=== Guest73201 is now known as mfisch === mfisch is now known as Guest18735 [06:46] Good morning [07:43] g'day mate [09:14] The following packages have unmet dependencies: xserver-xorg-core : Depends: libgcrypt20 (>= 1.6.1) but it is not installable [09:14] mlankhorst: ↑ [09:17] worksforme? [09:17] seems more like a libgcrypt20 issue though :p [09:18] doesn't work for the image build [09:18] no [09:18] http://people.canonical.com/~ubuntu-archive/component-mismatches.svg [09:20] odd [09:20] nah [09:20] you need to get it put into main somehow [09:21] how does it even grab that as dependency.. [09:22] it should have used libgcrypt11-dev [09:22] E: Package 'libgcrypt-dev' has no installation candidate libgcrypt-dev is a virtual package provided by: libgcrypt20-dev libgcrypt11-dev E: Using libgcrypt20-dev (no default, using first one) [09:23] yeah but xorg-server is a main package, it should not be able to build from universe [09:23] ci train innit [09:24] I think if you want 11 then build-depend on 11-dev [09:24] the CI train stuff doesn't respect components :( [09:25] or nettle :P [09:26] is there any planned transition to libgcrypt20-dev? [09:27] I'd just do a rebuild if I were you [09:27] yeah [09:28] mlankhorst, just do a non change rebuild [09:32] the ci train broke my build! [09:33] or rsalveti did [09:33] I guess most people don't build-depend on pure virtual packages so it's probably quite unusual? [09:33] * Laney shrug [09:33] are you doing the upload? [09:34] yeah [09:36] ✔ [09:58] larsu, Laney: https://launchpadlibrarian.net/165133766/evince%20with%20no%20menu%20in%20xfce.png [09:58] current comment on bug #1131664 [09:58] Launchpad bug 1131664 in rhythmbox (Ubuntu) "The default apps should have standard menubars under Unity" [High,Triaged] https://launchpad.net/bugs/1131664 [10:02] seb128: http://ur1.ca/gkpgm [10:32] xnox, hey, what are /usr/share/upstart/xdg/autostart/ for? [10:32] xnox, how is that different from /etc/xdg/autostart? [10:33] seb128: it's an overlay added to XDG_CONFIG_DIRS, if and only if, the given user session is managed by upstart. [10:33] what's the point? [10:34] seb128: typically if we convert a .desktop file to an upstart job, we want to override the xdg-autostart one to avoid race/clashes so one drops the override into /usr/share/upstart/xdg/autostart/ such that only upstart job is used [10:34] xnox, I see, thanks [10:34] seb128: and if one happens to not have an upstart managed session, normal xdg autostart files are used. [10:35] maybe i should have just patched all xdg/autostart implementations in the archive to check for the upstart jobs and launch that instead.... [10:35] xnox, thanks for the explanation, that makes sense to me now ;-) [10:36] xnox, is "start on started dbus and starting gnome-session DESKTOP_SESSION=ubuntu" a valid start line for unity-settings-daemon, i.e. is DESKTOP_SESSION defined for gnome-session? [10:36] robert_ancell: yes, sounds reasonable. [10:36] xnox, also, can we do the inverse for gnome-settings-daemon, i.e. "DESKTOP_SESSION != ubuntu"? [10:37] robert_ancell: i believe we did push back hard on making sure, e.g. Ubuntu Kylin, keeps using DESKTOP_SESSION=ubuntu instead of a custom string. [10:37] but might be worth checking that. [10:37] pretty sure we got them to de-brand that [11:04] Laney, http://paste.ubuntu.com/6884550/ wdyt? [12:00] something broke alt+tab not showing the mouse cursor anymore [12:00] my main desktop doesn't have that bug, but a recently upgraded laptop does [12:01] filed as bug 1272284 [12:01] just thinking where to reassign it [12:01] tjaalton: Error: Could not gather data from Launchpad for bug #1272284 (https://launchpad.net/bugs/1272284). The error has been logged [12:01] meh [12:01] tjaalton, new xorg? [12:01] no [12:01] bug 1272284 [12:01] it was filed way before it landed [12:02] yeah i heard, but I don't