[06:12] good morning desktoppers [06:17] good morning [06:21] salut didrocks [06:26] hey oSoMoN [07:18] Morning oSoMoN, didrocks, seb128 [07:23] hey duflu [07:23] good morning desktopers [07:23] duflu, how are things? thanks for the trello/work summary email! [07:24] hey duflu, re seb128 [07:24] seb128, you're welcome I think. Just nervous about that information getting duplicated outside on Trello now. So some of it will soon be out of date [07:25] duflu, I don't think we are duplicating anything? [07:25] seb128, only in email, if you count that [07:26] duflu, the emails are just to make sure that everyone is aware of what they have to do and also have an estimate of how much work they think they have/how long each feature will take, the trello isn't good at helping us with that [07:27] that helps me to see if team members could take on extra work or already have more than enough for the cycle [07:27] Yes, fair enough. I will stop short of suggesting Gantt charts :) [07:28] haha [07:29] planner is on the list of things to remove since it still uses gconf [07:29] that's the only gantt tool I tried on debian/ubuntu :) [07:35] seb128, is the difference between pending and current some smoke tests to pass? http://cdimages.ubuntu.com/daily-live/ [07:36] duflu, yes [07:37] I never remember where the log for those test is though :/ [07:37] same [07:39] seb128, jibel would know... [07:39] ? [07:39] yes === maclin1 is now known as maclin [07:43] duflu, logs are there https://platform-qa-jenkins.ubuntu.com/ but I didn't finish enabling smoke tests for cosmic yet [07:44] jibel, no worries. I will come back to it at a later date [07:45] duflu, no koza yet - so I assume no BT meeting [07:46] willcooke, not sure. Once or twice he went straight into the meeting without logging into IRC. Although my only relevant comments are in the status report [07:47] duflu, I have nothing at this point either. Want to try the bluez snap path again, and heber is lined up to look in to that some more [07:47] until then, nothing from me [07:48] willcooke, also I will wait for cosmic smoke tests to pass before confusing the situation with a proposed update [07:48] The most vocal users in favor of 5.49 already have it via the PPA, and are happy [07:49] +1 thx [07:49] willcooke, in other news upstream PulseAudio fixed BT audio skipping this week (after some years) [07:49] I don't think we need to wait for the iso to be current to update bluez [07:49] Will backport that too [07:49] but as you prefer [07:50] seb128, I prefer it to be usable before my dev machine uses it :) [07:50] we also have no hurry to update, it's not like cosmic had lot of users and that they needed it [07:50] right [07:54] seb128, I should have just said that I won't propose an update to cosmic until I have tested it on cosmic :) [07:54] that makes sense :) [07:56] bluetooth always on on startup while you disable it in the previous boot is a feature/bug? [07:56] Trevinho, those gdm fixes, they seem a bit much for a SRU no? do you think there would be an easier way to fix the issue for bionic? [07:57] didrocks, I would say it's a bug but I'm unsure [07:57] or lack of feature to restore the status [07:57] I would expect the previous enablement state to persist as well [07:58] seb128: hey, the pure fix is just two lines... Or one. It's in the 1st commit [07:58] I'll cherry pick them once landed [07:58] Trevinho, good morning :) [07:58] Trevinho, great, thanks! [07:58] hey Trevinho [07:58] Trevinho, how is the hackfest going? [08:02] yo [08:02] hey Laney [08:03] hey Laney, how are you? [08:03] didrocks, https://bugzilla.gnome.org/show_bug.cgi?id=638117#c36 sounds like it should work [08:04] Gnome bug 638117 in general "Remember off/on state between sessions" [Normal,Resolved: notgnome] [08:04] salut french [08:04] alright, it stopped being nice weather though [08:04] would be great to see if someone else reproduce, to see if it's config related or not [08:04] still unnice weather here [08:04] went out for the daily sawfly removal session and it was windy [08:05] Laney, what are the flies eating? [08:05] the gooseberry and redcurrant bush leaves [08:06] https://www.rhs.org.uk/advice/profile?pid=517 [08:06] these jerks [08:06] so they strip the leaves and the plant dies? [08:06] little buggers [08:06] didrocks, https://github.com/systemd/systemd/issues/8590 [08:06] systemd bug 8590 in systemd "Bluetooth state not respected between reboots (is on even if on previous boot I turned it off)" (comments: 2) [Needs-Reporter-Feedback ❓, Rfkill, Open] [08:06] bug 8470 in linux-source-2.6.15 (Ubuntu) "duplicate for #8590 Oops caused by ipw2200 driver" [Medium,Fix released] https://launchpad.net/bugs/8470 [08:06] I never let it get that far enough, but I guess so [08:06] seb128: nice finding foo! :) [08:06] :) [08:06] actually this year we caught the first lot quite early on so it's only been a few leaves [08:09] ok, keeping the tab opened [08:09] logged what happens in the journal when shutting down bluetooth [08:09] tomorrow, at next boot, I'll file the rfkill info + anything I can find about bluetooth [08:23] good, I'm going to try that as well/see if it's persistant on my machine, I let you know next time I reboot [08:26] th! [08:26] thx* [08:49] back in a bit, got some errands to do [08:51] :) [08:57] seb128: sorry I was walking to the venue... [08:57] seb128: pretty good, lots of things planned more than done, but we've good plans at least [08:58] nice [08:58] good improvements? [09:02] hopefully :), a good one which is almost done is the one about killing the g-s on gdm when logged in, so that it won't stay there for no reason [09:04] Trevinho, ah, is somebody working on that? [09:04] that would be great [09:04] they start catching up with unity :p [10:40] 'they' I mean, isn't this pretty much 'we' given Didier is a Board member ;) [10:49] Oops. Drank afternoon coffee and missed dinner. [10:49] Later all [11:38] Wimpress: by any chance, have you ever tried to POST manually without ubuntu-report to the database? We have 4 records which are a little bit weird. distro is "ubuntu-mate" (and that doesn't exist, as it's the ID in os-release, which is ubuntu whenever flavor you are running), variant is hardcoded "desktop" for now, but it's set to "ubuntu", and version isn't the version id (like 18.04) but "desktop" [11:39] they were ran in a mate session as well :) [11:41] (+ another record with correct variant, but distro still manually set to ubuntu-mate) [12:01] good morning! [12:02] hey andyrock! [12:03] hey andyrock, how are you? had a good flight? [12:03] yeah long but all good [12:04] where are you exactly? [12:16] medellin [12:23] He follows his teacher 😂 [12:25] jbicha, seb128: btw I'm preparing an MP for gnome-control-center 3-28 [12:25] we need a modified (and simpler patch) [12:25] cool, great [12:33] andyrock, thanks [12:34] andyrock, hope you enjoy south america :) [12:38] didrocks Almost certainly me (the human) dicking around with ubuntu-report but not interaction between Welcome and ubuntu-report [12:40] jbicha: https://gitlab.gnome.org/GNOME/gnome-control-center/merge_requests/49/diffs [12:40] GNOME bug (Merge request) 49 in gnome-control-center "night-ligth-dialog: Avoid dereferencing invalid pointer" (comments: 0) [Opened] [12:40] I'm also preparing a new to fix the random warnings [12:40] Wimpress: ok, I think those were some direct curl usage, as in https://BASE_URL/distro/variante/version, you can only change BASE_URL from the tool [12:41] andyrock: ok, I'll wait for that one then :) [12:51] jbicha: and this one https://gitlab.gnome.org/GNOME/gnome-control-center/merge_requests/50/diffs?commit_id=475548d0c9ea5814d8aabc11317466011146c767 [12:51] GNOME bug (Merge request) 50 in gnome-control-center "night-light-dialog: Ignore G_IO_ERROR_CANCELLED errors" (comments: 0) [Opened] [12:52] the second MP includes both commits to avoid rebasing [13:40] jbicha, seb128 also if you could take care of this: https://code.launchpad.net/~azzar1/gnome-shell/fix-1768786/+merge/345203 === pstolowski is now known as pstolowski|lunch [13:47] andyrock, I don't know much about this patch but the commit message is confusing "Inhibit suspend if...screen lock on suspend is enabled " [13:47] why would we inhibit suspend? [13:47] to lock before [13:48] once it's locked we unhibit it [13:48] the problem is that if you disable automatic login we still want to inhibit the suspend before [13:48] because in ubuntu the lock on suspend is managed by another gsettings key [13:49] k, so "Inhibit suspend" is "Inhibit suspend until the screen is locked"? [13:59] andyrock, ^ [14:02] andyrock, jbicha, the fix makes sense to me, I'm a bit unsure why we have that patch though. Upstream