[03:19] morning desktopers [04:12] hi [04:13] current gnome3-team/py-gi breaks software-center and therefore ultimately ubuntu-desktop. [04:14] The following packages have unmet dependencies: python-gi : Breaks: software-center (< 5.6.0-0ubuntu3) but 5.6.0-0ubuntu2 is installed. [04:14] The following actions will resolve these dependencies: [04:15] ubuntu raring + gnome3-team ppa [04:18] one way to fix that problem is for the sru for bug 1173249 to be accepted [04:18] Launchpad bug 1173249 in aptdaemon (Ubuntu Quantal) "update-software-center AttributeError during upgrade from 12.10 to 13.04" [High,Triaged] https://launchpad.net/bugs/1173249 [04:49] Good morning === dpm-afk is now known as dpm [06:50] good morning [06:53] hey jibel! [06:54] * pitti donne une accolade à didrocks [06:54] * didrocks donne une accolade à pitti [06:56] Bonjour didrocks === greyback|away is now known as greyback [07:18] didrocks, I searched why I couldn't limit swap usage in containers yesterday evening. It's actually enabled in the kernel but resource control must also be enabled on the command line with swapaccount=1 [07:18] didrocks, with that parameter on the kernel stops the process that tries too allocate too much memory [07:18] didrocks, I'll set it on our platform and maybe we could retry the hud in otto ? [07:18] jibel: oh, so nice! doesn't need any recompile then! Those has to be set on the host as well, or just the guest? [07:19] didrocks, the boot parameter is on the host. [07:19] jibel: sure, but we'll have to trick is to have the mem leak version. I'm running the fixed one from tonight right now monitoring [07:19] and it seems the mem leak is indeed fixed [07:19] didrocks, there is a price to pay for resource control, that's why it's disabled by default I guess [07:20] right, we don't really care on our tests machines though :) [07:20] didrocks, control of 1G of swap costs 512K of memory [07:20] jibel: affordable for us seeing what can be the result otherwise :p [07:21] jibel: btw, the -check seems to either have one wrong computation set or we had one failure on hud where we accept 0, we'll need to check but the message isn't informative [07:21] (this is not otto, just the collect result) [07:22] didrocks, I also disabled log rotation so log files do not vanish under our feet [07:22] ah, excellent! [07:22] is that on the new container? [07:23] (ok, we passed the point where the machine memory was totally out of control yesterday for hud tests) [07:23] I guess there are some slight leak still [07:23] but we are at ~650M total ram size, including swap [07:25] I wonder why we have some regular: [07:25] tail: /var/lib/lxc//raring-i386-20130603-2203/run/delta//var/local/autopilot/autopilot.log: file truncated [07:25] especially has we have it through an exec now [07:26] didrocks, ah, the redirection is misplaced [07:27] * didrocks looks again for the test branch :p [07:27] didrocks, r25 of the testsuite [07:27] jibel: do you have the lp: address? [07:27] jibel: my laptop is closed [07:28] didrocks, http://bazaar.launchpad.net/~otto-dev/otto/testsuite_autopilot-unity/revision/25 [07:28] * didrocks branches [07:29] jibel: needs to be exec >>$AP_LOGFILE btw [07:29] jibel: if we restart the session… [07:30] didrocks, done [07:30] jibel: ok, tell me once you are ready for continuing :) [07:30] hud passed [07:31] didrocks, ready [07:31] so fast? ok, coffee and coming :) [07:41] RAOF: hey Chris, how are you? [07:45] pitti: Pretty good. [07:45] pitti: Yourself? [07:45] RAOF: quite fine, thanks! I'm glad that the two weeks of continuous rain seem to be over at last [07:45] RAOF: do you plan a colord 1.0 upload in the next time for Debian? [07:46] pitti: Yeah, it's mostly staged in git. [07:46] RAOF: I was looking at the two failures in https://jenkins.qa.ubuntu.com/view/Saucy/view/AutoPkgTest/job/saucy-adt-colord/25/ARCH=amd64,label=adt/ [07:46] RAOF: missing valgrind test dependency, and importing the upstream commit for dropping g_type_init() [07:46] could we include