[04:59] Good morning === highvolt1ge is now known as highvoltage [08:04] hey! [08:08] Laney, hey [08:08] good morning desktopers [08:09] hey seb128 [08:09] good weekend? [08:09] yes, weather was not as good it is today, but the w.e was still good ;-) [08:09] you? [08:10] very good thanks, lots of bike riding [08:11] heheheh [08:12] not much bike riding here this weekend! [08:12] salut seb128, bon week-end? [08:13] mondays, perfect time to do paperwork [08:13] hey Laney, bonjour seb128 [08:14] ça va didrocks [08:14] lut didrocks, pitti [08:14] * pitti waves to mlankhorst, too [08:14] didrocks, nickel, et toi ? [08:14] hey mlankhorst [08:14] guten morgen pitti! [08:14] heya :) [08:14] I mean.. [08:14] seb128: pas fait grand chose vu qu'il pleuvait ;) [08:14] bonjour! [08:17] didrocks, weather was not so nice here either, and I still have some remaining bits of cold, so I spent quite some time on the TV (watching tennis and playing video games) as well ;-) [08:17] heh :) [08:28] good morning everyone [08:30] chrisccoulson, hey, how are you? had a good w.e? [08:31] seb128, yeah, i'm good thanks. my w.e wasn't too bad. i managed to take lots of junk to the recycling centre this weekend [08:31] how about you? [08:35] hey chrisccoulson! [08:35] hi didrocks [08:46] chrisccoulson, heh, mine was a lazy one, which is good ;-) [08:46] chrisccoulson, before travelling next w.e ... [08:47] heh, yeah, i'm not looking forward to travelling [08:49] I'm looking forward seeing everyone though ;-) [08:56] when will you guys arrive? [08:57] * pitti will arrive at SFO on Sat at 16:25 [08:58] ogra_: ah, you are on the same one [08:59] on sunday though [08:59] some time around then [09:00] ogra_: oh [09:00] * didrocks on Sun 12:50 [09:01] sat 17:15 [09:02] seb128, yeah, i'm looking forward to seeing everyone. just not the travelling part ;) [09:02] i don't look forward to spending 10 hours sat in a tiny seat in a tin can :) [09:02] hehe [09:02] * Laney has been considering premium economy [09:02] what's the difference? leg room? [09:02] basically yeah [09:02] I don't mind much the flight there; I fear the flight back [09:03] * ogra_ guesses todays strike will actually even affect next weekend [09:03] or perhaps I'll try and take 3 Melantonins :) [09:03] ogra_: you think they'll continue to strike? [09:03] pitti, easier to hack in a small seat than to sleep?K [09:03] (lufthansa shot down all flights today) [09:03] seb128: yeah [09:04] pitti - i never sleep on flights. i think i need something a bit stronger [09:04] seb128: and there isn't a night supposed to happen [09:04] ogra_, did they? shrug [09:04] * didrocks doesn't mind flying as well, just waking up at 5AM is a little bit too much ;) [09:04] pitti, nope, but 1700 flights is a lot, they will have to book people around all the week [09:04] chrisccoulson: yeah, I can hardly sleep too; maybe an hour or two [09:04] ogra_: yeah [09:04] so all flights out of germany will be full [09:05] and that will have some effect through the week ... and imho until the weekend [09:06] ogra_, well, I'm sure we will manage it, rebooked if needed, maybe with some promotion :p [09:06] (overbooking etc) [09:06] ogra_, not sure how they handle stuff, but I would assume that frequent flyers have a better chance to not be rebooked [09:09] * ogra_ lost his silver status this year :? [09:09] you cool kids with your status [09:43] Laney: FWIW, I'd treat not having a status against not having to fly so much any time :) [09:44] (says a mere-mortal wood class passenger) [09:44] amen [09:48] hmm.... can't some of those airlines re-route via paris/amsterdam/london ?! [09:48] extra hops are a pain, but if it gets more flights off the ground ..... [09:49] xnox, sure, they usually do their best to re-route you but it produces fallout throughout the week [09:50] vClientSprint [09:50] vBeerDrinking: error: plugin not found [09:50] true, vTrappist just wouldn't cut it [10:31] All 203 tests passed [10:31] phew [10:46] moin [10:50] desrt, good morning [10:50] desrt, had a good W.E? [10:50] yup [10:50] super-productive [10:50] cool [10:51] well, nice it was productive, though it's not W.E are for :p [10:51] we're spriting, dude :p [10:51] i managed to land icon serialisation, finally [10:51] \o/ [10:51] and got alex to review that gobject performance work [10:51] and dbus activation of apps is now working in a branch on my computer [10:52] gonna polish that up today and try for a review [10:52] push that somewhere [10:52] it's actually pretty trivial [10:52] good [10:52] the real hard stuff was done in GApplication over 2 years ago [10:52] this is just some small changes there plus some changes to GDesktopAppInfo [10:53] I still remember daniels getting his laptop stolen with months of xorg work on it without copy :p [10:53] heh [10:53] i have dejadup to protect me :) [10:53] ;-) [10:53] anyway... no more fork()/exec() in unity [10:53] or anything, for that matter [11:06] seb128: any opinion on bug 1085169? has seen ~1000 installs from ppa, seems to be fixed, should we possibly SRU that one? [11:06] Launchpad bug 1085169 in indicator-appmenu (Ubuntu) "LibreOffice Menus Stop Working even with libreoffice>=1:3.6.2~rc2-0ubuntu4 and indicator-appmenu>=12.10.3-0ubuntu2.1" [Undecided,Confirmed] https://launchpad.net/bugs/1085169 === MacSlow is now known as MacSlow|lunch [11:30] Sweetshark, hey [11:31] Sweetshark, what's the diff? seems like a valid SRU candidate to me yes [11:45] this is the relevant patch: http://anonscm.debian.org/gitweb/?p=pkg-openoffice/libreoffice.git;a=blob;f=patches/lp-1085169-we-overactivate-here-as-it-does-not-seem-.diff;h=998f0a10ea58f530a55642e525aae845717505e4;hb=193cbdb33935f59f4ab88a17304eef72cbfa4f2d [11:47] mmmm sru's [11:52] Sweetshark, seems fine for a SRU, though the description seems like a workaround rather than a fix for the issue...? [11:58] seb128: a bit. In LibreOffice, when you open a menu it would then populate the menu and link it to the dispatchers (thus getting the states, checkboxes, disabled stuff). What afernandez tried was to do the same with appindicator-menu. It seems there are a few cornercases were this 'activate-as-late-as-possible' doesnt work properly. [11:58] ok [11:58] Sweetshark, well, if the change works, ship it ;-) === tkamppeter_ is now known as tkamppeter [12:01] seb128: Activating proactively however has another advantage. a 'normal' menu would open quick enough, but we have a dbus roundtrip inbetween, so the menu would be slow as you open it -> (dbus-roundtrip populates the menu) -> then you could do stuff, so activating more should give a bit of a performance boost. [12:02] and we are not that 1980-ties memory constrained that this hurts us to have addtional listener for more than one submenu. [12:04] I will run the shiny new subsequentcheck against a version running in unity (and not headless as in in the build) though, just to see if there are any tricky lifecycle issues hiding here. === MacSlow|lunch is now known as MacSlow === dednick is now known as dednick|lunch === ritz__ is now known as ritz === seb128_ is now known as seb128 [12:52] Sarvatt, woh, thanks god === dednick|lunch is now known as dednick === m_conley_away is now known as m_conley [13:56] hey qengho, how are you? [13:57] didrocks: good day! [13:57] qengho: do you some time today for a short meeting on webkit/v8/blink and all that fun party? :) [13:58] welcome to 1996 [13:58] ;) [13:58] didrocks: Any time in the next 7 hours is yours. [13:58] qengho: excellent, let me try to get everyone ready! Thanks :) [14:14] How do I get ps-jenkins to re-run on a branch? It says it's waiting in queue but that doesn't seem to be true from the index on s-jenkins ... [14:15] "Publishing status: Waiting in queue" [14:16] chrisccoulson: mind joining now to this meeting? ^ [14:16] Laney: I think you should ask mmrazik or fginther about it [14:16] what's this? [14:16] ok, hopefully fginther got pinged about that [14:17] fginther: talking about this MP, BTW: https://code.launchpad.net/~laney/lightdm/logind/+merge/160089 [14:17] chrisccoulson: meeting to discuss maintainance of webkit/v8/blink [14:17] or maybe it doesn't matter if it doesn't get a jenkins rebuild since I obviously did fix it [14:18] Laney, looking... [14:18] Laney: and you never make any mistake isn't it? If only there was the jenkins "please jenkins, merge right away, I'm good!" :) [14:18] tag* [14:18] heh heh [14:18] the trust tag! :p [14:18] didrocks: jdstrand as well?! =) [14:18] xnox: jdstrand isn't available, that's why he told me to ask chrisccoulson :) [14:19] Laney, this will not get rebuilt since it was just a commit message. Jenkins will only auto rebuild if a new version is committed [14:19] fginther: So it will have run the rest of the tests? [14:19] didrocks: darn =) [14:19] chrisccoulson can speak for our team-- he is up on everything and would be the person from our team helping out with the maintenance story [14:20] (primarily) [14:20] chrisccoulson: available then? ;) [14:20] didrocks, yeah ;) [14:20] Laney, the build