=== shengyao_afk is now known as shengyao [05:43] Good morning [08:44] good morning desktopers! [08:47] hey seb128! [08:47] good week-end? [08:47] didrocks, hey! yes, you ? [08:48] same here, quiet but nice :) [08:51] quiet is good ;-) [09:00] bonjour seb128, ça va / [09:00] ? [09:02] pitti, salut, ça va bien et toi ? [09:03] seb128: je vais bien aussi, merci ! nous avons eu un bon week-end [09:03] pitti, le Stollen est très bon, encore merci ! [09:04] seb128: de rien ! nous avons mangé le premier stollen hier aussi *yummy* [09:04] hey ho [09:04] Laney, hey, how are you ? [09:04] hey Laney, how are you? nice we? [09:04] "weekend" [09:04] pitti, *yummy* indeed ;-) [09:04] hey seb128 pitti [09:05] seb128: il était très bien après 4 heures de travail dans le jardin :) [09:05] was nice thank you, played my new games a bit [09:05] also got a grand new christmas jumper [09:06] err, "jumper"? that must mean something different than what I think [09:06] pulli [09:06] oh [09:06] right on time for the frost and snow :0 [09:07] not the electrical component :P [09:34] so I can't install updates through update-manager anymore [09:35] "You are not allowed to perform this action" [09:35] I'm still in the admin group [09:35] seb128 says it might be a problem with my polkit?! [09:36] larsu, pkcheck might help to see if things are alright there [09:37] hm, that string is nowhere to be found in update-manager :-/ [09:38] pkcheck seems buggy, it always opens me the manpage [09:38] * seb128 tried pkcheck --action-id org.debian.apt.upgrade-packages [09:38] or --help [09:39] I tried with --process , but that makes it crash [09:39] larsu, do you have policykit-desktop-privileges installed? [09:40] seb128: yes: 0.16 [09:40] seb128: FTR, running with gtk 3.10 here since Sat, and the only regression that I noticed is that the ssh/gpg password dialogs are now very wide (not word wrapped any more) [09:40] passing action id and pid make pkcheck return 0 [09:40] pitti, thanks, that's a known issue/mclasen fixed it in git this w.e [09:40] pitti: solved last night ;) [09:41] nice [09:41] pitti, thanks for testing it ;-) [09:41] larsu, I tried [09:41] pkcheck --action-id org.debian.apt.upgrade-packages --process `pidof update-manager` [09:41] but that opens the manpage still [09:41] heh, not for me [09:41] with the reverse order of arguments [09:41] but that hopefully doesn't matter… [09:41] seb128: then pidof update-manager fails and returns nothing [09:42] seb128: try with $$ [09:42] larsu, [09:42] $ pkaction --verbose --action-id org.debian.apt.upgrade-packages [09:42] ah right, I pasted the actual pid [09:42] seb128: that opens the man pagew [09:42] seb128: if it's running, use pidof -x update-manager [09:43] pitti, larsu: yeah, my fault, of course the process is python... [09:43] pitti, danke [09:43] larsu, what are you trying to do exactly in update-manager? [09:43] hm, update-manager also shows me packages to update that apt-get upgrade doesn't [09:43] seb128: update... [09:43] normal ones? [09:43] yep [09:44] just the window that pops up when updates are available [09:44] I didn't even start it manually [09:44] I wonder if the environment of that process is wrong [09:45] can you "strings /proc/`pidof -x update-manager`/environ" and pastebin that? [09:45] larsu, ^^ [09:45] the packages it wants to install are "fakeroot" and "libfakeroot" [09:45] apt-get upgrade doesn't want to do that [09:45] seb128: oh, and telepathy-gabble is broken now (immediately crashes at startup), not sure whether that's GTK; filing/looking at trace now [09:45] pitti, I don't know about that one, I doubt it's gtk though, telepathy connectors are non UI bits [09:45] seb128: http://paste.debian.net/68834/ [09:46] right [09:46] larsu, env seems alright :/ [09:46] seb128: I think the problem is the packages it wants to install [09:47] seb128: control center seems to have a similar (no word wrap/too wide) issue, try security&privacy [09:47] larsu, what action does it try to do? [09:47] larsu, can you screenshot the error? [09:47] seb128: it's just a message dialog with the string I pasted earlier [09:47] pitti, yeah, that issue is all dialogs [09:47] larsu, seeing the dialog would help to say if it's polkit or update-manager [09:47] seb128: ah, mclasen fixed it in gtk git, not in the keyring dialog