[00:13] ubiquity installer didn't get the memo about almost everything transitioning away from gconf === Quintasan_ is now known as Quintasan === Quintasan_ is now known as Quintasan [07:36] Good Morning [08:02] good morning [09:13] good morning, team!! [09:13] hello [09:14] how's it going, Laney ? [09:14] good === rvr is now known as rvr_ [09:15] hi Laney, robru! [09:15] chrisccoulson, did you have a good breakfast? [09:15] can i join in with this talking on IRC to people across the table? :) [09:15] robru, yes, breakfast was good thanks. how was yours? [09:15] _ [09:15] | |__ ___ _ _ __ _ _ _ _ _ ___ [09:15] | '_ \ / _ \ | | | / _` | | | | | | / __| [09:15] | | | | __/ |_| | | (_| | |_| | |_| \__ \ [09:15] |_| |_|\___|\__, | \__, |\__,_|\__, |___/ [09:15] |___/ |___/ |___/ [09:16] chrisccoulson, no! go home! Laney and I are having a serious conversation here. [09:16] lol [09:16] Laney: When you do cowsay on a sprint you have to do in real life. So where is your cow costume? [09:17] * Laney ebays one [09:23] hey desktopers [09:24] hi seb128 [09:24] * chrisccoulson pokes seb128 (literally) [09:24] ouch [09:24] =))))))))))) [09:24] hi xnox [09:24] * xnox is a safe distance away around the corner =) [09:30] thinks missing in the canonical office: quadcopters capable of autonomous flight able to pick up a can of Coke at the kitchen and delivering it to the workplace (orders taken via IRC bot obviously). [09:32] xnox: yesterday we were discussing buying these http://www.amazon.co.uk/Zuluu-76330-Vuvuzela-25/dp/B0038DKHKE/ref=sr_1_1?ie=UTF8&qid=1361265787&sr=8-1 [09:32] you wouldn't be a safe distance away from that [09:33] =)))))))))00 [09:33] * xnox has nerf guns at home ;-) [09:33] * Laney sees that is prime eligible ... [09:33] Anyway, my laptop's desktop is started and managed by upstart user session ;-) [09:33] it's only tuesday, going to be a long week! [09:34] xnox, do you know when you guys plan to email ubuntu-devel@ about that? [09:34] Laney, btw just fwded you the emails about the platform test ppa [09:34] seb128: well stgraber was meant to blog about it on monday?! [09:34] got it [09:35] seb128: I can forward you the email.... it had a gnome-settings-daemon race between upstart & gnome-session which I have fixed. [09:35] seb128: but currently I have gnome-session respawning and failing to spaw compiz, when upstart has already started a compiz. [09:35] Laney: see also https://bugs.gentoo.org/show_bug.cgi?id=27727 ... [09:35] bugs.gentoo.org bug 27727 in Other "Larry can NOT be a Cow." [Trivial,Resolved: wontfix] [09:36] wow, who actually cares about calling it a cow or a bull? [09:37] xnox, please do and Cc Laney on those, he's going to be one of the braves who will play with it this week ;-) [09:37] I am currently blaming /usr/share/gnome/wm-properties/compiz.desktop [09:37] (as that's the only compizy thing I can tell gnome-session is trying to launch.....) [09:39] do you have a desktopy devel mailing list I can email this?! [09:43] seb128: https://bugzilla.gnome.org/show_bug.cgi?id=656565 [09:43] Gnome bug 656565 in general "Signalify (already existing) GtkMenuShell.insert()" [Normal,Resolved: fixed] [09:47] Laney: sent you a mail about upstart user sessions ;-) =) [09:47] good [09:47] i'll try it later [09:47] awesome. [09:47] you can watch how badly it breaks in my xmonad session :-) [09:48] well you'd want to write an upstart job to spawn xmonad first. On par as the compiz job, currently shipped in the upstart package in the ppa I linked. [09:49] every session will need to do this? [09:51] Laney: well. we should come up with a plan such that we don't need to. [09:51] Laney: on the other-hand compiz & window decorators are crashy and we do want upstart respawing those. [09:53] Laney: upstart works by itself just fine and stgraber simply glued Xsession.d and a minimal amount of upstart jobs to launch a default unity desktop session with login/out compatability. [09:55] right then [09:55] i got the email, ta [10:04] attente, https://bugs.launchpad.net/ubuntu/+source/indicator-application/+bug/608219 [10:04] Ubuntu bug 608219 in gtk+2.0 (Ubuntu) "Submenus not added when done so with Glade" [Low,Fix released] [10:13] Does anyone think the unity messaging menu would be a good place for a nagios monitor? [10:19] SuperMatt: I would love that! [10:19] SuperMatt: after all hostgroups is just like spam mailboxes ;-) [10:22] y'see, the problem is that nagstamon is broken in 13.04 due to the indicator whitelist being removed [10:22] so I've suggested on a bug report for nagstamon that the unity indicator api be used, but also allow the option of dropping it in to the messaging menu [10:46] seb128: looks like upstream 3.6 is not