=== nudtrobert1 is now known as nudtrobert === Guest39753 is now known as dad98253 === nudtrobert1 is now known as nudtrobert === Guest22203 is now known as dad98253 [04:31] Is this the correct place to ask a question about re-compiling the unity desktop on ubuntu? [05:49] Hello [05:56] good morning [07:13] Bonjour tout le monde ! [07:16] bonjour pitti ! [07:16] comment ça va ? :) [07:16] Laney: FYI, the repeated "restart all workers and 5 of them die of quota issue" is due to the undeletable instances in bos01 (you see the ~ 100 h old instances in cron mail), RT has been pending since Friday to clean them up [07:17] but it's been remarkably quiet over night otherwise [07:42] good morning desktopers [07:42] hey didrocks pitti [07:42] ça va seb128 ! [07:42] oui ! et toi ? [07:42] je vais très bien, merci ! [07:43] j'ai joué au padel et au tennis hier, c'était bien [07:43] bonjour seb128! [07:43] lut didrocks :-) [07:43] seb128: "padel" n'est pas dans mon dictionaire.. [07:44] seb128: c'est similaire à squash et tennis ? [07:44] pitti, https://en.wikipedia.org/wiki/Padel_%28sport%29 [07:44] pitti, c'est un nouveau sport, oui un peu entre le squash et le tennis === Guest43325 is now known as fredp === fredp is now known as Guest88016 [08:12] yawn [08:12] good morning everyone! [08:14] hey larsu! how are you? [08:14] tired [08:15] but good otherwise, thanks! And you? [08:15] I'm good thanks! [08:15] you didn't sleep well? [08:16] good morning larsu! [08:17] hey larsu! [08:17] * pitti hands larsu a nice cup of tea [08:19] morning didrocks and pitti [08:19] and thanks :) (I just made one in fact) [08:19] seb128: ya, woke up a lot. But nothing bad [08:20] nothing that a tea cannot fix! [08:20] good :-) [08:36] happyaron, hey, you said you would look at https://code.launchpad.net/~brunonova/ubuntu/trusty/ibus/lp1240198_2/+merge/278226 right? [08:56] morning all [08:56] hey willcooke [08:57] seb128, I have my kids Christmas play at school until ~ 3:15pm. Could you stand in for me if I don't get back in time for the meeting. I'm sure I will, but just in case. [08:57] hey didrocks [08:57] hey willcooke [08:57] willcooke, sure, no problem [08:57] thx [09:01] hi hi [09:01] Morning laney [09:02] what's up willcooke [09:03] hello Laney [09:05] hey didrocks!!!! [09:05] what's up [09:05] hey Laney [09:06] ooh la la, a seb128 [09:06] :-) [09:07] Oh, it's public holiday in Italy today so Trevinho is off. Didn't find out until after I'd sent the email though. [09:07] bonjour Laney ! [09:07] Laney: finally figured out some black magic from yesterday! :) [09:08] just need monkey updating all plymouth community themes now [09:08] (but that will be a task for tomorrow, enough typing/updating) [09:09] hey pitti [09:09] how's the lyon visit? [09:09] is your host treating you well? [09:09] Laney: I have to eat snails and raw mussels all day! [09:10] *facts* [09:10] yum yum [09:10] Laney: nah, it's lovely to hang out with didrocks, I enjoy it [09:10] and they have wine taps instead of water [09:10] just a bit sad that the light festival got cancelled [09:11] yeah, shame [09:11] Laney: well obviously it's in the jungle the mighty jungle that the lyon sleeps tonight, so I would imagine it involves hacking through huge swathes of green stuff ;) [09:13] awoooooo [09:15] Laney: close it's awimaway awimaway awimaway :D [09:16] ah morning Laney & willcooke [09:17] hey! [09:18] bah, of course new nautilus has a pile of segfaults :-/ [09:18] * seb128 upstreams some [09:21] hi larsu! [09:21] how's it going? [09:22] slowly waking up :) [09:22] and you? [09:22] willcooke: had to reboot this morning, something took out dbus and therefore network manager [09:22] sweet [09:22] willcooke: nm apport auto reported [09:25] larsu: good! went climbing last night for the first time in a week-ish, couldn't go for various reasons before then [09:25] * Laney raaargh [09:26] how was the quiz? ;-) [09:26] ah cool [09:26] fun! What's the most eastern state of the EU? [09:26] this was surpisingly hard for us last night (but we got it right) [09:26] and we were 2nd place \o/ [09:27] oh good question [09:27] I