git? [09:48] seb128: ah okay [09:48] larsu, I wonder if you are simply seeing an update-manager error because e.g those packages are un-authentificated for some reason [09:48] pitti: yeah, there was a global "don't make windows wider than 640px" in there which Company removed [09:48] larsu: ack, thanks [09:48] pitti, right, https://git.gnome.org/browse/gtk+/commit/?h=gtk-3-10&id=2436627eb4644234a9e577538ec334d224e3d2be [09:49] pitti: which is obviously a correct fix, but we should wait to fix all the things it breaks before applying it :) [09:49] seb128: http://imgur.com/CUuWFrv [09:50] larsu, good that I asked, that's not a polit dialog :p [09:50] larsu: right, like https://git.gnome.org/browse/gtk+/commit/?id=b53b5578f [09:51] seb128: really? I can't find this string in update-manager's source either [09:51] larsu, it's in aptdaemon [09:51] that's the backend update-manager is using [09:52] ERROR_NOT_AUTHORIZED: _("You are not allowed to perform this action"), [09:52] ah! [09:53] does it have something to do with the fact that apt-get upgrade doesn't want to install the new fakeroot? [09:53] let me try installing that manually [09:53] ya, that totally worked [09:53] my bet is that it's going to come from an unauthentificated repository or something [09:53] weird [09:53] is aptd running? [09:54] yes [09:55] do you still have the issue now that you apt-get installed it? [09:55] no, there's nothing more to install [09:55] I'll have a look again later [09:55] seb128: thanks for your help so far! [09:56] larsu, yw [09:57] seb128: the commit pitti linked to doesn't fix the issue for everything. I'm running the same since Friday. [09:57] many apps have custom message dialogs, for example gedit [09:57] larsu: no, there's another one for a different standard dialog, and similar fixes need to be added to every dialog which has that issue [09:57] larsu, see the one I listed [09:57] that was just one example [09:57] just before pitti [09:58] ah got it [09:58] the commit seb128 pointed to fixes it for everything, but AFAIUI it's just a bandaid for the stable 3.10 branc [09:58] that isn't in 3.11.x [09:58] right [09:58] makes sense [09:58] 3.10 is what we target for the LTS though [09:58] so if that works that's going enough for now [09:58] we should still fix apps though [09:58] famous last words [09:58] lol [09:59] it's only code, what could go wrong [09:59] ;-) [10:00] haha === Sweetsha1k is now known as Sweetshark [10:47] pitti, did you get the gabble bt? [10:48] seb128: I sent it to errors, but I don't see it on "my reports", werid [10:48] pitti, e.u.c/whoopsie seem buggy [10:48] ah, again missing .uploaded; go whoopsie [10:48] didrocks reported some rb issue on friday that never made it to e.u.c either [10:49] https://errors.ubuntu.com/oops/53d057e6-5b3f-11e3-91f0-2c768aafd08c [10:49] one needs to restart whoopsie, its inotify handling is broken [10:49] what do you mean "missing .uploaded"? is whoopsie failing to report issues when tell it to do so? [10:49] shrug [10:49] right [10:49] seb128: above is my report, it has the trace [10:50] yeah, looking at that [10:50] "Source ID 4 was not found when attempting to remove it" [10:50] seb128: so, certainly not GTK related, more likely the new glib becoming stricter [10:51] pitti, do you make glib abort on errors? [10:52] seb128: not that I know of; it's a CRITICAL [10:52] so far we didn't abort on those, but maybe telepathy sets that [10:53] pitti, seems like https://git.gnome.org/browse/glib/commit/?id=a919be3d39150328874ff647fb2c2be7af3df996 [10:53] that shouldn't sigabrt though [10:53] (doesn't for me) [10:53] (or maybe I just didn't run into that error with gabble) [10:54] #ifdef ENABLE_FATAL_CRITICALS [10:54] /* make critical warnings fatal */ [10:54] fatal_mask = g_log_set_always_fatal (G_LOG_FATAL_MASK); [10:54] fatal_mask |= G_LOG_LEVEL_CRITICAL; [10:54] g_log_set_always_fatal (fatal_mask); [10:54] #endif [10:54] seb128: ah, so gabble at least supports that mode [10:55] seb128: and according to configure.ac it defaults to yes [10:55] hum, k [10:56] pitti, can you report the gabble issue to https://bugs.freedesktop.org/enter_bug.cgi?product=Telepathy ? [10:56] seb128: yes, will do [10:56] pitti, danke [10:56] we should probably turn that flag off in the package as well [10:56] larsu, oh, found