=== beidl_ is now known as beidl === oCrazyLemn is now known as CrazyLemon === maclin_ is now known as maclin [08:01] good morning desktopers [08:02] hey there [08:03] Laney, hey, how are you? [08:03] i'm good [08:03] friend just dropped in for a "surprise" breakfast with me [08:04] aka eat all my food before taking his flight back to grenoble [08:04] can't trust the french... [08:04] did he bring croissants and stuff? ;-) [08:04] haha [08:04] lol [08:04] sounds like vuntz [08:04] nothing [08:05] also on the topic of the French, I saw this yesterday: http://www.theguardian.com/money/shortcuts/2014/apr/09/french-6pm-labour-agreement-work-emails-out-of-office [08:05] you guys have it good [08:05] yeah [08:07] so you guys stop IRC pinging after 6pm! [08:07] didrocks, ^ \o/ [08:08] agreed! [08:08] I'll start finishing at 6pm from today! [08:10] hey seb128, Laney didrocks [08:10] Laney, btw Trevinho apparently found the issue with the lockscreen [08:10] darkxst, hey [08:10] seb128: nice [08:10] what was it? [08:10] hey darkxst! [08:11] Laney, an unitialized variable [08:11] hey darkxst, wie gehts? [08:11] HA [08:11] my favourite [08:12] Laney, good, why the german! [08:12] multiculture [08:12] fighting back against the french [08:12] how do I upload an updated our seed? [08:13] (read my earlier lines) [08:13] Laney, I see [08:13] push to the branch [08:13] mvo, hey! today is supposed to be release freeze, any chance you have a look to that polkit exception update-manager issue? [08:13] apt-get source ubuntu-gnome-meta; cd ubuntu-gnome-meta*; ./update.sh(?); debuild -S; debdiff; sbuild; test; upload [08:13] Laney: the battle is lost, don't fight, but embrace it! [08:13] mvo, if not I'm going to try having another go to understand the magic in that code [08:13] :) [08:14] alright alright, pass me a chocolate croissant [08:14] nooooooooo [08:14] stop this threat! [08:14] Laney, thanks, got it! [08:14] :) [08:15] darkxst: basically the meta package pulls from the seed branch and runs a program on it to calculate the dependencies for you [08:15] so it's a two step process to get it updated in the archive [08:16] Laney, right, I new the meta package was auto-generated, just didnt know how! [08:16] * Laney nod [08:23] seb128: does the locking timeout fix work for you? [08:23] oh crap [08:24] Laney, dunno, Trevinho just went to bed I think and his branch merge conflicts :/ [08:24] I have a "can't type on the lockscreen" bug [08:24] he had other fixes duriong the night [08:24] focus issue? [08:24] seb128: ok, sorry for that, once I fnished wrestling with apt I will look at this [08:24] or no password prompt [08:24] mvo, thanks [08:24] I think a password prompt had a grab [08:24] then it timed out and locked the screen [08:26] interesting bug [08:27] mvo, https://code.launchpad.net/~brian-murray/aptdaemon/bug-1266844/+merge/207276 as well please, still topping the trusty errors :/ [08:27] mvo, sorry for being nagging about those [08:28] can I unlock it somehow or do I need to restart the session? [08:30] Laney, can you use the session indicator to go the greeter and log back from there [08:30] no, same issue with mouse [08:30] Laney, go to a vt and kill the password prompt that has the lock? [08:30] yeah... what would that be? [08:31] it's a 'please give your google password' one that I get at every login [08:31] indicator-datetime-service? [08:31] or e-d-s [08:31] evolution-calendar [08:32] otherwise I get you can try to kill unity I guess [08:32] I got it, was an e-d-s thing, I think evolution-source-registry [08:32] k [08:33] can you open a bug about it please? [08:33] "it" being "shouldn't lock when a grab is active" [08:33] ya [08:33] thanks [08:33] is that what g-s does? [08:34] yeah, it doesn't lock if it can't get a grab [08:35] Laney, that's the reason for the famous bug #49579 [08:35] Launchpad bug 49579 in gnome-screensaver (Ubuntu) "screen doesn't lock when some menu is open" [Low,Triaged] https://launchpad.net/bugs/49579 [08:36] ah yes, that one === vrruiz_ is now known as rvr [09:01] bregma, I've got a fix for the launcher suddenly appearing (launcher was force-hidden). should I base the fix on my alt-tab-viewport-bias patch or should I base it on trunk? [09:35] seb128: bug #1202754 is the one, right? [09:35] Launchpad bug 1202754 in update-manager (Ubuntu Saucy) "update-manager crashed with SystemExit in exit(): 0" [High,Confirmed] https://launchpad.net/bugs/1202754 [09:35] mvo, yes [09:35] mvo, easy to reproduce, get an update that prompt you for a password (e.g new kernel) and wait on the prompt or deny it === oCrazyLemn is now known as CrazyLemon [10:03] mvo, sorry, got sidetracked but I had an extra comments, there are probably different issues/changes there [10:04] mvo, the bugfix is to handle the denied case without triggering an exception, but we should also change the timeout delay [10:04] mvo, e.g it doesn't make sense to deny the update after 30?s of prompt waiting === vrruiz_ is now known as rvr [10:44] seb128: could you please test http://paste.ubuntu.com/7230205/ ? [10:45] seb128: this should fix the issue, but I get some other strange behivor around the polikit stuff that is probably some local problem. but I think its the fix [10:45] seb128: I'm not sure the error handling is correct that leads to the window_main.exit(), but that is a entirely different issue :) [10:46] mvo, I can test next time I've the situation where update-manager prompts me [10:46] mvo, does that mean update-manager exit? [10:47] mvo, that also doesn't fix the problem that if you walk away for 5 minutes and come back you should still be able to enter your password, rather than finding things closed/timeouted right? [10:48] seb128: I think you enter your password and then update-manager exits :) but no crashreport anymore [10:48] seb128: this is what I meant with "error handling not quite correct" [10:48] k [10:48] do you know how to force update-manager to prompt for auth? [10:48] or how to create a situation that leads to that codepath [10:48] seb128: but that is something that mterry may know about maybe, looks like he added the error cases [10:49] seb128: give me a sec to commit a branch [10:49] seb128: I guess you could change the defaults for the desktop permissions for polkit too [10:49] mvo, do you know where the polkit auth is done? aptdaemon? [10:49] seb128: but I forgot where the file is [10:49] mvo, changing the default timeout value for the prompt should be a one liner, I just didn't manage to find where that code is [10:49] seb128: yeah, aptdaemon asks polkit [10:50] seb128: oh, there is a timeout? that is good to know [10:51] beidl, please base on trunk unless you're modifying the same code and there are conflicts [10:51] bregma, already done, based on trunk and proposed [10:52] seb128: do you have a example for the timeout? [10:52] seb128: I mean, some example