guess ... Romania [09:27] WAIt [09:28] I DONT [09:28] cyprus? [09:28] nice! [09:28] we were debating between bulgaria and cyprus [09:28] almost forgot about those guys down there [09:28] and were quite embarrased when we saw how far east cyprus is :) [09:28] it's not even a contest [09:28] malta too but I guess they are pretty west [09:29] * Laney looks at a map [09:29] yeah Malta is South of Italy [09:29] haha [09:29] that's actually closer to Turkey than I thought [09:29] and you're right, Romania goes a fair bit further east than Bulgaria [09:30] https://en.wikipedia.org/wiki/Romania#/media/File:EU-Romania.svg [09:30] the question I scored in was "name at least 5 of the Canary Islands" [09:31] it helps to have been there a few times :) [09:31] it's going to be annoying for some wikipedian to have to recolour the UK grey in all of these maps [09:31] man I would have no idea about that [09:31] good knowledge [09:31] Laney: don't leave us :'( [09:31] Laney: ya, got all of them [09:32] would have probably got lanzarote and tenerife [09:32] but only because he said there are 7 - would have missed el hierro otherwise [09:32] haha [09:32] it's the christmas quiz at our pub tonight [09:32] * Laney has his new jumper on in celebration [09:33] la gomerra, la palma, gran canaria, fuerte ventura, tenerife, lanzerote, el hierro [09:33] Laney: pics or it didn't happen! [09:33] ... or it won't have happened?! [09:35] larsu: Don't be daft he won't have it on now there are no video calls, Laney will still be in his coding pants :D [09:35] haha [09:35] I sent it via a private messaging platform often used to subvert governments [09:37] * davmor2 scourers the darknet and now wishes he hadn't looked for Laney dresses up :D [09:39] for the record, it looks amazing [09:39] yeah... never scour the dark net for anything [10:11] open telgram app. switch to 2× hidpi. switch back to 1×. Feel happy about mickey mouse sized hand cursor (only on Telegram app) [11:15] Laney: doesn't += add a space as well? [11:15] I wonder what changed to make this a problem now [11:17] apparently not [11:17] or not any more, something obviously changed :( [11:18] can't imagine... [11:21] Laney: hm, according to the docs this only works in makefiles, no in shell [11:21] *not [11:21] 'this'? [11:22] += adding whitespace [11:22] maybe $LIBM was "" before [11:22] in any case, thanks for the catch [11:22] that seems unlikely [11:22] any other idea? [11:22] maybe glib had a trailing whitespace in its pcfile [11:22] or something [11:22] ah that could be as well [11:23] https://launchpad.net/ubuntu/+source/libindicator/12.10.2+14.10.20140922-0ubuntu2 ... [11:23] top approve or do you want to wait for seb? [11:23] * Laney screams [11:23] lol [11:23] xnoooooooooooooox [11:23] yeah feel free to top approve [11:24] will upload it and clobber this upload [11:24] thanks! [11:30] good morning [11:30] was i meant to be on IRC before uploading things that block s390x bootstrap or some such? =) [11:30] larsu, Laney ^ [11:30] probably at least check merge proposals / propose your changes [11:31] hi xnox! [11:31] Laney, i like your solution better. [11:31] but i am shorter. [11:32] 2015-12-08 11:32:07,182 WARNING ============== POINT OF NO RETURN ============= [11:32] CI train is funny [11:34] larsu: do you know if there's a convenient way to run gvfs out of jhbuild? [11:34] I have 10 or so gvfs-* processes running [11:35] if I kill them and run gvfsd is taht enough? [11:35] in a jhbuild shell^ [11:37] yes [11:37] it dbus activates them iirc [11:38] that will get the right libglib? [11:38] ah wait, not sure jhbuild adds dbus activatable stuff [11:38] the jhbuilt one [11:38] heh [11:38] but you can run them manually [11:38] the helper proceses, I mean [11:38] and then it will definitely take the already-running ones [11:38] dunno which one I want [11:38] probably -trash [11:38] start them all :) [11:38] but this has some -spawner argument [11:39] * Laney wibbles [11:39] just run it in the same way that the dbus service file does [11:39] ugh that doesn't exist :/ [11:40] * Laney tries what happens just running