[05:29] good morning desktoppers! [05:30] o/ [05:30] morning [05:36] hey koza, jibel [05:39] good morning [05:41] salut didrocks [06:13] Morning oSoMoN, koza, jibel, didrocks, seb128 [06:20] hey duflu [06:22] koza, did we cancel today's meeting? the calendar says not === maclin1 is now known as maclin === Guest12885 is now known as fredp === fredp is now known as Guest11681 [06:59] seb128: what's your plan for fixing the libcdio ftbfs? the debian BTS has a patch, but that looks like an ABI change too. otoh there is a new upstream ... === JanC is now known as Guest86327 === JanC_ is now known as JanC [07:29] oSoMoN, hi, would be good to get libreoffice 5.4.2 in artful [07:31] (given the large amount of bug fixes https://wiki.documentfoundation.org/Releases/5.4.2/RC1#List_of_fixed_bugs) [07:46] Trevinho: I think you did a booboo in the theme change [07:46] Trevinho: have you tested it with gnome-terminal? [07:46] didrocks: yeah, I posted the screenshots here [07:47] didrocks: what you mean? [07:47] Trevinho: https://imgur.com/a/qPgwA [07:47] Trevinho: for me, it's even darker than previous proposal [07:47] like a big dark bar [07:47] and between top bar and menu [07:47] oh, let me see what it has been pushed then [07:48] Trevinho: ironically, it looks better when not selected :p [07:50] weird let me check again... [08:03] duflu, sorry for lack of re. i have changed network in between and this makes the ssh connection to whee the irssi is running stall [08:03] duflu, i believe we have not, it is only will who is in NY [08:04] Yeah, I read the email [08:11] hello all [08:22] didrocks: mh, looks fine here [08:22] https://usercontent.irccloud-cdn.com/file/KU6XXyEi/ [08:28] didrocks: are you maybe running some other version? Or hacked theme? as here it seems to go well, in any case introspect the terminal and check what line causes the menubar to be like that and what the decoration please [08:28] Trevinho: no, I just have your package installed, let me get the line [08:29] Trevinho: hum, I can't select the title bar in gnome-terminal [08:29] Trevinho: as I think it's drawn by the Shell, no? [08:29] didrocks: in wayland you can [08:29] Trevinho: I'm on Xorg [08:30] Trevinho: as I'm hacking the Shell [08:30] didrocks: mh, ok.. I've tested both to be fair [08:30] but. [08:31] but? ;) [08:31] didrocks: ... but let me double-check [08:31] and so I did [08:31] and it works fin in both wayland and xorg here [08:32] hum [08:32] strange [08:32] so, I need to switch to wayland to pick the color and rule applied? [08:32] didrocks: no, restarting the shell and setting/unsetting ambiance should be enough [08:32] from tweaks or cmdline [08:33] Trevinho: interesting, after restarting the shell, it's way better [08:33] I mean evne with alt+f2 r should do it (so I did all the times) [08:33] ah, ok... fair [08:33] but the previous value was weird, it wasn't exactly the one with previous theme [08:33] I wonder if some values are partially picked [08:33] the theme doesn't apply to mutter without restarting (bug known, I will fix it one day :)) [08:33] and not all until restart [08:34] Trevinho: ok, to be picky, just a little bit too bright this time too quickly [08:34] (which makes the top dark becoming quickly light) [08:34] want a screenshot to confirm? [08:34] let me see [08:35] if you can move the gradient to become this level of light, like 10% more down [08:35] I guess that would perfect to me [08:35] I like the backdrop state [08:35] and headerbar as well [08:35] I tried moving the gradient but I wasn't too happy, but let me se it again [08:38] you said you want 10% more down to be come lighter or darker? [08:39] ah, saw it.. so later to be light.. [08:39] The problem is that with apps with no menubar... [08:39] It won't look too nice in these [08:39] didrocks: ^ [08:39] check xterm [08:40] even with few % later to be light? [08:40] like 5-10? [08:40] let's see [08:40] if you point me to the change, I can just have a try with those values [08:40] but yeah, I confirm that we need to take the xterm case into account [08:46] https://usercontent.irccloud-cdn.com/file/B3YXBNaV/ [08:46] https://usercontent.irccloud-cdn.com/file/xLpDbLNZ/ [08:46] didrocks: ^ [08:46] that's with lighter using 0.2 more [08:46] err, sorry, darker stay 0.2 more [08:47] 20% i mean [08:48] Trevinho: I like it better, do