[03:54] ::workspace-bugs:: [1304787] startkde qdbus invocation is broken with the 4:4.11.8-0ubuntu4 change @ https://bugs.launchpad.net/bugs/1304787 (by Joseph Yasi) [05:48] ::qt-bugs:: [1304805] KDE desktop login failure: "Could not start d-bus. Can you call qdbus?" @ https://bugs.launchpad.net/bugs/1304805 (by meldroc) [06:21] two folks have filed a bug about something similar to your first problem just while I was away [06:22] oops, wrong chan [06:36] Good morning. [06:52] I must say the postgres backend does help with kmail speed :) [06:52] (need to make sure you install the package, then restart - it appears to migrate itsself [06:55] ::qt-bugs:: [1304805] KDE desktop login failure: "Could not start d-bus. Can you call qdbus?" @ https://bugs.launchpad.net/bugs/1304805 (by meldroc) [06:56] Hehe, just the bug I'm facing. [06:57] lordievader: ouch [06:57] I hope I don't face it again [06:57] lordievader: oh, and good morning [06:57] because it is fixed for me now [06:57] o/ [06:57] lordievader: set sources to main repos and update [06:57] :D [06:58] Funny thing is project neon5 works fine. [06:59] I'm thinking I won't update/upgrade tomorrow morning until checking in here [06:59] :-) [06:59] I thought I had the main repos, will check in a bit. Saw somewhere that qt4-dev-tools might work for some reason, so trying that. [06:59] ;) [06:59] good morning [06:59] o [06:59] o/ [07:00] I gave that advice earlier, but it didn't help that person [07:00] unfortunately [07:10] morning soee [07:14] jussi: Sources have been unchanged. [07:20] !info kde-workspace-bin [07:20] !info kde-workspace-bin trusty [07:21] kde-workspace-bin (source: kde-workspace): core binaries for the KDE Plasma Workspace. In component universe, is optional. Version 4:4.11.5-0ubuntu0.1 (saucy), package size 1824 kB, installed size 8742 kB [07:21] kde-workspace-bin (source: kde-workspace): core binaries for the KDE Plasma Workspace. In component universe, is optional. Version 4:4.11.8-0ubuntu4 (trusty), package size 1826 kB, installed size 8761 kB [07:38] https://errors.ubuntu.com/problem/fd73219a6c413c7360e223fe56b7cceb35ceb421 [07:38] people still have synaptiks [07:38] Riddell: what do we do about it :O [07:39] could be a post-upgrade-before-reboot-crash, but who knows [07:40] https://errors.ubuntu.com/problem/b674d33a11fddc5b8e414df042f41bb5a9b047af [07:40] printer-applet [07:40] much madness [07:42] $ apt-cache policy synaptiks [07:43] N: Unable to locate package synaptiks [07:43] I never removed it [07:43] and used to have it [07:51] Anyone else affected by bug 1304805 [07:51] bug 1304805 in qt4-x11 (Ubuntu) "KDE desktop login failure: "Could not start d-bus. Can you call qdbus?"" [Undecided,Confirmed] https://launchpad.net/bugs/1304805 [07:55] ::qt-bugs:: [1304856] QDbus fails to start on login @ https://bugs.launchpad.net/bugs/1304856 (by [Po]lentino) [08:24] apachelogger: hmm, make a transitional empty package? [08:25] lordievader: meh, I thought I'd fixed that yesterday [08:25] lordievader: Yes, me. A few days ago it worked by installing qdbus-qt5, but not since the updates of (yesterdag/thi morning) [08:25] Riddell: Perhaps the fix is still in proposed or something? [08:25] .... [08:25] Riddell: you broke it [08:25] again [08:25] hmm, maybe I created it yesterday instead [08:26] alvin: Can you mark that you are affected by the bug? [08:26] sure [08:26] Riddell: Hihi, that is a possibility. A less comforting one though. [08:27] yeah I did [08:27] doh [08:27] Riddell: https://launchpad.net/ubuntu/+source/kde-workspace/+changelog [08:27] see 4:4.11.6-0ubuntu3 [08:27] 4:4.11.6-0ubuntu5 [08:27] 4:4.11.8-0ubuntu2 [08:28] if $qdbus >/dev/null 2>/dev/null; then [08:28] your change broke it entirely [08:28] you cannot do exec assignments as part of unescaped execs [08:29] I know (now) [08:29] it works if I remove the quotes from the qdbus="QT_SELECT=qt4 qdbus" line [08:31] even so [08:31] it is a broken fix [08:31] how? [08:31] it breaks qt5 [08:31] how? [08:31] equally if I were to remove qdbus4 it would be broken [08:31] Riddell: oh, actually it might not [08:31] its only set locally to the startkde script, a previous version had an export QT_SELECT which did break qt5 [08:32] rohan's probably did twiddle it in the env [08:32] Riddell: yes, that is nifty [08:32] hopefully something depends on qdbus-qt4 so it won't get removed [08:32] let's check [08:34] Riddell: looks good, nice fix, only the quotes get