=== maclin1 is now known as maclin [04:03] bye all [04:48] * Trevinho did it again... :/ I'll be here again a little late in the morning. [05:28] hello [05:42] Morning hikiko [05:49] * thumper feels pangs of nostalgia [06:08] good morning [06:22] bonjour~ [06:23] hey larsu! [06:23] morning didrocks! How goes? [06:23] hey didrocks larsu desktopers [06:23] larsu: still sneezing a little, but near the end, and you? [06:23] re seb128 [06:23] hello seb128! What's up? [06:23] didrocks: started to sneeze yesterday :/ [06:23] but doesn't seem *that* bad [06:24] larsu, I need coffee, a bit tired this morning ... what about you, feeling better than yesterday? [06:24] seb128: a little bit better. Had a good night's sleep [06:24] I hope it doesn't get worse over the day [06:24] good [06:24] bonjour tout le monde ! [06:25] bonjour pitti! [06:25] ça va seb128 et didrocks ! comment allez-vous ? [06:25] hey pitti, et toi, comment vas-tu ? [06:25] salut pitti, comment ça va ? C'est bien à Budapest? [06:25] je vais bien, merci ! je suis allé courier le matin [06:26] et on a eu des bonnes heures de "hack sessions" hier soir [06:26] vous avez hacké sur quoi ? [06:28] seb128: helping people with some beaglebone issues, and I taught britney another corner case [06:28] there was free beer :) [06:29] good :) [07:11] moin all [07:13] hey Sweet5hark [07:16] hey Sweet5hark [07:17] didrocks, seb128: hey there ... [08:00] morning [08:00] yoooo [08:01] hey all [08:03] hey willcooke, Laney! You guys are both in sync :) [08:03] evening darkxst [08:03] hi Laney, willcooke, darkxst [08:03] what's up [08:03] man I hate icons [08:03] * larsu puts another item to the sprint [08:04] spent the weekend and a bit up in the mountains enjoying the last of the snow, before our first heatwave arrives this weekend ;( [08:04] darkxst: meanwhile, we are enjoying the last rays of sun over here [08:05] hey willcooke Laney darkxst [08:05] as usual, I'm glad launchpad sends me an email when I edit a blueprint [08:05] I had almost forgotten! [08:09] looking like it's going to be a nice day! [08:11] willcooke: BUD? [08:11] Budapest [08:12] Snappy sprint [08:12] finally: http://s-jenkins.ubuntu-ci:8080/job/udtc-trusty-tests/1966/ [08:12] willcooke: ah! [08:12] hum, public url once synced to our jenkins mirror :) [08:12] willcooke: that will be an intersting discussion - will we have a snappy packagekit backend? [08:12] and done! https://jenkins.qa.ubuntu.com/job/udtc-trusty-tests/1966/ [08:12] larsu, it's rather looking like.... "if we build it we will have" [08:12] still have to have the distro fixed. /me looks at our python maintainers… [08:13] larsu, but I will be having a meeting with those guys when they are back from Budapest to work out exactly what the requirements are [08:13] larsu, right now it looks like we have a couple of options: [08:13] 1) Add Snappy support to USC [08:13] 2) Add Snappy, and paid-for-apps etc to G SC [08:14] usc is basically unmaintained, isn't it? [08:14] larsu, I guess I missed the start of the discussion, or is that backlog from yesterday? [08:14] "that will be an intersting discussion" [08:14] * seb128 tries to find what "that" is [08:14] seb128: no willcooke just moved and item around on the sprint planning blueprint and I wondered what BUD stands for [08:14] ah ok [08:15] seb128, Laney ok to upload gsettings-desktop-schemas 3.18, its just one bug fix, on typo fix and lots of translation updates [08:15] (didn't have "cities" in mind, but "business units") [08:15] darkxst, no objection from me if that's what you described [08:20] indeed [08:20] will go to the queue for someone to double check anyway [08:22] Laney: https://code.launchpad.net/~larsu/humanity/eog-fullscreen-toolbar/+merge/272879 and https://code.launchpad.net/~larsu/ubuntu-themes/eog-overlay-buttons/+merge/269482 fix eog's theming [08:22] reviews/landing appreciated ;) [08:24] larsu: who is the person you asked for a review from? [08:25] you [08:26] LIES [08:26] ? [08:26] danrabbit [08:26] Laney: do you have approval chops for icons? [08:26] * Laney doesn't understand icon themes really either [08:27] apparently so [08:27] seb128, Laney k, uploaded [08:27] Laney: ah thanks for the catch. I reassigned [08:27] I had accidentally proposed to merge into lp:humanity [08:28] this whole project is a mess [08:28] anyway, will look shortly [08:28] thanks [08:28] didrocks, could you have a look to https://bugs.launchpad.net/ubuntu/+source/oneconf/+bug/1421884? do you remember if oneconf use the apt cache without taking a lock? (can it take one if it's not root?) [08:28] Ubuntu bug 1421884 in oneconf (Ubuntu) "/usr/share/oneconf/oneconf-query:SystemError:/usr/bin/oneconf-query@245:async_update:update:update:compute_local_packagelist:__init__:open" [High,Confirmed] [08:28] did you give the toolbar a background? [08:29] (don't have the new theme on the laptop atm) [08:29] (so can't check right away) [08:29] Laney: yes, that's the 2nd mr I linked [08:29] Laney: actually it's missing that commit, let me push it [08:29] oh boy - looks like today is fsck day [08:30] haha [08:30] willcooke: no ssd? [08:30] the theme branch is the one we already looked at [08:31] shall I do a new one? [08:31] seb128: ah, I may not take a lock, let me look at the api [08:31] larsu, yeah this has got an SSD, but suddenly errors. This laptop has been telling me it's getting old for a few months now. [08:31] didrocks, thanks [08:32] nah [08:32] ok I'll push then [08:32] https://apt.alioth.debian.org/python-apt-doc/library/apt.cache.html -> doesn't mention any lock, apart from a lockexception :p [08:32] * didrocks continues looking [08:34] seb128: nothing on the doc about lock if I'm right [08:34] seb128: it's doing on your behalf supposively [08:34] didrocks, let me comment reply to that email [08:35] lockfile = apt_pkg.config.find_dir("Dir::Cache::Archives") + "lock" [08:35] thanks! [08:35] it's done for you when you update the archive [08:35] (which is where it's failing) [08:35] seb128: wait [08:35] I found it another class [08:35] * seb128 waits [08:36] let me test that's possible as a user [08:40] seb128: yeah, if you can ask how we can get a system lock as this is running during user's session… [08:40] seb128: getting the exception due to this [08:48] what's the "quiet volume down" key? [08:49] u-s-d grabs XF86AudioLowerVolume twice because of it (and confuses unity in the process) [08:49] larsu: the one that doesn't rattle? [08:49] haha [08:50] oh got it: it changes the volume without making those plop sounds [08:50] (which we don't have anyway in unity, do we?) [08:51] larsu: I do get plopps when I press the vol up/down keys [08:51] I don't :/ [08:51] * ogra_ too and i would be unhappy to not have them [08:51] haha [08:51] * Laney strokes ogra_ [08:51] *PLOPP* [08:51] :) [08:51] DON'T CHANGE MY DESKTOP [08:52] if you want quiet, use mouse scroll on the indicator [08:52] ah! I guess this is because of the same race in unity? [08:52] hey what is speech-dispatcher? [08:52] becasue the same key is bound to both actions [08:52] * darkxst make xomnad point to gnome-shell, just for Laney ;) [08:52] (volume up and quiet volume up) [08:53] how do you bind quiet volume up? [08:54] you don't [08:54] it's hardcoded [08:54] if you want quit volume up, set org.gnome.settings-daemon.plugins.media-keys volume-up to something other than XF86AudioRaiseVolume [08:54] and the press XF86AudioRaiseVolume [08:55] this is amazong [08:55] zooooong [08:58] Laney: ah, I was wrong: this is when pressing as well [08:58] (but hardcoded to that) [08:58] oh nice, that's a cool trick [08:59] * seb128 doesn't understand what's going on with those keybindings [09:00] not sure I want to though [09:00] Trevinho: are there any other duplicate key grabs? (the volume one turns out to have modifier) [09:00] I'm happy to let larsu deal with it :) [09:00] and Trevinho [09:00] seb128: what's the problem? [09:01] Laney: biggest issue right now is in