[00:00] ochosi, ah, it's not safe to call anything in the sigterm callback [00:00] that might do something [00:00] oh [00:00] ok [00:00] you should use (me looks it up) [00:01] ochosi, you should use g_unix_signal_add if you can use glib 2.30 [00:01] robert_ancell: this is the current version: http://bazaar.launchpad.net/~lightdm-gtk-greeter-team/lightdm-gtk-greeter/trunk/view/head:/src/lightdm-gtk-greeter.c#L1814 [00:02] I'd switch the daemon to using that but it came out after we started [00:02] and then just call g_main_loop_quit from there [00:02] and clean up after that [00:02] actually, yeah gtk_main_quit is the right thing [00:03] so that code should be just fine then? [00:03] except replace signal() with g_unix_signal_add() [00:03] because it's not safe to call pretty much anything from a signal handler [00:04] g_unix_signal_add runs from the main loop [00:04] ok, could that be causing the issue? [00:04] possibly? [00:04] but worth trying [00:04] yup, definitely worth trying [00:05] because lightdm would have sent you the signal, then the gtk_main_quit might have done something weird with a thread [00:05] humm, i see [00:05] so just connect the gtk_main_quit directly as callback to that signal? [00:05] you probably can as a shortcut [00:12] robert_ancell: sorry to be so verbose (i've never used g_unix_signal_add before), like this? g_unix_signal_add(SIGTERM, gtk_main_quit(), NULL); [00:12] instead of the current signal (SIGTERM, sigterm_cb); [00:12] yes, except drop the () from the gtk_main_quit [00:12] ah sure [00:13] I haven't actually used it myself but that looks right [00:13] well i can at least push it to a branch and ask ppl for testing [00:13] thanks a lot for the pointer, robert_ancell! [00:13] np, hope it solves it! [00:14] me too :) === duflu_ is now known as duflu === FJKong_afk is now known as FJKong [06:03] Good morning === soren_ is now known as soren [09:02] good morning desktopers [09:02] hello! [09:02] hey Laney, how are you? had a good w.e? [09:03] pitti, hey, do you know why we didn't get a langpack update for ~1month? would be nice to have one for beta this week [09:03] we had a few regular ones in february [09:03] but nothing since [09:04] seb128: err, we didn't? [09:04] # disabled for 14.04 beta 1 preparations [09:04] eek, I forgot to re-enable the cronjob, sorry [09:04] seb128: I'll build them now [09:04] no worry [09:04] pitti, danke! [09:05] pitti, where is the cronjob (do I have access in case it happens once while you are not around to do the edit)? [09:05] seb128: germanium.canonical.com [09:05] langpack:x:2520:pitti,dpm [09:05] seb128: I'll send an RT to get you added [09:06] seb128: no seb128 user on germanium, I'll include that [09:07] hey seb128, good thanks - climbing and hanging out in the sun during the parts where it was sunny [09:07] you? [09:07] & hey pitti ;-) [09:07] pitti, thanks, but both dpm and you might be enough? well, your call, I guess it doesn't hurt having an extra backup to the list ;-) [09:07] seb128: yep [09:08] Laney, good thanks, some shopping, some house cleaning, went to the spa and mostly been lazy otherwise ;-) [09:08] nice! [09:08] I've been hearing about French elections on the news this morning ;-) [09:09] oh right, local elections [09:09] but as usually, some people use those to show they are unhappy about the government [09:10] yeah, because of the Front national winning in some towns [09:10] right [09:10] hey Laney, how are you? [09:10] doing good thanks pitti, sunny day today! [09:11] although still quite cold [09:11] and you? good weekend? [09:11] Laney: heh, here as well; just the right time after a dreadfully wet and cold WE.. [09:11] (oh no, the panda has crashed for the first time since dist-upgrading to trusty without being rebooted) [09:11] (#willitboot) [09:11] Laney: yeah, we changed the planned BBQ to a board/card gaming afternoon with some friends yesterday, and we did some gardening on Sat morning [09:11] Laney: wow, still using your panda? [09:13] nice [09:14] yeah, it does some graphing for my currentcost and runs debmirror [09:14] & test builds of small stuff [09:14] not