working with our themes... [10:46] desrt, how so? [10:47] i just did a build of upstream 3.6.1 (from the tag) and it has the same issue [10:47] i'm checking our patchset now [10:48] git_cssimage_load_when_needed.patch is the only one touching css stuff and i guess it's unrelated? [10:49] desrt, right, that's recent and a backport from 3.8 [10:49] that's a performance improvement thing [10:49] well, that's probably not in then [10:49] since 3.8 is broken :) [10:49] i'm checking 3.6.4 now [10:50] maybe there is a fix since 3.6 only on the 3.6 branch [10:50] desrt, did you build using the same libdir sysconfdir etc? [10:50] no [10:50] well it needs to find the engine on disk [10:50] the module dir might be different [10:50] ya... obviously :p [10:50] * desrt symlinked the things [10:50] k [10:50] dunno then [10:50] everything is fine except this menu issue [10:50] weird [10:51] can we blame Will? [10:51] ;-) [10:51] we can blame will for anything, really [10:51] if he's awake... [10:51] attente: hi! [10:51] * ogra_ blames will [10:51] yup :) [10:51] who is will btw ? :) [10:51] * didrocks blames as well [10:51] 3.6.4 not working :( [10:51] * desrt mumbles [10:51] ogra_, attente, one of the most recent desktop hires [10:51] ah, cool [10:52] * ogra_ doesnt blame will then .... puppy bonus [11:16] do you have a stash of usb keyboards? [11:16] (onboard is not yet present in initramfs) [11:17] there's some down here [11:22] =) [11:42] hi.. is this the channel for ubuntu-papercut team ?? [12:00] can I just point out for the record TheMuso has the patience of a saint! [12:00] honshu: yes [12:01] seb128, oh, i forgot to tell you, the dailies for nx7 work again since 20130218.1 in case people need to do fresh installs [12:03] ogra_, \o/ [12:03] ogra_, xnox told us this morning, thanks! [12:03] compiz dies often though ... seems its related to a11y [12:04] (every time i focus a test input field the doecorationas (and focus handling) vanishes [12:04] *text [12:07] TheMuso, i think the latest pyatspi upload started to try to talk to SessionManager over dbus, which is not started when we are in Ubiquity. [12:07] Do you happen to know something about that? [12:07] * xnox ponders if we should experiment and disable a11y to see if that is a culprit. [12:08] xnox, tricky if you want to test onboard [12:08] xnox, it doesnt happen with kbd and mouse attached [12:09] i also see a lot of: [12:09] (oem-config:1138): Gtk-WARNING **: gtk_widget_size_allocate(): attempt to allocate widget with width 846 and height -39 [12:09] in oem-config.log (i guess thats the reason for the off screen drawing of the windows) [12:10] ogra_: nah, we see those in normal ubiquity as well. I was given tips by folks from Cambridge to make gtk warnings critical such that they crash the app and then go one by one fixing them. [12:10] k [12:11] it's mostly to do with with width-for-height management in gtk3 and us failing to provide those calls for some (custom) widgets. [12:11] ah [12:19] Hey guys when I start up my lenovo y580 with raring I here the drums but the screen is blank turns out the contrast is on minimum is there anything I can do to get more info for you guys for a bug if there isn't one? [12:22] s/contrast/backlight/ ? [12:26] xnox: pyatspi is only needed by orca et al, do you mean at-spi2-registryd? [12:27] xnox: I checked the latest daily as of yesterday in virtualbox, and Orca worked properly when enabled at the try/install ubuntu screen. [12:27] TheMuso: yeah, i think at-spi2-registryd is talking on dbus more than it did before. [12:30] xnox: Nothing jumps out at me in the git log as to why this may be the case... Is this causing problems in Ubiquity? [12:32] TheMuso, in oem-config ... we switched to compiz recently and it dies if you focus an input text field [12:32] yeah, since we are not staring / running full session and don't have SessionManager dbus name running, hence it tries talking to it and times out. [12:32] it may or may not be related to other input problems we are seeing on the nexus. [12:32] (and that uses a11y by default for touch input) [12:34] the crashing compiz is definitely not related to input issues of the nx7 [12:35] you dont have it crashing the decorator plugin on the desktop ... [12:36] (even if it dies it still has all elements in place, they just dont react ... while in ubiquity/oem-config the whole of compiz vanishes completely) [12:36] Ok, so what if we rip out the code in ubiquity-dm that calls at-spi2-registryd, and let it get brought up via GTK's use of libatk-bridge? [12:36] Has anybody tested to see if that makes any difference? [12:36] I can try that. [12:37] Certainly worth trying... [12:37] (the point of ubiquity-dm