gvfsd [11:41] oh yeah that worked [11:41] didn't do what I expected though, RATS [11:41] ah, it has special mount files and always mounts the trash backend [11:41] /usr/share/gvfs/mounts/trash.mount [11:41] so you'll need to restart it manually [11:42] it seems to spawn the backends from the main gvfsd [11:42] or kill it and hope that the gvfsd from jhbuild uses the correct one [11:42] just starting this worked [11:42] ah cool [11:42] that's my cue: going to lunch D [11:43] enjoy! [11:44] thanks :) [11:46] hm, lunch date just moved [11:46] * larsu finishes his patch then [11:55] hey, is there already a public development repository for 14.04.4 HWE? (wily kernel & stuff) [11:56] try #ubuntu-kernel? === alan_g is now known as alan_g|lunch === hikiko is now known as hikiko|ln [13:42] happy tuesday [13:42] hey desrt [13:42] hey desrt! [13:42] good morning, EU residents [13:45] happy tuesday desrt [13:47] hihi didrocks [13:49] hey desrt [13:49] greetings seb [13:49] * desrt makes some coffee === alan_g|lunch is now known as alan_g === hikiko|ln is now known as hikiko [14:45] willcooke, Could you spare me a moment? [14:46] He said he was out this afternoon [14:46] You should probably just ask [14:48] I'd like to include GNOME Software + Ubuntu plugins in Ubuntu MATE. [14:49] I'd like to understand what packages require installing so I can test this and integrate into the seeds. [14:50] I appreciate it is early days and not everything will be ready yet. [15:09] flexiondotorg, we don't even have gnome-software in the archive yet, it requires packagekit 1.0 which is blocked on other work [15:09] seb128, OK. Understood. [15:20] ARGH [15:20] WHY IS MY COMPUTER BEEPING [15:20] Alarm Tue, 15:30 Desktop team weekly meeting [15:20] where did that come from? [15:21] gnome-calendar? :) [15:21] e-d-s? ;) [15:21] no I know what the eds ones look like [15:22] could be gcal but then why didn't it happen on the laptop too? [15:23] also I doubt that has a service [15:24] yeah… [15:24] apart if it's opened [15:24] not [15:28] no beeping on my xenial [15:28] any ideas? [15:29] gnome-calendar doesn't have a service either [15:29] do you have it open? [15:29] no [15:29] k, so not it [15:29] indicator stack didn't change, so not likely it either [15:29] some phone thing? [15:29] I doubt it [15:29] as said there were no recent changes there [15:29] what then [15:29] e-d-s 3.18 maybe? [15:29] didn't get it last week [15:30] they added notifications [15:30] maybe you enabled some settings when testing gnome-calendar? [15:30] or maybe it did it for you [15:30] but for now it's meeting time! [15:30] * Laney makes another event [15:30] #startmeeting [15:30] Meeting started Tue Dec 8 15:30:48 2015 UTC. The chair is seb128. Information about MeetBot at http://wiki.ubuntu.com/meetingology. [15:30] Available commands: action commands idea info link nick [15:30] hey [15:31] hello [15:31] Roll call: andyrock, attente, desrt, dgadomski, didrocks, fjkong, happyaron, hikiko, laney, larsu, qengho, seb128, sweet5hark, themuso (out), tkamppeter, trevinho (out), robert_ancell (out) [15:31] aye [15:31] seems like willcooke is not back yet, so I'm going to start this one [15:31] .... .. [15:31] thanks seb128 :) [15:31] word up [15:32] hey [15:32] ok, let's get started [15:32] no andyrock so he might be out [15:32] #topic attente === meetingology changed the topic of #ubuntu-desktop to: attente [15:32] hey attente [15:32] hey seb128 [15:32] gtk-mir backend. [15:32] (eof) [15:33] bah no beep that time [15:33] thanks attente [15:33] #topic desrt === meetingology changed the topic of #ubuntu-desktop to: desrt [15:33] desrt, hey [15:33] gsettings. [15:33] (eof) [15:33] :) [15:34] * larsu is looking forward to didrocks this week [15:34] but seriously... did some good work on the dconf confinement stuff last week in montreal with xclaesse [15:34] biggest change here is what i am working on now: DConfChangeset is being imported into glib as GSettingsBackendChangeset [15:35] which is something that i've vaguely wanted to do for a couple of years now... it's ugly as all hell, but it's going to make a lot of