you mind pastebin the diff so that I can try live? [08:48] (thunderbird, terminal & xterm) [08:48] or just tell me when changing the % [08:49] didrocks: http://paste.ubuntu.com/25718782/ [08:50] didrocks: feel free to play with it, as it's quite late here... :) [08:50] I'm turning off my machine soon :P [08:53] Trevinho: go to bed :) [08:53] Trevinho: I prefer 0.1 [08:53] BUT [08:53] the only last remaining one [08:53] is thunderbird [08:53] mh [08:53] it looks ugly [08:54] oh [08:54] let me pastebin a screnshot [08:54] true [08:54] I see it [08:54] ah ;) [08:54] good [08:54] so thunderbird fix + 0.1 instead of 0.2 IMHO [08:54] (and good to ship :p) [08:54] (there is still tomorrow for this ;)) [08:54] well, firefox is the same... eh, that's why I did a different vesion before [08:55] I guess the issue is the menu bar being too dark [08:55] it should be the same color than gnome-terminal menu for instance [08:55] but really, go to bed :p [08:55] we are nearly there! [08:57] eh, this issue was the reason for http://bazaar.launchpad.net/~3v1n0/ubuntu-themes/use-inverted-decoration-gradient-on-maximized-windows/revision/581 [08:57] but we can't do it with three stops, it's too visible [08:57] so, let me find alternatives [08:58] tomorrow... [08:58] yep! [08:58] have a good night Trevinho :) [09:01] didrocks: mh, I guess thunderbird / ff doesn't expose the .maximized class... that's why [09:02] seb128, hanging out? [09:04] duflu, sorry, forgot to decline the meeting, I was out until now and I'm sick/have no much voice [09:04] good morning desktopers [09:04] k [09:05] doko, libcdio? I've no clue about it and no plan [09:05] doko, is that something we are supposed to look at? [09:05] ricotz, oSoMoN, it seems a bit late to land a libreoffice update, freeze is tomorrow [09:06] night Trevinho [09:09] Morning seb128 [09:13] didrocks: it seems that selecting these maximized windows isn't possible... [09:13] hey flexiondotorg, how are you? [09:13] So.... [09:13] only thing I can think of is something like [09:13] https://usercontent.irccloud-cdn.com/file/Zp6iiElG/ [09:13] not that I like much... [09:14] seb128: yes, according to the owned packages ... [09:14] Wishing I hadn't been up until 2am "testing" gaming on Ubuntu 17.10 ;-) [09:14] doko, where is that list again? [09:14] anyway bed for real [09:14] cheers [09:14] flexiondotorg, haha [09:14] Trevinho, night! [09:15] Trevinho: isn't the issue in the menu color? [09:15] seb128: enoclue ... but desktop-team is the bug subscriber [09:15] see you later Trevinho :) [09:16] doko, I see [09:17] seb128: I feel that the even/odd lines aren't too cool on rows with only one time, and even when you start a row… [09:17] for g-c-c- [09:18] https://imgur.com/a/Sd4zq [09:18] see the 3 panels [09:19] didrocks, yeah, looks better without the alternance imho [09:19] I would though impact the odd instead of even [09:19] could work out, maybe? [09:23] yeah, well I think it looks better all white personally and it seems a bit late to change it now for this cycle, wdyt? [09:28] Did someone say all white?... :) https://bugs.launchpad.net/ubuntu/+source/ubuntu-themes/+bug/1721693 [09:28] Ubuntu bug 1721693 in ubuntu-themes (Ubuntu) "Settings list separators conflict visually with list item backgrounds" [Low,New] [09:29] seb128: agreed, and following more adwaita [09:29] seb128: let's see, the build is almost done [09:29] cool [09:29] duflu, adwaita has sort of the same issue, separators are not very visible [09:30] seb128, yeah but I don't know of anyone who prefers the Ambiance list over Adwaita [09:30] seb128: yeah, clearly not looking nice: https://imgur.com/a/I2jqm [09:30] didrocks, weird looking indeed [09:31] didrocks, seb128, I have two identical monitors... one for 16.04 and one for 17.10 and it actually looks worse (higher contrast) on the 17.10 [09:31] ok, let's drop this, I have the fix for the highlight part and the theme GTK_THEME respect [09:31] Maybe something weird with 16.04 colour management when I view the screenshot [09:31] great [09:31] thanks didrocks [09:31] duflu: waow [09:31] * didrocks bzr uncommit; bzr revert [09:32] I mean I can't tell what you're seeing... because I am seeing the list two different ways on two different monitors [09:34] ... all the more reason to recommend everyone who does graphical things should calibrate their