unity, indeed [09:01] u-s-d code is convoluted, but seems to do the right thing [09:01] larsu, just trying to understand what you wrote about the double actions on the same key [09:02] seb128: if you ask unity to grab the same key twice, it only sends you a signal for one of those grabs [09:02] which is the bug we're seeing [09:02] can you bind the same key to 2 different actions? [09:02] there was some confusion about u-s-d doing that in normal operation [09:02] seb128: no [09:02] the u-c-c ui doesn't let me [09:02] k, what I though [09:03] so how did you end up there? [09:03] the problem happens when u-s-d restarts after a crash [09:03] it ask unity to grab the same keys again [09:03] and unity sends signals about the old grab [09:03] shouldn't unity ungrab the first one if asked for an already assigned one? [09:03] so u-s-d doesn't see them and does ... nothing [09:03] seb128: yes. that's the bug [09:03] also didrocks said in his case there was no u-s-d crash or restart [09:04] seb128: also it should ungrab anything that someone who just crashed had grabbed before [09:04] darkxst, do you know if anyone is working on bug #1432098? it's ranked high on e.u.c issues [09:04] bug 1418771 in gjs (Ubuntu) "duplicate for #1432098 gjs-console assert failure: *** Error in `/usr/bin/gjs-console': free(): invalid next size (fast): 0x00007f74a804b240 ***" [Low,Confirmed] https://launchpad.net/bugs/1418771 [09:04] seb128: I'm hoping he's wrong on that. If he isn't, I'll need a dbus log of his session's startup [09:05] seb128: I spent quite some time trying to reprocduce [09:05] k [09:05] maybe u-s-d restarts for some reasons [09:05] (and I told it multiple times :p) [09:05] * larsu changes location. bbiab [09:05] so no, no crash, can restart for some reason [09:05] but clearly no crash [09:06] let's fix the restart issue and see if that resolves it [09:06] didrocks: if that happens, Trevinho's fix will solve this issue for you [09:06] yeah [09:06] but still would be great to have a 100% understanding of the issue [09:10] didrocks, I guess having a bustle log of your login when getting the issue would help [09:11] yep, how to get this already? [09:11] the upstart ted's post? [09:11] right [09:11] didrocks, https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=800501#22 btw [09:11] Debian bug 800501 in python-apt "SystemError on missing files in /var/lib/apt/lists" [Normal,Open] [09:12] seb128: thanks, so nothing I can do for now on the apt thing [09:12] seb128: ok, will add that and relog later [09:12] right [09:13] didrocks, I reported bug #1323586 as well but that got wontfixed and should be handled in oneconf (or declared a buggy system) [09:13] bug 1323586 in oneconf (Ubuntu) "oneconf crashes if one apt list file is unreadable" [High,Confirmed] https://launchpad.net/bugs/1323586 [09:13] I think that one is less common [09:15] seb128: hum, yeah… I don't think that client should have to deal with it, but I'm happy to just ignore it [09:15] I think it's best [09:15] one user just commented saying it currently apport prompts every hour [09:15] seb128: note that we need the lock fix first, and the crash is at the same place [09:15] so probably better to just catch it [09:15] well, he decided to put into that state and knows that he did it [09:16] yeah, but I guess some people don't realize they added a file which is not world readable [09:16] like they copy from another disk with different uid or something [09:16] seb128: can be… anyway, let's get the lock fix in, then, I can handle that one propertly [09:16] properly* [09:16] or a backup from a vfat drive [09:16] right [09:16] and I'll just catch this exception [09:16] thanks ;-) [09:16] yw [09:17] keep me posted if you are subscribed on the python-apt thingy [09:17] yeah, I reported those so I can do that === hikiko-lpt is now known as hikiko [09:17] thanks! [09:18] yw! === greyback__ is now known as greyback [09:24] pitti, let me know when you are around and have a minute to