[00:01] bryceh, anything useful on that bug? [00:05] robru, nothing abnormal in those logs, just the GPFs in gnome-panel [00:05] robru, anything in /var/crash ? [00:05] there are two for gnome-panel [00:05] bingo [00:06] no, wait, one's a .upload, so I uploaded it already [00:06] robru, so, use apport-retrace to turn those into stack traces - see http://www.piware.de/2011/08/apport-retrace-made-useful/ [00:07] doesn't seem to be doing anything... [00:09] is apport-retrace known for running silently for a long time? it's just sitting there with no output. [00:09] no it should give something back relatively quickly. what's your command line? [00:09] $ apport-retrace -s -S system /var/crash/_usr_bin_gnome-panel.1000.crash [00:09] just like on that link you sent me [00:09] yep [00:09] hmm [00:10] maybe it's downloading symbol packages, we can give it a few minutes [00:11] running it with -v would probably clue us in on what it's doing [00:11] ok, yes, in the meantime suddenly unity has decided to draw the orange 'maximize' box over the right half of my screen. I was not dragging a window when this happened, and it won't go away. [00:12] weird, made it go away by dragging a window [00:12] -v is showing that it's doing an apt-get update with all kinds of errors [00:13] example? [00:13] Err http://ddebs.ubuntu.com quantal/multiverse Translation-en [00:13] [00:13] Err http://ddebs.ubuntu.com quantal/restricted Translation-en [00:13] [00:13] Err http://ddebs.ubuntu.com quantal/universe Translation-en [00:13] [00:13] Err http://ddebs.ubuntu.com quantal/main Translation-en [00:13] x100 [00:13] hmm [00:13] still going though [00:14] sorry, I mean, it's not ALL errors, just a lot of them. [00:14] there are many successful hits scrolling past the screen now. mostly errors at the beginning [00:14] ok, not sure what causes those, maybe they're not relevant [00:15] looks like you can get apport-retrace to cache stuff via '-C ~/.cache/apport-retrace' [00:16] I'll let it go for a bit I guess. [00:19] bryceh, ok, it's spewed a load of errors at me now, want me to pastebin it? [00:19] yep [00:20] http://paste.ubuntu.com/1159968/ there we go [00:20] ah nice [00:21] does this look in any way related to gnome-shell running in 2d mode? or is this a whole other thing? [00:22] interestingly it looks like it's faulting in gtk code [00:22] ok [00:23] so, maybe unrelated to the X update after all, but still looking [00:25] let me know if you need any more info [00:25] bugs #948667 and #1036403 appear to show the same fault [00:25] Launchpad bug 948667 in gnome-panel "gnome-panel crashed with SIGSEGV in gtk_widget_pop_verify_invariants()" [High,Confirmed] https://launchpad.net/bugs/948667 [00:26] 948667 has a shedload of dupes [00:26] seems like a good one to fix then ;-) [00:28] robru, what's your steps to reproduce it? [00:29] step 1. log in. step 2. crash. [00:29] any thoughts on the gnome-shell 2d fallback? gnome-shell doesn't use gnome-panel does it? I thought gnome-panel was the old gnome2 thing. [00:31] I'm not a gnome dev, I'm not sure exactly how everything's wired together, but I believe it's using metacity rather than the gnome-shell renderer. [00:32] oh, thunderbird is crashing too! [00:32] man, my system has really gone to hell today [00:32] apport isn't even noticing thunderbird crashing. nothing in /var/crash either. [00:33] (thunderbird:19741): e-data-server-ERROR **: Error calling StartServiceByName for org.gnome.evolution.dataserver.Sources0: GDBus.Error:org.freedesktop.DBus.Error.Spawn.ExecFailed: Failed to execute program /usr/lib/evolution/evolution-source-registry: Success [00:33] Trace/breakpoint trap (core dumped) [00:33] robru, welcome to running the development version of ubuntu ;-) [00:33] There's a weird error. "Thunderbird has crashed because of Evolution." well, I don't really want Evolution if i'm running thunderbird, do I? [00:35] installing EDS seems to have fixed thunderbird. impressive [00:37] bryceh, do you know why the launchpad apport retracing service thing always tells me that my