is not locking the screen on suspend? that seems a bit weird ot me [14:02] or is that about allowing to disable the locking on suspend? [14:03] FYI, automated smoke tests for cosmic are activated and first desktop image promoted to cosmic [14:03] nice! [14:03] and upgrades from bionic too [14:03] duflu left but maybe he noticed tomorrow :) [14:03] I'm sure he will [14:03] seb128: guess we just need to replace or with and: "I"nhibit suspend if automatic screen lock is disabled and screen lock on suspend is enabled [14:04] right now if automatic screen lock is disabled we don't inhibit the suspend [14:04] we should if screen lock on suspend is enabled [14:04] right [14:04] I was just speaking about the wording of the changelog entry [14:04] it's a bit confusing as it is atm [14:05] seb128: so replacing or with and is fine for you? [14:06] that's what I did in the first place and for some reason I changed it later [14:07] andyrock, let me read it again :p [14:07] use more words [14:07] andyrock, that patch doesn't inhibit suspend, it only does until the screen is locked no? [14:08] I would stick to what I suggested [14:08] "Inhibit suspend until the screen is locked" [14:08] if that's what it does [14:08] then you can add the context [14:08] "Inhibit suspend until the screen is locked also in the case where <...> is used" [14:08] or something around those lines [14:10] popey, hey, can you approve https://community.ubuntu.com/t/canned-response-to-forwarding-files-bugs-upstream/5829/3? [14:12] seb128: kk [14:12] andyrock, thx [14:18] anyone know about oem-config-prepare? [14:18] Laney, ^^ maybe? [14:19] bit open ended :-) [14:19] it's blowing up because there is no oem user, wondering how that should be created [14:19] the question to be more specific, is does anyone know what creates the oem user/session that is used on first boot after installation? [14:19] Created symlink /etc/systemd/system/default.target → /lib/systemd/system/oem-config.target. [14:19] sudo: unknown user: oem [14:19] sudo: unable to initialize policy plugin [14:19] the context is that custom microsoft image, which is not built using ubiquity [14:20] the way it works in the ubiquity workflow is that it boots first into a temp oem user where the prepare icon on the desktop is clicked on after tweaking [14:20] we want to do that in the image build automatically there [14:20] should the build just adduser oem and start the prepare command then? [14:20] yeah, which prepare seems to be trying to sudo oem [14:20] or is there more magic to set up things? [14:21] i was surprised there was nothing in livecd-rootfs and ubiquity related to the oem user [14:21] that i could find [14:22] I think ubiquity creates it in oem_config mode [14:22] it pilots that page of the installer [14:22] one second, let me find the right bit [14:24] https://git.launchpad.net/ubiquity/tree/ubiquity/plugins/ubi-usersetup.py#n219 [14:25] seb128: done [14:26] popey, thanks [14:27] Laney, thx [14:27] so i guess i just need to create the user myself [14:28] this thing is using the oem mode somehow? [14:28] I thought it was just supposed to boot to a preinstalled desktop [14:28] Laney, yeah, but we want oem-config to run [14:28] to setup the user, etc [14:29] i had it running g-i-s via gdm, which worked well [14:29] but then we didn't get ubuntu welcome [14:30] alright [14:31] Laney, so the oem user should be a regular user, probably uid 1000 right? [14:31] then oem-config changes properties on it? [14:32] don't think it should be 1000, that's the real user's uid [14:32] ok [14:32] look at ok_handler [14:33] i see it now [14:33] thx [14:33] no problemo === pstolowski|lunch is now known as pstolowski [14:45] kenvandine: you could maybe hack g-i-s to still run the welcome wizard after logging in after completing new user mode [14:46] it might be a little annoying for GNOME users but we don't really have GNOME users for the new user mode [14:47] kenvandine: or — I forget if I mentioned it to you — but have you seen bug 1762287? [14:47] bug 1762287 in gnome-initial-setup (Ubuntu) "Offer Ubuntu customizations in new user mode too" [Wishlist,Triaged] https://launchpad.net/bugs/1762287 [14:48] LocutusOfBorg, hey, do you know how much libayatana-appindicator has been tested in other distros/desktops and where? and did they still use a different dbus namespace that