launching all deamons by itself, instead of using activation, is such that it can kill them upon exit and clean up after itself) [12:37] so while it can help identify the cause, we will also need to think about cleaning up. [12:37] I understand that, but if at-spi2 stays around for the desktop session, surely tat isn't a big deal... [12:38] xnox, btw, do we plan to possibly use upstart session management for that in the future ? [12:38] Well the funny thing is, I suspect that if GTk was cleaning up properly as it normally should, at-spi2-registry d probably would get shut down. [12:38] =) [12:38] But, due to the crazy mainloop bug, GTK3 in raring and quantal is curently patched such that GTk accessibility stuff is not properly cleaned up... [12:39] TheMuso: before or after we reverted the "improvement" where it does it at each main_loop exit?! [12:39] =)))) [12:39] there is an api call to clean it up? maybe we should just try calling that at the end. [12:39] xnox: Bug 1056300 [12:39] Launchpad bug 1056300 in GTK+ "If a GTK application quits the main loop and restarts it again, accessibility is lost." [Medium,New] https://launchpad.net/bugs/1056300 [12:40] Is the bug I am talking about, and we patched it out in GTk so what a11y would continue to work in ubiquity. [12:40] s/what/that/ [12:40] yeah, i remember that one. [12:41] So if that weren't patched out, I suspect at-spi2 would be shut down. [12:41] But even then I'm not sure. [12:43] At the very least, libatk-bridge may load at-spi2 with different command-line arguments to what ubiquity-dm uses, so it may just require chanignt eh command-line arguments in ubiquity-dm in the end. [12:43] changing the even [12:44] ogra_: upstart user-sessions is not production ready yet, but yeah it would make ubiquity-dm so much nicer. [12:44] yeah [12:44] well, we did plan u-u-s for 13.04, no ? [12:45] at least for starting desktop components on demand [12:45] ogra_: 14.04 production 13.04 upstart should be in the archive with user-session support and some DE launching capability (optional) [12:45] yeah [12:45] if we are good we do want it by default, but there are quirks to work out. [12:46] well, i wouldnt do it by default yet for everything ... but i guess ubiquity-dm would make a great testbed [12:46] e.g. my ssh agent didn't get the correct environment and we should figure out the story of how to override/not-override converted stuff for xdg-autostart desktop files, Xsession.d snippets and other pieces in *-session launchers. [12:47] ogra_: i reflashed root, but I want to keep my userdata partition but ac100 tarball is shouting at me. [12:48] any way to skip tarball installer now? =) [12:48] shouting ? how ? [12:49] well, the ac100 side of the installer is pretty sctrict, it uses a builtin md5 sum from the initrd to make sure bootimg and tarball match [12:49] dropping me into initramfs shell saying that it can't find root.img - did you use the right image? [12:49] well the tarball is not there any more ;-) i want it to just boot. [12:49] mount --bind /dev /root/dev [12:49] chroot /root [12:50] mount -t proc proc /proc [12:50] update-initramfs -u [12:50] (assuming there is no ac100-tarball-installer instaled in your rootfs) [12:50] awesome thanks. [12:50] (else remove that package) [12:51] * xnox really did not want to recompile packages I just compiled for testing ubiquity. [13:49] mterry: I didn't actually do that merge [13:49] jdstrand, I realized after talking with Robert. His fault, and he fixed it up. [13:49] jdstrand, thanks [13:50] np [14:13] seb128: bonjour monsieur, comment se fait le sprint? [14:18] pitti, salut, le sprint est bien, on travaille et on s'amuse également [14:18] seb128: I couldn't help grin this morning about [14:19] systemd | 44-10ubuntu1 | raring/universe | source [14:19] ;-) [14:19] but it's great to at least have the libraries now [14:19] I hope we can make logind work somehow [14:19] I heard that someone called it a franken-init ;-) [14:20] well, that's what it is really [14:20] "wash my hands, but don't make me wet" [14:20] but yeah, if we can get the stuff through salami tactics only, I'll take it :) [14:21] pitti, slangasek said he would look at it, but he has tons to do as always, so not sure about the when [14:21] he said that logind is mostly a matter of figuring what to do with cgroups [14:21] * pitti currently has lots of fun blowing the dust off pygobject 2.28 === m_conley_away is now known as m_conley === Sweetsha1k is now known as Sweetshark [15:16] mterry, https://bugs.launchpad.net/ubuntu/+source/harfbuzz/+bug/1130165 [15:16] Ubuntu bug 1130165 in harfbuzz (Ubuntu) "[MIR] harfbuzz" [Undecided,New] [15:36] has anyone else on raring had any issues with