monitors [09:35] duflu: with or without calibration, it doesn't change that the 2 colors next to each other are merging with the background color [09:35] which looks weird [09:35] (as they are exactly the same) [09:36] Yeah I guess if you were choosing new alternating colours then I would be concerned about the monitor of the person choosing the colours [09:36] Without alternation it's fine [09:37] yeah, let's follow Adwaita and fix the other issues [09:39] In other news... I had a happy thought today. I suspect we're already seeing fewer unique crashes in gnome-shell than we did with unity8 development. Just a suspicion. [09:40] changing location, brb [09:46] duflu, agreed, that's what errors.u.c seems to confirm too [10:09] is gnome-boxes working for anyone? it crashes on start for me [10:16] works here [10:16] back [10:16] launches here [10:16] jibel: have you messed with the vm manually? [10:17] what are you guys trying? [10:17] didrocks, no [10:17] like removing in libvirt [10:17] didrocks, bug 1721258 [10:17] bug 1721258 in gnome-boxes (Ubuntu) "gnome-boxes crashed with SIGABRT in g_assertion_message()" [Critical,Confirmed] https://launchpad.net/bugs/1721258 [10:17] not alone [10:18] when I got crash on startup, it was when I removed the metadata in ~/.config ou something like that [10:18] I needed to clean all vms [10:18] using virsh directly [10:18] I ran it clean, never run it before. So not the same experience as you. [10:18] I don't have any VM [10:19] hum, I have one, but popey has 0… [10:19] s = 0x55e8f49758e0 "assertion failed: (default_connection != null)" [10:19] wonder if that's related [10:20] virsh list is empty? (maybe the VM was created via virt-manager a while ago or so) [10:21] virsh list --all is empty [10:21] I never use it [10:22] well, rarely [10:22] sounds like something to ping upstream about maybe? [10:23] teuf is responsive to some of our boxes issue (ofc, as any french guys ;)) [10:27] bien sûr [10:39] didrocks seb128 While debugging an issue with the Atom snap I've seen that gvfs-bin binaries regress on 17.10 [10:39] flexiondotorg, how so? [10:40] Executables for things like `gvfs-trash` still exist but don't do anything other the print an notice stating that you should now use `gio trash` instead. [10:41] So any applications shelling out to gvfs-* no longer function correctly. [10:42] Ah, I take that back. gvfs-trash is also deleting the file. [10:42] flexiondotorg, the gvds command still work, they just display the notice to tell you using the new binary [10:43] Thanks seb128. Just tested that. Sorry for the noise. [10:43] no worry [10:58] seb128: hey I saw you reverted that patch, as soon as I get my multimonitor setup situation fixed I'll take a quick look to see if I can fix it [10:58] maybe we don't need to sru it but we can discuss with upstream [10:59] seb128: O [10:59] bah [10:59] to see if they can accept it and maybe have that feature for 18.04 [10:59] seb128: I've got a patch for bug 1720331 -- the gnome-c-c patch alone should help, but the whoopsie-preferences one would make things a lot more solid [10:59] bug 1720331 in gnome-control-center (Ubuntu) "Whoopsie continually relaunching" [Medium,Triaged] https://launchpad.net/bugs/1720331 [11:01] andyrock, thanks [11:01] jamesh, ah, nice [11:05] seb128, can you merge/sponsor https://code.launchpad.net/~gunnarhj/gdm/language-switching/+merge/331980 ? [11:07] oSoMoN, so the workaround is what you recommend for us/this cycle? [11:10] seb128, yes, that's what I would recommend [11:11] oSoMoN, wfm, thanks [11:12] seb128, I'll continue discussing the issue with halfline to hopefully come to a solution that can be upstreamed, but given final freeze is tomorrow, if we want this fixed for 17.10 we have to go for that workaround [11:12] agreed [11:12] thanks [11:12] (which appears to be solid enough, given the limited testing Gunnar and myself did) [11:15] didrocks, should I just upload that or did you want to have a look/do something about the fallback session thing today [11:15] ricotz, will try to (but time is running out…) as soon as I'm done with bug #1718446 (and sorry for the delayed answer) [11:15] bug 1718446 in gdm3 (Ubuntu) "Ensure wayland -> xorg fallback to the corresponding session" [High,In progress] https://launchpad.net/bugs/1718446 [11:16] seb128, didrocks : I'm still