things easier... not least of which is the current work at hand about confinement [15:35] the import itself is done (as a patch) but i'm currently porting various backends to using it [15:35] so i'll probably have that done by next meeting [15:35] (real eof) [15:35] great [15:35] looking forward to see that land ;-) [15:36] thanks desrt [15:36] me too [15:36] np [15:36] #topic dgadomski === meetingology changed the topic of #ubuntu-desktop to: dgadomski [15:36] dgadomski, hey [15:36] hey [15:36] I'm chasing one new issue - Unity (and Gnome 3) crash on a VM running under PowerKVM. No useful info in the logs, waiting for /var/crash data. Direct rendering is on (according to glxinfo), so I guess it's some kind of a bug. I'll create a lp bug after getting more input. I guess there's nothing preventing Unity from running on ppc64? [15:37] * created https://wiki.ubuntu.com/DebuggingPolicykit, will work on putting more info there. Will appreciate any feedback, ideas [15:37] * still running Xenial in several VMs tests to verify fix for bug #1337873, no problems so far [15:37] bug 1337873 in ifupdown (Ubuntu Vivid) "Precise, Trusty, Utopic - ifupdown initialization problems caused by race condition" [Medium,Confirmed] https://launchpad.net/bugs/1337873 [15:37] EOF [15:37] not sure about ppc64 graphical stack... [15:37] thanks dgadomski [15:37] seb128: do you know who can know more about ppc64? [15:38] try asking on #ubuntu-devel maybe [15:38] seb128: will do, thanks [15:38] but I doubt we have many people who know about unity/graphical stack on ppc64el [15:38] it's not a focus I would guess [15:38] yw [15:38] ok, next [15:38] #topic didrocks === meetingology changed the topic of #ubuntu-desktop to: didrocks [15:38] didrocks, hey [15:38] manage living with a german guy here. [15:38] (eof) [15:38] :p [15:39] * Most of the week went into the plymouth merge (we never merged with debian and the package wasn't common). Fixed and dropped some patches. Tested with passphrase with plymouth in initramfs. After some firefighting, will do flavors theme updates (all themes needs to be refresh for a /lib vs /usr/share transition) [15:39] * Dist-upgraded to xenial, found some issues after dist-upgrade and either pinged maintainers or upload fix/revert + discussion on bug report. [15:39] * Went to change passport for Ubucon [15:39] * Doing some personal development stuff [15:39] * Wrote summary & speaker note for scale conf [15:39] * Pushed language default strategy change + another dictionary. Got a golden image with only russian (not even en!) on the image and fixed it :) [15:39] . [15:39] do you feel developed personally? [15:39] I guess as you are Laney ;) [15:40] I experienced a film style montage [15:40] haha [15:40] thanks didrocks ;-) [15:41] #topic FJKong === meetingology changed the topic of #ubuntu-desktop to: FJKong [15:41] FJKong, hey [15:41] youdao dictionary code review [15:41] work on build cef in progress [15:41] a little lag by my vps [15:41] eof [15:41] you don't seem to lag much [15:41] thanks FJKong ;-) [15:42] #topic happyaron === meetingology changed the topic of #ubuntu-desktop to: happyaron [15:42] happyaron, hey! [15:43] no happyaron I guess? [15:43] #topic hikiko === meetingology changed the topic of #ubuntu-desktop to: hikiko [15:43] hikiko, key! [15:43] hello :) [15:44] I am still doing the shadows, I am close to finish the effect.. eof [15:44] thanks hikiko [15:44] :D [15:45] #topic Laney === meetingology changed the topic of #ubuntu-desktop to: Laney [15:45] hey Laney! [15:45] oh hi [15:45] • More work on appstream charm, supports in place upgrading now, filed RT to deploy to prodstack [15:46] • patch pilot [15:46] • New upstream of g-i, for some reason we have a delta now [15:46] • Some work on rebuilds / test smoothing for poppler, openscenegraph, gdal which got largely invalidated by s390x enablement [15:46] • some administration of autopkgtest [15:46] • backport gstreamer to the overlay ppa for touch guys [15:46] • start looking at trashing bug maybe caused by my patch from vivid, didn't manage to reproduce the behaviour people said they