code or docs how the polkit timeout is increased? [10:53] mvo, let me look [10:55] seb128: lp:~mvo/update-manager/lp1202754 [10:56] seb128: then you can run PYTHONPATH=. python3 UpdateManager/backend/InstallBackendAptdaemon.py [10:56] seb128: which will fail for me, but this failure is no longer causing a crash [10:57] mvo, let me try that [10:58] mvo, I might be wrong for the timeout thing, can't find any reference for it ... I wonder how other prompts are handled [10:58] seb128: polkit auth is currently entirely broken for me [10:58] seb128: lets talk after lunch [10:58] mvo, like I'm trying with unity-control-center, I went to the user panel and did "unlock", the dialog stays open without hitting an error [10:59] entered a password, worked fine [11:00] seb128: yeah, I check in a wee bit [11:01] seb128: after lunch, one step at a time :) [11:01] mvo, [11:01] UpdateManager/backend/InstallBackendAptdaemon.py [11:01] Traceback (most recent call last): [11:01] File "UpdateManager/backend/InstallBackendAptdaemon.py", line 238, in [11:01] b = InstallBackendAptdaemon(None) [11:01] TypeError: __init__() missing 1 required positional argument: 'action' [11:02] mvo, lunch sounds like a good idea to me as well ;-) [11:04] mvo, that branch also seems to not have recent commits? like the most recent one is the archive upload you did on monday [11:19] dpm, ping? [11:22] tseliot: kgunn, RAOF suggested that you might have info on switching from nouveau to intel graphics on my macbook pro w/ ubuntu 14.04. Any pointers? [11:23] camako: if you install the nvidia binary driver, you can easily switch back and forth using a UI [11:24] (nvidia-settings) [11:24] I'm not sure about nouvea [11:24] *nouveau [11:24] * tseliot -> lunch [11:24] tseliot: oh...one secret, camako will be on the mir team [11:24] so no good with proprietary driver atm [11:24] and...well, nouveau not so hot [11:24] since he has an integrated intel gpu...just easier to turn off [11:25] kgunn, camako: ok, then I guess the vga switcheroo is your best bet to disable nvidia [11:25] kgunn, camako: https://help.ubuntu.com/community/HybridGraphics#Using_vga_switcheroo [11:26] mhr3_, I'm about to go, would you mind sending an e-mail and I'll reply? Or if it's quick, I'll be around for the next 5 mins [11:26] dpm, ok to do a quick hangout when you're back? [11:27] mhr3_, I'll be away for the afternoon and I won't be back probably until 20:00, which might be a bit late. Would tomorrow early morning work? [11:27] tseliot: Cool.. Let me dig into that... [11:27] dpm, nah, i'm on holiday tomorrow :) i'll send a mail then [11:28] mhr3_, ok, cool :) === dpm is now known as dpm-afk [11:29] tjaalton: mlankhorst ...would you guys happen to know how to disable the nvidia gpu altogether on a macbook and use solely the intel integrated gpu ? [11:29] camako tried switcheroo already and it didn't work.... [11:30] ( trying to let tseliot go to lunch :) [11:30] camako: maybe share how switcheroo didn't work ? [11:31] kgunn: I entered the commands on that link but saw no difference === MacSlow is now known as MacSlow|lunch [11:37] seb128: pushed now, sorry [11:38] mvo, k [11:38] kgunn: no idea about nvidia gpu, but on my macbook killing lightdm/plymouth and doing echo IGD > /sys/kernel/debug/vgaswitcheroo/switch is enough.. [11:38] cat that file to see for sure [11:40] seb128: I have the prompt now, do I just need to wait for the timeout? [11:40] mvo, yes, like less than 1 minute [11:40] seb128: cool [11:40] and here we go [11:41] mvo, I can confirm that the deny prompt case doesn't trigger apport with that update [11:41] ! [11:42] thats a start isn't it [11:42] yeah [11:43] mvo, so yeah, with your fix, update-manager silently exit [11:44] after 30s timeout [11:48] seb128: phonecall, bbiab [11:48] mvo, k [12:06] Laney: your changelog says "* debian/patches/window-height-700px: Set window height to 700px; 900 is too [12:06] + small and causes the window to start off screen." [12:06] but if 900 is too small how can 700 be better? [12:06] big [12:07] seb128: hi, I don't know why it merge conflicts... If I take the last brancfh of the queue here and merge on it all the others, it just goes well [12:07] big, small, so easy to get confused :) [12:07] seb128: try to put them in revers order if you can [12:07] Laney: want to fix it or shall I accept? [12:07] Trevinho, hey, it says "criss-cross merge", which is usually bzr slang for "things are weird" [12:08] seb128: haven't you able to compile it anyway, right? [12:08] Riddell: it's up to you [12:08] Laney: accepted! [12:08] Trevinho, the morning has been crazy enough that I didn't try a local build, do you want me to try? [12:08] seb128: mh, yeah I know... but it's strange lp:~3v1n0/unity/lockscreen-prompt-activation basically contains all the others [12:09] merci [12:09] seb128: I don't want you to lose much time, let's see if using the inverse order is fine [12:09] Trevinho, did you mark the other ones as pre-required? [12:09] seb128: I couldn't mark two as prereq [12:10] Trevinho, you could chain those ;-) [12:10] seb128: as the last one has both the osk and sjakthol as prereq [12:10] e.g mark osk as prerequire of the fix and mark the fix as prerequire of your update [12:11] I think you have to actually merge them into each other to do that [12:11] or should, at least [12:11] yeah, and its done into lp:~3v1n0/unity/lockscreen-prompt-activation [12:11] it contains everything [12:11] so if you use it as base and then you merge on it it will work [12:11] at least locally it works [12:12] seb128: http://pastebin.ubuntu.com/7230493/ [12:12] err http://pastebin.ubuntu.com/7230493/ [12:12] uff, http://pastebin.ubuntu.com/7230494/ [12:14] seb128, larsu: what do you say about adding also https://code.launchpad.net/~3v1n0/indicator-session/unity-session/+merge/215076 ? [12:18] hi seb128, can you help me land some of my MPs? [12:18] uff, launchpad just hangs here [12:18] attente, hey, sure, which one? I put the "don't show keyboard properties entry on the lockscreen" one in the landings earlier [12:18] and can't push [12:19] Trevinho, larsu is out for lunch, let's wait for him to be back [12:19] seb128: ok [12:19] seb128: is that the switch layout on lock one? https://code.launchpad.net/~attente/indicator-keyboard/switch-layout-on-lock [12:20] attente, no, https://code.launchpad.net/~attente/indicator-keyboard/1291962-2/+merge/213346 [12:21] attente, which other ones do you need? [12:21] lp seems under attack, damn it! :) [12:21] seb128: https://code.launchpad.net/~attente/indicator-keyboard/switch-layout-on-lock/+merge/213737 [12:22] seb128: https://code.launchpad.net/~attente/ubuntu-themes/update-indicator-keyboard-icons/+merge/213359 [12:22] seb128: https://code.launchpad.net/~attente/gnome-control-center/upstream-xkb-option/+merge/214538 [12:22] attente, g-c-c ... do we need that patch in u-c-c as