in the way [08:34] sorry for the raging :P [08:34] Riddell: I still should point out that this is not much of a fix to the actual problem [08:34] kde-workspace-bin depends on qdbus [08:35] qtchooser is generally broken upstream [08:35] everything else calling qdbus will still fall over dead when qt5-defaults is installed but no qdbus5 [08:35] Riddell: I agree [08:35] I don't understand why they can't rename like everyone else [08:35] I also don't understand why debian can't make hard dependencies between chooser and the choosable binaries === vinay is now known as Guest63683 [08:36] that would mean always having qtcreator installed wouldn't it? [08:36] qtc is not choosable [08:36] only moc/uic/qdbus etc. [08:37] mind you, if it only deps on those three it might be enough [08:38] it's just a very big mess :'( [08:40] seems to be getting bigger [08:41] which one? the qt mess or the overall kubuntu mess? [08:42] qt [08:42] because before, it was solveable [08:42] the folks today can't fix it [08:42] nah [08:42] invocation was just rubbish [08:42] cool [08:43] there is no way to recover from that other than fixing the invocation itself [08:43] an easy fix is awesome [08:43] but do we have a long-term answer? [08:43] on the plus side Riddell figured out a nice way to prevent the problem [08:46] Riddell: I think we need a different way to manage our own translations [08:46] shit is not getting imported into launchpad [08:46] and getting it into langauge packs for verification is also a drag^2 [08:47] apachelogger: and as I said with split up upstream tars we won't have a very useful kde-l10n-xx package to just include [08:47] so we might have to look at generating our own language-packs somehow [08:47] "FAILED: kubuntu-notification-helper (The source kubuntu-notification-helper - 14.04ubuntu11 is already accepted in ubuntu/trusty and you cannot upload the same version within the same distribution. " [08:47] * valorie gets off IRC so an important email can be written [08:47] dual upload of kubuntu-notification-helper? [08:55] Riddell: yeah [08:55] Riddell: I am not getting mails for notification-helper [08:56] ::qt-bugs:: [1241859] package libqtgui4 4:4.8.1-0ubuntu4 failed to install/upgrade: lectura insuficiente en buff... @ https://bugs.launchpad.net/bugs/1241859 (by Lucas Germán Burgos) [08:56] ::qt-bugs:: [1242028] samplebuffer qt multisampling doesn't work after upgrade to 13.10 @ https://bugs.launchpad.net/bugs/1242028 (by Damian Obernikowicz) [08:56] didn't find any mails from yesterday's upload, so I uploaded again... [08:56] hum [08:56] Riddell: I still do not get the problem with not having kde-l10n-xx btw [08:57] we'll have to install all languages for KF5 and Plasma packages which will take up lots of disk space [08:57] that's an upstream problem [08:58] if kde doesn't want 1/4 of the size of a plasma next workspace to be translations, then they cannot put translations in the source tarballs [08:58] if they don't care, then we do not care [09:04] ::workspace-bugs:: [1304805] KDE desktop login failure: "Could not start d-bus. Can you call qdbus?" @ https://bugs.launchpad.net/bugs/1304805 (by meldroc) [09:04] apachelogger: hmm, how else would you like upstream to do it? [09:05] cos upstream is kindae me [09:06] Riddell: put in one tar? [09:06] oh wait [09:06] Riddell: aren't kf5 and plasma different anyway? [09:06] kf5 would have l10n bundled [09:06] plasma would have it in kde-l10n-xx [09:16] apachelogger: kf5 have lots of tars and plasma has lots more tars [09:16] apachelogger: plasma gets released separately from applications [09:16] agateau: you might have input into this ↑ [09:17] * agateau reads [09:17] or maybe not, you're not the release dude :) [09:18] heh [09:18] not much input indeed [09:18] but you made me realize our current tarballs do not include translations, which is something i got wrong, so thanks nevertheless [09:21] Riddell: then I guess we'll have to life with translations in each package? [09:23] thing is, you can look at it from the other side as well, if I install the application kittenpix, why would I need to get all l10n of all kde packages in the archive [09:23] just as silly [09:27] mm [09:28] * Riddell out for some hours today [10:05] oh noes [10:05] I installed with german keyboard layout [10:05] * apachelogger lost -.