an issue with your theme change [10:57] larsu, if you search in nautilus, the default/first returned item is wrongly colored [10:58] seb128: ah, also when the window is unfocussed [10:58] right [10:58] I noticed alt-tabbing [10:58] are you sure this is an issue with my change? It seems unrelated to backgrounds [10:58] pitti: so... how's them suspends? [10:59] desrt, good early morning! [10:59] seb128: good .. afternoon yet? [10:59] 1 more minute? :) [10:59] desrt, not sure I would describe 12 as "afternoon" [10:59] the afternoon is after lunch :p [11:00] seb128: it is indeed. I just tested. [11:00] * larsu hates css [11:00] larsu, :-( [11:00] this seems totally unrelated [11:00] seb128: time for an english lesson... ;) [11:00] haha [11:00] the etymology of the word afternoon is "after noon". [11:01] woh [11:01] TIL: forenoon is a thing [11:01] s/is/was? [11:02] it has a wikipedia page, so is...? [11:02] sounds like something Queen Vicoria would have said when she was a child [11:02] *Victoria [11:02] ;-) [11:02] and... "forenoon" != "fore noon" and "afternoon" != "after noon" as wikipedia is happy to point out [11:02] if you put the space in there then it goes all the way to midnight (either side) [11:02] without the space it typically refers only to daylight hours [11:03] so the afternoon is shorter in the winter? [11:04] and the evening starts at 4? [11:04] it also notes that the term has no precise definition :p === jjardon_ is now known as jjardon [12:38] seb128, Laney ... is that known ? http://people.canonical.com/~ogra/evo-theme.png ... got it with msy latest trusty upgrade last week [12:38] ogra_, seems like a theme issue, I think Laney mentioned it before uploading evo 3.10 [12:39] k [12:53] this looks like the issue I had in gedit [12:53] seb128: are my theme changes already uploaded? [12:53] larsu, no, neither is gtk 3.10 [12:54] ah [12:54] so that's not it then :) [12:54] not likely no ;-) [12:55] larsu, do we still need gtk patches for o-s btw? (#gtk+) [12:56] seb128: thanks [12:56] larsu, thank you ;-) [13:02] seb128: are you okay if I put the nautilus fix on the same MR? [13:02] larsu, nautilus fix for? [13:02] oh, the icon color [13:02] seb128: the black icon when unfocussed [13:02] no problem [13:03] mmm, I see a suspicious change in evo [13:03] Laney, one that requires to adapt themes [13:03] ? [13:11] * mterry stares at seb128 to review welcome-wizard [13:11] seb128, I'm in (near?) your timezone! [13:11] mterry, hey, how is London ? [13:11] seb128, grey [13:12] stating the obvious there... [13:12] seb128: it's up if you want to have a look. https://code.launchpad.net/~larsu/ubuntu-themes/dont-set-all-bgs/+merge/197234 [13:12] larsu, danke [13:18] * seb128 stares at mterry [13:19] * mterry goes back to fighting Mir [13:19] mterry, you are speaking about https://code.launchpad.net/~unity-team/ubuntu-system-settings/welcome-wizard/+merge/186862 ? [13:19] seb128, yeah [13:19] mterry, I reviewed that a week ago, it's Needs fixing [13:19] seb128, ? [13:19] mterry, I don't see anything happening since? [13:19] seb128, huh! I didn't get an email for that [13:20] mterry, well, basically I'm waiting on your guys to reply to my question/address my first review comments... [13:20] seb128, makes sesne [13:20] mterry, that's probably because you down own the branch, it's a team one [13:20] down->don't [13:20] seb128, fair [13:21] I can't seem to subscribe to just one merge request [13:21] :( [13:21] mterry, you can add you to the reviewers list as workaround I think [13:21] ah, I go to the branch itself [13:26] good morning! [13:27] cyphermox, hey, how are you? [13:30] seb128: fabulous, and you? [13:31] it's going to be even more of an awesome day if I receive my SensorTag today, so I can start to work on 4.0/LE bluetooth [13:34] cyphermox, I'm good thanks ;-) === jibel_ is now known as jibel === Ursinha-afk is now known as Ursinha [14:07] seb128: can I keep pushing stuff into my gtk 3.10 repo or did you make any further changes? [14:07] I've finally fixed the settings install thing [14:08] larsu, I did tweaks for the ppa upload [14:08] seb128: did you upload them into a bzr repo? [14:09] kenvandine: hello :) [14:09] larsu, same lp:~ubuntu-desktop/gtk/ubuntugtk310 [14:11] seb128: so can I consider my repo as merged and do new MRs on