looking too hopeful [09:24] ochosi, hey, I noticed that the theme change to restore the buggy symlinks made the gnome/unity-control-center keyboard icon wrong [09:25] seb128: oh, i'm very sorry about that regression, i was obviously not aware of that. can you show me a quick screenshot of what it looks like and what it should look like and i'll try to fix it [09:26] (most likely an icon from Humanity that would have to be included in ubuntu-mono. in case that's impossible due to licensing issues, we'd have to revert the patch, but in that case i'd suggest to also drop the broken symlinks in the process) [09:26] ochosi, what it looks like, /usr/share/icons/ubuntu-mono-light/status/22/input-keyboard.svg [09:26] ochosi, what it should be, /usr/share/icons/Humanity/devices/48/input-keyboard.svg [09:27] so is the icon in unity-control-center scaled up? [09:27] or is it 22px [09:27] scaled up [09:28] ochosi, http://people.canonical.com/~seb128/keyboard.png [09:29] hmpf, yeah, well actually that's a problem of how icon-themes inherit in ubuntu (iirc there's a ubuntu specific patch) [09:29] I don't think it is [09:29] our themes are in priority order: mono > humanity > gnome > hicolor [09:29] on other distros if an icon isn't found in the demanded size, it looks in the inherited theme [09:29] in ubuntu it scales up/down any icon with that name it finds in the icon theme [09:30] i'm not saying which version is better in general [09:30] well, you have flags in GTK to specify if you want to enforce the size or not I think [09:30] but that's how things work [09:30] hum, are you sure it's the theme? [09:30] larsu, do you know about ^? [09:30] so anyway, to work around this problem, the easy thing is to just include a 48px version of input-keyboard.svg in ubuntu-mono [09:31] that way, the correct icon gets used in the settings-manager [09:31] well, the icon is not mono :p [09:31] I think we should rather drop the 22x from it [09:31] (there is also the license issue) [09:31] yup, that's the other solution [09:31] ochosi, https://code.launchpad.net/~jbicha/ubuntu-themes/drop-input-keyboard-icon/+merge/181793 [09:32] so, what i'd propose: [09:32] not the first time we have that issue [09:32] :) [09:33] * larsu reads scrollback [09:33] so i think an ok workaround would be to just rename the input-keyboard.svg to ibus-keyboard.svg [09:33] that way there's still support for the ibus and nothing is broken [09:35] shoudln't it be input-keyboard-symbolic? [09:35] nah, it's not a symbolic icon [09:35] it's just a mono icon [09:36] shouldn't it be input-keyboard-panel? :p [09:36] yeah, that's another easy solution :) [09:36] * larsu is confused [09:36] although that name doesn't get used anywhere ;) [09:36] larsu: how can i help? shall i re-summarize? [09:36] yes please [09:36] ok [09:37] 1) there were broken symlinks in ubuntu-mono to keyboard.svg and ibus-keyboard.svg (pointing to input-keyboard.svg, which wasn't there anymore) [09:38] 2) i saw that, brought the icon back, symlinks fixed [09:38] 3) that causes unity/gnome-settings.manager to pick up that icon for the keyboard settings [09:38] 4) so there are several scenarios to resolve this problem [09:39] 4.1) add a 48px colored version (from Humanity?) of input-keyboard to ubuntu-mono so that one gets used in the settings manager [09:39] 4.2) drop input-keyboard and all the symlinks [09:40] 4.3) rename input-keyboard to either ibus-keyboard or something new to keep the symlinks alive [09:40] 5) decide what to do (this is where we are now) [09:40] ochosi: thanks. 