and test already passed. the failure was only due to the missing commit message [14:21] fginther: Fair enough then - all I could find on Jenkins was a build log, not the test results [14:21] http://s-jenkins:8080/job/lightdm-raring-amd64-ci/17/console [14:22] Laney, there appear to be no tests configured as part of the pdebuild [14:23] hmm [14:23] Laney, hmm as well, I see "Enable tests: yes" [14:24] qengho: inviting you as well :) [14:26] didrocks: applying raring updates. one sec while config and reboot. === alan_g is now known as alan_g|afk [14:58] cyphermox, libhud-qt trunk has Ubuntu.HUD 1.0, but it looks like all the apps import 0.1 [14:59] cyphermox, the api bump was a month ago... but the apps aren't ported yet [14:59] is there a transition planned? === alan_g|afk is now known as alan_g [15:00] didrocks, this is why we should have soname like versioning of the qml packages :) [15:03] kenvandine: it reminds something I advise at first :p [15:03] didrocks, indeed, we did talk about that :) [15:03] ;) [15:04] didrocks, what confuses me is that the -app packages are building in the daily-build-next ppa [15:04] kenvandine: because we have the latest hud in it [15:04] right [15:04] so the autopilot tests aren't running with that ppa [15:05] kenvandine: yeah, they are failing [15:05] i ported share-app to that on friday, because it i installed libhud-qt from that ppa [15:05] not because of that... [15:07] i think we need to discuss versioning of those packages again... who was against versioning those? [15:07] kenvandine: what I understood is that today they are still keeping HUD 0.1 [15:07] kenvandine: for quantal [15:07] or is it that they aren't parallel installable? [15:07] and tomorrow, we need to transition to latest HUD [15:08] kenvandine: mind helping on that tomorrow? ^ [15:08] certainly :) [15:08] sil2100 has some knowledge on this as well [15:08] thanks kenvandine :) [15:08] i already did share-app :) [15:08] kenvandine: awesome! :) [15:08] so are they going to branch for quantal and them port trunk? [15:09] right ;) [15:09] well branch == release a latest iso [15:09] s/them/then [15:09] and then having isos on raring [15:09] ah [15:11] * kenvandine wonders what's the point of versioned imports in qml if they aren't parallel installable... [15:12] kenvandine: add me to the list! :) [15:12] kenvandine: I even don't see where the version is defined [15:12] like the .so name doesn't have any version in it [15:12] kenvandine, don't we have the same issue with versioned libraries that aren't parallel installable too ;) [15:12] chrisccoulson: we never do that… :p [15:12] (ie, libraries that depend on translations or gsettings schema, which form part of the ABI too) [15:12] chrisccoulson: hum, maybe for the projects we are upstream for, but shhhh :p [15:12] lol [15:13] hehe [15:13] right, another one ;) [15:13] * didrocks needs to go out for a few [15:13] didrocks, bzoltan said we can [15:14] kenvandine: can we apply that to our libs then? :) [15:14] would be good to have a place with best practice for the qml bindings at some point [15:14] like guidelines, it never killed anybody AFAIK ;) [15:19] kenvandine: transition not planned yet, no. I think we'll need to do the porting [15:19] shouldn't be huge [15:19] not huge [15:20] it's just that they all have to land together [15:20] yeah :/ [15:20] i'll figure out how to version qml bindings today :) [15:31] Laney: I'd like to work with Robert and you next weed on landing all the logind bits; does that sound ok to you? === racarr_ is now known as racarr [15:36] pitti: yes, sounds like a good thing to work on to me [15:39] pitti: is there a list somewhere of rdeps of the UPower D-Bus API that's going away? [15:39] Laney: I don't think we have that one yet; sounds like another archive grep [15:39] * Laney nods [15:39] but upower won't disappear that quickly [15:39] yeah [15:40] it can be done as a "while you're in there" thing for now [15:40] but still good to know IMO [16:02] Laney: oh, I was confused; the shutdown/suspend stuff will drop from 1.0 indeed [16:02] right - but we can keep 0.9.x if necessary [16:03] IIRC it'll be maintained for some time as RHEL ships it === jasoncwarner__ is now known as jasoncwarner [16:34] The nvidia-current package provides a conffile that disables nouveau kernel module. SO, it's possible to have removed nvidia packages, but not purged conffiles, and have a broken system. I'm not sure how to handle that or assign the bug. Ideas? https://bugs.launchpad.net/ubuntu/+source/nvidia-graphics-drivers-304/+bug/1171532 [16:34] Launchpad bug 1171532 in nvidia-graphics-drivers-304 (Ubuntu) "nvidia-current shouldn't add conffile that configures 'modprobe'" [Undecided,New] [16:35] presumably it should check if it is installed before doing the disabling [16:36] Its mechanism so far of disabling is aliasing other package's module to "off" in /etc/modprobe.d/ [16:43] no idea what the best way of doing that is [16:43] ask mlankhorst [16:54] erm modprobe has a mechanism for blacklisting drivers, 'blacklist nouveau' in /etc/modprobe.d would be enough [16:59] * didrocks waves good evening [16:59] mlankhorst: the problem is package A blacklists B's package in a conffile, and provides B's functionality in package contents. So removed, but not purged A breaks things. [17:02] ideally the conffile would be outside of /etc then === alan_g_ is now known as alan_g [17:03] it's not a conffile then, which solves the problem [17:03] is there another location? [17:03] running through strings I'm guessing /usr/local/lib/modprobe.d, seems wrong though [17:04] mmm, modprobe.d does list a few alternatives [17:04] hm maybe /lib/modprobe.d too [17:05] man modprobe.d that is [17:06] at least if I look at the source it would appear that /lib/modprobe.d would be the correct destination then [17:11] however, it looks like mkinitramfs does not respect /lib/modprobe.d currently [17:13] presumably nobody knew it existed, ah well :) [17:20] kenvandine, hey [17:20] hey seb128 [17:21] kenvandine, cjwatson asked earlier if friends-identica should go to universe or be seeded [17:21] universe [17:23] kenvandine, thanks [17:23] kenvandine, had a good w.e? ;-) [17:23] yup, and you? [17:24] excellent, thanks ;-) [17:33] yay... i have 2 working versions of Ubuntu.HUD installed in parallel [17:33] it is possible! [17:34] one from the top the other from the bottom ? [17:36] ogra_, :-D [17:42] kenvandine: lxc containers do not count, btw =)))))) [17:42] :-D === ayan_ is now known as ayan === m_conley is now known as m_conley_away === m_conley_away is now known as m_conley [19:53] Laney: well I made up, biked 42 km or so today [20:01] jasoncwarner: around? [20:05] tedg, hi :), maybe you can comment on https://bugs.launchpad.net/plank/+bug/1162851 if you have an idea what is causing it, thanks [20:05] Launchpad bug 1162851 in Plank "GLib-GObject Warnings and Fatal LIBDBUSMENU-GLIB errors when starting Plank from the CLI" [Low,New] [20:42] desrt, hey, got a question for you [20:43] desrt, what could cause gdbus to stop responding to any requests (and also to not send them) [20:43] mhr3: did you fork()? [20:43] desrt, no [20:43] maybe the worker thread got wedged somehow [20:44] it's super weird, i'm seeing this in unity from time to time [20:44] and the worst thing is that it seems to be working otherwise quite ok [20:44] just dbus is ignored [20:45] need to check top next time it happens to see if something keeps spinning the cpu [20:46] desrt, i suppose a GSource could be the cause of that? [20:46] there is not very much that happens in the worker thread [20:47] anyway... you know the routine [20:47] check for spinning and get a backtrace (of all threads) in any case [20:47] i wish i could reproduce it when i actually have debug symbols [20:48] desrt, what prio does the worker use to send it back? [20:48] DEFAULT or something lower? [20:49] mhr3, when you see that, do you have stuff blocking in dbus sync calls? [20:49] seb128, i didn't see a stacktrace from it yet, it's what i've been trying to get the whole day, but i just couldn't trigger it [20:50] mhr3, just wondering, I've a bug where sometimes things "block"/"timeout" in dbus sync calls [20:50] like clicking on any link freezes xchat-gnome [20:50] or firefox blocks when using webapps [20:50] and sometime compiz freezes as well [20:50] i don't think unity uses any sync calls [20:50] i hope at least [20:51] well, my issue is probably different, it seems like dbus stop working as it should [20:51] like xchat-gnome hangs on gvfs dbus calls [20:51] when opening urls [20:51] seb128, that is somewhat what i see too [20:51] but I've no clue how to debug that [20:51] any dbus calls don't get dispatched and also nothing is received [20:52] it's not specific to unity [20:53] i haven't seen it outside of unity [20:54] then again, i'm not running 13.04 just yet :) [20:54] and even when i did see it was on 13.04 [20:56] back (timeouted) === seb128_ is now known as seb128 [21:21] morning === m_conley is now known as m_conley_away