discuss the recent apport changes [09:44] seb128: I'm on-and-off ; what's the issue? [09:44] seb128: apart from the dkms hook crash in bug 1500450 [09:44] bug 1500450 in apport (Ubuntu) "/usr/share/apport/package_hook:FileExistsError:/usr/share/apport/package_hook@64:make_report_file" [High,Triaged] https://launchpad.net/bugs/1500450 [09:44] pitti, you dropped make_file_path from apport.utils [09:44] seb128: ah [09:45] is that wanted? [09:45] wasn't there a way to keep it and make it at least write some warning? [09:46] seb128: ish; we need to fix all hooks for the recent security issue, so /usr/share/apport/package-hooks/dkms_packages.py needs to be adjusted too [09:46] pitti, I just uploaded a fix for bug #1499842 but I guess we need to do security update with that change as well? [09:46] bug 1499842 in dkms (Ubuntu) "dkms_packages.py crashed with AttributeError in __main__: 'module' object has no attribute 'make_report_path'" [High,Fix committed] https://launchpad.net/bugs/1499842 [09:46] seb128: we can keep the old interface, but we'd then hide the vuln [09:46] seb128: oh, you already did? thanks! [09:46] pitti, k, it would have been nice to mention the api change in the changelog [09:46] it puzzled me a bit [09:47] pitti, I just did for wily, I guess we want to do that for other series but it should be done through -security since that's where apport went? [09:47] mdeslaur, ^ [09:47] pitti, yw! [09:47] seb128: yes, I agree; that, or we need to put the old insecure API back [09:47] but as we need to upload the other hook users anyway, we might just as well use the new API [09:48] right [09:48] or replace it by a function that raise a error explaining the issue [09:48] like "is replaced by ... because of ..." [09:49] pitti, anyway, I handled dkms for wily and [09:49] it seems it was the only hook using that function [09:49] seb128: ah great -- I was about to ask to check other hooks [09:49] :-) [09:50] pitti, thanks for the reply [09:52] seb128: I'll release the current packages in -proposed to "clear the way" [09:52] pitti, danke [10:03] darkxst, could be worth raising with robert_ancell but https://errors.ubuntu.com/problem/27941445d596fb71be692b9008c0847d2ce6c428 (xorg abort issue) is ranked high on e.u.c for wily and the recent report all look like gdm ones [10:12] bug #1499508 seems similar [10:12] bug 1499508 in xorg-server (Ubuntu) "Xorg crashed with SIGABRT in OsAbort()" [Undecided,New] https://launchpad.net/bugs/1499508 [10:12] https://launchpadlibrarian.net/218756042/XorgLogOld.txt has [10:12] "(EE) Cannot establish any listening sockets - Make sure an X server isn't already running(EE) " [10:45] seb128: do you have a vcs version of your eds upload? [10:47] Laney, let me check [10:47] otherwise i can commit the diff [10:50] Laney, I don't understand, the 1ubuntu1 version from the archive has different content from the vcs [10:51] -#ifeq ($(DEB_HOST_ARCH_OS),linux) [10:51] -# DEB_CONFIGURE_EXTRA_FLAGS += --with-phonenumber=/usr [10:51] -#endif [10:51] in the rules [10:51] and some .symbols difference [10:52] Laney, I forgot that eds was in a vcs so I don't have a stacked/ready to push change but I'm happy to fix it, unsure what to do with those divergences though? [10:57] Laney, if you do an upload can you include https://git.gnome.org/browse/evolution-data-server/commit/modules/ubuntu-online-accounts/e-signon-session-password.c?h=gnome-3-18&id=1765be5703f4e129ecc5c821dbffac4e18d26e68 ? (I can also commit to the vcs once it's sorted out) [10:59] I probably forgot to bzr bd -S again after merging some more stuff in from Debian [10:59] those changes make sense [11:02] so I let them in the vcs, commit 0ubuntu2 different from the archive and you build a 0ubuntu3 that includes them and reconsiliate archive and vcs? [11:02] ya [11:03] I would mention them in the changelog of ubuntu3 [11:04] seb128, pitti: I'll take care of the dkms update [11:04] mdeslaur, thanks, I did the wily one [11:06] Laney, ok, you