well? [12:22] seb128: no, just g-c-c [12:22] k [12:23] is our input switcher/indicator even used on g-c-c sessions in trusty? [12:23] seb128: not at all, it's using upstream gnome's [12:24] k [12:24] seb128: i'm also trying to fix the sound buttons and input switching on the lock screen, not sure if i can make the freeze though... [12:24] attente, ok, so indicator-keyboard ... the one I have is good to land to right? [12:25] attente, don't worry about those, that seems fine to SRU [12:25] attente, we are going to have a stack of bugfixes that are not going to make it for release and than we are going to land as SRUs for .1 [12:25] oh ok [12:26] like multimedia keys on the lockscreen never worked [12:26] so that's a nice bonus, not an important issue for release [12:26] multimedia keyts.... on my laptop the +vol -vol change resolutions, also! [12:26] changing the layout is easily done with the indicator as well [12:26] seb128: nobody's annoyed about not turning off the sound while locked? :P [12:27] I mean, they get interpreted like the SUper+p or something similar [12:27] attente, did we use to do that? [12:27] Trevinho, weird [12:28] seb128: i don't know. i might be overestimating the importance of that working though [12:29] attente: I think they actually are quite important === MacSlow|lunch is now known as MacSlow [12:30] Trevinho, btw Laney found an important issue with the lock screen this morning, not sure if you saw [12:31] Trevinho: ok, let me deal with it. not sure if i can fix it for freeze, but .1 should be fine [12:31] Trevinho, if the lock timeout is reached while some is grabbing the input (e.g a keyring prompt), the screens locks but the input is still blocked/redirected and you can't type your password to unlock [12:32] Trevinho, https://bugs.launchpad.net/unity/+bug/1305586 [12:32] Launchpad bug 1305586 in Unity "Lock screen is unusable when a client has a keyboard/mouse grab" [Undecided,New] [12:34] seb128: let me retry it should be fixed now... and when there's grab, we can't do anything but not locking [12:34] Trevinho, right, which is what g-s was doing [12:34] Trevinho: creating a proxy seems to be a bit overkill to call these two methods... [12:34] that's not great but better than locking without giving a way to unlock later [12:34] seb128: we have some few cases... at least we blank the screen [12:35] larsu: yaeah for now, but I'd like to add also for reboot and restart soon, to popup the right dialog [12:35] larsu: and well also gs has 2 methods (1 really used) [12:35] Trevinho: I'd still just use g_dbus_connection_call(). I won't block the merge on that issue though [12:36] larsu: but I need to check if it's there, doing async calls [12:36] larsu: otherwise I need to move to sync [12:36] Trevinho: but please use an if/else condition in lock_current_session() instead of returning early [12:36] larsu: ok [12:37] Trevinho: yeah that's a good point - this way you only need to check on startup whether the session name exists [12:39] larsu: pushed the fix [12:39] seb128: do you know what I can use to try that Laney issue? [12:40] seb128: as I've mostly used menus for now [12:40] Trevinho: looks good to me now, but I'll leave the top-approving to charles (if only so that he knows what's going on)( [12:40] Trevinho, set the idle/lock timeout low, ssh something, when gnome-keyring prompt for the key, wait [12:41] mh, I've saved my pwd... :P [12:42] Trevinho, kill gnome-keyring-daemon [12:42] restart it [12:42] export the right env when you run ssh [12:42] if you run it by hand it prints the env on stdout [12:45] Trevinho, seems better with the new branches order, https://ci-train.ubuntu.com/job/landing-006-1-build/17/console [12:45] Trevinho, build ongoing for silo 006 [12:45] seb128: cool [12:47] Oh, how I love running alphas and betas. Each and every apt-get upgrade is so fulfilling! I <3 you guys. [12:48] BigWhale, thanks ;-) [12:48] :) [12:50] seb128: screen is grabbed, lock starts but never actually locks... after some time the user go back closes the reason why the screen was not locked: should we lock at that point? [12:50] seb128: as it's what we're doing [12:50] Trevinho, no strong opinion on that [12:50] either way can be seen as buggy [12:50] yeah, both cases have troubles [12:51] doing extra locking rather than not enough seems on the safe side [12:51] seb128: mh, I've tried the ssh dfialog, but for some reason compiz doesn't show it as grabbing [12:52] seb128, you hit build on the lockscreen silo? [12:52] do you get the keyring one [12:52] bregma, yes, reconfigured and tried again, as said earlier this morning the branches had merge conflicts [12:52] bregma, I put Trevinho's one as first this time, seemed to work [12:52] bregma, it's building in https://launchpad.net/~ci-train-ppa-service/+archive/landing-006 [12:52] seb128, yes, I'm just making sure things are moving along because we have a backlog of unity bugfixes to land too [12:53] bregma, right, I'm with you there, we got the compiz/bamf/nux fixes from yesterday in trusty/merged in [12:53] so things are moving ;-) [12:53] i have no problem with you hitting build, I just want to make sure it got done [13:05] pitti, hey [13:05] ça va seb128 [13:05] pitti, ça va bien, et toi ? [13:06] seb128: je vais bien aussi, merci [13:07] pitti, do you know about any recent langpack issue? [13:07] $ dpkg -L language-pack-gnome-fr | grep gedit [13:07] $ [13:07] $ dpkg -L language-pack-gnome-de | grep gedit [13:07] $ [13:07] seb128: no, not really; except for the botched rebuild for beta-2, which got replaced with a new set [13:07] translations vanished from my eog or evince or ... [13:07] seb128: yes, they are in -base [13:08] seb128: we refreshed -base for b2 [13:08] hym [13:08] seb128: tomorrow we'll get fresh ones for the final, BTW [13:08] hum [13:08] $ dpkg -S gedit.mo [13:08] language-pack-gnome-en: /usr/share/locale-langpack/en@shaw/LC_MESSAGES/gedit.mo [13:08] language-pack-gnome-en: /usr/share/locale-langpack/en_AU/LC_MESSAGES/gedit.mo [13:08] language-pack-gnome-en: /usr/share/locale-langpack/en_CA/LC_MESSAGES/gedit.mo [13:08] language-pack-gnome-en: /usr/share/locale-langpack/en_GB/LC_MESSAGES/gedit.mo [13:08] $ [13:08] is that a case of "replaces got used, things got upgraded, and files are missing"? [13:08] * seb128 tries a --reinstall [13:09] seb128: actually, I suspect that my approach of rebuilding the -base packages without a -base export from LP was somehow botched [13:09] the delta might not have applied exactly to what I used as full export [13:09] pitti, works after a --reinstall [13:09] curious [13:09] I guess Replaces: and upgrade order bitting [13:10] I think we had issues like that in the past [13:10] language-pack-gnome-fr Replaces language-pack-gnome-fr-base [13:10] so if you get base first [13:10] then install a version of language-pack-gnome-fr that ships the mo [13:10] it