stack traces are invalid because a handful of packages are out of date, but then I go to do an update&&dist-upgrade and there are no updates to install? that has been infuriating me for a while now [00:37] I have reported a *lot* of invalid bugs over the last month. [00:37] robru, that annoys me as well. Happens all the time with gnome bits. [00:38] is there any way around that? any way to say "no, I have all the updates, don't mark my bug as invalid." [00:39] robru yeah there is an environment variable you can set [00:39] APPORT_IGNORE_OBSOLETE_PACKAGES=1 [00:39] THANK YOU [00:40] well hold on, my experience has been it's kinda iffy [00:40] I still get those errors. Maybe it makes them less frequent though [00:40] gonna give it a shot [00:40] also gonna try and log back into gnome-shell and see what it does. [00:41] I ended up just learning to use apport-retrace. It gives more control over the stack trace generation that way, and lets you even get into gdb to poke around, so it's a lot more powerful [00:41] yeah [00:42] y [00:45] ok, gnome-shell seems to be working now. not sure what happened there before. [00:47] thanks bryceh [00:49] heh [00:50] robru, did you do an apt update / dist-upgrade between it being failly and working? [00:50] bryceh, yeah. [00:50] ah [00:50] bryceh, I guess the first dist-upgrade I did that caused the problem didn't get everything? some kind of inconsistent state / bad timing [00:50] not really sure. [00:51] possibly, or could be a legitimate bug that someone slipped a fix in for [00:51] that happens quite often when running the devel version pre-feature-freeze [00:52] you can review your /var/log/dpkg.log to see what got updated if you're curious [00:53] * bryceh wanders off to debug a monitor/kvm problem... [00:53] Yeah, gnome-shell's in there. [00:53] bryceh, no worries, thanks again. [03:12] hey robert_ancell wondering if you think you might have some time in the next few days/next week to help seb out a bit...he has quite a bit on his plate and was hoping you might have a bit of slack to help him here or there. [03:12] jasoncwarner_, what does he need help with? [03:37] robert_ancell, i think jasoncwarner_ was referring to the gnome update [03:37] he is also helping port apps using libindicate to libmessagingmenu [03:37] and helping to land that [03:37] kenvandine, yeah, he confirmed that over pm [03:37] ok [03:38] kenvandine, are there bugs on the libindicate stuff? [03:38] no :) [03:38] so just rdepends basically? [03:38] he and larsu just started that stuff today [03:38] ok [03:38] indicator-messages with libmessagingmenu is in the ubuntu-desktop PPA now [03:38] not built yet [03:40] ftbfs now [03:40] urgh [03:40] they depend on the new gtk... [03:40] which is still in -propose, i think [03:40] do the builders know about that? [03:40] desrt, i fixed that... i set the PPA to use proposed [03:40] now they ftbfs [03:59] sucks :( [03:59] got a log? [04:24] desrt, you back [04:24] ? [04:24] back from what? [04:24] desrt, I think you headed out earlier [04:24] yes. that's true. [04:24] i am back. [04:25] desrt, anyway, could I get you to have a quick look at gcalctool master - it crashes on startup (inside a gtk_widget_show) and it seems to be related to the radio buttons in the app menu (i.e. disabling them makes it work). The backtrace is deep inside GTK+... [04:27] it doesn't crash for me [04:27] desrt, are you running Ubuntu? I was thinking it might be a dbus menu issue [04:27] and the appmenu is working [04:27] and Unity? [04:27] no [04:27] the window is very small though [04:27] ok, so at least the release I made is only borked for Ubuntu [04:27] oh [04:27] i should make install :) [04:27] heh [04:28] let me try in unity [04:29] no crash [04:29] desrt, 12.10? [04:29] yes [04:29] weirdnes [04:29] weirdness [04:30] desrt, updated? [04:30] let me do an entire unity session [04:30] i just did unity --replace [04:31] still working totally fine [04:31] hmm, do I just release it and assume my box is weird? [04:31] i don't think so :) [04:31] oh! [04:31] now i see