] Hi marcustomlinson [07:26] goood morning desktopers [07:26] o/ seb128 [07:27] hey duflu , how are you? [07:27] salut seb128 [07:28] seb128, I'm well, you? [07:28] lut jibel, comment ça va ? [07:28] duflu, I'm alright, it was a bit less warm this night and I slept better [07:38] seb128, ça va bien, osomon est venu passer quelques jours à la maison, c'était sympa. Et le temps est redevenu plus supportable. [07:42] morning seb128 [07:44] hey marcustomlinson, how are you? [07:45] jibel, c'est sympa une halte comme ça :-) [07:45] seb128: yeah alright, you? [07:45] seb128, tu es le bienvenu aussi si jamais vous passez en bretagne [07:55] jibel: that LGTM, I can sponsor it if you want [07:58] jibel: your test build is for 1.6.1 [07:58] the current package is 1.6.0 [07:58] do you want me to review/sponsor the fix on top of 1.6.0 or is there a full 1.6.1 to come? [07:58] in the latter case you might want someoen with more context knowledge [07:59] arr [07:59] this is native - so the cahnge makes it 1.6.1 [07:59] * cpaelzer crawls out of his own trap :-) [08:00] the very noisy diff made me nervous https://launchpadlibrarian.net/493259121/autopilot-gtk_1.6.0_1.6.1.diff.gz [08:00] but the fix https://git.launchpad.net/autopilot-gtk/commit/?id=00d7a3d7111cdca00b949bfa9c315781e6701328 LGTM and I'd sponsor that if you want [08:00] do you ? [08:00] cpaelzer, that's weird [08:01] cpaelzer, can you just sponsor the fix then, i'll have a look at this strange diff [08:01] yes [08:01] and I'll do my own check build before sponsoring [08:01] looks like a direct upload not committed t o git [08:01] just to be sure [08:03] marcustomlinson, I'm alright, slept better this night! [08:04] moin [08:05] jibel, merci pour l'offre, je la garde sous le coude si jamais on passe dans ta région ;-) [08:05] hey Laney , how are you? [08:05] seb128: good to hear! [08:05] hey Laney [08:06] hellsworth: you typoed the commandline, --apt-pocket=proposed:src:libreoffice -> --apt-pocket=proposed=src:libreoffice [08:06] moin seb128 marcustomlinson [08:07] I'm alright [08:07] some jerk of a mosquito bit me though, itchyyyyy [08:07] jibel, cpaelzer, the diff noise seems to be mostly .git being included in the update for some reason and then one commit which was probably missing from the vcs [08:07] Laney, we have been dealing with those for a few nights, really annoying [08:08] cpaelzer, actually it's sample git hooks that shouldn't have been uploaded. But there is still one diff in hello_color.ui that shouldn't be there [08:08] waking every hour to the bzzzz sound near your head [08:08] jibel: a test build wit hjust the fix worked welll [08:08] jibel: I'll upload it [08:08] cpaelzer, great, thanks [08:08] jibel: maybe set a marker to check tomorrow if everything worked out as planned? [08:08] cpaelzer, yes, I'll do that. [08:19] Morning Laney [08:34] Laney: I worked out why my "GDM on Ubuntu Core" prototype was running the greeter under X11: I didn't have Xwayland installed [08:34] hey duflu jamesh [08:34] ah! [08:34] how come that wasn't more obvious in the log? [08:35] because it doesn't log anything for this: https://gitlab.gnome.org/GNOME/gdm/-/blob/master/daemon/gdm-local-display-factory.c#L200 [08:37] oho, this shortcuts the "try wayland, fail, try X" path completely [08:37] This is the commit that introduced the check: https://gitlab.gnome.org/GNOME/gdm/-/commit/9ab95809ddd201c2c8b030a814287c3be73d03be [08:40] fair enough, although a log message might be warranted [08:40] still, good find [09:21] mutter pristine-tar 7fa8c24 Simon McVittie mutter_3.36.5.orig.tar.xz.delta mutter_3.36.5.orig.tar.xz.id * pristine-tar data for mutter_3.36.5.orig.tar.xz * https://deb.li/3oAuM [09:21] mutter upstream/3.36.x 8456674 Simon McVittie * pushed 21 commits (first 5 follow) * https://deb.li/3zQaN [09:21] mutter upstream/3.36.x b1d4582 Jonas Ådahl src/backends/meta-screen-cast-window-stream-src.c * screen-cast/window-stream-src: Fix indentation * https://deb.li/3g6JV [09:21] mutter upstream/3.36.x 92db890 Jonas Ådahl src/backends/ meta-screen-cast-monitor-stream-src.c meta-screen-cast-stream-src.c meta-screen-cast-stream-src.h meta-screen-cast-window-stream-src.c * screen-cast/src: Add flag to maybe_record() * https://deb.li/iOE4g [09:21] mutter upstream/3.36.x cf88d64 Jonas Ådahl src/backends/ meta-screen-cast-monitor-stream-src.c meta-screen-cast-stream-src.c meta-screen-cast-stream-src.h meta-screen-cast-window-stream-src.c * screen-cast: Let the reason for recording determine what to record * https://deb.li/NQnv [09:21] mutter upstream/3.36.x 2d89959 Jonas Ådahl src/backends/ meta-screen-cast-monitor-stream-src.c meta-screen-cast-stream-src.c meta-screen-cast-stream-src.h meta-screen-cast-window-stream-src.c * screen-cast/src: Make record functions return an error when failing * https://deb.li/igzdn [09:22] mutter upstream/3.36.x 449fa7b Jonas Ådahl src/backends/meta-screen-cast-stream-src.c * screen-cast/src: Fix signedness of timestamp field * https://deb.li/2jOc [09:22] mutter tags e43f76d Simon McVittie upstream/3.36.5 * Upstream version 3.36.5 * https://deb.li/SaK1 [09:27] mutter upstream/latest 8456674 Simon McVittie * pushed 74 commits * https://deb.li/3MQvR [09:27] gnome-shell upstream/latest 4abb56a Simon McVittie * pushed 138 commits * https://deb.li/Ij4x [09:51] Hi all! [09:51] Question: Who is responsible for the selection of fonts and the fonts configuration choices on a default Ubuntu install? I'm asking due to the discourse discussion about Arabic fonts. Somebody mentioned Kubuntu, and I noticed that they prefer Noto Sans over DejaVu Sans and with that make Noto Sans fonts the default for most scripts of significance. While that approach probably affects the bulk of Ubuntu users, it makes it [09:51] much easier to provide improved font rendering for quite a few non-latin scripts. [09:51] I can't help wondering if that configuration has been considered for standard Ubuntu but dismissed. Or has nobody considered this topic in several years? [09:53] GunnarHj, hey, I don't think we have any 'font person' in the team or had one for a while [09:54] seb128: Then how would you go about proposing a significant change? [09:58] good luck? ;-) [09:59] GunnarHj, I think a discourse post with the proposed changes, a summary of the pro/con and maybe screenshot showing the impact [10:06] seb128: Sounds reasonable. Even if the Kubuntu approach looks interesting, I'm probably not ready to start such a project for groovy. Think I will stick to fixing Arabic 'the clumsy way' (i.e. a language/locale guarded improvement) for now. [10:12] seb128: Btw, any chance can you sponsor the latest seed MR? (The rest is pure language-selector fixes.) [10:12] https://code.launchpad.net/~gunnarhj/ubuntu-seeds/+git/ubuntu/+merge/389180 [10:15] seb128: Repeating since you were temporarily disconnected: [10:15] Btw, any chance can you sponsor the latest seed MR? (The rest is pure language-selector fixes.) [10:15] https://code.launchpad.net/~gunnarhj/ubuntu-seeds/+git/ubuntu/+merge/389180 [10:16] GunnarHj, what's the purpose? unsure what difference supported makes there, I'm not really familiar with such changes [10:17] I've started a board for 3.38 updates so we can synchronize work better [10:17] https://trello.com/b/sdyPskaS/gnome-338 [10:17] Laney, Trevinho , marcustomlinson : ^ [10:18] you are probably the most interested [10:18] seb128: If I understand it correctly, including those packages in supported means that if you install Ubuntu using the Arabic language, the installer pulls them. Otherwise not. [10:18] GunnarHj, k, I