were seeing yet, asked on the bug for info [15:46] • fix nautilus rename popover placement [15:46] • libindicator build fix [15:46] ︘ [15:46] Laney: which trashing bug is this? [15:46] * desrt is so sick of all the trash lately :p [15:46] 748248 [15:46] thx [15:46] apparently my patch created a regression [15:47] but I can't work out what that is :) [15:47] oh. this one. [15:47] POST MEETING! [15:47] yup. [15:47] thanks Laney [15:47] #topic larsu === meetingology changed the topic of #ubuntu-desktop to: larsu [15:47] larsu, hey [15:47] hey! [15:47] bugs [15:47] (eof) [15:48] lol [15:48] no nautilus menubars? :-( [15:48] i like this new style [15:48] good job, attente :) [15:48] (was anyone under 13 bytes?) [15:48] haha [15:48] "hey!\nbugs\n" vs. "gsettings.\n" [15:48] i started a trend [15:48] ok: reworked the screenshot patch for gtk because Company [15:48] do you have the updated one somewhere? [15:48] I looked at the bugzilla earlier when I reverted the SRU [15:49] but it seems there is no followup yet? [15:49] found a riduculous workaround for nautilus having all actions in view. instead of win. [15:49] (requires a gtk patch because desrt) [15:49] larsu: did you take a look at that thing we discussed at the sprint at all? [15:49] seb128: not uploaded, but almost done (one regression left to figure out) [15:49] desrt: that thing wouldn't fix this [15:50] i dunno. i see them as related :) [15:50] that was about dynamic menus [15:50] export-all-the-menus and export-all-the-actiongroups *shrug* [15:50] this is aboutthe actions not being reachable [15:50] desrt: we never wanted to export all action groups?! [15:50] anyway [15:50] also some smaller fixes. notify-osd. some more logs work [15:50] some reviews [15:51] (actual eof) [15:51] great reviews [15:51] would request again a++ [15:51] Laney, larsu, btw gnome-logs MIR got approved, so we can swap gnome-system-log for it when we want [15:52] cool [15:52] awesome, thanks [15:52] thanks larsu [15:52] should file calendar one [15:52] * Laney notes that down [15:52] do you still plan to reply on the list? [15:52] that's the same task ya [15:52] I tried the tb-eds integration earlier [15:52] it hangs tb for me as well [15:52] :( [15:52] also I find the calendar UI (it works without eds) too complex [15:53] gnome-calendar+1 [15:53] cool [15:53] ok, next [15:53] thnaks [15:53] #topic qengho === meetingology changed the topic of #ubuntu-desktop to: qengho [15:53] qengho, hey [15:54] Hey! [15:54] * New Chromium, v47. distpatches--. Tested & gave to #security for upload. [15:54] * In progress: Precise C++ abi backporting. [15:54] * To-do this week: More snap work. More Cr/Ozone/Mir work. xdg-patch patch again. [15:54] * Maybe: My desktop machine needs an overhaul, so in the process I'll try to get some ZFS integration into Ubiquity. [15:54] * Gripe: Nautilus crashy. [15:54] EOF [15:55] qengho, nautilus on what serie? feel free to open a bug with bt and I can have a look [15:55] Not xdg-patch. xdg-utils [15:55] seb128: xenial. I will. [15:55] thanks qengho [15:56] #topic seb128 === meetingology changed the topic of #ubuntu-desktop to: seb128 [15:56] • uploaded nautilus 3.18! [15:56] • fixed some small issues in the gtkappchooser dialog (missing icon, label wrap mode) [15:56] • some Debian merges [15:56] • looked at some translations issues (desktop & phone) [15:56] • spent quite some time trying to debug evolution-data-server/uoa/oauth issues, fixed one problem due to an unseeded package but the auth is using password instead of oauth and that's wrong, didn't figure that one yet though [15:56] • some sponsoring [15:56] • review of incoming bugs/errors, upstream several new segfault showing on e.u.c after xenial GNOME (gedit, nautilus, n-m, ...) updates [15:56] [15:56] random note: I'm testing my patch on 2× and firefox is still 1×. Clicking links with the large pointer is *hard*. [15:56] but looks funny [15:57] firefox gets 2× for me [15:57] seb128: what's the bug # for the need-to-restart-eds one? [15:57] for me too if I restarted [15:57] *restart it [15:58] Laney, I didn't file it yet, thanks for the