have that issue [12:02] still? [12:02] I'm on radeon though [12:02] hmm ok.. [12:02] I'm all intel [12:03] seems to be intel issue, then :P [12:03] but it was reported on nvidia [12:03] so not that [12:03] hm [12:03] tjaalton, mlankhorst: compiz/unity didn't change for a long time [12:03] iirc tseliot had the same issue [12:04] but maybe Trevinho knows about the issue [12:04] what's the issue? [12:04] disappearing mouse cursor on alt tab [12:04] tjaalton: not showing or not working [12:05] tjaalton: there's a ccsm option for that, maybe it got changed [12:05] both [12:06] for sure I have it... on trunk [12:07] mouse disappears if I explicitly disable it in CCSM :P [12:07] where's the option? [12:07] unity [12:07] switcher [12:08] yes, or [12:08] unchecked here [12:08] real_size_ = Style::Get()->MenuItemNaturalSize(entry_->label()); [12:08] ops [12:08] gsettings get org.compiz.unityshell:/org/compiz/profiles/unity/ disable-mouse [12:08] check it's not true === MacSlow is now known as MacSlow|lunch [12:09] false on both [12:09] although [12:10] I've ran the updates already, but not booted the machine where I can't repro this [12:10] mh, weird... I don't know if it was a regression, that is fixed in trunk... and trunk is much different from archive version [12:11] mlankhorst: I've never seen it here [12:12] huh, another machine that was last updated on jan 17th is also affected [12:12] guest session works [12:12] so it's session specific [12:13] yup, also on a machine updated today [12:14] so it's not X/hw specific :) [12:16] toggling the disable-mouse setting on and off fixed it... [12:16] of course that "solution" doesn't scale [12:20] moved the bug over to unity [12:23] tseliot: ok was tjaalton then :) [12:24] I just bumped into the bug and tried :) === alan_g is now known as alan_g|lunch [12:48] xnox: are upstart jobs in /etc/init automatically updated when a package is updated? Or do I need to take care of that in the preinst script? [12:48] tseliot: they are handled normally, as any other conffile. So by default you don't need to do anything. [12:48] cyphermox_, reported bug 1277057 [12:48] Launchpad bug 1277057 in network-manager (Ubuntu) "network-manager doesn't respond after waking from suspend" [Undecided,New] https://launchpad.net/bugs/1277057 [12:49] tseliot: but if you want to remove/delete one forever, you need to use dpkg-maintscripts-helper [12:49] tseliot: see man $ dh_installdeb, package.maintscript -> rm_conffile [12:49] tseliot: ditto mv_conffile to rename one [12:50] xnox: I just want to make sure that the old job (same path, same name) is upgraded to the new one [12:50] tseliot: not sure if .maintscript was already support in precise, otherwise there is a wiki page in debian how to do it with pre/post inst/rm scripts. [12:50] tseliot: yes, it will be. [12:50] xnox: ok, problem solved then. Thanks [12:50] if it's part of the same binary package. [12:50] yep [12:50] all good then. [13:06] mlankhorst: sorry, but indeed, the ci train is not respecting the right components [13:06] all I did was a bump :-) === MacSlow|lunch is now known as MacSlow === alan_g|lunch is now known as alan_g === Ursinha-afk is now known as Ursinha === sarnold_ is now known as sarnold === Ursinha-afk is now known as Ursinha [14:42] seb128, my tiny issue that i'd love being fixed https://bugs.launchpad.net/ubuntu/+source/bash-completion/+bug/568404 :) [14:42] Launchpad bug 568404 in bash-completion (Ubuntu) "dpkg completion should support ddeb" [Wishlist,Fix released] [14:42] the status lies [14:43] mhr3_, sudo dpkg -i *.ddeb [14:43] ;-) [14:45] mhr3_, I see how to fix that, I can do it [14:46] seb128, \o/ [15:18] sil2100, didrocks: can I get a setttings silo? ;-) === Guest18735 is now known as mfisch === mfisch is now known as Guest93890 === alan_g is now known as alan_g|tea === greyback is now known as