mumble, you need to restart it twice for it to connect to the server? [15:36] seb128: what do you think about gnome-games 3.8 for raring? [15:47] jbicha, does it require gtk 3.8? [15:48] seb128: no [15:49] the big change is that each game is a separate source package now [15:49] jbicha, what changed? I don't have a strong opinion on games, if there are no new depends feel free to update [15:52] jbicha: well it's split into many packages now, isn't it? and did sudoku get my python3 patch merged? [15:55] xnox: no your bug is still open and GNOME is freezing now so unless we rush it in, it will need to wait until next time [16:00] desrt: hi! === zz_jackyalcine is now known as jackyalcine [16:24] czajkowski: yeah, same for me, the first time it just seems kind of stuck somewhere [16:24] lalaland [16:46] mterry: ping [16:47] teemperor, hello [16:48] mterry: sorry for interrupting you, ancell forwarded me to you as i wanted to get informatins whether we should mlock in our lightdm greeters ( i found out we don't use it atm in elementary OS) [16:49] teemperor, ah [16:49] teemperor, probably. Security team recommends it [16:49] teemperor, *however* [16:50] teemperor, we did a naive mlockall in lightdm in 12.10. And it made the lightdm process balloon in memory. We fixed that in raring. But we continue to mlockall in unity-greeter, because we figure that since the greeter is short-lived, it's fine [16:50] teemperor, you may be OK with it as well, but just wanted you to be aware that your greeter will consume a lot more memory after this change [16:51] teemperor, you can also use more selective secure memory via libgcrypt or some such [16:51] teemperor, but if you're OK with large memory usage, a simple mlockall at the beginning of the greeter is good enough [16:54] cyphermox: I find I start it up, kill it straigh away and then re launch it then it connects no issues [16:54] mterry: ok, thank you for that nice overview over the problem! [16:54] yeah === francisco is now known as Guest89363 [16:57] teemperor, yw! [17:11] sil2100: hi [17:12] sil2100: how can i help? [17:14] desrt: I already asked on gnome-hackers, thanks! [17:26] seb128: http://thehawksmoor.com/locations/guildhall [17:39] mterry: another question, is it possible that there is no vala-binding for mlockall in precise? [17:42] teemperor: A quick glance at the unity-greeter vala source shows me that a psudo binding for mlockall etc was manually added pointing to the correct header file. If you look at the unity-greeter source, have a look in src/fixes.vapi. [17:43] TheMuso: ah, ok, i was searching for a vapi on the top of the branch-directory [17:43] thank you [17:43] np. [17:44] teemperor, yeah I don't think so. See unity-greeter for an example of manually adding the vapi [17:45] oh, TheMuso got it first :) [18:09] attente, https://bugs.launchpad.net/ubuntu/+source/rhythmbox/+bug/1031449 [18:09] Ubuntu bug 1031449 in rhythmbox (Ubuntu Quantal) "rhythmbox leaking memory" [Undecided,Confirmed] [18:35] hey all... [18:35] hey jasoncwarner_, how are you? [18:35] didrocks good, except for the whole being in constant pain bit... ;) [18:36] didrocks: you? and everyone else there? [18:36] jasoncwarner_: getting well as you can see on g+ :) in addition, we are getting a sunny London weather! [18:36] jasoncwarner_: hope your back is fixing gradually [18:37] didrocks: yeah...kenvandine is a jerk for posting the good food on g+ :/ [18:37] jasoncwarner_, yo yo [18:37] jasoncwarner_: did you see my post? We keep a piece for you :p [18:37] jasoncwarner_: hey. what's up? [18:37] jasoncwarner_, you know we miss you! [18:37] kenvandine: :) [18:37] hey desrt [18:37] jasoncwarner_: going here on thursday: http://thehawksmoor.com/locations/guildhall [18:37] jasoncwarner_, we might go for another round of burgers tonight [18:38] tower 42! [18:38] kenvandine: not likely :) [18:38] bah [18:38] * desrt wants to go to the theatre, you bums [18:38] desrt, loser! [18:38] desrt: I'm holding down the fort in Canada for you, robru and attente [18:38] desrt, you're so artsy [18:38] jasoncwarner_: :) [18:38] kenvandine: i could practically be a designer with this scarf and how i wear it around inside and stuff [18:38] haha [18:38] desrt: v-neck, scarf and glasses without lenses [18:39] * desrt tosses it for flair [18:39] v-neck! [18:39] * desrt gets on that [18:39] holy shit [18:39] i can't see!!! [18:40] attente has strong glasses, it turns out [19:08] * didrocks waves good evening === jackyalcine is now known as zz_jackyalcine === zz_jackyalcine is now known as jackyalcine === jackyalcine is now known as zz_jackyalcine === zz_jackyalcine is now known as jackyalcine === m_conley is now known as m_conley_away