experimenting with a less intrusive solution for the fallback bug, will keep you posted if that bears fruits [11:16] k, let's get that locale issue fix in the queue [11:17] we can always stack another revision on top later today if needed [11:17] yup [11:23] oSoMoN, et uploadé/mergé, merci ! [11:24] cheers [11:30] seb128: "that" being? I only see a language switching MP here? [11:31] ah, I see what you meant [11:31] that being gdm [11:31] without the other fix for the fallback session [11:31] (sorry, was answering on the hub) [11:31] didrocks, right [11:34] now we have a stack in the queue [11:34] we should perhaps try to poke reviewers :p [11:34] oSoMoN, ok, the packages are basically there in the ppa already [12:03] chrisccoulson, please push your packaging branches [12:26] quick question: if I a g_assert(false) is reached, should apport collect the crash info in /var/crash ? [12:27] yes [12:27] it doesn't work here [12:27] I'm on artful [12:27] you can look at /var/log/apport.log to see why it didn't, there might be a reason in there [12:27] kk thanks [12:28] there is not such a log here [12:28] apport should be enabled [12:28] andyrock, systemctl status apport? [12:29] seb128: maked as dead [12:29] that's your issue I guess [12:29] *marked [12:29] any info in the log at the bottom? [12:29] andyrock, what do you have in /etc/default/apport ? [12:30] http://paste.ubuntu.com/25719810/ [12:30] yep [12:30] it's enabled in /etc/default/apport [12:30] sudo systemctl start apport ? [12:30] I think I'm not the only one [12:30] then status? [12:30] I'm asking because I sent a ppa to collect more debug to some users but no report is collected [12:31] let me check [12:31] it started [12:31] maybe it dies after the crash [12:31] and if you trigger the assert now? [12:32] does it work? [12:32] checking... [12:32] you can ask the user if they have apport enabled in /etc/default/apport and what's the systemctl status [12:32] because we have some issues that don't get collected for some reasons [12:32] like XWayland segfault don't trigger report in my experience [12:33] which is pretty annoying and I don't know why, would need to debug [12:34] andyrock: hey, I couldn't ping you on the hub, are you subscribed? [12:35] didrocks: yes I am [12:35] seb128: now it worked [12:35] andyrock: ah, @and… didn't give me anything [12:35] andyrock, good! [12:35] andyrock: can you respond to https://community.ubuntu.com/t/status-of-the-top-icons/495/8 ? [12:35] try with @azzar1? [12:35] ah, that was it :) [12:35] you are hiding! [12:36] I've to change my irc nickname [12:36] ;) [12:36] edited! [12:36] I need a way to handle the switch [12:37] hidding on IRC as well? ;-) [12:39] ahaha [12:39] btw it's weird that apport crashed before [12:39] yeah [12:39] Trevinho: you broke the theme again! [12:39] let me rebot [12:39] *reboot [12:39] to check if apport is still alive after a reboot [12:40] andyrock, can you "journalctl -u apport"? [12:41] Trevinho: I ensured that the suggested action in disabled state was ok, your additional styles give something really confusing and ugly: https://imgur.com/a/6f3U9 [12:41] mine is all black to show it's disaled [12:41] disabled [12:41] yours it green with weird text [12:41] andyrock, btw https://bugs.launchpad.net/~azzar1/+assignedbugs is long, you might want to review that an unassign yourself from things you don't plan to work on anymore [12:42] apport is dead again [12:42] Trevinho: I really think there is no need for additional suggested-action state [12:42] andyrock, cat /etc/default/apport ? [12:42] seb128: sure, let me debug apport before [12:43] seb128: you don't believe in me [12:43] :D [12:43] https://www.irccloud.com/pastebin/zehmuIom/ [12:44] nothing in the logs [12:45] weird [12:46] I mean I got this one [12:46] https://www.irccloud.com/pastebin/6BRSQMSC/ [12:46] ah [12:46] maybe that's the issue? [12:46] not sure, because of the last line [12:47] time to do some python [12:48] it's weird, maybe try to sh -x /etc/init.d/apport [12:48] and see what is done [12:48] with sudo [12:49] systemctl is confusing to me still [12:49] unsure why it's showed as inactive without any log explaining why [12:57] seb128: soooo I think the problem is my locale [12:57] https://www.irccloud.com/pastebin/AJGyHahb/ [12:57] andyrock, ? [12:57] why LANGUAGE and LC_ALL are empty [12:57] I'm