4.2 sounds best to me, but I guess that would mean breakage in other parts? [09:40] * seb128 votes 4.2 [09:40] 4.1) has licensing issues, so i'd not advise to go with that [09:40] it's also not a mono icon [09:40] right [09:40] so it shouldn't be in ubuntu-*mono* [09:40] seb128: I like your logic ;) [09:41] I doubt it's going to create new issues [09:41] it sounds ... sound [09:41] the icon was missing/symlink buggy for a while [09:41] that's what I as thinking. Let's just do that, then [09:41] so it would just send us back to that state, with the buggy symlinks cleaned [09:41] 4.2 is possible, ibus will simply have another icon then (a colored one i suppose) and look less integrated for those that use it over indicator-keyboard (not sure many will, but just mentioning it) [09:41] oh [09:41] hum [09:41] ibus in the panel, you mean? [09:41] yup [09:42] there's most likely no harm in keeping input-keyboard as ibus-keyboard [09:42] right, that works for me as well [09:42] * larsu is okay with that as well [09:43] ochosi, ok, let's do that then, rename it ibus-keyboard [09:43] okeydokey [09:43] not sure about the "keyboard.svg" in that status folder [09:43] no idea what that would be used for [09:43] you can either make it a symlink to ibus-keyboard or drop it [09:43] * ochosi is a bit clueless about that one [09:44] drop it I would say [09:44] fine by me [09:44] we can add it back if somebody find an use and request for it [09:44] less surprise this way [09:44] (the broken symlink from before didnt work anyway, so it's not a regression in 14.04 so far) [09:45] yup [09:45] sounds sane [09:45] seb128: do you want/need a MR from me or are you handling this directly? [09:45] ochosi, mp would be nice so I can ack it and land it [09:45] if I do it I need to find another lander [09:46] hehe, bureaucracy :) [09:46] yeah... [09:46] ochosi, thanks ;-) [09:46] no problem ;) [09:46] * seb128 finds https://bugs.launchpad.net/ubuntu/+source/gnome-control-center/+bug/1189382 which has a good point [09:46] Launchpad bug 1189382 in gnome-control-center (Ubuntu) "Misleading icons for “Keyboard” and “Keyboard Layout” in Control Center" [Undecided,New] [09:46] we should invert the Icon= of those 2 .desktops [09:55] seb128: hm, yeah, that is somewhat true. so does that affect my MP and what we just discussed though? [09:56] ochosi, no, the icons are right, the .desktop in u-c-c should be updated [09:56] ok, good [09:56] * ochosi is a bit distracted and just wanted to make sure [09:56] thanks for checking ;-) [09:57] np [10:08] seb128: done [10:09] ochosi, thanks [11:00] * ogra_ sighs ... [11:00] so my XPS13 still doesnt turn off the screen if i close the lid ... [11:01] it should suspend? [11:01] and i get a black screen about 30sec after unlocking when it was closed over night [11:01] seb128, no, it shouldnt [11:01] seb128, but the backlight should turn off [11:02] * ogra_ had hopes that was fixed with the last upgrade (which i did yesterday) [11:02] is that the "screensaver doesn't turn the screen off" issue? [11:02] but seems it wasnt [11:02] might be [11:02] when did the issue start? [11:02] when we got the new lock screen? [11:02] two weeks ago perhaps [11:02] yeah [11:02] right [11:03] the unity guys are supposed to work on that [11:03] it seems to turn it off ... but only after i unliocked [11:03] you can probably go back using gnome-screensaver as a workaround meanwhile [11:04] i.e. i open the lid in the morning, see the lock screen, type my PW, see the desktop for 30sec, then everything goes black with a clock in the top right ... when i then wiggle the pointer i get the desktop back [11:04] it doesnt hurt, i can wait ... just want to be sure we dont release with it [11:06] yeah, don't worry [11:06] ogra_, https://bugs.launchpad.net/ubuntu/+source/unity/+bugs?field.tag=lockscreen [11:06] oh, thats a few :) [11:07] yeah, I'm pondering if we should go back to the old lockscreen, need to check with bregma what's the status on getting his team fixing those issues [11:07] ogra_, I think your issue is bug #1292935 [11:07] Launchpad bug 1292935 in unity (Ubuntu) "New Lockscreen inhibits putting the screens to sleep." [High,Triaged] https://launchpad.net/bugs/1292935 [11:08] right [11:08] jst reading that [11:08] *just [11:08] seb128, thanks ! [11:08] * ogra_ clicks "me too" and moves on === ayan_ is now known as ayan === Sweetsha1k is now known as Sweetshark [11:15] seb128: there are a few very similar to that I don't know if they have all been merged [11:16] davmor2, I don't know, I'm waiting for some round of fixes to land to go through the bug and triage them/ask if the issues are resolved [12:15] Laney, Trevinho: what was the conclusion