greyback|away === alan_g|tea is now known as alan_g [15:41] seb128: uh oh! One moment ;) [15:42] robru: maybe? ^ [15:42] seb128, didrocks sil2100 : yeahhhh what what? [15:43] didrocks, I buy my slots from sil2100 usually: [15:43] ! [15:43] ahah! [15:48] seb128: do you think you or someone on your team could update pidgin to 2.10.9? [15:49] mdeslaur, trusty? [15:49] seb128: yes [15:49] mdeslaur, yeah, it's one of those cases where I would have done it if we didn't get a sponsoring request doing the update [15:50] mdeslaur, sponsoring request which is buggy, it's taking me longer to get the contributor to fix those stuff that it would have taken me to do the update [15:50] hrm [15:51] mdeslaur, anyway, I've it on my list since monday, I planned to finish the update myself today, going to do that in a bit [15:51] mdeslaur, thanks for the ping :-) [15:51] seb128: great, thanks! [15:51] yw [15:51] ah, I see the bug now, sorry for not checking first [15:52] mdeslaur, no worry [15:53] ChrisTownsend, hi, do you have time to help debug that failing AP test? [15:57] larsu, can you do a no-change commit to lp:~larsu/unity-control-center/wrap-keyboard-label and see if jenkins notices? [15:58] attente_: I plan on looking at it again this afternoon as I'm a bit tied up working some other AP failure issues? [15:59] attente_: It's possible I did something wrong, so I'll check it again since you and Trevinho don't have failures on that test. [15:59] ChrisTownsend, ok, no problem [15:59] bregma, hey, what's up with unity landing in trusty? can we get one? [15:59] seb128, I would like that more than anything [16:00] attente_: have you tried to get an unit test instead? As you know... AP is like our enemy right now blocking the releases... So I'd prefer to have more stable tests. [16:00] bregma, what's blocking? [16:00] seb128: ap ^ [16:00] seb128, ongoing issues with autopilot [16:00] Trevinho, bregma: are those regressions? [16:00] Oh boy... [16:01] seb128, not in Unity [16:01] the current trusty version doesn't even build [16:01] Trevinho, a unit test for testing the keyboard accelerators? [16:01] we need an update plan, which is not "need autopilot to be fixed" [16:01] attente_: I mean you can test the registration of compiz actions... (as they're tested in compiz code to work)... and so just faking the dbus server and trying to register them [16:02] checkin you get what you expect [16:02] attente_: see the gdbus server tests or gnome-session-manager ones [16:04] seb128: I've been working on these AP failures for weeks now trying to get them in check. This is the latest bug entered for this: https://bugs.launchpad.net/unity/+bug/1276516 [16:04] Launchpad bug 1276516 in Unity "unity trunk failing autopilot tests (02/2014)" [High,New] [16:05] seb128: The vast majority of those failures are due to either machine issues or a bug in Autopilot itself. There are a few still tied to the AP test themselves in Unity which I will get fixed very soon, but as you can see from my comments in that bug, the results are not deterministic. [16:06] seb128: I also have a proposed fix for the AP bug, but it's still waiting to be reviewed. [16:07] or, we could have them ignore the AP failures and release into distro anyway, but I can feel the burn already [16:08] Trevinho, i don't really understand what you mean by faking the dbus server, the key grabber is the server in this case [16:09] bregma, ChrisTownsend: thanks for the update, I hope you guys get those tests in shape soon, we could use an unity update, blocking the update might be creating more issues than it's solving === gatox is now known as gatox_lunch [16:30] attente_: yes... But you have to use it in test-mode (that you supported) so basically using a test name... THen you just write a glib::proxy that calls it and you can verify that the actions are actually registered [16:32] Trevinho, this is just to test the accelerator registration/un-registration, right? how should we test the triggering of the actions? [16:33] what time's the team dinner? [16:34] attente_: well, what you did is fine... but it's not testing that the whole thing works... So I'd suggest to open a window (there's a mocking python feature to create windows with the menus you want, but you can use also a very know window such as gnome-calc)... then you can check if the relative menu will open when an action is triggered [16:34] Trevinho, i don't mind adding those extra unit tests, but i think it'd be good if we can keep these AP tests anyways just to make sure the whole thing works [16:34] attente_: all this is possible with AP [16:35] attente_: fine for me, but they don't have to be unstable... [16:35] attente_: so let's wait Chris results [16:35] attente_: anyway for testing the whole thing, doing what i said is checking more [16:35] Trevinho, those AP tests already exist in the test_panel.py file [16:36] attente_: ah... Cool, I forgot I wrote them :D [16:36] :P [16:36] attente_: so... yes that's fine :) [16:37] ok, cool. i'll try to figure out with ChrisTownsend why they seem unreliable on his side [16:37] when he can spare some time [16:47] jasoncwarner: what time is the team dinner? [16:47] 7pm [16:47] probably meetin in lobby at 6:30pm [16:48] mlankhorst: ^^ [16:48] steak? [16:48] chrisccoulson: NO STEAK FOR YOU! [16:48] chrisccoulson: errr of course! [16:48] hi jasoncwarner :) [16:48] hi chrisccoulson ;) === m_conley_away is now known as m_conley [16:52] can I have chrisccoulson's honorary steak? [16:52] oke [16:52] ty [16:52] Laney, only if you have it uncooked [16:52] geez, brits [16:52] *you don't coock steak!!* [16:52] :) [16:53] steak can come cooked? [16:53] I can ask for a well done steak >:D [16:53] heh [16:54] I know someone who did that and then got thrown out of the restaurant by the chef [16:54] and then put curry and mayo on top [16:55] lol [16:58] mlankhorst: get the kangaroo steak! :) [16:59] they're a pest in australia, anyway [17:07] if you go after pest status in .au you need to eat frogs i guess [17:18] at least you don't have problems running over frogs in the shimmering light [17:18] can't say the same for kangaroos.. [17:20] heh === greyback|away is now known as greyback === gatox_lunch is now known as gatox [17:35] lp:~laney/ubuntu-system-settings/panel-loader === alan_g is now known as alan_g|EOD [18:23] off to the lobby.. === vila_ is now known as vila === m_conley is now known as m_conley_away [19:26] attente_: Hey, I'm re-running your AP test and still get the one failure. [19:26] attente_: Here is the output: http://pastebin.ubuntu.com/6887016 === Guest93890 is now known as mfisch === mfisch is now known as Guest22857 === chrisccoulson_ is now known as chrisccoulson === Guest22857 is now known as mfisch [22:08] ChrisTownsend, hey, can you do: dbus-monitor "member='AcceleratorActivated'", then press shift+ctrl+alt+a? [22:12] attente_: D'oh, give me a few minutes to get my test machine back to running your branch. I'm about to push a fix for something else, so it'll be a few minutes. [22:13] ChrisTownsend, no worries, thanks for your help :) [22:22] attente_: Ok, I'm running your branch and run I run dbus-monitor, I get nothing. [22:23] attente_: I'm wondering if Terminator is eating the key combo... [22:23] attente_: I see it loose focus and come back when I hit that combo. I'll try this in gnome-terminal. [22:25] attente_: Nope, that didn't help. [22:25] so dbus-monitor reports nothing... weird [22:25] attente_: Right. [22:38] ChrisTownsend, ok, it seems i'm also in a state where none of the tests work [22:39] so i'll try to figure it out on my side tomorrow and ping you again if i need help [22:39] attente_: Ok, sounds good. Have a good night! [22:40] ChrisTownsend, thanks!