not completely sure but that should confuse python3 [12:58] I mean "UnicodeEncodeError: 'ascii' codec can't encode characters in position 978-1094: ordinal not in range(128)" this does not make so much sense [12:58] andyrock, seb128: I haven't followed the whole conversation, but apport is supposed to be active (exited) [12:58] there is no daemon which runs all the time, the init script just pokes the path into the kernel sysctl [12:58] pitti: hey! my apport is marked as died [12:58] hey pitti [12:59] pitti, right, that's how it is for me, but it's inactive for andyrock but has no error/log includes in the systemctl status nor in the journal [12:59] (there is no real service file, it's just the synthesized wrapper around the init.d script) [13:00] andyrock: "systemctl status apport.service"? is it enabled? [13:00] pitti, http://paste.ubuntu.com/25719810/ [13:00] https://www.irccloud.com/pastebin/8wWNzIBk/ [13:00] ls -l /etc/rc*.d/*apport* [13:01] there should be some `/etc/rc5.d/S01apport -> ../init.d/apport` [13:01] nothing [13:01] andyrock: there you are :) [13:02] so this is a fresh install [13:02] and my locale is broken [13:02] and apport too :D [13:02] /etc/rc*.d/*apport* [13:02] err [13:02] sudo update-rc.d apport defaults [13:02] nice! [13:03] this should be done by the .postinst, but apparently wasn't [13:03] andyrock, your locale doesn't seem too broken, LC_ALL not defined is fine and LANGUAGE missing shouldn't be an issue [13:05] kk [13:05] still apport should not crash when logging [13:06] andyrock, well, that's the ui and that hits an unicde error problem one one of the files you have in /var/crash no? [13:06] maybe a weird filename or something? [13:06] pitti, nice to see you around btw! [13:06] pitti, comment ça va ? [13:07] seb128: ça va bien, merci ! j'attends avec impatience aller à Berlin la prochaine semaine ! [13:07] python3 should support unicode out of the box [13:07] pitti, une conférence ? [13:07] je vais revoir Lars à nouveau, et des amis [13:07] they're using some fixed encoding [13:08] seb128: oui, https://all-systems-go.io/ and a two-day Cockpit hackfest before [13:08] btw it's not in the UI the crash [13:08] the affected strings are info['ExecutablePath'], info['ProcCmdline'] [13:09] https://bugs.launchpad.net/ubuntu/+source/apport/+bug/1368911 [13:09] Ubuntu bug 1368911 in apport (Ubuntu) "Apport crashes when writing to stderr: UnicodeEncodeError: 'ascii' codec can't encode character '\xc1' in position 130: ordinal not in range(128)" [Medium,Triaged] [13:10] we can use the quick fix suggested in comment #1 [13:10] andyrock, can you ping bdmurray about that on #ubuntu-devel? [13:10] sure thing [13:10] thanks [13:12] seb128: et toi, comment vas-tu ? [13:13] pitti, ça va bien mais j'ai un rhume :-/ [13:13] urg, file-roller is weirdly themed when unfocussed [13:13] seb128: argh, le vieux ubuflu [13:13] pitti, oui, la variante enfant [13:13] kindergardenflu [13:20] I just told Steve about the missing rc links :-) [13:21] Laney, can you open a bug about it? [13:21] Laney, apport didn't change but I guess it's dh_installinit behaving differently [13:22] seems similar to https://bugs.launchpad.net/ubuntu/+source/apport/+bug/361579 [13:22] Ubuntu bug 361579 in apport (Ubuntu Jaunty) "does not register init scripts rc.d symlinks any more" [High,Fix released] [13:23] Laney, oh, and hey and thanks for pinging Steve about it :-) [13:26] didrocks: answered [13:28] thanks andyrock :) [13:29] andyrock: hum, I need to amend your answer [13:29] andyrock: extensions as part of a mode aren't in enabled-extensions [13:31] didrocks: I need to check if she/he has other extension enabled [13:31] that can conflict with our extension [13:32] andyrock: ahhhh, yeah, could be interesting [13:32] I guess she/he still has one extension dealing with appindicator [13:32] let me edit :) [13:33] andyrock: edited, yeah, I think we should add something around it in the apport hook [13:33] didrocks: ? [13:34] andyrock: I edited my answer on the hub. I think we should make so that `ubuntu-bug gnome-shell` reports enabled extenstions [13:34] we've the GSettingsChanged.txt [13:34] ? [13:35] or something like that [13:35] does this work [13:35] seb128: https://bugs.launchpad.net/ubuntu/+source/apport/+bug/1722801 [13:35] Ubuntu bug 1722801 in apport (Ubuntu) "apport isn't started