reminder, wanted to check with mardy first [15:58] * seb128 does try to ping him [15:58] I let you know when it's registered [15:58] ok, next [15:58] thanks [15:58] just want to subscribe [15:58] yw! [15:58] #topic Sweet5hark === meetingology changed the topic of #ubuntu-desktop to: Sweet5hark [15:58] Sweet5hark, hey [15:58] - LibreOffice/GNOME Content Apps Hackfest Madrid [15:58] -- much of mentoring and review [15:58] -- wow spanish community has lots of potential [15:58] -- such patches being merged [15:58] - trying to behave: TDF board election ongoing [15:58] - playing with snappy/snapcraft for LibreOffice [15:58] EOF [15:58] efficient ;-) [15:58] back, sorry all [15:59] hey willcooke, no worry [15:59] * willcooke reads log [15:59] willcooke, want to take it from there or should I do the few remaining updates? [15:59] seb128, would you mind finishing, then I can read the log [15:59] #topic tkamppeter === meetingology changed the topic of #ubuntu-desktop to: tkamppeter [15:59] willcooke, no problem [15:59] tkamppeter, hey [16:00] - cups-browsed: Completed development of filtering printer lists. Now if there are very many printers in the network, making only the important ones available is now possible by mean of server's IP addresses, queue name, and practically all printer properties broadcasted via DNS-SD. [16:00] - Will not be on next Desktop Sprint, date equals the date of next OpenPrinting Summit. [16:00] - Bugs. [16:00] 'sup [16:00] thanks tkamppeter [16:00] `sup? [16:01] #topic TheMuso === meetingology changed the topic of #ubuntu-desktop to: TheMuso [16:01] * Submitted orca gsettings backend initial patch upstream, bug# still a little work to do, no feedback given yet. [16:01] * Started looking into merging/uploading new alsa package versions, will do in the new year if no movement on Debian's side before then. [16:01] * More work on the a11y profile manager library, started populating the source tree, nothing pushed yet. [16:01] Hope you all have a pleasant break. [16:01] #topic robert_ancell === meetingology changed the topic of #ubuntu-desktop to: robert_ancell [16:01] - GNOME Software Ubuntu support work [16:01] - LightDM SRU verification [16:01] #topic Trevinho === meetingology changed the topic of #ubuntu-desktop to: Trevinho [16:01] · Prepared unity/compiz SRU for HiDPI cursor support [16:01] · Meeting with Kylin guys https://wiki.ubuntu.com/Ubuntu [16:01] Kylin/Meeting/2015/20151203 [16:01] · New review with kylin lockscreen, proposed a branch to cleanup the [16:01] code so that we can share the most of the logic. [16:01] · Continued working in launcher/trash/volumes/filemanager integration. [16:01] Focusing in main nautilus management [16:01]  [16:01]  [16:02] bah, those chars display as [TAB] in my IRC client [16:02] they do display as Ubuntu logo in gedit though ;-) [16:02] works here [16:02] great ;-) [16:03] #topic other topics? === meetingology changed the topic of #ubuntu-desktop to: other topics? [16:03] ok, anything else to discuss? [16:04] probably come up to ldn on thursday if you want me to [16:04] that would be good [16:04] might not get the 0630 train this time though :p [16:05] no need to arrive too early [16:05] * desrt reads "gsettings backend", get nervous [16:05] * desrt has visions of compiz [16:05] haha [16:05] * desrt tries to find that bug [16:06] ok, let's wrap the meeting there [16:06] willcooke, or did you have an update/something you wanted to discuss? [16:06] *** desert will need to write the gsettings replacement for the gconf backend in LO [16:06] nothing at the moment [16:06] lol. not touching that. [16:06] my laptop is made of aluminium, which has too low of a melting point for LO work [16:06] desrt: hrhr [16:07] haha [16:07] #endmeeting [16:07] !!! [16:07] best part about old jokes: they're already old [16:08] seems not I guess [16:08] #endmeeting === meetingology changed the topic of #ubuntu-desktop to: Home of the Desktop Team, https://wiki.ubuntu.com/DesktopTeam | For help or questions, try #ubuntu | Amaterasu watches over you benevolently | Tag "gtk318" for gtk 3.18 bugs please and thanks [16:08] Meeting