can pull [11:07] thx [11:08] yw, sorry for forgetting the vcs in the previous upload [11:16] * larsu goes for lunch [11:19] Trevinho, andyrock, if any of you look at ups/menu things could you add bug #1501289 to your list as well while you are at it? [11:19] bug 1501289 in unity (Ubuntu) "unity-panel-service "Source ID was not found when attempting to remove it" warnings" [Undecided,New] https://launchpad.net/bugs/1501289 [11:22] mdeslaur: thanks; sorry, not much time this week on the sprint [11:26] mdeslaur, pitti, there seems to be some regressions from the apport update btw, e.g bug #1500450 [11:26] bug 1500450 in apport (Ubuntu) "/usr/share/apport/package_hook:FileExistsError:/usr/share/apport/package_hook@64:make_report_file" [High,Triaged] https://launchpad.net/bugs/1500450 [11:31] yes, I'll handle this [11:31] seb128: mh, oh... Is that UPS? :o [11:31] Trevinho, yes [11:32] * Trevinho just up... Long night hacking again :P [11:32] that's weird, I mean couldn't be an indicator? [11:34] larsu: have you seen the branch I proposed? [11:35] although it crashes on CI, while testing (nothing should be related to the keys)... seb128 is there any way to get a core file from ci on jenkins? [11:36] pitti: I would be really greatful if you get bored during one of the meetings if you can have a look at the jayatana vivid SRU (it's a one line change and the issue got quite some press yesterday…) [11:37] larsu: it should also address the issue you were saying this morning, but let's see :) [11:38] didrocks: sorry, I can't -- I misspelled jajatana like three times now :) [11:39] pitti: ahah yeahatana :p [11:39] Trevinho, unsure core on jenkins, maybe jibel can help you? [11:39] don't those machines report to whoopsie? [11:39] or what team maintains those? [11:39] * Laney thought they did [11:39] but maybe not [11:39] didrocks: fait [11:40] accompli [11:40] pitti: merci ! [11:40] lunch, bbiab [11:40] enjoy seb128 [11:41] Laney: I'm not sure as they'd also report issues that are only on a proposed branch, wouldn't that be distracting? [11:42] maybe it's a different set of machines which does that [11:42] probably best to ask whoever runs the machines :P [11:46] I've no idea who... fginther is the process owner, I'll ask him when back. [11:49] seb128, Trevinho jenkins just attach any file the test gives it in a location defined in the job. I don't know these jobs but they must be configured to export the core or crash from the testing environment to a predefined location that jenkins has access to. fginther can surely help with that [11:50] jibel: ok thanks === hikiko__ is now known as hikiko|ln [12:06] good morning. [12:11] hey qengho [12:35] Trevinho: no I haven't seen it (not subscribed to unity) [12:40] larsu: am I missing a fix for the fullscreen button? [12:40] "Leave fullscreen" [12:40] it has a black border on the lfeft [12:41] oh I think it's transparent [12:41] ah is it a separate button from the toolbar? [12:50] larsu: it's on https://code.launchpad.net/~3v1n0/unity/gnome-keygrabber-refactor, I'm trying to understand why the CI crashes, but it works fine here. Adding support for only sending the signal to who has requested it makes it a little more complicated, but should work [12:56] Trevinho: hm okay, I'll have a look [13:12] good morning [13:12] hi andyrock [13:17] hey andyrock [13:28] pitti, thanks for fixing the apport issue, but I'm not sure to understand, does it means the new apport can't write crash file when one is existing? [13:28] or in which case do those trigger? [13:34] good morning #ubuntu-desktop [13:35] hey desrt [13:35] how are you today? [13:35] good [13:35] last day at the sprint [13:35] more file monitor fixing today, i think :) [13:36] good :-) === mzanetti- is now known as mzanetti [13:37] hey desrt andyrock [13:37] hello willcooke [13:39] morning desrt! [13:42] hihi [14:00] seb128: right; that's what the main apport does for unseen crashes; we don't have this fine-grained logic right now for