on new installs" [Undecided,New] [13:35] and np, and hi [13:35] Laney, thanks, I think it's xnox's fault [13:35] andyrock: ah? maybe, need to look for an example [13:36] so two bugs in apport spotted today by me \o/ [13:36] :D [13:37] andyrock: features, not bugs ;) [13:38] Laney, I commented on the bug, I think the removal from the upstart job by xnox made dh_installinit act differently and drop the needed lines [13:40] think it's going to get assigned to bdmurray, but thx [13:40] guess someone else could get in there first and fix it if they wanted ;-) [13:40] bdmurray wfm [13:40] I don't intend to debug more, I just got curious of what changed since the rules and postinst didn't change since xenial [13:41] andyrock, thanks for spotting the issue :-) [13:42] we coincidentally found that here too at the same time [13:42] * didrocks curses andyrock for spotting the issue :) [13:42] serendipity [13:42] seb128: now we'll get a bump in errors report [13:45] yeah, which is good [13:45] shame that we didn't get it before beta [13:45] we are probably missing reports :-/ [13:45] hey Trevinho, I subscribed you to this bug: https://bugs.launchpad.net/ubuntu/+source/gnome-shell/+bug/1722811 [13:45] Error: ubuntu bug 1722811 not found [13:46] willcooke, is that a segfault? [13:46] willcooke, he's sleeping [13:46] Anthony Wong [13:46] oops [13:46] Anthony reported it [13:46] yeah, segv [13:46] k [13:46] I guess it hasn't been retraced yet [13:46] it's not accessible [13:46] ack [13:46] I've added you too seb128, just fyi [13:47] thx [13:47] If you have hidpi, and it auto scales to 200%, when you set it to 100% then it seg faults [13:47] having steps to reproduce is useful [13:48] confirming the crash here [13:49] andyrock, that's good, make easier to work on it for whoever has a look [13:55] mmm no symbols in the bt [13:56] shouldn't debug symbols be installed by default in dev releases? [13:56] no [13:56] they are bigs and not useful to most users [13:56] and you can use apport-retrace afterwards to get the symbols and a debug bt for you [13:56] btw "sudo apt-key adv --keyserver keyserver.ubuntu.com ---" fails [13:56] ah ok [13:56] I used directly apport-unpack [13:57] that's the manual way [13:57] root@cosmic-terrapin:~# [13:57] lxd's machine names ♥ [13:57] hehe [13:57] seb128: ERROR: report file does not contain one of the required fields: Package [13:58] andyrock, did you meant to add a --recv ? [13:58] using apport-retrace [13:58] :-/ [13:58] seb128: i used this command "sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys 428D7C01 C8CAB6595FDFF622" [13:58] andyrock, well I guess enabled the ddeb source and install the packages you need manually [13:58] it's stucked [13:58] andyrock, what error did you get? [13:58] hmu [13:58] it can be that it fails to write the package name because of the unicode/ascii issue [14:00] could be [14:01] andyrock, it's your day of hitting issues :-) [14:01] usually that's you [14:01] ;) [14:02] yeah [14:02] most of my days feel like that [14:02] I start debugging one problem [14:02] and mid-day I'm 3 issues down the stack that I hit on the way [14:02] trying to figure them in reverse order before getting back to what I meant to do [14:03] seb128, I marked bug 1715330 as fix released. I cannot reproduce the problem on an up to date artful. [14:03] bug 1715330 in Mutter "gnome-shell crashed with SIGSEGV in _cogl_boxed_value_set_x()" [Critical,Fix released] https://launchpad.net/bugs/1715330 [14:05] mmm I fixed that ascii fixed but I still don't get the Package file [14:06] jibel, great, thanks [14:06] I'm using gnome-shell installed from the repos [14:19] apport-retrace ... -R ... fixes this [14:19] my fault [14:19] why your fault? [14:20] btw I've a gnome-shell report and it has the same issue [14:20] so might be another apport bug [14:20] seb128: https://bugs.launchpad.net/ubuntu/+source/apport/+bug/1315530 [14:20] Ubuntu bug 1315530 in apport (Ubuntu) "apport-retrace needs initial data collection" [High,Invalid] [14:21] look at Martin's comments [14:21] looks like it's supposed to work like that [14:22] confusing, retrace should do the -R for you [14:24] andyrock, but yeah, technically not a bug [14:31] Laney, seb128 - that is sad. I am surprised we didn't notice this yet =/ [14:31] i guess we are happy that there is no poop-up on boot?! =) [14:31] right [14:32] heh [14:32] it's only for new installs though I think [14:32] right [14:32] seb128: Thanks for merging/sponsoring! [14:32] upgrades have the postinst which acted on install when it was still working [14:32] who cares about those guys! [14:32] GunnarHj, hey, yw! [14:32] seb128: I'm really going to write that paper we talked about yesterday; I'm ambivalent to the "replace with their solution later/next cycle" idea you mentioned at the MP. [14:33] GunnarHj, k, thanks for writing that [15:44] seb128, Laney : I attached a patch to gnome bug #788552 (probably not suitable for upstream but rather for a distropatch), your opinion welcome [15:44] Gnome bug 788552 in general "Ensure wayland -> xorg fallback to the corresponding session" [Normal,New] http://bugzilla.gnome.org/show_bug.cgi?id=788552 [15:44] oSoMoN, thanks [15:45] note that I attached it to the upstream bug report because ideally I'd like to get halfline's opinion on it too [15:46] right [15:46] I'm about to go back and spend some time away from the computer [15:46] but I'm going to let upstream/Didier/Iain time to comment and I have a look tomorrow morning to see what's the status :-) [15:54] I'll prepare merge requests against gdm3, ubuntu-session and gnome-session in case my approach gets approved [16:05] oSoMoN: if you're going to distro patch anyway, why not use your patch from comment #11? [16:05] that doesn't have that bug where you don't get the wrong session selected I don't think [16:05] anyway, fine with whatever you think is best [16:09] Laney, right, I'll test again the patch in comment #11, it has the advantage of not requiring to patch gnome-session [16:10] with that other patch in comment #25 I was trying to conform to the expectation that fallback sessions have the same desktop file name [16:10] which according to Ray is how the fallback mechanism is supposed to function [16:10] nod [16:10] makes sense [16:10] I think the final patch is probably a combination of both of those [16:11] well, reading those keys instead of looking at the filename [16:11] but you get it [16:13] yes, but I don't think I can realistically implement that and have it reviewed and merged upstream before tomorrow (national holiday in Spain tomorrow, I won't be around much), so it's gonna have to be a simple distro patch [16:14] ok [16:14] Hi, After new design of the website I can not access any of the api doc for [16:14] indicators and messaging menu. For example, [16:14] https://developer.ubuntu.com/api/devel/ubuntu-13.10/python/MessagingMenu-1.0#MessagingMenu.Message [16:14] now gives 404. :/ [16:14] that's fine, thanks for working on it! [16:15] Doc team's Gunner pointed me to this: https://lists.ubuntu.com/archives/ubuntu-community-team/2017-October/001567.html [16:15] Does any one know what happened here ? [16:20] So the recent half-tiling change in mutter is causing some really weird, screwy behavior. [16:23] dmj_s76: are you using gtk 3.22.21 or .24? did you see that there was a mutter update today? [16:31] jbicha: I'll test [16:31] We were seeing this bug for a few days including yesterday. [16:39] popey: Anything on the issue? [16:39] hmm? [16:40] Read my comments above... [16:42] I don't understand. What's the issue I can help with? [16:42] popey: I raised the issue here: https://lists.ubuntu.com/archives/ubuntu-doc/2017-October/020562.html [16:43] davidcalle: [16:43] ^ [16:56] jbicha: Okay, I think I understand this a little better now. [16:57] So the mutter half-tiling resize *is not* present on Wayland, but it exists and is very buggy on X11 [16:58] Is half-tiling an intended feature in Ubuntu (17.10) or is its presence on X11 an anomaly. [16:59] *half-tile resizing rather [17:07] 1) resizing the half-tiling alignment looks really bad. (Slow, desynchronized redraw) One window lags behind the other, so you see the desktop through a gap between the windows as you drag. [17:07] 2) Since the mutter change that enables changing alignment, also makes half-tiled windows raise to the foreground in weird and inconsistent ways. === Guest11681 is now known as fredp === fredp is now known as Guest18603 [17:17] here we go: https://bugs.launchpad.net/ubuntu/+source/gdm3/+bug/1718446/comments/4 [17:17] Ubuntu bug 1718446 in gdm3 (Ubuntu) "Ensure wayland -> xorg fallback to the corresponding session" [High,In