them, please? [07:46] Certainly. [07:46] I'm pretty sure I've added valgrind to the test dependencies locally :) [07:47] Pushing it just requires (a) finishing checking that everything actually works and (b) marrying inclination to opportunity. [07:49] heh [07:50] good morning desktopers [07:52] bonjour [07:54] salut [08:00] morning! [08:02] seb128: continuing from yester-night [08:02] $ dpkg -L poppler-data | grep -i janan [08:02] /usr/share/fonts/cmap/adobe-janan2 [08:02] Laney, good morning to you too [08:03] someone typoed in there [08:04] tsdgeos, what would be the correct name/path? [08:04] japan [08:04] japan2 [08:05] but not janan2 [08:05] tsdgeos, ok, got it [08:05] debian/links:usr/share/poppler/cMap/Adobe-Japan2 usr/share/fonts/cmap/adobe-janan2 [08:05] tsdgeos, thanks, I will get it fixed in both Debian and Ubuntu [08:06] seb128: tx [08:07] tsdgeos, do you have a bug report/example of issues created by the broken name (just to know if we should get the fix in raring as well) [08:07] seb128: i don't [08:07] it was brought to my attention yesterday by someone that would not share his file [08:08] tsdgeos, ok, I will see if I can get a SRU for it still, but it usually help to have a testcase for verification of the fix [08:08] though the error/fix is obvious so that might not be required there [08:09] tsdgeos, btw we got a slightly less outdated poppler in saucy (the current stable one) [08:09] just for info [08:09] awesome :-) [08:10] we might go for 0.24 later in the cycle (once the private abi stop changing ;-) [08:17] hmm no jbicha? was wondering why realmd was not updated all the way to 0.14.2 [08:19] ah, https://bugs.freedesktop.org/show_bug.cgi?id=65325 would explain it :) [08:19] Freedesktop bug 65325 in General "Can't build realmd without systemd" [Normal,New] [08:59] tjaalton, mlankhorst: hey, what ppa do you guys use as a staging for the new xorg aimed at saucy's upload? [09:01] seb128: canonical-x/x-staging [09:03] tjaalton, do you have an estimate of the level of testing that got so far? [09:04] seb128: we've used it [09:04] number of users that tested it, was it tested on all main drivers (ati, intel, nouveau, nvidia, fglrx) with unity, barrier pointer, etc? [09:04] yes [09:04] :) [09:04] ok [09:04] actually, fglrx only got abi support recently, so it's not there yet [09:05] sorry to ask again about that, some people want to make really sure the update is not going to create issue for the unity team [09:05] worth us testing it? [09:05] do you have any estimate when it will be? do you plan to do a call for testing on ubuntu-devel@ or something when you get close from upload? [09:05] aiui the blocker is unity [09:06] what about it? didn't you have a patched version in the ppa? [09:06] getting an official upload with the new pointer barrier work [09:06] the ppa has a hacked version [09:06] that is kept uptodate so it'll install [09:06] guess mlankhorst has the most recent info about the status [09:11] tjaalton, do you know if there is a bug report/merge request for the new pointer barrier? [09:15] seb128: hmm no [09:15] there should be I guess [09:15] that would be useful indeed [09:16] bug 1097991 [09:16] Launchpad bug 1097991 in unity (Ubuntu Raring) "Port to new barrier event API" [Undecided,New] https://launchpad.net/bugs/1097991 [09:16] this is what the ppa version has https://code.launchpad.net/~brandontschaefer/unity/move-pointer-barrier-to-xi-1.6.99.1/+merge/150175 [09:17] tjaalton, hum, that got merged in Unity trunk? what happens if you try running unity trunk on saucy with xorg 1.13? [09:18] oh, no, it's not merged [09:18] ignore me [09:18] it'll likely fail [09:18] not sure [09:19] afk -> [09:19] tjaalton, thanks === vrruiz_ is now known as rvr [09:33] desrt: It looks like the conversion to installed tests has made the gsettings tests use the installed glib-mkenums unless $GLIB_MKENUMS is set to something else in the environment [09:34] shouldn't it prefer the built path if it's available to work with make check and as an installed test? [09:34] woah, wait [09:35] desrt: I think you already fixed this in trunk :P [09:41] seb128: yeah it will land with the new unity, the unity testing will be done with the x ppa enabled [09:41] wow! [09:42] how did I use git for so long without knowing about git format-patch ^! [09:42] such a luddite [09:42] or a whole series, ... [09:42] yes I knew that one [09:42] that's /all/ I knew [09:42] but more often I just want to take one commit [09:42] I hate that you need to know stuff like "^" in git [09:43] I don't know such a shortcut in bzr [09:43] I only know how to do ranges there [09:43] what does it do in git? [09:43] it's the ID of the previous commit (probably in reality it's more complex, but that's how I think of it) [09:44] and ! gets you the path or something, so ^! is like n-1..n [09:44] my git knowledge isn't at a very technical level [09:44] yeah, mine either [09:45] bzr is nicer :p [09:45] shame everyone wants to use git :/ [09:46] Laney, in bzr you can do -c-1 or -c-2 [09:46] to see the diff of n-1 or n-2 [09:46] I usually want to say "give me the diff of n" to cherry-pick it [09:47] so I guess that's bzr diff -c-1 -r n [09:47] bzr -c [09:47] ? [09:47] bzr diff -c [09:47] for the diff of commit [09:47] ah [09:47] I've been doing something with ranges ¬_¬ [09:47] where can be a number of a a relative position (-1, -...) [10:04] try #2 of getting nouveau running on this laptop === Sweetsha1k is now known as Sweetshark [10:08] oh crap, there are even more registers! [10:24] tsdgeos, http://lists.debian.org/debian-devel-changes/2013/06/msg00386.html btw [10:25] tx [10:26] thank you for pointing the issue ;-) [10:29] uh, any idea guys why firefox in saucy complains about an unreachable proxy when using system-wide proxy settings with _no_ proxy configured? [10:55] free and open source has finally moved to the dark side: Lawrence Lessig is guest at this years Bilderberg meeting. [11:19] blurg [11:32] pitti, you're not affected by the flooding in europe, are you? === MacSlow is now known as MacSlow|lunch [12:04] chrisccoulson: my parents are, they live right at the bank of the Elbe, and watching it like a hawk [12:05] chrisccoulson: personally I'm not === alan_g is now known as alan_g|lunch [12:37] Laney: ya...wasn't too happy with the original installed tests work, so i spent most of last week fixing it up [12:37] seb128: i notice activity on https://bugs.launchpad.net/ubuntu/+source/linux/+bug/922968 [12:37] Ubuntu bug 922968 in OEM Priority Project precise "shouldn't queue a second suspend if the machine is already suspending" [Medium,Confirmed] [12:37] desrt: yeah, seems nicer still on trunk [12:37] seb128: this is fixed now, no? [12:38] Laney: thanks for the ping :) [12:38] desrt: currently seeing async-create-delete failing; do you know of anything like that? [12:38] GLib-GIO:ERROR:/build/glib2.0-cCE0t2/glib2.0-2.37.1/./gio/tests/file.c:478:test_create_delete: assertion failed (data->monitor_changed > 0): (0 > 0) [12:38] hmm. [12:39] could be an inotify/fs thing so I uploaded to a PPA [12:39] Laney: you're running uninstalled or uninstalled? [12:39] desrt, it should, can you comment on the bug saying that? (nobody did a status update there it seems) [12:39] make check [12:39] Laney: that's news to me. i wonder if the change notification simply comes too late [12:39] GFileMonitor has some built-in delays [12:39] in order to group changes [12:40] seb128: we fixed it in sassy, right? no backport to older releases, afaik... [12:40] keep an eye on https://launchpad.net/~ubuntu-desktop/+archive/ppa/+build/4640842 [12:41] * Laney pops out to have lunch in the park [12:42] desrt, right, it's not going to be trivial to backport and I'm not sure how much they need/want a backport or if they are happy to have it fixed for the next lts [12:42] desrt, first step is to say that and see what they reply ;-) [12:43] seb128: i just write some info [12:43] desrt, thanks [12:43] *wrote [12:43] * desrt needs coffee :) === MacSlow|lunch is now known as MacSlow [12:57] pitti - yeah, it looks