a segfault [04:31] i was running it under jhbuild [04:32] so i was getting upstream gtk [04:32] this looks highly fascinating [04:33] I'm glad you find it so [04:33] i blame larsu :) [04:40] i think i understand the issue [04:41] there's at least a gtk bug here... [04:44] yup. plain as day at this point. [04:44] gtk bug [04:46] desrt, upstream or ubuntu? [04:47] upstream, unfortunately [04:47] brb [04:56] robert_ancell: okay. i pushed a fix to gtk [04:56] robert_ancell: please test [04:56] other than that, i think gcalctool is fine [04:56] thanks for the poke [04:57] desrt, thank you. I was struggling to debug it and figured you'd have some idea :) [04:57] ya [04:57] larsu and i just turned over a huge amount of code in this area [04:57] please shout again if you see more problems [05:31] desrt, fix confirmed [07:35] seb128: bonjour, ça va? [07:36] hey desktopers [07:36] pitti, hey, ca va bien! und dir? [07:38] seb128: un peu fatigué, je n'ai pas dormi bien [07:39] pitti, il faut trop chaud ? [07:39] yeah, that and the thunderstorm [07:39] we had > 30 degrees over the day, and then got clouds in the evening so it didn't cool down [07:40] seb128: btw, bonnes nouvelles: I dropped gksu from apport in trunk [07:40] so it's only update-notifier, and that doesn't sound too hard really [07:40] ah, nice for the cool down [07:40] pitti, @gksu: great, finally! ;-) [07:41] ./src/gdu.c: invoke_with_gksu(cmd, "/usr/share/applications/gnome-app-install.desktop", FALSE); [07:41] mvo: ^ that doesn't seem too relevant any more? [07:42] oh, there's some ./data/backend_helper.py which uses it, too [07:42] and then of coruse invoking apport for system crashes, but that sounds pkexec-able === tkamppeter_ is now known as tkamppeter [08:08] pitti: yeah, that is not relevant anymore, this was used for addon-cds for edubuntu a long time ago [09:42] i'm feeling a renewed hatred towards flash [10:09] chrisccoulson: oh really? renewed? not like this long sustained hatred? [10:09] heh :) [10:30] hey jasoncwarner_, how are you? [10:32] hey pitti! good, thanks. You? [10:32] pitti: just dealing with the normal FF stuff :) means getting people to nail down what is "important" etc ;) [10:32] jasoncwarner_: ça va bien, merci! [10:33] jasoncwarner_: I really enjoyed my holidays in the last two weeks [10:33] https://www.piware.de/fotos/Bodensee-Aug12/ [10:58] thanks, pitti , I'll check those out! === MacSlow is now known as MacSlow|lunch [12:09] seb128, hi :), maybe you have a moment for http://paste.debian.net/plain/185052 [12:10] ricotz, hey, I can sponsor that, thanks [12:10] seb128, thanks [12:12] seb128: hey [12:12] seb128: i pushed a pretty important fix to gtk last night [12:12] seb128: you're gonna want to pick that up [12:12] desrt, hey, how are you? [12:12] sleepy :) [12:12] * Sweetshark is proud of: https://bugs.launchpad.net/ubuntu/+source/gcc-4.7/+bug/1017125/+attachment/3271642/+files/lp1017125.cxx proving a bug in boost 1.49 ... [12:13] Ubuntu bug 1017125 in boost1.49 "boost::unordered_multimap<>::erase(iterator, iterator) broken on quantal" [High,Confirmed] [12:13] desrt, https://launchpad.net/ubuntu/+source/gtk+3.0/3.5.12-0ubuntu3 [12:13] desrt, that one? [12:13] desrt, g-o-a are working ;) [12:13] seb128: sounds right :) [12:13] robert_ancell++ === MacSlow|lunch is now known as MacSlow [13:00] seb128, good morning [13:01] kenvandine, hey, how are you? [13:01] good, and you? [13:02] seb128, last night i noticed indicator-messages was dep wait so i enabled proposed in the PPA [13:02] but then i-m ftbfs [13:02] i didn't look at why, it was 1am :) [13:02] kenvandine, right, I'm looking at that next, seems like a missing build-depends on gtk-doc-tools [13:02] cool [13:03] so i didn't even start porting gwibber to lmm :) [13:03] i keep using lmm as shorthand for libmessagingmenu [13:03] but when i see it i think -lm [13:03] :) [13:04] hehe [13:06] kenvandine, btw https://code.launchpad.net/~robertcarr/bamf/webapps/+merge/120432 failed to merge on [13:06] "No package 'libunity_webapps-0.2' found" [13:06] kenvandine, so