takes over it [13:10] then upgrade language-pack-gnome-fr [13:10] you end up without the files [13:11] pitti, sorry for the noise, I guess it's going to be resolved for everyone with the next base update [13:11] (didrocks was having the same issue) [13:12] seb128: as for that dialog it seems to do something weird, as at x level we get the grab event but not the ungrab one :o [13:12] seb128: yes, it will be, but more or less just "by accident" [13:12] seb128: so we indeed need to think about the more fundamental problem; probably by extending our eglibc patch to look in two dirs instead of one, and preferring an -updates dir [13:12] so that we can do without the bidir replaces [13:13] that'll waste hd space again, though [13:13] (and space on images with non-empty update packs) [13:13] need to think this through more thoroughly [13:13] perhaps we can force an order with conflicts: or so [13:13] pitti, the other solution would be to force the unpack order through pre-depends or something [13:14] yeah [13:14] so -base pre-depends: update pack (>= $base_version) [13:15] so that the update pack is always unpacked first [13:15] yes [13:15] we'll probably confuse apt a lot with a gazillion pre-depends :) [13:15] better to check with mvo though [13:15] could be yes [13:15] so, it's not something which I'd like to change at this point, perhaps for the first SRUed langpacks and in U [13:24] seb128: I think I have polkit prompt vanishing under control now, needs a bit of cleanup though [13:26] mvo, great! [13:44] seb128: is 5min acceptable? [13:45] mvo, do we need to pick a time, what's the issue with no timeout? [13:45] mvo, with the "I walk away for lunch and come back to find things in a weird state" scenario, 5 minutes would still be an issue [13:46] seb128: there is code in aptdaemon to prevent a "denial-of-service", i.e. if something just creates transactions and then does nothing with them. but I guess we could relax this [13:46] seb128: lunch is a good point [13:48] mvo, how are other softwares handling that DoS case? e.g software-properties [13:49] mvo, I'm unsure how the timeout prevents a DoS, you could DoS with requests in less than 5 minutes anyway? [13:50] seb128: s-p does not create transactions, its just providing a method call interface [13:50] seb128: but I guess the risk of DoS is pretty small [13:51] mvo, well, if you want to keep that, set it to 30 minutes at least [13:51] ok [13:52] mvo, ideally when the timeout kicks you would come back on an error messages telling you why the update didn't happen [13:52] mvo, but we can do that later [13:53] seb128: yeah, lets fix the immediate bug and then we can tweak it, unfortunately there is some more on my list :/ === Ursinha is now known as Ursinha-afk [13:54] mlankhorst: Screen goes blank when I enter "echo IGD > /sys/kernel/debug/vgaswitcheroo/switch"... Any idea how to debug that? [13:59] camako: ssh ;-) [13:59] mvo, yeah, no worry, I think fixing the exception on auth denied + increasing the timeout is going to already be a good improvement/fixing most of the cases [14:05] is it just me, or is rhythmbox *really* crashy with mtp? [14:06] I don't use a mtp device, I didn't see high ranked issues on e.u.c [14:06] could be device specific? [14:06] doesn't seem to be [14:07] do you have a stacktrace/bug reference? [14:07] https://bugs.launchpad.net/bugs/1304543 [14:07] dobey: Error: launchpad bug 1304543 not found [14:09] Sarvatt: did you try new unity? [14:10] Trevinho, bregma: +1 from me to land silo 006, my issues of screen turned off are resolved, double locking as well [14:10] yay [14:10] seb128: cool [14:10] seb128, OK, I'm just finishing up my retesting here [14:11] Laney: I'm also working on the grab for that dialog [14:11] seb128: did you try going back to g-s? [14:11] bregma, let me know when I should press the publish button [14:11] the one time I don;t is when something goes wrong [14:11] Laney, how do I do that? [14:11] seb128: enable osk [14:11] yay [14:11] seb128: or reader [14:12] Laney: I'm also working on that grabbing dialog, but that guy seems quite annoying, since he doesn't notify compiz that is taking the grab, for some reason [14:12] Trevinho, Laney: works fine, enable osk in the settings, ctrl-alt-L -> g-s, disabled it back, ctrl-alt-L -> unity lock [14:12] great [14:12] \o/ [14:12] Trevinho, good job (as usual) ;-) [14:12] Trevinho: what guy? [14:12] andyrock: done the most, to be fair [14:13] too kind [14:13] andyrock: I've worked on his codebase, cleaning things [14:13] Laney: the gnome keyring dialog [14:13] andyrock, Trevinho: good team work guys ;-) [14:13] ah, dunno what you should do [14:13] yeah i had to switch on other bugs and had to stop working more that 4 hours a day [14:13] see how gnome-screensaver handles it [14:14] seb128: especially when you think that most has been done on busses and planes... :D [14:14] university exams :/ [14:14] Trevinho, that's quality programming time: no interruptions [14:14] Laney: I mean, I can check if we have a grab, but only trying to grab... [14:15] seb128, double locking is gone too? \o/ [14:15] Trevinho, offline work can be productive, less ping crazyness ;-) [14:15] andyrock, yes [14:16] nice [14:16] bregma: yaeh, indeed... unless when you figure out that your girlfriend is looking at at you with crazy eyes, since few minutes, and the queue to get into the plane is going to be cosed in minutes; but you still have to push a fix :D [14:17] Trevinho: yeah so it tries to do the grab then doesn't lock if that isn't successful [14:18] Laney: yes, I can do that, and probably it should be at general level (since also the dash would try to show otherwise), but I'm figuring out the cleanest way [14:19] l3on: what was the reason again for abandoning the theme fix and make a gtk patch instead? [14:19] l3on: for the resize issue [14:20] larsu, is it regarding gtk+3.0 patch ? [14:20] ya [14:20] well... in gtk happens the follow: [14:23] when you use set_titlebar function on gtk_window, gtk itself try to check if screen can handl with __GTK_FRAME_EXTENTS (see function gtkwindow.c:gdk_window_supports_csd) if it return FALSE gtk set this variable priv->custom_title = TRUE; [14:23] when it tries to decor windows does the follow: [14:23] else if (priv->custom_title) [14:23] gdk_window_set_decorations (gdk_window, GDK_DECOR_BORDER); [14:24] which means: window is going to be decorated only with borders by WM [14:24] that sounds exactly right to me [14:24] as gtk will put a close button in the header bar [14:25] exaclty [14:25] but some WM (like compiz and fluxbox) can't resize window [14:26] they can't resize the window if it is only decorated with borders?! [14:26] it looks like if titlebar is missing, those MW are unable to understand how to resize them... I had took a look at compiz and trusty code, but I did not understand how to enable __GTK_FRAME_EXTENTS [14:27] this is the patch applied in clutter: [14:27] s/clutter/mutter [14:27] https://mail.gnome.org/archives/commits-list/2013-August/msg03783.html [14:28] l3on: mh, I thought this is for the case when frame extents is _not_ available [14:28] which seems really simple.. If I receive a __GTK_FRAME_EXTENTS atom, I just set properly windows borders [14:29] but without any documentation on my hand I was unable to understand how unity works behind scenes... so I gave up and patched GTK [14:29] which, to my point of view, is making things in the wrong way (why only borders if __GTK_FRAME_EXTENTS is not supported?) === Ursinha-afk is now known as Ursinha [14:30] l3on: you need borders if frame extents is not supported. You don't need a title bar though [14:30] as far as I know.... [14:30] but apparently those window managers are buggy even then? [14:30] Trevinho: ^ ? [14:31] larsu, gtk puts resize stuffs in a ".window-frame margins" which are some "magin and invisible" borders for the window. If WM can't handle with those, you cannot resize window... [14:32] mh, I didn't check that issue in full details yet, but I believe we can easily check if a __GTK_FRAME_EXTENTS atom is set, other than motif ones... [14:32] see Adwaita css code con .window-frame [14:32] l3on: the css shouldn't influence borders added by the wm [14:33] Trevinho: it's freeze day.. [14:33] I asked about this a couple of weeks ago but you were too busy :) [14:33] larsu: I know, and a full of stack of fixes to do [14:33] larsu: yep, and still I am :°( [14:33] seb128: new aptdaemon uploaded [14:33] larsu: we'll handle that on a SRU, I suppose [14:33] http://git-man-page-generator.lokaltog.net/ [14:33] mvo, \o/ [14:34] Trevinho: if we do that, then I won't merge l3on gtk patch [14:34] in somehow is wrong, gtk-3 theme should set a margin on .window-frame (this is way Ambiance does not work also in Gnome-Shell) [14:34] seb128: are you okay with that? ^^ [14:34] larsu: I mean, we can merge it for now [14:34] larsu: if we want to ship a more usable state [14:35] l3on: Ambiance is broken in shell anyway. I'd be fine merging your shadow branch though if that fixes things [14:35] Trevinho: it will be insanely ugly, as you'll have a titlebar on top of the header bar [14:35] seb128: actually, please reject, I upload a new with https://code.launchpad.net/~brian-murray/aptdaemon/bug-1266844/+merge/207276 [14:35] larsu: ah, yeah... [14:36] larsu, I tried to stay away from that those CSD discussions since I don't know enough about the topic ... can you summarize the sugestion? [14:36] mvo, ok [14:36] seb128: Trevinho suggests adding frame extents support to compiz as an sRU [14:37] seb128: I'd hold off from taking l3on's gtk patch in that case [14:37] larsu, how much code change are we talking about? [14:37] seb128: uploaded again, this time with both fixes [14:37] seb128, my lockscreen-landing testing has passed sanity checks, I've marked the silo as Testing Done, can you push Publish or shall I do the usual request on #ubuntu-ci-eng? [14:37] bregma, I'm doing it [14:37] seb128: I've not checked that yet, but if it's just about checkin an atom, and updating compiz, it shouldn't be that match [14:37] be carefull: __NET_FRAME_EXTENTS and __GTK_FRAME_EXTENTS are two different atoms [14:37] seb128: Trevinho knows better than me, but from what he said earlier it doesn't seem to be a lot [14:38] if compiz already support the first one (which from seems to) it should be easy add the gtk related extents [14:38] * bregma goes off to prepare for the next Unity landing [14:38] larsu, Trevinho: in any case etoomuch for release today, so GTK or compiz, let's revisit as a SRU next week [14:38] or when higher importance issues are resolved [14:38] seb128: makes sense. Sorry for the additional ping.- [14:38] mvo, rejected [14:40] l3on: I don't have programs here exporting that atom btw [14:40] tweak tool doesn't seem to have it [14:40] it does [14:41] anyway for testing purpose: http://python-gtk-3-tutorial.readthedocs.org/en/latest/layout.html#id5 [14:41] just a window using set_titlebar [14:42] xprop on it returns me no atom about frames... [14:45] bregma, ok, lock screen in unapproved, up to Laney&co next [14:55] seb128, Trevinho, i've got this working decently ok: https://code.launchpad.net/~attente/unity/enable-sound-buttons-on-lock-screen/+merge/215194 [14:56] attente, nice [14:57] the two main problems is that the sound notify osd bubbles don't appear and modifier-only input switching doesn't quite work perfect [14:59] kenvandine, larsu: do you know about https://jenkins.qa.ubuntu.com/job/autopilot-testrunner-otto-trusty/4101/console? [14:59] "/var/local/autopilot/autopilot.log: /usr/bin/system-settings: symbol lookup error: /usr/lib/x86_64-linux-gnu/qt5/qml/GSettings.1.0/libGSettingsQmlPlugin.so: undefined symbol: _ZN10QGSettings17isSchemaInstalledERK10QByteArray [14:59] /var/local/autopilot/autopilot.log: }}}" [14:59] ugh, no.... [15:00] seb128: mhr3 added isSchemaInstalled the other day [15:01] did we update the bindings without the qt lib or something? [15:01] looks like the qml plugin links agains an older version of the qt lib? [15:01] I don't think so... [15:01] no mhr3 around [15:01] * larsu checks [15:02] seb128: the function's there [15:04] fginther, om26er: do you know what's up with the ubuntu-system-settings CI? [15:04] the otto logs are weird, they have dbus timeout and permission denied issues [15:05] seb128, is it normal that the network icon on the panel is duplicated? Or is it a bug? [15:05] xclaesse, that should be fixed in today's update [15:05] good [15:05] xclaesse, it's also because you got indicator-network (maybe unity8) installed [15:06] and I have to delock twice as well, that's known bug? [15:06] xclaesse, the fix just got uploaded, should be available later today [15:06] hmm [15:06] seb128, perfect, thanks :) [15:06] xclaesse, yw [15:06] there's certainly not a good enough dependency from qtdeclarative5-gsettings1.0 to libgsettings-qt1 [15:07] hum, shouldn't shlibs ensure that? [15:07] that were the 2 most visible issues I have in 14.04, since we are close to release I wanted to be sure that was known :) [15:07] btw, do we have a release day already? or is it still "some day in April" ? [15:07] xclaesse, did you install unity8/indicator-network on purpose/to test, or did that get pulled it for you (we had a few bugs where upgrades would pull those in) [15:07] xclaesse, you can probably uninstall indicator-network if you don't use it [15:07] seb128, AFAIK it got pulled [15:08] xclaesse, https://wiki.ubuntu.com/TrustyTahr/ReleaseSchedule [15:08] I upgraded to 14.04 a while ago [15:08] xclaesse, we have a release day since the start of the cycle [15:08] oh [15:08] ok, thanks :D [15:09] it probably should get symbols [15:09] xclaesse: indicator-network is pulled in by unity8 stuff, so if