progress] [17:22] * Laney won't have time to review that, hopefully someone else can [17:22] can probably do it in the queue once uploaded though [17:22] thanks [18:02] dmj_s76: it's an intended feature but it doesn't work at all yet since gtk 3.22.24 is stuck in artful-proposed [18:03] I agree that it looks bad during resizing https://bugzilla.gnome.org/show_bug.cgi?id=788598 [18:03] Gnome bug 788598 in general "tiling in mutter 3.26.1 looks glitchy when resizing" [Normal,New] [18:04] jbicha: You're saying that the current version is buggy but *should* work well in the proposed package? [18:05] I'm saying that resizable half-tiling requires the proposed gtk3 package (otherwise it won't be enabled at all) [18:05] but the glitchiness hasn't been fixed [18:06] jbicha: There are also some bizarre behavior issues re: focus/raising issues, handling more than two half-tiled windows, some windows not letting you resize. [18:06] please file those bugs upstream, it's not an Ubuntu-specific feature and we didn't write the code for the feature :) [18:16] jbicha: the gtk3 package specifically fixes gtk3 windows not allowing resize (but non gtk3 apps working)? [18:19] I don't know, could you file detailed bugs? [18:20] working on it [18:21] that just sounded familiar [18:37] https://usercontent.irccloud-cdn.com/file/eoUHxHdC/lp.png [18:38] ops wrong channel [19:16] jbicha: https://bugzilla.gnome.org/show_bug.cgi?id=788846 [19:16] Gnome bug 788846 in general "window focus regression with resizable half-tiling" [Normal,New] [19:34] why oh whyyyyyyyyyyyyy [20:55] jamesh: online? [20:58] robert_ancell: seb128 wanted your input on LP: #1721359 see also his uploads today for https://launchpad.net/ubuntu/+source/gnome-orca/+changelog [20:58] Launchpad bug 1721359 in simple-scan (Ubuntu) "Simple Scan translations are not synced with upstream" [Undecided,New] https://launchpad.net/bugs/1721359 [20:58] good morning :) [20:58] jbicha: hi :) [20:59] jbicha: I'm not sure about that one, since we've moved hosting to GNOME aren't the Launchpad ones obsolete? [21:00] everything in main uses LP language packs unless we explicitly opt out [21:00] So I set the simple-scan project in Launchpad to do translations, but I don't think I've ever changed any settings (or know of any) for the packaging translations. [21:01] Any idea who knows how the importer works? [21:01] seb128 ;) [21:02] I guess you 2 should discuss it [21:02] Comparing against another project... I'll play with some settings [21:08] hi robert_ancell [21:11] willcooke: heya [21:14] jamesh: we need to allow polkit auth for cancel, i.e. http://paste.ubuntu.com/25722399/. This allows all operations to cancel, so not sure if this needs to have different polkit actions depending on the change. Can you add that one to your list? [21:14] I'm fixing the cancelling code in snapd-glib and we can't cancel an install/remove with polkit... [22:38] seb128: I'm surprised you're awake, I pinged robert_ancell about simple-scan translations but maybe he wanted to discuss that with you? [22:38] jbicha, seb128 is awake? [22:38] maybe :) [22:39] shrug [22:40] I've a cold and didn't manage to sleep so I was checking if it was going alright for the guys in N-Y [22:41] I don't understand those translation sharing properties well, best to check with wgrant [22:41] It seems like Launchpad doesn't handle un-upstreaming very well. Some of the options didn't seem to be resettable. [22:41] but basically I think that if sharing with an upstream branch is enabled then it uses the translations for that vcs and doesn't import the files from the package [22:42] the issue is that sharing is usually with trunk, which not be the serie you package [22:42] also the imports use to bug on submodules so some were outdated [22:43] I guess it depends of the project if it makes sense to share translations or not [22:44] oh and I don't know what it means for packages that have distro specific strings, if sharing is enable does it use the upstream template? [22:45] robert_ancell, ^ basically what I know, many things I'm uncertain about so I recommend you try asking wgrant or somebody who understands that part of launchpad [22:45] I think I changed enough things that it might work now, will ping wgrant if now [22:45] not [22:47] sounds good [22:49] on that note try to go back to sleep [22:49] night