about the apps scaling UI? [12:16] Trevinho was reworking it to look more like the design [12:16] the stuff in the landing-011 is cool (just tested it), having the current slider changing the text and GTK scales as fit [12:16] seb128: nice :) [12:16] Laney: yeah, I've mostly done it, but i've to add the 2nd slider [12:16] like 1.5 sets unity and text to 150% and keep gtk to 1 [12:16] cool [12:17] 2 does the right set and put everything to 2 [12:17] nice to see also the menus match the panel text ;-) [12:17] Trevinho, why a second slider? the current behaviour seems fine to me [12:18] seb128: we want to control things on multimonitor [12:18] what things? [12:19] seb128: the scaling of text or gtk is global, so we need to define what to do if there are multiple monitors attached [12:19] right [12:19] didn't we speak about having a combo that let you select which screen you want to use as a basis for scalling? [12:19] seb128: or I've done a combo for now [12:19] seb128: http://i.imgur.com/4REmE0H.png [12:20] seb128: that's what I've done for now [12:20] if that's fine, it's ready === MacSlow is now known as MacSlow|lunch [12:20] I'm sure mpt is going to have to say on labels alignement at least [12:21] having the "1" on the side rather than under the scale as well [12:21] I can make them to stay on left, not sure... As they don't look so nice [12:21] yeah, these are small things to tune :P [12:21] the logic thing on the combo is the main one [12:21] right, I like that screenshot on principle [12:22] though I'm still unsure the combo is something users are going to understand [12:22] in any case, get it in the shape you want, then ask for a ffe/uife, you are going to need those (or recycle the one from hikiko, though the solution changed slightly so the description needs to be updated) [12:24] yeah... === FJKong is now known as FJKong_afk [12:25] Laney: any thought on that screenshot ^ [12:26] The idea is reasonable, but I still find 'Highest scale across displays' to be unintuitive [12:27] Laney: yeah, any proposal? [12:29] Not as such ... [12:36] Something like 'Largest available size' [12:36] ? [12:36] (brb, early lunch) [12:46] attente, hey, you are around that this time? [12:46] * seb128 losts track of timezones [12:48] attente, I'm looking at landing your u-g-m fixes, they seem mostly good but that makes "_" show back in e.g inkscape or gvim, like "_Open" (the underscore is on its own char rather than being an underline) [12:56] attente: ah, about that.... I've noticed that in inkscape, if you've loaded svg files with "_" in the name, then they shows as underlined inside the reopen menu === alan_g is now known as alan_g|lunch === MacSlow|lunch is now known as MacSlow === alan_g|lunch is now known as alan_g === greyback is now known as greyback|lunch [14:23] bregma, hey, I tested the unity landing ppa, +1 for me to land it [14:24] kenvandine, hey, well done on landing the content-hub work! ;-) [14:24] seb128, thx :) [14:26] seb128, thanks [14:26] oh, andyrock is there! [14:27] andyrock, hey, how are you? still working on those lockscreen issues? [14:28] seb128, yes [14:28] they are pretty complicated [14:29] andyrock, do you have any estimate when you are going to get fixes? beta freeze is today, it starts looking like we should maybe considering reverting to gnome-screensaver if the bugs are too difficults to be fixed for the release [14:29] gah! /me cries [14:30] i can have a fix for the double lock today [14:30] for the lockscreen not invoking idle screen not sure [14:30] andyrock, that would help maintaining the feature in place, that one is a real user experience blocker [14:30] ok [14:30] can't we just keep gnome-screensaver handle the idel? [14:31] seb128, that what we do [14:31] the problem is that g-s requires the grab to do that [14:32] oh [14:32] i managed to get the fade starts [14:32] but as soon as it finishes for some reasons it starts again [14:33] ok [14:33] one bug at the time, let's fix the double lock one, then the idle one [14:33] ok [14:34] than we have another issue [14:34] you switch user [14:34] than you move back to the other