you've got unity8 installed, you'll have it as well as nm-applet [15:09] or a version at the very least [15:09] seb128, Good job with 14.04 btw, I'm using it since before beta1 and my system never broken completely. Pretty stable even while being on dev. :D [15:10] dobey, xclaesse: anyway the upstart job got fixed to not start the new indicator on unity7 (since it's not feature complete yet) [15:10] xclaesse, thanks ;-) [15:11] dobey, ok, let's do a "sudo apt-get autoremove unity8" [15:11] fginther, seb128 I saw that, it seems otto is acting up [15:11] (otto is the desktop test runner in the lab) [15:11] om26er, I know what otto is ;-) [15:11] the bus driver in the simpsons [15:12] THAT [15:12] the plane driver in "Airplane!" [15:13] bregma, is it desired to show all regular desktop apps under unity8 preview? [15:13] charles: I've fixed the tests, can you check this please https://code.launchpad.net/~3v1n0/indicator-session/unity-session/+merge/215076 ? [15:13] bregma, it's not like they worked under mir [15:13] seb128, oh and my wife is still on 12.04 LTS and didn't see why she would need to upgrade... until I told her that the volume can go over 100% with the keyboard in 14.04. Now she can't wait for the upgrade :D [15:13] mhr3: all the ones that should be shown; but if it's under mir, there will be problems, yes [15:14] little details matters... [15:14] mhr3: but it is a "preview" after all [15:14] xclaesse, that's making quite some people happy, including me (I use it often in hangouts ;-) [15:14] speakers are really crap on those laptops [15:14] I just tried the maximum volume on my speakers [15:14] was intense [15:15] mhr3, it's not desirable to show apps that won't run... but right now it also does not show apps thate *will* run [15:16] there some european laws that forbid devices to go over some dB, wondering if ubuntu allowing to go over 100% could infringe that... [15:17] bregma, so how is the scope supposed to know which ones to show [15:17] xclaesse, you could do that before, but only from gnome-control-center [15:18] bregma, it's not like all apps were tagged with SupportsMir=true [15:18] and it's far too late in the cycle to go through the entire archive and start adding it [15:18] Trevinho, sure [15:19] charles: thanks, so we can put in the last landing silos... :P [15:22] mhr3, if an app does not do the right thing, it's broken and I'm OK with that: we need to make sure 'the right thing' is clear in the docs and have a handy answer prepared for when folk complain their app doesn't who up [15:24] bregma, sorry lost context, last saw this [15:24] and it's far too late in the cycle to go through the entire archive and start adding it [15:24] hey guys! is it too late to push a last fix regarding gestures? [15:25] beidl, no, it's in the review queue, there's probably time to get it in still [15:25] bregma, great! a tiny little question before I push: does the 4 finger tap gesture to show the dash require that the last activated scope is shown? [15:27] I'm just asking because it might be required by design [15:27] beidl, no, I think it's supposed to show the "home scope", just like pressing the super key or the BFB icon [15:27] either choice is still better than nothing [15:27] as in, what we currently get, an empty dash [15:27] bregma, perfect, then I'll push it :D [15:29] Trevinho, nice work [15:29] approved [15:29] charles: thanks [15:29] bregma: ^ can we use our silo for that ? [15:32] Trevinho: do you have time to try out https://code.launchpad.net/~attente/unity/enable-sound-buttons-on-lock-screen/+merge/215194? [15:32] attente: I'm checking the code right now [15:32] attente: it looks good [15:32] Trevinho: oh, ok, thanks [15:32] Trevinho, yes, theoretically, since we don't have one assigned yet [15:33] bregma: then... https://code.launchpad.net/~3v1n0/indicator-session/unity-session/+merge/215076 :) [15:44] attente: if you can do a quick push, there are small things that it would be nice to have http://pastebin.ubuntu.com/7231289/ but if you can't in time, it's not mandatory [15:45] Trevinho: is it expensive to create dbusproxy objects every time the key is pressed? [15:46] attente, you want https://code.launchpad.net/~attente/unity/enable-sound-buttons-on-lock-screen/+merge/215194 in today? [15:46] attente: no, we're geenrally doing that quite often, when we don't need signals [15:46] bregma: I'm looking at it now [15:47] bregma: i'd like it in, but have no problem leaving it out if it's too risky [15:48] Trevinho: ok, i'll make the changes then [15:48] seb128, you're doing a separate landing for https://code.launchpad.net/~3v1n0/indicator-session/unity-session/+merge/215076 ? I'll remove it from the Unity one then [15:48] attente: ah, rebase your branch against last trunk [15:49] bregma, I just put it because that seemed a different unit set and I didn't know you were planning to include it, I'm happy to have it in the unity silo though [15:49] bregma, let me delete my line [15:50] * bregma doesn't want all the silos taken before he gets one [15:51] bregma, assignement is usually in order so you should be next === tkamppeter_ is now known as tkamppeter [15:55] attente: on test run, it works well... so... once you've cleaned things just ping me [15:57] Trevinho, bregma: didrocks just had a case of "input was going to the win under the lockscreen :/" [15:57] seb128: was that grabbing, or after ? [15:57] seb128: as for the grab I've a fix, I'm trying to finish for the last landing [15:58] no grabbing, a test user, logged in, went to settings to change the "lock after ..." setting to 30s [15:58] ctrl-alt-t [15:58] waited 30s [15:58] move cursor [15:58] it doesn't seem to be reproducable easy though [15:58] easily [15:59] weird thing is that both his session and the test one had the issue [16:00] in fact he can reproduce [16:00] * seb128 tries here [16:03] yeah, I can reproduce as well [16:03] you need to wake up the screen by moving the cursor [16:03] not by using the keyboard === gatox is now known as gatox_lunch [16:04] mhr3_, isn't it the X-Ubuntu-Touch entry in the .desktop file that indicates the app will run on Touch? [16:05] it used to [16:05] mhr3_, shouldn't it still? [16:06] bregma, Trevinho: can be reproduce in one session, just do [16:06] seb128: so it works with mouse not with key? as they mostly use the same code, unless nux doesn't give us events (not that strange= [16:06] - change "lock after ..." to 30s [16:07] seb128: done [16:07] - ctrl-alt-L [16:07] don't touch anything then [16:07] the screen dim/turn off [16:07] - touch the touchpad/mouse [16:07] - see that the password entry is not focussed [16:07] bregma, yes, i think sdk adds it automatically still [16:07] - try to type -> nothing [16:07] - click on the password entry [16:07] -> no way to focus it [16:07] Trevinho, ^ [16:07] try those steps [16:08] bregma, how can i get