- [10:30] 'Morning folks [10:42] Uploading muon_2.1.70-0ubuntu4_source.changes: done. [10:42] Successfully uploaded packages. [10:42] more upstream commits, fixing some notifier issues, throwing out old unused notifier and replacing it with the new one [10:46] Riddell: phonon-backend-gstreamer1.0 is still around? [10:46] uhhh oh, something is broken. [10:47] no doubt you already know though. [10:47] is there a solution yet? [10:47] sudo nano /usr/bin/startkde [10:47] somewhere in the first 30 lines or so there's a line starting with qdbus= [10:47] remove the quotes in that line and it should work [10:47] or wait for update [10:48] okies [10:48] qdbus=QT_SELECT=qt4 qdbus [10:48] instead of [10:48] qdbus="QT_SELECT=qt4 qdbus" [10:48] yeah [10:49] thankyou sir [10:50] all order is restored to the world [11:00] apachelogger: qdbus="QT_SELECT=qt4" qdbus [11:00] would be less obfuscated [11:04] allee: very much point of view dependent [11:04] apachelogger: might be [11:12] jussi: how the hell do channel notices work, can't seem to send any [11:12] or rather, they don't arrive ^^ [11:12] apachelogger: why do you want to notice the channel? [11:13] not this one, work channel needs feeding with build status information [11:13] it appeared to me that channel notices would be more noticable then your regular msg [11:15] apachelogger: in my client its simply /notice #channel message [11:15] perhaps the channel flags block it [11:15] apachelogger: which channel? [11:15] can't say :P [11:15] heh [11:15] oh [11:15] work channel [11:15] what channel flags are there? [11:16] oh, not even registered it seems [11:17] Kalidarn: careful, noticing the channel is not nice for peopl [11:18] is that what he meant? [11:18] or something else [11:18] yes [11:18] jussi: how else am i supposed to express my love for everyone in #kubuntu-devel [11:31] Kalidarn: 102 cans of coffee for free would do the trick :D [11:34] ghostcube: 102 might do for one developer... [11:34] hmm i just took the people in here .. maybe we should count it up i++ [11:34] ::D [11:45] apachelogger: muon accepted. [12:00] Riddell: I reverted doko's plasma-scriptengine-ruby change after confirming the issue he was worried about is one I fixed. It's in bzr for the next -workspace upload. [12:12] ::workspace-bugs:: [1304805] KDE desktop login failure: "Could not start d-bus. Can you call qdbus?" @ https://bugs.launchpad.net/bugs/1304805 (by meldroc) [12:25] kubotu: i got that too [12:25] i was able to login in xfce [12:25] but not kde [12:26] * jussi hugs jarkko (next tiem you can talk to a human, not a bot) :D [12:26] lol [12:27] didnt notice [12:27] jarkko: Should be fixed in the latest package update [12:27] when was it? [12:28] i went to console and updated some packages over there, meanwhile logged in to xfce [12:28] those big updates are always so insecure [12:32] jarkko: kde-workspace 4:4.11.8-0ubuntu5 [12:33] i read the launchpad report === apachelogger is now known as apakhelogger === apakhelogger is now known as apaKhelogger [12:59] oh, apport crash fixed [13:18] ghostcube: coffee comes in cans? [13:18] you must live in a strange land [13:21] isnt this dbus bug reltated to not installed dbus 64 bit package? [13:21] i remember this happened some times ago too [13:21] -s [13:23] ah no forget it.. [14:25] Uploading dolphin-plugins_4.12.97-0ubuntu2_source.changes: done. [14:25] Successfully uploaded packages. [14:25] upstream bug fix import === apaKhelogger is now known as apachelogger [14:57] ScottK: ping === jono is now known as Guest1478 [15:02] Uploading baloo_4.12.97-0ubuntu4_source.changes: done. [15:02] Successfully uploaded packages. [15:02] upstream bug fix import [15:12] ScottK: we have the possibility to include pam-kwallet (i.e. wallet gets automatically opened on login, without weird password dialogs when network manager wants to connect etc.)... enabling this will require a package on the seed that previously was not seeded, optional loading of the pam module in the lightdm pam config, as well as a patch for kde-workspace [15:12] does the release team approve? :P [15:14] http://quickgit.kde.org/?p=kde-workspace.git&a=commit&h=f3d4a7ef1dcbe4cb9116736888cef1c36eae0295 is the workspace change [15:17] what an incredible socket name... [15:23] Riddell: thanks for uploading gpgme. could you do another upload to fix a build failure: http://paste.kde.org/pvjmauvkc [15:26] yofel: did dist-upgrade and installed some debug symbols today, can't login anymore :-( ... it can't call qdbus ... although on a virtual terminal it works (complains