user [14:34] the lockscreen starts to fade [14:35] but it's an issue with all the lockscreens [14:35] let's not worry much about things are not regressions [14:45] morning all [14:45] anyone else seen an issue where unity background is completely gray but controls work fine (left side bar) [14:45] I think gnome and Unity aren't playing nice together [14:45] bregma, do you keep the upstream unity bugs on "fix commited" rather than "fix released" on purpose? if so, it might be nice to roll a tarball (or do whatever you require to close those) to clean the noise from the buglist [14:45] jmadero, hey, use #ubuntu for user questions [14:45] lol dang Sweetshark [14:46] seb128: thanks - I was sent here by Canonical;) [14:46] lol [14:46] Sweetshark, we don't do user support on that channel, we have launchpad for bug reports :p [14:49] seb128: sure, but its a long journey from #libreoffice-qa to launchpad ... [14:49] seb128, we keep upstream bugs as 'fixed committed' until we do an upstream release -- which will be soon so we can branch for maintenance [14:50] Sweetshark, well, that question is an unity one... [14:50] bregma, you should roll a beta tarball or something so you can close them ;-) [14:50] we'd release more often except we keep getting process curveballs thrown at us [14:50] jmadero: would you have a filed a bug on launchpad, if I would have told you in the #libreoffice-qa channel? or would you have just walked away? [14:51] bregma, everybody else just close their upstream bugs when stuff land in Ubuntu nowadays (or just use one buglist, the ubuntu one) [14:51] jmadero, can you share a screenshot showing your issue? (it doesn't ring a bell but I'm unsure to understand your description) [14:51] Sweetshark: lol trying to get me to take sides knowing we're colleagues on LibreOffice ;) [14:52] but to be honest - I would have walked away - I in general use a different DE but like Unity to work [14:52] so that I can test LibreOffice bugs - this is a minor inconvenience [14:52] let me boot into Unity - one minute [14:52] okay more than one minute - Okular is terrible and takes about 5 minutes to save :-/ [14:53] but I can describe better [14:54] after I installed Gnome3 from ppa - Unity was completely broken, I reinstalled Unity and it seems to now work fine but the background is this terrible gray and I cannot change it [14:54] everything else seems to work fine -- although there were some tricks I had to search for (like the mouse pointer goes invisible when you install gnome3) but that's been fixable [14:57] jmadero, weird, gnome3 shouldn't impact on Unity sessions [14:57] Trevinho, I don't understand what is your new migration script in unity doing? [14:57] seb128: yeah I never had the issue until this round of install - very strange [14:57] okular is still frozen so can't screenshot it right now [14:58] I try not to hate on free software but okular can be troublesome ;) [14:58] hehe === vrruiz_ is now known as rvr [15:03] still frozen, dangit [15:05] seb128: maybe you'll know more, maybe it's already resolved - in the morning I upgraded my desktop system and apt forced the uninstall of ubuntu-desktop because checkbox-ui is unavailable [15:05] seb128: is that a known thing? [15:06] sil2100, yeah, I think it's a fallover from the python transition === greyback|lunch is now known as greyback === alan_g is now known as alan_g|tea === alan_g|tea is now known as alan_g [15:45] Laney, thanks for looking at activity-log-manager, if you feels like working a bit more on it there are a few other issues (the lock screen values are not translated, you also get a segfault if you open online-account then security&privacy, ...) [15:46] seb128: oh, interesting, is that how you do it? [15:46] the segfault? [15:46] no, doesn't happen [15:46] yes, there is one libgee unhappy when you do that [15:46] I tried to reproduce it [15:46] but couldn't [15:46] run u-c-c, go to online accounts, go back to the grid, click on privacy? [15:46] yeah [15:46] happens every time with those steps for me [15:47] aha I got it now [15:48] Laney, I've a guess at why it's happening [15:48] uoa and