top of that one? [14:11] larsu, yes [14:11] s/repo/branch [14:13] good morning sil2100 [14:13] hey kenvandine, how are you? had a good w.e? [14:13] yup [14:13] and you? [14:14] same ;-) [14:20] seb128: https://code.launchpad.net/~larsu/gtk/properly-fix-settings-hacks/+merge/197379 [14:21] that should fix the problem :) [14:23] grrrrrrr [14:23] grrrrr [14:24] I can't even commit changes today, gedit keeps hitting those gdk_window_has_native assert errors [14:24] larsu, ok, merged, thanks! [14:25] seb128: that only happens with overlay-scrollbars, right? [14:25] * larsu can't reproduce it [14:25] larsu, I guess, I didn't try disabling those [14:25] if you have steps to reproduce, please let me know [14:26] larsu, EDITOR=gedit bzr commit is doing it half of the time here [14:26] like it took me 4 tries to be able to commit your gtk merge [14:27] .oO(obvious fix: EDITOR=gvim) [14:27] Sweetshark, you don't edit with writer? ;-) [14:27] seb128: doesn't happen for me, but I do see warnings [14:27] GtkScrolledWindow 0xc67880 is mapped but visible child GtkScrollbar 0xc62ed0 is not mapped [14:27] larsu, right, those are not new, they are there half of the time since saucy [14:28] seb128: only when I have a table of contents in my commit message. [14:28] larsu, the assert is happening when I close gedit btw [14:28] Sweetshark, so... never? ;-) [14:28] seb128: ah, before it saves I assume [14:28] I'm trying for the 20th time or so, works [14:28] :-( [14:29] I'm back on gtk 3.8 from trusty though, I wonder if that makes a difference [14:29] let me try with 3.10.5 [14:29] seb128: Indeed I cant remember having the need for that. [14:30] seb128: can you try that and also with GTK_MODULES= to disable o-s [14:30] just to check if that'S the problerm [14:30] *problem [14:31] larsu, sure can try, but knowing it started the day you made the o-s native, it's too much of a coincidence to be something else [14:31] ah okay [14:31] sigh [14:32] larsu, shrug, can't reproduce, let me try with gtk 3.8.7 again to see [14:32] larsu, got it on the second try with 3.8 [14:32] didn't get it with 10 tries with 3.10 [14:32] so maybe it's fixed in 3.10 [14:34] larsu, it seems to have something to do with the warnings you see [14:34] larsu, with 3.8 if I get the warning I get the segfault on closing [14:34] I never got it without having the warning before [14:34] with gtk 3.10 I get the warnings but no segfault [14:35] seb128: okay, I'll investigate the warning. [14:35] not having it is better anyway :) [14:35] thanks [14:35] right, those are annoying me for a while [14:35] they flood my stdout [14:35] also, warnings usually warn about something that might be a problem :) [14:35] right [14:36] larsu, so I'm not able to reproduce with GTK 3.10 either, which is good [14:44] Laney, not sure I found your translations more intuitive :p [15:00] seb128: I didn't know how to do the scaling without it [15:00] Laney, I'm trying to understand what you are doing/learning the canvas rotations and scaling stuff :p [15:00] you don't have to invert the y coordinate now! [15:03] kenvandine: could you sponsor a package for meee? [15:03] :) [15:04] sil2100, sure [15:05] kenvandine: https://code.launchpad.net/~ubuntu-unity/zmqpp/ubuntu <- there's a new version here (new ubuntu one) [15:05] kenvandine: would be cool if you could take a look and sponsor if ok :) [15:11] pitti: I don't have apport installed and want to get a core dump of a program. So I set ulimit -c unlimited, right? That results in a very weird issue: the command never returns after crashing, I can't ^C it, and I can't do anything in the directory I did this in (no ls, no rm) [15:11] pitti: any idea what might be the cause of this? Kernel holding a lock? [15:11] * larsu must reboot to make this go away [15:11] larsu: is it actually writing the file? [15:12] it might just be very big [15:12] pitti: I don't see the file after a reboot, so I guess not [15:12] when I set it to 10M I have the same problem... [15:12] argh, now did it in my home dir [15:13] and can't tab-complete anymore without the terminal locking up [15:13] $ cat /proc/sys/kernel/core_pattern [15:13] larsu: ^ what does that say? [15:14] core [15:14] that sounds fine [15:14] I saw a kernel core dump when rebooting just now [15:15] larsu: I got several kernel crashes