would prefer someone understanding the installer to confirm that [10:18] No, that is ship-live [10:19] confirm with e.g. fonts-nafees on http://cdimage.ubuntu.com/ubuntu/daily-live/current/groovy-desktop-amd64.manifest [10:20] Laney: But then how would you make the standard Ubuntu installer pull those packages? Or is it sufficient to include them in pkg_depends? [10:25] Laney: To clarify: I'm not talking about including the packages on the ISO. I'm talking about making the installer pull them from the archive if the user is contected to the net when installing. [10:25] I see, then maybe having them in main is enough [10:25] jibel: autopilot-gtk built fine on all arches and entered the autppkgtest queues [10:26] I hope the results will be as green as the test history on it :-) [10:29] Laney: A comment on that file says: "Fonts installed on demand by language-selector:" [10:29] https://git.launchpad.net/~ubuntu-core-dev/ubuntu-seeds/+git/ubuntu/tree/supported#n72 [10:29] Laney: I thought that the way to make them be considered to be in main is to add them there. [10:30] Laney: Why would we else have that list? [10:34] It is, that's what it is for [10:34] so perhaps it is OK if you only care about people who are online [10:35] I don't think I've tried that facility myself, can't speak for how/if it does that [10:37] Laney: Well, either on line when installing of opening language-selector after install. That's the possible way I see short-term to respond to the request for better Arabic rendering. Doing something else would require big changes (see above). [10:37] Laney: I was pretty sure (before you objected) that it works the way I claim. :) [10:38] I didn't know from your initial question that you only wanted it to work when online [10:38] Perhaps it does [10:40] Laney, seb128: So, does any of you have the guts to sponsor that MR? [10:41] seb128: cool thanks [10:51] GunnarHj: Not right now [10:51] I dunno if we need MIRs for fonts or anything [10:51] Laney: The fonts-noto source package is in main already. [10:52] ok [10:52] still no, sorry [10:52] busy with other things [10:54] Laney: It's no urgency really. I'm fine for now if the conclusion is that it's the right thing to do. Then I can proceed with discussing some details with some Arabic users. [11:03] GunnarHj: Why don't you upload the language-selector bits to groovy now and see if people are happy with it? [11:03] Then if so we can do the seeding thing [11:09] Laney: Sure, that may be a better order of things. So far it has been a discourse discussion only, and I want to have a couple of things clarified from the Arabaic users before uploading. [11:31] GunnarHj, sorry I disconnect during lunch, I don't understand enough those changes to be comfortable uploading atm but document what you propose and we should try to get the changes reviewed one way or another at some point [11:31] would also be useful to know what Debian does in that regard [11:37] seb128: While you were off line, I agreed with L_aney that I'll upload the language-selector changes first. Sure, I can take a look at Debian too, but I don't expect it to help much. Debian tends to just provide fonts packages and leave it to the user to handle the configuration, while we make some effort at least to provide a sensible default for the most important scripts. [11:38] k [11:38] +1 from me to start with language-selector [14:02] Hum, why is gnome-software still in main? [14:03] I would have expect it to fall into universe when we switched to snap-store [14:03] I only see gstreamer1.0-packagekit which recommends 'gnome-software | apper | packagekit-installer' is that enough to make it stay there? [14:08] seb128: yeah, i was wondering about that [14:08] I could try to demote and see how that goes :p [14:10] awesome.... what