activity-log-manager link to different libgee versions [15:48] so I wouldn't be surprised if that's a symbol conflict [15:49] we should try to have only 1 libgee version in there [15:50] Laney, you can maybe have a go at try to build alm with the new libgee (I would start by trying that) [15:52] I'll look into it, thx for the steps [15:52] yw! [15:52] do you want me to write that on the bug? [15:53] Laney, sometime writing about thing helped, I just though about the libgee versions mismatch now, where I had looked at the bt before [15:53] it just occurred to me while trying to think about what the online account had which was special [15:57] seb128: where do you see gee coming from online-accounts? [15:59] Laney, hum, I tried to uninstall libgee-0.8-2 and it was removing it, but that's through indirect depends [16:00] here goes that theory :/ [16:01] could be loaded into the process through plugins [16:03] yeah [16:03] the empathy stuff use libgee [16:03] it would still be good to have a go at building alm with the new gee, cleaning the old lib and maybe resolving that issue on the way [16:14] tedg: any reason https://code.launchpad.net/~a-j-buxton/libindicator/remove-timeout/+merge/198070 hasn't been reviewed? [16:14] Laney, We've talked about it, it needs all the startup patches to land to make it safe. [16:15] Laney, It's more in holding than "not reviewed" [16:15] seb128: ah, I missed the mention. It's keeping the user-setting text-scaling, copying it from gnome to unity [16:16] Trevinho, how is it doing so? is there a new key for unity? I don't see it mentioned there [16:18] seb128: yes there's a new key for unity... That will update the gnome one accordingly, if needed [16:19] Trevinho, the script doesn't mention any unity key though, how does it work [16:20] the Popen(("dconf load "+GNOME_UI_SETTINGS_PATH).split()..) and "p.communicate(input="[/]\n"+GNOME_TEXT_SCALE_FACTOR+"={}".format(text_scale_factor).encode('utf-8'))" is a bit magic to me [16:21] seb128: mh, yeah... I must have been too tired when I wrote it.. I probably committed it when not finished -_- [16:21] seb128: let me fix it [16:21] damn it [16:21] Trevinho, that landing in trusty so you need a new merge request [16:21] and when I tried it worked only because thanks to unity [16:21] mh, ok [16:22] landed* === mjohnson15_2 is now known as mjohnson15 [16:47] seb128: http://people.canonical.com/~laney/package-junkyard/ please to test [16:48] Laney, you want me to install junk?! ;-) [16:48] sure, doing that in a minute [16:49] there's always some gems in the heap [16:49] we have a whole tv show about that, "scrapheap challenge" [16:59] Laney, seems to work fine and fixes the segfault [16:59] nice [16:59] a correct theory then [17:00] excellent [17:00] if you still feel like fixing mores issues there, there are the missing translations and the whoopsie unlock control not behaving as it should [17:02] yeah, looking at the lock one [17:02] why is bzr randomly different from the packagE? [17:02] like someone imported the quilt packages there but they don't actually apply in the branch [17:03] I'm unsure, that project is a mess atm [17:03] Seif started it iirc, but they stopped being active [17:03] and it has the generated c code in bzr [17:04] weird stuff [17:04] robert_ancell has been fixing a bunch of issues and mps fixes but they are still waiting for review I think [17:04] we should probably take over it and put it under CI/train landing, etc [17:05] Laney, https://code.launchpad.net/activity-log-manager/+activereviews ... [17:05] seems like robert_ancell tried to get it under CI [17:07] happyaron, hey, could you review https://bugs.launchpad.net/indicator-keyboard/+bug/1290881 ? [17:07] Launchpad bug 1290881 in indicator-keyboard (Ubuntu) "indicator-keyboard suppresses IBus candidate window" [High,In progress] [17:07] mmm [17:07] happyaron, there is an ibus patch in there, I assigned it to you some days ago, that's blocking the indicator work === maclin__ is now known as maclin [17:18] seb128, can you remind me how i find what crash reports have been submitted from a particular