recently too [15:15] go 3.12 [15:15] sigh [15:15] larsu: did anything make it to /var/log/kern.log before you rebooted? [15:16] larsu: did you reboot with sysrq, or just powercycle? [15:16] pitti: I didn't check. Is kern.log.1 the one from the previous boot? [15:16] larsu: no, should be in kern.log; .1 is from yesterday (logrotate) [15:16] pitti: sysrq? [15:16] kern.log seems clean [15:17] larsu: Alt+SysRq (usually on PrintScreen) plus "s"ync, "u"nmount, "b"oot [15:17] the direct keyboard → kernel link if everything in userspace is FUBAR [15:17] pitti: my system was usable, I restarted normally [15:17] that'll sync() and remount everything r/o, so that your disk stays clean [15:17] larsu: ah -- then you *can* check in another terminal what happens to "core"? [15:18] larsu: I thought you said even your terminal froze [15:18] pitti: any terminal in which I try to do anything to that directory freezes [15:18] like, ls [15:18] so I can't have a look whether core gets created [15:19] ouch [15:19] ah, found the kernel's bt further up in kern.log [15:19] larsu: I'd say, high time to reboot with a live CD and run fsck [15:19] if you care about your $HOME [15:19] * larsu cares about his $HOME [15:19] or init=/bin/bash and do it from there [15:19] (with unmounted or readonly /home) [15:20] ah, I think we even have this nice rescue mode in grub [15:20] okay thanks. I'll do that [15:20] we do? [15:20] that'll throw you into some curses dialog which has fsck, apt repair, and other goodies [15:21] ah nice [15:22] pitti: http://paste.debian.net/68870/ [15:22] gedit was the program I tried to get the core from [15:22] not that I understand much kernel stuff, but if I see BUG and ext4 in a kernel core I get the shivers.. [15:23] larsu: fsck o'clock! [15:23] sil2100, sponsored [15:23] kenvandine: thanks! [15:30] seb128: Do you have one of my poppler debdiffs locally? Looks like I deleted it from both my pc & LP :( [15:31] mitya57, http://people.canonical.com/~seb128/ubuntu-ubuntu.debdiff [15:31] seb128: merci! [15:31] mitya57, btw I just commented back on the bug [15:31] mitya57, de rien ;-) [15:32] mitya57, we have the usual "need to rebuild libreoffice, and for that needs to deal with the new toolchain and the issues it creates" [15:32] I would suggest waiting until Sweetshark is ready to upload 4.2 to trusty [15:32] there is nothing that looks like it's important in that poppler 0.24.3 -> 0.24.4 [15:33] tsdgeos, not nice of you of changing sonames in a stable update :-( [15:33] seb128: I'm fine with doing it later, but I'll upload updated diffs anyway [15:33] seb128: you know [15:33] when i write an email warning about it [15:33] i don't expect people to be surprised about it [15:33] tsdgeos, the issue is not being surprised, it's being able to roll out the update [15:34] well, i prefer the arch users getting the bugfix tha noone getting the bugfix [15:34] which is what would happen, no? [15:34] right [15:34] pitti: thanks for your help. fsck says everything's alright [15:34] except that now we get none of the other fixes either [15:35] or we need to revert that api change [15:36] i'm sorry, it's how we manage the project [15:36] we can argue again about it [15:36] but it is how it is [15:37] right, it's just an unfortunate situation [15:37] the Ubuntu archive rules mean that to be able to roll out that update with the new soname we need to rebuild everything with the new version, which includes libreoffice [15:37] which is not an easy thing to rebuild [15:37] oh well [15:38] no point arguing about it, we can delay until the day we are ready to rebuild the rdepends in trusty [15:39] larsu: does that dir with the core file work again? [15:40] pitti: yes, but the core file it not in it... [15:41] larsu: right, it already crashed in open(), not surprising [15:42] * larsu nods [15:42] if only this crash also happened in gdb, then I wouldn't have to deal with this problem :) === shengyao is now known as shengyao_afk [15:42] I blame seb128! [15:42] err, it doesn't happen in gdb? strange [15:43] larsu: apport doesn't catch it either? (i. e. with a core pipe) [15:47] pitti: "start: unknown job: apport" [15:47] 'apport' is installed [15:47] larsu: you have /etc/init/apport.conf ? [15:48] yes [15:48] larsu: your system sucks! [15:48] haha [15:48] * larsu hides [15:48]