bad :( i hope it improves for them soon [12:58] this summer is certainly turning out to be quite dreadful :/ === tkamppeter_ is now known as tkamppeter [13:01] chrisccoulson: it's 22 down here today, summer arrived! [13:01] czajkowski, yeah, here too. but in context, it's still been pretty bad. and i suspect the nice weather for us this week will be short-lived ;) [13:03] oh up till now it's been awful, but the 10 day forecast has a lot of BBQ weather in the future. [13:03] i don't trust 10 day forecasts :P [13:03] * chrisccoulson is feeling pessimistic ;) [13:07] although, perhaps i should chance it and invite people round for a BBQ. i have more beer at home than i could possibly drink on my own :) [13:24] chrisccoulson: spoken like a true englishman === alan_g|lunch is now known as alan_g [13:24] ( between the distrust of the weather, the pessimism, and the beer ) [13:27] heh :) [13:28] chrisccoulson: did you see. http://www.buzzfeed.com/expresident/best-ways-to-annoy-a-british-person [13:31] lol @ #7 [13:36] hah :) [13:36] i was looking through some of the comments too [13:36] "I know the easiest way to annoy a brit is give them a hersheys bar and tell them its good chocolate" [13:40] haha [13:41] LOL [13:41] chrisccoulson: I find the luke warm water for my tea more annoying [13:41] or been given ice tea! and told it's nice! [13:41] [FULLYBUILT] Successfully built on chindi11 (arm ppa builder) [13:41] grr [13:42] Laney, trying to debug an issue and it's not hitting the bug? [13:42] test fails in sbuild here but not on the ppa [13:42] * Laney tries a bare metal (well, lxc) build [13:53] I suspect this is bug #882147 [13:53] Launchpad bug 882147 in coreutils (Ubuntu) "overlayfs does not implement inotify interfaces correctly" [Undecided,In progress] https://launchpad.net/bugs/882147 === m_conley_away is now known as m_conley [14:19] seb128: ping? [14:40] can we have a libunity-webapps rebuild for the packagekit transition? [14:41] kenvandine: ↑ maybe? [14:42] (if that's your baby) [14:46] 8. Violate the sanctity of a queue. [14:46] my favourite. [14:47] Laney, sure [14:47] robru tends to those, but if you need it before he wakes up i can do it [14:48] doesn't seem massively urgent [14:48] ok [14:48] there's still quite a few packages waiting according to http://people.canonical.com/~ubuntu-archive/proposed-migration/update_output.txt [14:49] jbicha: you're making the needed rebuilds happen? === davidcalle_ is now known as davidcalle [14:52] Laney: I think everything else that needs a rebuild has already been started [14:53] ok [15:11] Sweetshark, hey (back from exercice) [15:12] Sweetshark, saw your /msg, no worry if you are not back for the meeting, I will paste your summary for you [15:13] seb128: Im still around, was a bit delayed and had no reply, so will stay for the meeting now (and fly out of the door right after it ;) ) [15:13] Sweetshark, ok [15:17] whoo dates for FOSDEM are out already for 2014 folks [15:18] 1&2nd of Feb [15:23] czajkowski: that is not exactly unexpected ;) [15:23] Sweetshark: well it kinda is as it's usually the following weekend [15:24] I spent whole day looking at register dumps, I'm exhausted :/ [15:27] czajkowski: Really, isnt it the first weekend of Feb? [15:27] (usually) [15:28] no it used to be valentines weekend practically, and then got brought forward more reently [15:30] Sweetshark, qengho, Laney, mlankhorst, tkamppeter, attente, desrt, larsu: hey, it's meeting time [15:30] moin! [15:30] yo [15:30] hey [15:31] meeting \o/ [15:31] let's get started [15:31] Sweetshark, hey [15:31] - annoyed everyone with a LibreOffice upload, tweaks to fix against the moving baseline [15:31] - upstream LibreOffice 4.1 hassle [15:31] - LibreOffice 4.1 beta1/2 packaging (not yet published) [15:31] - improving tinderboxing for libreoffice.org [15:31] EOF [15:31] hi [15:32] larsu: i love meeting [15:32] Sweetshark, glad to see that we got libreoffice to build in saucy, thanks to you (and to Laney for fixing the remaining issues) [15:32] seb128: oh, indeed. [15:32] Sweetshark, thanks ;-) [15:32] * Sweetshark bows to Laney. [15:32] qengho, hey [15:32] I was violently sick when I saw it regenerated its control file during a build [15:33] ...the debian is strong with this one [15:33] hum, no qengho (again)? [15:33] 3 times in a row, /me notes to check out on qengho [15:33] Laney, hey [15:33] • The week of the archive transition™ [15:33] • gtkspell3 transition; patch and upload some packages [15:33] • eds transition; patch a few packages, submit patches upstream, test and upload [15:34] • update evolution [15:34] (check up on*) [15:34] • update empathy (rebasing both on Debian) [15:34] • review and sponsor a d-conf split to Debian for Seb [15:34] • help out on some packages for the poppler transition (libreoffice ... dear god); get that migrated [15:34] • update glib2.0; testsuite failures; debug, now testing before uploading later [15:34] • update glib-networking; similarly testing & will upload along with glib [15:34] • now upgrading n7 to saucy images & will start to look at system settings stuff [15:34] eof [15:34] seb128: Sorry, late again. [15:34] Laney, thanks [15:34] - preparing chromium 27 for testing and release. [15:34] +-- awaiting #webapps updates. [15:34] +-- already testing armhf/precise for specialty installations. [15:34] - will package an installer for new AMZN extension later this week. [15:34] EOF [15:35] qengho, ah, no worry, thanks for the update ;-) [15:35] mlankhorst, hey [15:35] i think lts-raring is in archive now, not sure if it's moved to main yet or still in universe. I've made nvd7 acceleration work with nouveau (external firmware, going to try nouveau firmware tomorrow), some other stuff but my brain is offline from comparing tedious stuff all day to get nvd7 to work.. [15:35] oh and birthday \o/ [15:35] mlankhorst, thanks [15:35] oh, today? [15:36] mlankhorst: happy level-up [15:36] mlankhorst, happy birthday! [15:36] * mlankhorst has gained no new skill, but a working laptop! [15:36] lol [15:36] ...oh what i would give for a working laptop [15:36] some cold hard canadian dollars [15:37] anyway ^D [15:37] ;-) [15:37] mlankhorst, thanks [15:37] tkamppeter, hey [15:37] - Ghostscript developers have added PWG Raster support to MuPDF, Mike Sweet told that he wants to work on mupdftoraster wrapper filter [15:37] - Got patches to start CUPS daemon on-demand via xinetd and systemd [15:37] - Answered and triaged printing-related bug reports [15:37] - Caught up on e-mail after vacation [15:39] tkamppeter, cups on-demand patches ... didn't you have some for systemd from fedora already? getting any for upstart? [15:39] phwoar, xinetd [15:40] seb128, these are the ones from Fedora which I got. Unfortunately, there are none for Upstart. I have to work wwith an Upstart expert on them. [15:40] Laney: no. it's new!! [15:40] and ... different! [15:40] argued about patches all morning! [15:41] tkamppeter, thanks, if you need help with upstart maybe ask on #ubuntu-devel to jdoh/xnox/stgraber [15:41] er. wrong channel. sorry. [15:41] attente, hey [15:41] seb128, hi! [15:41] finished up the shortcut widget for the region panel [15:41] indicator-keyboard and region panel i18n [15:41] double menu bar thing [15:41] no luck debugging an ibus/g-s-d issue, will try again [15:41] started writing some code to generate svg files for all of the keyboard layout icons [15:42] EOF [15:42] seb128: mrahw [15:42] attente: also, macos gtk menubar integration work [15:42] desrt, wrong channel again? [15:42] haha [15:42] attente: no. this is something you did this week... [15:42] desrt, not on work time [15:42] o O [15:42] I was going to say :p [15:43] attente, I saw your gtk3 merge request, I will try to get to it today [15:43] seb128, cool, thanks [15:43] :) [15:43] attente, thanks [15:43] desrt, hey [15:43] hey [15:43] started out by chasing some test failures... that lead me to figure out that the state of our build system wrt. testing is pretty sore [15:44] so i spent most of the week redoing the test framework in glib