could possibly go wrong :) [14:11] you get told off by do_ko :> [14:16] lol [14:30] fml [14:30] how does update-manager defeat focus stealing protection so well :( [15:49] good morning desktopers [15:50] Laney: thank you for catching my typo!!! i so didn't see that... [15:55] hey hellsworth [15:55] no worries, the errors you got were *quite suspicious* :) [15:55] hey Laney autopkgtest is running on my old ass pi! [15:56] it's funny that my pi 4 is just sitting by while the 3 takes the show :) [15:57] and armhf builds fail consistently on LO 7. so no magical fix and the build failure sounds potentially related to the autopkgtest failure [15:57] could be that if you rebuilt 6.4.5 it would now fail in the same way [15:57] possibly [15:59] Laney: is there a way to get maximum debug output to sdtout from the autopkgtest? [16:00] you can get autopkgtest *itself* to give you debug output, but that's not going to make the test it is running do anything different I'm afraid [16:01] oh i see [16:01] you'll want to modify the debian/tests/foo script & other bits of the package for that [16:01] okey dokey [16:01] we'll see how this goes paired with the system logs before i do that though [16:08] 👍 [16:09] * hellsworth switches gears to giving the libreoffice snap some much needed tlc... [16:21] Laney: hmm now my autopkgtest is failing because no space left on device.. but i used a 32Gb sd card in this old pi [16:21] could this be because there is only 1Gb of ram? [16:21] hey desktopers [16:21] i don't have any newer armhf hw... [16:22] maybe marcus' qemu path is the way to go then [16:22] I dunno, how did you configure lxd? [16:22] you can run armhf on arm64 fwiw [16:22] lxd init with all of the defaults except i chose to make the conatiner accessable over the network [16:22] hi ricotz [16:22] oh Laney that is a game changer.. [16:22] I don't remember what the defaults are, does it make a weird small partition?!?!?! [16:22] let me get groovy on my pi4 then [16:22] heh [16:22] Laney, hellsworth, hey [16:22] hey ricotz [16:24] hellsworth, do you have an armhf build to run autopkgtests with? [16:24] i do and that was my goal but i'm running out of space.. [16:25] 6.4.5 [16:25] ah I see [16:25] trying to troubleshoot the 6.4.5 issue so we can maybe stop holding up so many things in proposed [16:25] so the current binary in groovy [16:25] yep [16:25] ok [16:26] my 7.0 armhf build failed after 22 hours.. [16:26] the actual gcc 10 ICE look bad still [16:26] /<>/dbaccess/source/core/api/StaticSet.cxx: In member function ‘virtual com::sun::star::uno::Any dbaccess::OStaticSet::getBookmark()’: [16:26] did something change? [16:26] /<>/dbaccess/source/core/api/StaticSet.cxx:47:1: internal compiler error: Segmentation fault [16:26] that look slike the same failure [16:26] yeah, same thing [16:27] ricotz: can you file a bug for that and ping dok_o? better if you have a smaller reproducer [16:27] i'm working on the 6.4.5 deb to hopefully fix that but also the 7.0 snap so at least we can provide 7.0 in _some_ form [16:27] and if gcc-9 works that's an option imo [16:27] while 10 is worked on [16:28] can i just do lxd init again to reconfigure? [16:29] "No manual entry for lxd" [16:29] why not? [16:33] don't know the answer to these questions! [16:34] no problem.. still didn't hurt to ask :) [16:34] seems silly that there's no man page for lxd though [16:34] can snaps have manpages? [16:36] (not really, https://forum.snapcraft.io/t/support-for-man-pages/2299) [19:27] jibel: https://github.com/openzfs/zfs/commit/221e670 [21:37] kenvandine, so yeah, I demoted gnome-software and it wants to come back according to https://people.canonical.com/~ubuntu-archive/component-mismatches.txt [21:37] [Reverse-Recommends: gnome-software, gstreamer1.0-packagekit (MAIN)]