machine? [17:18] (i'm sure you've told me before) [17:20] chrisccoulson, you can find the link in system-setting->security&privacy [17:20] chrisccoulson, or webbrowser 'http://errors.ubuntu.com/user/'$(printf $(sudo cat /sys/class/dmi/id/product_uuid) | sha512sum) === Fudus2 is now known as Fudus [17:26] seb128, thanks [17:29] chrisccoulson, yw! === alan_g is now known as alan_g|afk [18:00] has anyone on the desktop team got an account on bugzilla.mozilla.org? [18:01] that is the most trappy question I have ever heard :P [18:01] HERE, HAVE FIREFOX, LALALA [18:01] (no I don't, sorry) [18:02] :( [18:03] oh, seb128 is already offline [18:03] in that case, he volunteered [18:03] hah [18:03] he has one [18:03] using his debian email address [18:03] win === alan_g|afk is now known as alan_g [18:08] too bad robert_ancel isn't here [18:09] I just used simple scan, I am always impressed how well it works and how easy it is [18:28] It seems that rhythmbox now has a rhythmbox-plugins that includes the MPRIS plugin. [18:28] Is that installed by default? [18:28] It kinda breaks the sound menu to not have it installed [18:33] tedg, apt-cache show rhythmbox-plugins|grep ^Task [18:34] (no idea if MPRIS is enabled at package build time though) [18:34] ogra_, Ah, cool. I didn't know that info was there! [18:35] I think I must have just not gotten it on upgrade then. [18:35] there is also the "seeded-in-ubuntu" command [18:35] (but i'm to oldscool for that :P ) [18:36] Interesting though, seeded-in-ubuntu seems to only work with source package names? [18:36] Seems a bit odd. [18:38] ogra_, Hmm, they disagree. There is a task for ubuntu-desktop but seeded-in-ubuntu doesn't say that the plugins is in ubuntu. [18:38] it is in the ubuntu-desktop tsak, which we install during image creation [18:38] *task [18:40] ogra_, K, kinda curious what seeded-in-ubuntu uses then. [18:41] the manpage might know === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha [20:51] jasoncwarner, should be mark bug 1281058 as "ubuntu-desktop-trusty" [20:51] Launchpad bug 1281058 in systemd (Ubuntu) "The system shutdowns when multiple accounts are open" [Low,Triaged] https://launchpad.net/bugs/1281058 [21:27] robert_ancell: hmm, now i got a log for that bug i was telling you about (the greeter not exiting as it should), but i dont see anything strange/telling there, would you mind to take a quick peek? (https://bugs.launchpad.net/ubuntu/+source/lightdm-gtk-greeter/+bug/1290575) [21:27] Launchpad bug 1290575 in LightDM GTK+ Greeter "lightdm-gtk-greeter does not exit cleanly when logging in" [High,Confirmed] [21:28] ochosi, looking [21:28] ty! [21:30] ochosi, I'm guessing they edited the log files - I was wondering what weird bug they had that used "" as username :) [21:31] hehe [21:31] yeah, i'd presume so [21:32] ochosi, is the greeter that is left around interactive? [21:32] robert_ancell: from my single experience it's only shown in the taskbar like a minimized window [21:33] robert_ancell: i was able to right-click and close it and then everything worked fine, apparently not everyone is so lucky [21:33] robert_ancell: and i guess i should add that i don't use desktop-icons, so it might well be that the greeter overlays those [21:33] ochosi, ah, I think I know what it is [21:34] * ochosi feels more hopeful all of a sudden [21:34] ochosi, it might be due to XSetCloseDownMode (display, RetainPermanent); which means the X windows stay open after the greeter quits [21:34] This is so the background is shown before the session starts (otherwise it would revert to the default X background) [21:34] robert_ancell: yes, otherwise you get flickering on login, no? [21:34] yeah [21:35] I suspect the greeter is not closing the existing window / redrawing the background correctly [21:35] And that might be due to gtk_main_quit not working properly [21:35] ok [21:35] that makes total sense [21:35] so adding the g_unix_signal_add would tackle that most likely [21:35] yes [21:35] very very good [21:35] thank you very much! [21:35] np [21:36] Don't we all love X's rough edges :) [21:36] hehe === ajmitch_ is now known as ajmitch