keyboard to work in the preview session? [16:08] seb128: nothing I can't reproduce [16:08] would like to enter my u1 credentials so i could actually install some apps [16:08] mhr3_, you mean the OSK, or the physical keyboard? [16:08] bregma, either [16:08] Trevinho, the "don't touch anything" step is important, you need to touch nothing (no keyboard/mouse) between the time you ctrl-alt-L and the screen off [16:09] Trevinho, also really wait for the screen to be totally off [16:09] sure, otherwise there's no off [16:09] yep [16:09] you can't reproduce? [16:09] when you mouve the mouse the password entry is focussed? [16:09] seb128: not focused, but at soon I write it becomes [16:09] mhr3_, the physical keyboard has always just worked for me, you might need to click on the field to focus (no shortcuts) [16:10] Trevinho, what if you click on it before writing? [16:10] seb128: no, just typing [16:10] bregma, nope, doesn't do anything for me [16:10] Trevinho, try clicking? [16:10] seb128: as all the key events are acually taken by the field, unless they're not handled by the panel === maclin_ is now known as maclin [16:11] seb128: works here... Sometimes gnome power might not wake up us, so we monitor key/motion events on the dark window we draw... [16:11] bregma, anyway, try silo 012, has latest click and the session [16:12] mhr3_, unfortunately I have not seen that problem [16:12] Trevinho, yeah, I can't reproduce every time but I got it 3 times and didrocks as well [16:12] and I'm testing silo 12, it shows *too many* apps now [16:12] I think we need to filter and show only X-Ubuntu-Touch=true .desktop apps [16:13] I just tried to reproduce that a few times and it's always been focused here too [16:13] (not running what just got uploaded though) [16:13] I'm running what was in the silo earlier in the day/just landed to proposed [16:15] Trevinho, don;t want to distract you but what's the status with attente's MP? [16:15] bregma, in that case i'll just drop the preview-session from the silo [16:15] bregma, then the envvar won't be set, and it won't add all the regular apps [16:16] mhr3_, will it still show apps from .desktop files with X-Ubuntu-Touch=true? [16:16] * bregma goes to tst [16:16] will check with alecu [16:17] bregma: almost done integrating Trevinho's changes [16:17] Laney, Trevinho: hum, go it again: with switch to a test user, open ucc, change timeout to 30, lock, don't touch anything until screen turn off, mouve trackpad, try to type (seems to happen sometime when the password entry doesn't get focus/doesn't have a blinking cursor) [16:18] attente, vite! allez! allez! adiamo! vamos! [16:18] Laney, Trevinho: I wonder if "having several session opens" is a condition [16:19] seb128, could that bug possibly depend on the way the hardware/driver does dpms? [16:19] beidl, I guess it could [16:21] bregma, can you join us in hangout? [16:21] bregma, https://plus.google.com/hangouts/_/72cpj8q1td8fu8flmlc6vvdtvg?authuser=1&hl=en-GB [16:22] bah. bad merge with trunk [16:23] seb128: btw the fact is that it shouldn't lock, first of all.. and I don't see why that happens [16:23] seb128, om26er, I'm looking into the ubuntu-system-settings otto failures. unity8 is having similar problems and our usual methods for resolving otto issues aren't working this time [16:24] seb128, in the future, please ping the vanguard in the ubuntu-ci-eng channel [16:26] Trevinho, what do you mean it shouldn't lock? I do ctrl-alt-L [16:26] fginther, ok, thanks [16:27] seb128: oh, sorry... I got confused delayed lock activation. As I was touching that part [16:27] seb128: can you quickly compile unity? [16:29] Trevinho, "quickly" probably not [16:29] but I can compile it on my i5 ;-) [16:30] Trevinho, Laney: did you try with having 2 session open? [16:31] not yet [16:32] argh, User Accounts just shows a "My Account" section without the user in it [16:32] is that in guest? [16:32] ok, works after restarting it [16:32] no [16:32] was going to add a test user [16:32] weird :/ [16:33] Trevinho, btw it's weird that if you lock/doesn't touch anything the screen fade/turn off after like 10s while it doesn't do it if you move the cursor after locking [16:34] seb128: well, i used that as a compromise [16:34] seb128: basically if you intend to lock for turning the screen off, then it's done quickly [16:35] k [16:35] seb128: otherwise, it stays up unless you don't get next idle [16:35] seb128: so you can use music and stuff.. [16:35] seb128: since as soon as you turn off the screen, things go away [16:35] Trevinho, Laney: ok, simplified test "log into a test users from your session's indicator, ctrl-alt-t, don't touch anything, wait screen off, move trackpad nibble, try typing password" [16:36] I've not managed to make it not have focus [16:36] Trevinho: ok, amended the MP [16:36] Laney, if you just "ctrl-alt-l, wait power off, move mouse", you always get blinking cursor on the password entry? [16:36] * Laney waits for idle [16:37] sometimes I get it not blinking, but it has always taken the input [16:37] clicking or just typing [16:43] \o/ no more gnome-screensaver [16:43] hum, it happens like half of the time for didrocks and me, with the steps described there, not sure what is different [16:43] * mdeslaur hugs desktop team [16:43] mdeslaur, hug andyrock and Trevinho ;-) [16:43] * mdeslaur hugs andyrock and Trevinho [16:44] * Trevinho feels warm [16:44] :) === gatox_lunch is now known as gatox [17:02] Laney, seb128: ubuntu-system-settings doesn't run its ctests during build. probably should? [17:03] mterry, it doesn't? [17:03] yeah, it should [17:03] are you sure? [17:03] I though it did [17:03] seb128, Laney: override_dh_auto_test: [17:03] python tests/test_code.py [17:03] it certainly /did/ [17:03] bah [17:04] BAH I SAY [17:04] bah [17:04] it regressed in r650 [17:04] make it run dh_auto_test as well === 21WAABD33 is now known as sarnold [17:04] mterry, thanks for spotting it [17:04] even matter make the upstream build system run those python ones [17:04] mterry, do you want to send a mp or should I do it? [17:04] also if you're going to have random scripts in there please add nocheck handling [17:05] ifeq (,$(filter nocheck,$(DEB_BUILD_OPTIONS))) [17:05] seb128, you do it because some tests might have failed? tst-update and tst-update-manager? [17:05] mterry, they do?! [17:05] shrug [17:05] seb128, I'm not sure. They do in the wifi branch I'm reviewing. Not sure if that's wifi or trunk [17:06] mterry, it's not going to be before at least a week if I do it, but adding to my todo [17:06] mterry, I'm swamped with the LTS coming [17:06] seb128, sure. I'll get to it earlier if I can too [17:06] k, thanks === dpm-afk is now known as dpm [17:21] * seb128 waves to launchpad, hello? [17:22] Hrm [17:22] oh I was just going to say that [17:22] I can't get on http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html