in a way that can be copied into other projects (made some m4macros and a makefile for inclusion) [15:44] seems to be pretty well-received and should start appearing in other projects soon [15:44] did some more work on action descriptions meanwhile (but not too much) [15:45] and yesterday i did a lot of improvements for g_file_set_contents(), which should serve to reduce fragmentation when used on ext4 and improve performance by quite a lot everywhere (due to not calling fsync in cases where we don't need it) [15:45] also did some chatting with larsu about the qmenumodel stuff he's been working on [15:45] fin. [15:46] desrt, thanks [15:46] larsu, hey [15:46] hey, let me try the copy n paste thing that's so popular here... [15:46] landed gtk menu item patch (thanks seb128!) [15:46] landed ido custom menu item framework [15:46] hooked volume slider up to that and fix some related issues in the indicator-sound branch [15:46] helped renato with some indicator-messages fixes (which were essential for the dog food thing) [15:46] ported qmenumodel to use gtkmenutracker (thanks desrt!), so that we can have all the features in the new shell [15:46] upcoming: finish up qmenumodel; add media player widget into ido; finish up indicator-sound [15:47] oh .. errm ... EOF?! [15:47] ;-) [15:47] 'over and out' [15:47] seb128: your turn! [15:47] larsy: do we have a working indicator-backend -> qmenumodel -> panel rendering (or other rendering for testing) chain? [15:47] desrt, in a min ;-) [15:48] larsy is like larsu + Laney? [15:48] lol [15:48] seb128: well, I have a qmlscene thing that renders a menu. Need to talk to nick to integrate it into unity [15:48] larsu, ^ [15:48] grand larsy [15:48] seb128: instead of using qmenumodel-as-it-is-now [15:48] larsu: working except for submenus now, right? [15:49] desrt: and a couple of properties aren't exported yet, but that's a half-hour fix I plan to do after lunch [15:49] larsu: does it mean we can plug gmenumodels to qml (I'm asking for system settings in mind) [15:49] seb128: yes, this is exactly that [15:49] larsu: did you do the 'view' work yet? [15:49] desrt: no [15:49] or is someone else going to do that? [15:49] (or is it already done from before?) [15:49] larsu, pieces start coming together but it's hard for people who will work on system settings to know where to start [15:50] seb128: right. And as desrt says, we need someone to do the view part (mainly the custom menu items) for qml [15:50] ok [15:50] seb128: who's working on that, mardy? [15:50] larsu, thanks [15:50] desrt: "someone" will do that [15:51] larsu: it starts to feel like we have a lot of data shuffling going on here :) [15:51] apps -> gmenumodel -> dbus -> gmenumodel -> gtkmenutracker -> qtabstractlistthingy -> qtlistviewthingy -> custom renderers [15:51] larsu, mardy is working on the container app and the plugins loader, individual panels are being to be done by quite some people (see the Cc list on the email I sent recently) [15:51] seb128: right, thanks. [15:52] desrt: you're exaggerating [15:52] larsu: not by much [15:52] larsu, thanks, I will also set up a meeting about system settings later in the week, you will get an invite [15:52] ok, my turn [15:52] i'm not saying that i think it will be a problem [15:52] * some desktop merges and updates [15:52] * some work on the versions script (fixed some issues, updates some broken links) [15:52] * worked on the poppler transition [15:52] * some NEW reviews [15:52] * continued looking at the system settings work and qml, trying to figure out how the pieces fit together with indicator backends [15:52] * some bug fixing [15:52] [15:53] oh, poppler transition, it's that time of year :) [15:53] yeah :p [15:53] that's it [15:53] seb128, let me know if you get blocked figuring out anything with settings plugins [15:53] any other question/topic/comment? [15:53] weather :-) [15:53] * desrt wants to grill larsu a bit, but will do that privately :) [15:53] * larsu doesn't know the second meaning of grill. And is afraid [15:54]