ended Tue Dec 8 16:08:52 2015 UTC. [16:08] Minutes: http://ubottu.com/meetingology/logs/ubuntu-desktop/2015/ubuntu-desktop.2015-12-08-15.30.moin.txt [16:08] thanks everyone [16:09] thanks! [16:09] groovy meeting [16:09] * desrt finds old orca "let's use gsettings" bug with 40+ comments [16:09] * desrt finds very new orca "let's use gsettings" bug with a single comment/patch from luke [16:10] so, almost holidays and I wanted to land quite some feature work before that but I guess that's not going to be the case [16:10] we could land gnome-logs [16:11] desrt: we have builders on amazon ec2 these days with libreoffice precompiled ready to go though to eliminate the "my local hardware"-excuses ;) -- This mostly resulted in people being more honest about their opinions on "legacy" in legacy codebase ... [16:11] larsu, how likely is it that we can get nautilus menubar and unity-control-center port to geonames in the next week? [16:11] seb128: do it [16:11] LO builds in the cloud? sounds like climate change! [16:11] I'll do cal this week [16:11] bah, ssh is lagging [16:12] that is a sign to make tea [16:12] Laney, "do it" = logs? [16:12] yeah [16:12] right [16:12] I would like to land geonames as well [16:12] which is fine [16:12] * desrt waits for gigabit internet connection to do distcc in the cloud [16:12] woul dbe good [16:12] but not very useful if u-c-c doesn't use it [16:14] what else would want to be ported? [16:14] libtimezonemap/ubiquity + u-s-s? [16:16] u-s-s? [16:16] oh, touch one you mean [16:16] yeah [16:22] seb128, desrt: speaking about dconf/appservice, if you guys have apparmor people that can do that side of the work that would be awesome :D [16:23] *cough* [16:23] didn't attente did some work on that side? [16:23] so maybe we should chat about that now [16:23] sadly jjohansen's patches doesn't build, I fixed that, but it doesn't work neither :( [16:23] jjohansen: are you around? [16:23] otherwise somebody to talk to the security team about [16:23] attente: you as well? [16:23] i'm here [16:24] as far as i know, attente's patches did exactly what we needed [16:24] attente: do you know why jjohansen was taking a different approach? [16:24] desrt, when I asked, attente said their his patches are superseeded by jjohansen's [16:25] I personally don't care as long as I have an API that gives a strv of read/write keys [16:25] ya. exactly. [16:25] which is what we had from attente [16:25] via this impressively strange sequence of write()/read() syscalls on a pseudofile [16:26] (which, apparently, is how apparmor rolls) [16:27] Laney, https://bugs.launchpad.net/evolution-data-server/+bug/1523985 [16:27] Ubuntu bug 1523985 in evolution-data-server (Ubuntu) "evolution-source-registery needs a restart to pick newly activated uoa accounts" [High,Triaged] [16:28] xclaesse: for the time being, it might just make more sense to target against xdg-app [16:28] the apparmor guys will catch up to us eventually [16:28] we certainly shouldn't block on them [16:29] desrt, sadly xdg-app is more long term goal for us, we would like to confine apps with apparmor first [16:29] sure... but unless you know some kernel hackers, we're sort of blocked on that front, right? [16:29] but in all cases, we already missed our next deadline, so it's a feature postponed to next year anyway :p [16:29] desrt: i'm not sure the reason for the new patchset tbh, but when i tried building them, it seems a bit incomplete still [16:30] i think if we get to the point of "look... this is working already on our side... we're really just waiting for you" then it's a bit more of a compelling request [16:30] desrt, right [16:30] you could also just write a keyfile backend or something [16:31] desrt, I don't think it needs kernel side work, if I understood correctly it's only userpace apparmor patches [16:31] seb128: thanks! [16:31] i mean.. the appservice is already pretty strict about checking dbus name ownership... we sort of use that as the root of security... any sane system will enforce the same. [16:31] xclaesse: it does need kernel work [16:31] xclaesse: it's not possible to read out