I just added the Build-Depends to the packaging trunk, let's see if that makes the merge happy [13:06] thx [13:06] i was wondering where to fix that [13:06] looking at the log it was pulling lp:ubuntu/bamf [13:07] but ti seemed wrong to add it there [13:07] s/ti/it [13:07] kenvandine, yeah, not clean way, I just do a "prepare landing of next version" as UNRELEASED [13:08] ok [13:08] seb128, anything we need to do to trigger auto merge? [13:09] kenvandine, wdym automerge? [13:10] the auto landing stuff [13:10] whatever robot is out there moving code around :) [13:10] kenvandine, oh, talk to mmrazik or didrocks [13:11] it's trivial to add stuff to the setup, didrocks just did it for the video lens friday I think [13:11] i was talking about landing bamf [13:11] i assume it will retry automatically [13:12] BAMF! [13:12] !BAMF [13:12] Factoid 'BAMF' not found [13:12] haha [13:12] ubot2 joins in the action [13:13] kenvandine, oh, you mean to retry the merge? sorry I was not following you [13:14] kenvandine, the merge robot put the merge request to "Needs Review" when merging fails, so just putting back to "Approved" makes it retry [13:14] Yeah... sorry i often fail to make sense before my 3rd cup of coffee :) [13:14] cool [13:15] kenvandine, I've promoted libunity-webapps as well, the MIR was approved [13:15] cool [13:15] mterry, hey [13:15] seb128, hello! [13:15] mterry, https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-dummy/+bug/949600 ... is that considered ACKed from a MIR side? [13:15] Ubuntu bug 949600 in xserver-xorg-video-dummy "[MIR] xserver-xorg-video-dummy" [Undecided,New] [13:16] mterry, doko wrote in comment #1 "the package itself looks ok for main." [13:16] seb128, sigh... the unity branch failed to auto merge because no commit message [13:16] * kenvandine fixes [13:16] mterry, xorg-gtest wants xserver-xorg-video-dummy to work (can't run the ido test suite without that) [13:17] kenvandine, yeah, classic mistake, that's an annoying "feature" of the merger... [13:17] kenvandine, could you please rebuild MC without --enable-libaccounts-sso please? (sorry if it's not you who maintain that package) [13:17] kenvandine, that's the libaccounts plugin for maemo, not the one for new UOA [13:17] xclaesse, i do maintain it... [13:17] oh! [13:17] hehe [13:17] that's confusing :) [13:17] the UOA MC plugin is in empathy source [13:18] provided by mcp-account-manager-uoa binary package [13:18] seb128, I had assigned back to doko for double-confirmed approval after RAOF's comment. But I agree that a plain reading of doko's comment suggests it's already approved. Let me mark it as such [13:18] mterry, thanks, I pinged doko this morning but he hasn't replied in 5 hours and I'm sort of wanting to not block on him [13:19] kenvandine, yeah that's confusing, tbh the maemo stuff should be dropped or at least kept out of MC's tree [13:21] xclaesse, done [13:22] kenvandine, cool thanks :) [13:48] seb128, I noticed that nautilus got updated? [13:48] mterry, yes, is there an issue with that? [13:48] seb128, no, I just thought the plan of record was staying back [13:48] mterry, oh, by the time we had the discussion we were already on 3.5.3 [13:48] mterry, they started the cycle like a normal cycle [13:49] then changed the design like crazy and we were already on 3.5 [13:49] seb128, sure, but I thought we decided to do 3.5.3.really.3.4 or something [13:49] mterry, the plan of record is to have nautilus3.4 uploaded a different source [13:49] ah, a different source [13:49] OK [13:49] mterry, want to do it btw? or maybe robru? ;-) [13:49] I remember that now [13:49] hmmm, is there a way to make quantal not suck in a virtualbox? Im getting a fat repaint error galore ... [13:50] Sweetshark, install gnome-session-fallback and log into a gnome classic session? [13:50] seb128: k, but that would kinda defeat testing lomenubar, right? ;) [13:51] seb128, I would if I finish my various stuff before FF [13:51] Can't be too difficult [13:51] Sweetshark, you can add indicator-applet-appmenu to your gnome-panel [13:53] kenvandine, do you have a gwibber upload planned for soon? [13:53] Sweetshark: Is there a bug for that problem because i have the same problem? [13:53] kenvandine, ricots pointed to http://paste.debian.net/plain/185052 [13:54] Streamstormer: havnt checked yet. [13:54] Streamstormer, Sweetshark: do you have 3d enabled in your vm? [13:54] if not you probably fallback displaying unity-3d through llvmpipe which is known to be slow and buggy [13:54] that's going to be worked but not likely before ff [13:55] seb128, another small package likely to come through NEW today [13:55] seb128: no 2d or 3d acceleration activate yet. [13:55] mterry, ok, give me a ping once it's uploaded [13:56] Sweetshark: checking [13:56] seb128: I considered doing a demo video -- in which case I obviously want all the unity awesomeness. [13:56] Sweetshark, no luck for you in a vm atm, unity requires working 3d [13:56] or enable 3d in virtualbox [13:57] seb128, ricotz: uploaded, thanks! [13:57] kenvandine, thanks [13:57] the fix for llvm has been merged in compiz upstream, BTW [13:57] seb128: trying 3d [13:57] until then, -vm vmware provides somethign which is barely usable to start a program [13:57] kenvandine: Hey dude where would the sip accounts be now, I was using empathy in Q to access the confline but there is no sip option anymore :( [13:58] the default cirrus and -vga std doesn't work [13:58] pitti, "fix", not so much [13:58] davmor2, good times [13:58] account-plugin-sip [13:58] install that package [13:58] pitti, it's still too slow to be usable, they have plan to turn off effects in vms but that's on the unity side and not done yet [13:58] seb128: c'est ne pas bug 1021104? [13:58] Launchpad bug 1021104 in compiz "Severe damage artefacts and flickering when using LLVMpipe" [Critical,Fix committed] https://launchpad.net/bugs/1021104 [13:58] Sweetshark: Gallium 0.4 on llvmpipe but the problem is there since the last xserver update [13:58] seb128: ah, merci [13:58] davmor2, i wonder if it will migrate your account for you... [13:59] lunch [13:59] since you started without that plugin installed [13:59] pitti, that will improve things but I doubt it will be enough to be "good", but let's see ;-) [13:59] pitti, de rien [13:59] seb128: enabling 3d does improve much here it seems [13:59] Sweetshark, great [13:59] seb128: I still doubt it will make the armhf live images any useful, right [13:59] kenvandine: it broke all the others due to the fact that they all needed to authenticate [13:59] pitti, we would need to have llvmpipe working on arm to start with :-( [13:59] right [13:59] davmor2, but they worked after authenticating? [13:59] seb128: err, it doesn't? [13:59] pitti, there are ongoing discussions on the topic, I can fwd you the emails if you want [14:00] seb128: it seems removing -2d was even more premature then *grumpf* [14:00] kenvandine: I removed them all and set it all up afresh then it was fine [14:00] seb128: that's ok, thanks [14:00] pitti, no, llvmpipe is not enabled on arm atm and they are not sure they can, it's strongly using SSE instruction set [14:00] seb128: I'm frustrated either way :) [14:00] davmor2, oh... you shouldn't have done that :) [14:00] s/have/have needed/ [14:01] kenvandine: I'll reboot to be sure but there doesn't seem to be an icon for the sip account [14:01] seb128: yikes, no -- it does _not_ improve much here. Why do I always fsck up my negations? [14:01] pitti, well, it's not that easy, we didn't remove 2d to remove 2d, nobodo had the few weeks of engineering time to devote to port the 2d stack to gsettings and co [14:01] Sweetshark, :-( [14:01] davmor2, you're looking in online accounts right? [14:01] pitti, so choices were to block GNOME and unity3d another cycle or drop 2d [14:01] kenvandine: indeed, I get a black rectangle with a no entry sign [14:01] pitti, rock ... us ... hard place ... [14:02] well, PS could have chosen to drop 3D and keep only 2D; but anyway, we've been through this [14:02] * davmor2 lends seb128 and pitti some tnt and blasting caps, that should get you