either [17:22] He's dead, Jim. [17:22] that's great timing [17:22] not dead, just temporarily down [17:23] how do you know? [17:23] I'm waiting on some translation exports [17:23] bregma, what did you do to poor launchpad?! [17:24] it's feeling the freeze [17:25] "Oh, I'll just push these updates and then go away for the evening" [17:25] seriously, launchpad is making me break that french law [17:25] !!! [17:25] oh, that's right [17:25] launchpad is going to hear from my layer [17:26] lawyer even [17:26] seems back now [17:26] oh, it's [17:26] what you said ;-) [17:26] the lawyer threat worked? [17:27] bah, down again [17:29] pointless hanging around ;-) [17:29] * Laney will push stuff tomorrow [17:29] night! === qengho is now known as CardinalFang === CardinalFang is now known as qengho [17:34] Laney, night [17:49] seb128, https://code.launchpad.net/~system-settings-touch/ubuntu-system-settings/run-tests/+merge/215265 [17:49] seb128, I did the trivial bits [17:49] seb128, but the test fixes I wasn't sure how to proceed [17:49] seb128, but it's a shared branch [17:49] mterry, thanks [17:49] mterry, do you have the actual error from the tests? [17:50] they wfm [17:50] fascinating [17:51] seb128, http://paste.ubuntu.com/7231764/ [17:54] bbiab [17:54] mterry, oh, sorry [17:54] mterry, you wrote "tests" and I was thinking "autopilot tests", that's the one I've been running recently [18:11] Trevinho, bregma: oh, another fun bug ... open a guess session, pick "lock screen" by error [18:11] unity tries to lock the screen in loop it seems [18:12] it fades off, fail to lock, turn the screen back on, and cycle [18:12] it took me 15s to be able to go to logout [18:12] seb128: mh, it should not lock, only turn off the screen in that case [18:12] (likely going to happen in sessions where lock is locked down as well) [18:12] Trevinho, well, maybe I misinteprete what it's doing [18:13] it's just diming the screen in loop and waking it back up, making the session impossible to use [18:13] seb128: super weird [18:13] as using disable-lock-screen here is fine [18:15] Trevinho, do you get the issue in a guest? [18:15] seb128: let me try [18:16] seb128, do you get a "lock screen" choice in the session indicator menu? [18:16] I don't, in the guest session [18:17] bregma, no, I picked "logout" and clicked on the wrong icon [18:18] hmm, interesting, I don't see a loop but it doesn't wake up o keypress, only on mouse movement [18:19] hum, k [18:19] I filed https://bugs.launchpad.net/unity/+bug/1306211 [18:19] Launchpad bug 1306211 in unity (Ubuntu) "Trying to lock screen in a guest session leads to an unusable session" [Undecided,New] [18:20] bregma, andyrock, Trevinho: the screen seems to never turn off still btw, I'm sitting next to didrocks' laptop for almost 2 hours (he had to leave for an appointment), his lockscreen is still on with screen on [18:21] seb128: it works here on guest [18:21] I see the same behavior as bregma describes. [18:21] seb128: both calling the dbus method and using the dialog :/ [18:21] seb128, that's odd (not turning off) I tested that explicitly here [18:21] This is Unity from the archive. [18:22] Screen now turns off for me as well. [18:22] ChrisTownsend, seb128 please monitor gdbus monitor --session --dest org.gnome.SessionManager --object-path /org/gnome/SessionManager/Presence [18:23] Trevinho: It works for me. [18:23] and eventually gdbus call --session --dest org.gnome.ScreenSaver --object-path /org/gnome/ScreenSaver --method org.gnome.ScreenSaver.GetActive [18:23] ChrisTownsend: ah, sorry, [18:23] :D [18:23] it's the rule... The French guys, gets all the odds behavior.. It's story now. [18:23] ^_^ [18:23] Trevinho: The only thing is a keypress won't wake it the desktop, only mouse movement. [18:24] basically Murphy was nothing compared to seb128's machine :) [18:24] Trevinho: Well, Murphy is Irish, so... [18:24] new corollar: if a corner case bug can show up, it will happen on seb128's machine :D [18:25] and the funny thing is that the bug is really there... But hardly reproducible by everyone else :D [18:35] mh, I'm going off a bit, as I'm quite starving right now.. [18:44] Trevinho, you must starve! [18:47] Trevinho, ChrisTownsend, bregma: the screen turns off after locking if you don't touch the mouse, if you do it seems to turn off for ever [18:47] sorry it's almost diner time, might be on and off [18:52] Trevinho, ChrisTownsend, bregma: I had to try like 15 times to get the guest session issue again and it stopped when it touched the keyboard [18:53] I'm the only one here who has the launcher set to autohide, right? [18:53] After some time I see the hint shadow (that you usually get when moving the mouse pointer to the left edge) on top-right and bottom-right workspaces [18:54] actually, I don't even get the hint shadow when moving the pointer to the left edge. [18:55] beidl, I use autohide (hey, *I*'m sane) and I don't see the shadow either, but I rarely use the mouse to bring up the Launcher [18:56] bregma, we gots teh gesturez [18:58] beidl, true (we are special), but I usually use keyboard shortcuts anyway [18:58] oh, btw, my week of bug-hunting-without-being-asked resulted in finding a nasty issue with DND. bug 1304882 [18:58] Launchpad bug 1304882 in Unity "[Regression] Launcher doesn't reappear fully after hiding from DND" [Undecided,New] https://launchpad.net/bugs/1304882 [18:58] I already assumed you guys will at some point start to hate me for finding all of those :D [18:59] beidl:: I noticed that... and not at all, reporting and fixiing is very welcome [18:59] beidl: it might be caused by some events not being triggered all the times during dnd [19:00] my next work (SRU0) will about launcher dnd support btw, so... I hope to hit that as well [19:01] Trevinho: I poked around earlier today at this issue. what I found out is that this [19:01] animation::StartOrReverse(dnd_hide_animation_, animation::Direction::BACKWARD); [19:02] only gets called when moving a file/directory away from the edge before it hides completely [19:03] adding it in ProcessDndLeave() helps a little. [19:04] without it, only parts of the launcher is shown when dragging it out with the gestures or by hitting the super key [19:05] the remaining problem is that another try to DND an icon results in the launcher not showing up *while* trying the DND operation [19:05] (I hope that sentence makes sense) === FJKong is now known as FJKong_afk === mterry_ is now known as mterry [20:41] KombuchaKip, hi! [20:41] * KombuchaKip waves at robert_ancell === beidl_ is now known as beidl [22:36] is there any chance bug 1290228 will get fixed before finalrelease? seems like a big one to me. [22:36] Launchpad bug 1290228 in software-center (Ubuntu) "(14.04) Unable to install external deb packages "Package operation Failed"" [Critical,Triaged] https://launchpad.net/bugs/1290228