the contents of the FSA that apparmor compiles [16:31] you can currently only ask it yes-or-no questions [16:32] or rather, bitfield questions [16:32] ie: each node in the FSA has a (32bits?) mask that you can get by playing a string through it to take you to that node [16:33] desrt, a key file like this would do, indeed: [16:33] [apparmor label] [16:33] readable=key1;key2 [16:33] writable=key1;key3 [16:33] pretty much [16:34] for testing, at least [16:34] I already get the label from org.freedesktop.DBus, GetConnectionCredentials [16:34] i thought i had plumbed that through already [16:35] although there was some confusion on that point... there was an old crappy API and a new good API and i think i might have been waiting for the good API [16:35] desrt, that wasn't part of your code, but I added it [16:35] there is GetConnectionApparmorSecurityContext() as well, that's the old crappy one afaik [16:36] yes. i agree. [16:36] GetConnectionCredentials is the good one [16:36] but is the context wired up to the new call? [16:36] yes, I did it already [16:37] i get a dict which has pid and uid, but no context... but that may be because none of my things have contexts [16:37] oh. good. :) [16:37] and it gives meaningful reply, so that should be fine [16:37] ya. this is very good. [16:37] desrt, I do get a context here [16:37] on ubuntu wily [16:38] look at the unit test in appservice, it has an appormor rule you can load [16:38] i think i've uninstalled as much of apparmor as is possible without using equivs ;) [16:39] no surprise you don't get an apparmor context then [16:39] by default you should get "(unconfined)" string [16:39] ya. i don't even get that [16:39] i'm on testing, btw. [16:39] so probably my dbus is too old [16:42] {'LinuxSecurityLabel': [117, 110, 99, 111, 110, 102, 105, 110, 101, 100, 0], 'ProcessID': 25672L, 'UnixUserID': 1000L} [16:42] * desrt raises an eyebrow [16:42] if you interpret that byte array as ascii, that gives "unconfined\0" [16:42] did GVariant print that? [16:42] d-feet [16:42] figures :p [16:43] * desrt was about to go bug hunting [16:44] * desrt goes back to playing with changesets [16:53] xclaesse: we agreed that reset absolutely must always be successful, right? [16:53] * desrt notes that keyfile backend is not doing this now [16:53] desrt, yes and that's what I did in my implementation [16:54] * desrt tweaks [16:54] and indeed memory/keyfile backend can't do it at least for path reset [16:54] keyfiles are sort of tricky, in fact [16:54] because they may be readonly [16:54] for memory backend you get it for free if you use DConfChangeset afaik [16:54] well, nothing ever fails in memory backend [16:55] yep, but current implementation cannot do path reset, by using changeset that will fix that :) [16:55] i guess a readonly keyfile corresponds (roughly) to a default database with no user values present [16:55] amazing that we never really thought through all of this stuff when we first wrote it [16:56] the immense majority of our use cases doesn't need all those details tbh [16:56] nod [16:56] readonly keyfiles as a gsettingsbackend are probably not super common, either :) [16:57] not hard to fix, in any case [17:27] happyaron, sorry for the calendar spam [17:56] calling it a day, have a good evening! [17:56] see ya seb128 [17:57] happyaron, https://github.com/ibus/ibus-pinyin/blob/master/po/fr.po has some invalid tag, do you know who to ping to get that fixed? the project doesn't seem very active [17:57] #: ../setup/ibus-pinyin-preferences.ui.h:15 [17:57] msgid "UI" [17:57] msgstr "UI" [17:57] the "" should be "" [17:58] but that's for another day, just letting that in the backlog/for when you are up tomorrow === alan_g is now known as alan_g|EOD [18:06] right, see you! [19:12] g'night [21:13] lol. france wants to ban tor and shut down public wifi networks during emergencies [21:14] in other news, disney wants to ban people looking at the moon by making it invisible [21:31] desrt: Along with what I've heard coming from the US this morning, seems world leaders really don't understand modern communications/the internet.