package failures [14:00] seb128: we probably need to refine that, like remove an old report if it was seen already [14:01] pitti, is there a reason we need to make an user visible error out in those cases? [14:01] Trevinho: cool, works for me :) [14:01] seb128: sorry, WDYM? [14:01] right, I was going to say, what is cleaning old reports and making room for new ones? [14:01] seb128: normally the daily cron job [14:03] pitti, sorry I think I misread your change, I was expection some .warning, I though what you did would stop trigger apport prompt [14:03] but I guess it just make apport exit? [14:05] seb128: yes, the fix in trunk will now just log an error, and don't produce a "follow up apport crash: [14:05] " [14:05] k, sounds good [14:05] pitti, thanks [14:07] attente, hey, could you have a look to bug #1494331? or to talk to happyaron about it [14:07] bug 1494331 in unity-settings-daemon (Ubuntu) "fcitx can't use extra trigger key to activate" [Undecided,New] https://launchpad.net/bugs/1494331 [14:07] there is a suggested change (not using fcitx_input_method_activate) [14:10] seb128: sure [14:10] attente, thanks [14:11] larsu: cool, sorry I was otp :) [14:11] larsu: it should allow also multiple sources to grab a key, but it's released once the last left... [14:11] right [14:12] larsu: also if you call Ungrab with gdbus you're not allowed to do that... [14:12] Trevinho: right, because only the grabber can ungrab [14:12] man, this interface is bad [14:12] (not sure if I mentioned that before :D ) [14:13] yep [14:13] larsu: yeah, you didn't but it was quite obiouvs [14:13] hehe [14:14] I only need to have better names for variables (/me is so bad on that) [14:14] naming things *is* one of the biggest challanges [14:16] sometimes I'd like to be like vala.... [14:16] tmp1, tmp2... [14:16] :D [14:20] mdeslaur, you might want to include changes similar to http://bazaar.launchpad.net/~apport-hackers/apport/trunk/revision/3013 to the dkms change [14:21] seb128: hrm, ok, thanks [14:21] mdeslaur, yw! bonus point if you would do wily as well? ;-) [14:22] seb128: grr :) [14:22] sure, I'll do wily [14:22] * seb128 is going to pay back with fixing some other GNOME bug when he can [14:22] mdeslaur, 'ci [14:43] larsu: did you see my questions about the button? [14:44] Laney: no? [14:44] where? [14:44] 14:40 [14:44] oh, it scrolled off screen and I only saw Trevinho's ping sorry [14:45] Laney: I think you're missing it, yes [14:45] which bit? [14:45] the toolbar has the right background now [14:45] icons are the right size [14:45] should be in the menubar branch of eog [14:46] * larsu checks [14:47] Laney: ah no sorry, it's here: https://code.launchpad.net/~larsu/eog/fullscreen-button [14:48] neat [14:56] the first two pictures in my ~/Pictures make me angry [14:56] screenshots of the destruction of loads of big old trees done by the council to build a new tram line [14:56] less eog testing [14:56] test with /usr/share/backgrounds [14:57] that's what I did because the first pic in ~/Pictures was my profile picture [14:57] * larsu was tired of seeing his own face all the time [15:03] https://launchpadlibrarian.net/208152879/me-64.jpg [15:03] ya [15:05] seb128: did you find out what's going on with the theme? [15:05] can we just slip this eog branch in before the dual landing maybe? [15:05] but the one I have here is 960² [15:07] Laney, the landing you mean? [15:07] ye [15:07] no, pete was not online yesterday, but I added the eog bugfix branch to the silo [15:07] can't upload it if it a dual though [15:08] upload where? [15:08] ubuntu? [15:08] I don't understand [15:08] you can publish the silo [15:08] which part? [15:08] which is equivalent to an upload? [15:08] you think I should upload to the stable phone overlay? [15:09] no, I think you should add mps that we want to land to the silo [15:09] and we should publish then [15:09] those eog changes are noop for the phone [15:09] no reason they refuse them in the vivid overlay [15:09]