about missing display) [15:26] bug 1304805 - probably? [15:26] bug 1304805 in kde-workspace (Ubuntu) "KDE desktop login failure: "Could not start d-bus. Can you call qdbus?"" [Critical,Fix released] https://launchpad.net/bugs/1304805 [15:29] yofel: rebooting and trying dist-upgrade [15:29] debfx: yeah, in a minute [15:31] yofel: the afiestas says he's going to twiddle it still [15:32] ok ^^ [15:39] yofel: I am not seeing any update here ... so I manually fixed startkde [15:39] yofel: I am not on ninjas here [15:40] probably stuck in the queue somewher [15:46] kde-workspace 4:4.11.8-0ubuntu5 with the fix is in release, but maybe not on all mirrors [15:47] Riddell, yofel: okay thanks [15:57] apachelogger: will you file a FFe bug for that kwallet issue? [15:57] apachelogger: do you know the status of kwallet with gpg? [15:58] kwallet doesn't really need an FFe, the new pam-kwallet only fixes the bug that it was completely kaput [15:58] gpg rubbish was disabled [15:58] apachelogger: yeah I guess you could argue that, what needs done? [15:59] Riddell: all done, I just need to work out what configs we need [16:00] and well, I need an executive decision if the release team is ok with that sort of change [16:00] debfx: uploaded [16:13] much kitten [16:13] Riddell: did you see the latest comment on the qdbus bug? [16:13] the guy has a point you know ^^ [16:14] apachelogger: when compiling phonon-backend-vlc equalizer support should work in amarok right? [16:14] yeah [16:14] apachelogger: what about the analyzer in amarok? does not work with vlc-backend atm [16:15] no [16:15] kdeuser56: mind you, I think the EQ needs vlc 2.2 [16:15] apachelogger: yeah I actually meant that, sorry [16:16] apachelogger: forgot to say against vlc 2.2 [16:16] apachelogger: what is the problem with the analyzer? is it not implemented in libvlc? [16:16] something like that [16:17] apachelogger: I remember having read somewhere in a mailing list that it was implemented in phonon-vlc but broke something else and was therefore disabled [16:17] there's documentation about it somewhere in the phononvlc source [16:19] apachelogger: so this wont even be possible with vlc 2.2? [16:19] Riddell: thanks [16:20] apachelogger: what is this component called again? AudioDataOutput? [16:20] yeah [16:22] apachelogger: i think it was that mail I read: http://lists.kde.org/?l=kde-multimedia&m=134020496511242&w=2 [16:22] apachelogger: anyway thanks [16:24] apachelogger: hmm yes but do you agree it shouldn't export QT_SELECT? https://launchpadlibrarian.net/172351825/qdbus.patch [16:26] yeah, that breaks qtc [16:26] Riddell: I have a review of the changeroo on my board for tomorrow [16:27] don't really fancy getting sucked into that pile of madness this late [16:27] apachelogger: changeroo? [16:27] the change history [16:27] apachelogger: of what? [16:27] the attempts to unbreak qdbus resulting in more breakage [16:28] it might well be safest to simply change all $qdbus to eval $qdbus [16:28] that way qdbus can be defined as a string containing whatever rubbish needed [16:28] alas, more invasive, so maybe there's something more elegant [16:30] I am not sure why startkde uses that weird variable shit anyway [16:30] alias is a posix defined builtin of all shells [16:30] and what the var meddling does is try to mimic alias in a very crappy way [16:36] ah great, now you made me think about it :@ [16:37] expr might be suitable [16:38] or not ^^ [16:39] Riddell: why are we patching this? [16:40] bug #1180196 [16:40] bug 1180196 in qtchooser (Ubuntu) "qtchooser operation prevented by PATH forcing under KDE" [Undecided,Fix released] https://launchpad.net/bugs/1180196 [16:40] :S [16:45] apachelogger: kde-workspace [16:50] apachelogger: re the kwallet thing, I think (release team hat on) you need to file for an FFe. [16:50] ScottK: and would the FFe get accepted? [16:50] That said, I'm favorably disposed, but the paperwork needs to be done. [16:51] Compared to the already planned level of insanity for this cycle, this is a detail. Sure. [16:51] ^^ [16:57] debfx: Accepted the gpgme fix. [17:06] Riddell: for you consideration http://paste.ubuntu.com/7227078/ waiting for afiestas to approve [17:07] apachelogger: your shell script foo is impressive [17:07] turns out you cannot make the var expansion work because there's no eval involved at that point anymore, so it will bypass every builtin [17:09] (which is also why your first attempt