through the rock hope it helps ;) [14:02] but if we would employ rickspencer3's "roll back regressions" appraoch, it woudl have to come back [14:02] pitti, yeah, unity is only part of the issue though, totem, cheese, gnome-control-center, gnome-settings-daemon use clutter [14:02] because the current unity-2d was still miles better than what we have now [14:03] davmor2, ok, that one is missing an icon [14:03] pitti, that's an increasing trend, gtk4 will be GL only as will be Qt5 [14:03] kenvandine: and nope there is no sip account showing up [14:03] * kenvandine suspected so [14:03] i think we need the migration code to automatically install plugins for needed accounts [14:04] pitti, you can have a working unity-2d on arm but e.g totem video playing wouldn't work, so still not an acceptable user experience for a desktop [14:04] xclaesse, ^^ [14:04] anyway, unitymenus already looks rather nice and sessioninstaller integration also works like a charm. I can install libreoffice-templates (out of universe) with one click inside the LibreOffice dialog .... [14:05] kenvandine: can you add the need to pull in the accounts package it the telepathy package is installed? ie install telepathy-sofia-sip depends on account-plugin-sip [14:05] that's a good idea [14:09] goooooooooooooood morning desktoppers! [14:10] seb128: oh, great. this also prevents taking demo videos as recordmydesktop is unhappy with composition [14:10] kenvandine, what API should be used to install plugins? [14:10] Sweetshark, :-( [14:10] kenvandine, package-kit ? [14:10] kenvandine: I can confirm that adding my sip account and emapthy audio calls work again now woohoo! [14:12] woot [14:12] xclaesse, the packagekit api should work [14:13] hey robru [14:13] hi cyphermox [14:13] seb128, can you confirm? [14:14] seb128, thanks [14:14] kenvandine, thanks [14:17] kenvandine, it should yes [14:17] seb128: pour -sync, il faut le faire builder pour gtk2 et gtk3? [14:17] xclaesse, so that should work [14:18] cyphermox, hey, no, ido is only built for gtk3, the is no gtkswitch in gtk2 and -sync uses it [14:18] gotcha. [14:18] kenvandine, ok, commented on https://bugzilla.gnome.org/show_bug.cgi?id=682440 [14:18] Gnome bug 682440 in UOA "What if account plugin is not installed when doing UOA migration" [Normal,New] [14:19] xclaesse, thanks! [14:36] seb128, Sweetshark: the workaround in bug 1021104 is solving my problem Yeah :) [14:36] Launchpad bug 1021104 in compiz "Severe damage artefacts and flickering when using LLVMpipe" [Critical,Fix committed] https://launchpad.net/bugs/1021104 [14:36] kenvandine: because I like you https://bugs.launchpad.net/ubuntu/+source/empathy/+bug/1040091 [14:36] Ubuntu bug 1040091 in empathy "The sip account is missing an icon" [Undecided,New] [14:36] thx! === jibel__ is now known as jibel [15:01] seb128, ted named another beauty: lightdm-remote-session-uccsconfigure is in NEW [15:01] mterry, didn't we tell you that ted is not allowed to pick names? [15:01] kenvandine made it clear before ;-) [15:01] seb128, this one is at least over-verbose [15:02] mterry, opposite problem we've had in the past [15:02] but tedg is still not allowed to name packages [15:02] :-D [15:15] mterry, tedg: what is /usr/share/lightdm/remote-sessions/uccsconfigure.desktop used for? does it populate the list of sessions in the greeter or something? [15:15] seb128, Not the list of sessions, but the list of remote sessions. [15:16] seb128, It's new API that robert_ancell is adding. [15:16] seb128, So that it doesn't clutter the list of normal sessions. [15:16] tedg, we're talking after you login with your email, right? [15:16] tedg, and a list of remote sessions appears [15:16] tedg, where does it show? [15:16] tedg, worded differently [15:16] Name=UCCS Configure [15:16] Comment=Setup a UCCS Account [15:17] those are not translatable, do they need to be? [15:17] mterry, Well, it wouldn't ever show that list to the user, but we'd validate the list of servers against it. i.e. if you don't have a citrix client installed it would say that. [15:17]