did not work, it will not parse the command and export the variable but feed the should-be-variable to the fork like fork(QT_SELECT=4, qdbus) [17:09] ) [17:10] apachelogger or Riddell: Please update from bzr before uploading workspace again. I pushed some changes there last night. [17:10] yep, I noticed [17:11] Great. [17:12] ScottK: yay, and it even built :) [17:12] Bonus. [18:35] yofel: is that true: http://askubuntu.com/questions/365644/qdbus-cannot-be-called-in-kubuntu-saucy ? [18:36] kdeuser56: I have no saucy system to test one, someone else will have to verify that [18:36] *on [18:36] yofel: I will probably do it tomorrow [18:37] yofel: but if I understand correctly this would solve the qdbus issue, woudln't it? [18:37] I'm currently really busy with $LIFE, so I have no idea in what state our dbus issue is right now [18:38] and there's no bug report against the PPA, so THAT ^ issue is new [18:38] wait, that's on the 13.10 image, which was verified to work [18:38] so that askubuntu post makes no sense [18:40] yofel: the package qdbus-qt5 contains here in saucy the qdbus binary in the correct path (I downloaded the package) [18:40] yofel: as the package is also available in trusty: http://packages.ubuntu.com/trusty/libs/qdbus-qt5 this should be kinda solved [18:40] the point is that we don't want that installed [18:40] as that pulls in quite a bit of qt5 [18:40] and we have no need for qt5 [18:40] (yet) [18:41] yofel: yeah but if one installs qtcreator on kubuntu he will run into troubles currently [18:41] *sigh* [18:41] yofel: it will by default install qt5-default, but without qdbus [18:41] that's what we've been trying to fix the last few days... [18:42] yofel: have not tested for a few days so could be fixed [18:42] yofel: installing qdbus-qt5 together with qtcreator would solve the issue in theory .... [18:42] meh, I need to quit quassel. I need to get some work done and plasma keep steeling my keyboard focus everytime you ping me [18:42] sorry [18:43] sorry [18:43] I will simply stop pinging you so you can leave quassel open :-) [20:43] hello world [20:43] I'm back on IRC with a new and more reliable server for my quassel-core [20:43] and for my blog :D [20:43] currently under reconstruction [20:43] my previous server's hdd died [20:43] :( [20:44] and apparently I can't find my backups :( [20:48] hi ovidiu-florin! [20:48] gosh that sucks [20:51] yeah, it does... I have to rebuild my blog from google cache [20:51] to have my data recoverd would cost somewhere in between 15€ and 240€ [21:00] from the dead HDD I mean [21:03] ScottK, Riddell: bug 1305307 [21:03] bug 1305307 in pam-kwallet (Ubuntu) "FFe for pam-kwallet integration" [Undecided,New] https://launchpad.net/bugs/1305307 [21:03] won't be able to do anything WRT uploading until tomorrow though [21:06] apachelogger: Is the lightdm patch anywhere publich? [21:06] public even [21:15] ScottK: not yet created, haven't yet found out if my changes are correct ... but the change is pretty much a nobrainer, basically adds an option module like for gnome-keyring [21:15] ScottK: i.e. if you look at /etc/pam.d/lightdm* for everyline that has gnome-keyring we'd want the very same line for pam_kwallet.so [21:15] apachelogger: OK. Since that's the one package other flavors use, it's kind of important that gets a close review. [21:16] (i.e. I'd like someone non-Kubuntuish to ack that part of the change) [21:16] yeah, I plan to run that by robert [21:16] once I actually have a concrete patch I know works [21:16] Please do that and then post the patch in the bug and say he's agreed to it. [21:16] which I have been promised for tomorrow morning [21:16] ok [21:16] I'll be glad to approve the lot then. [21:18] Riddell: you still around? [21:36] ::qt-bugs:: [1305294] QT uses incorrect theme when GNOME_DESKTOP_SESSION_ID is unset @ https://bugs.launchpad.net/bugs/1305294 (by Øyvind Stegard) [21:46] ::workspace-bugs:: [1305307] FFe for pam-kwallet integration @ https://bugs.launchpad.net/bugs/1305307 (by Harald Sitter) [22:18] apachelogger: not really [22:25] apachelogger: approved! but needs thorough testing [22:28] looks like it might be safe to do the daily update/upgrades? [22:28] valorie: should be yes [22:39] Riddell: was just wondering when you'll check in tomorrow [22:39] Riddell: albert says we should have tarballs in the mornings [22:40] * littlegirl winks at Riddell [22:41] apachelogger: sounds appetizing for breakfast