libappindicator? if so how do things like unity or the gnome-shell extension handle the mix of client apps using one or the other? [14:48] sunweaver: ^ :) [14:49] jbicha, who is he? [14:50] The Debian maintainer [14:50] upstream. He is perhaps the Debian version of flexion (maintainer for Debian MATE) [14:51] k, thx [14:52] jbicha, do you know if webkit2gtk strings are being displayed in user facing UIs? seems like we don't have much translations for it/launchpad import didn't work in bionic (https://translations.launchpad.net/ubuntu/bionic/+source/webkit2gtk) [14:56] seb128: yes there are user strings. A simple test case is in Settings > Online Accounts and click to add a new Google account. The right-click options aren't translated in French [14:57] jbicha, ok, i'm adding that one to my list then :) [14:57] thx! [14:57] no templates available for xenial? [14:58] was that package source/name changed in between series? [14:58] or between xenial and xenial-security? [15:08] not really, but it only entered main in Feb 2016 [15:11] seb128: um, maybe it's as simple as that we don't built it --with translations or --with gnome [15:20] here's the proposed fix for cosmic: https://salsa.debian.org/webkit-team/webkit/merge_requests/1 [15:20] WebKIt bug (Merge request) 1 in webkit "Build with gnome-pkg-tools for Ubuntu language packs" (comments: 0) [Opened] [15:22] ubot5-ng: why do you capitalize webkit like that? [15:22] jbicha: I am only a bot, please don't think I'm intelligent :) [15:22] jbicha: Error: "why" is not a valid command. [15:56] it's not the bot; the project/team/thing is capitalised in that way [16:02] thanks :) === pstolowski is now known as pstolowski|afk [16:27] tsimonq2, did you try todays lubuntu image? I get bug 1754174 with automated ubiquity tests of Ubuntu. [16:27] bug 1754174 in ubiquity (Ubuntu) "[Lubuntu] "Install Lubuntu" fails with several commands not found and permission denied" [Critical,Fix released] https://launchpad.net/bugs/1754174 [16:57] jibel: That's because Lubuntu no longer uses Ubiquity. [16:58] We use Calamares now. [16:58] jibel: Unless... *please* don't tell me you were testing Bionic... [16:59] If that's present on Bionic, please sound alarms. [17:21] night all [17:40] tsimonq2, it's with cosmic and only with automated tests, manually it works fine [18:28] seb128, it is used in Debian since a lot of times and nobody reported issues so far, but I don't know.. let me check, did anybody complain so far? it should be a dropin replacement [18:39] jibel: OK [18:40] good night all [22:38] andyrock, how do I show the logout menu in g-s? [22:39] I'm on a details page but not seeing any change [22:46] robert_ancell: make sure you're running the built plugin (it could try to still load the old snapd plugin) [22:46] mmm it should appear like this: https://usercontent.irccloud-cdn.com/file/NxC87T1b/image.png [23:01] robert_ancell: btw I fixed all your suggestions [23:01] thx for the quick review [23:02] andyrock, np [23:03] That screenshot says "Sing out" :) [23:03] ops [23:03] yeah [23:03] I just noticed that too [23:03] :D [23:03] It should play some opera sound effect when you click on it [23:04] ahaha [23:04] I'm guessing the menu isn't showing because I haven't signed in? [23:04] it should show [23:04] but it should allow you to sign in [23:05] oh, ok. Must be something else... [23:05] ah, I stuffed up my git branch [23:07] andyrock, was there a discussion about why it only shows on the details page? [23:07] because login can also affect searches etc [23:07] nope but gnome-software can have more than one auth [23:08] at least in this way we can handle it defining a default auth [23:09] the default auth thing feels a bit odd - can't we just show all logged in accounts there (which will probably only be one in practice) [23:10] robert_ancell: that can be an option [23:10] I'll look into that [23:12] andyrock, something like gs_plugin_get_auths () that returns a GPtrArray (priv->auth_array). [23:12] Then you can iterate over that and show each one. [23:13] how do we distinguish one auth from the other? [23:13] e.g. doing like that we're going to have two auths now "snapd" and "ubuntuone" [23:14] I understand that "ubuntuone" is going away at some point [23:14] but for the moment it is still there [23:15] andyrock, you use gs_app_get_management_plugin which will return the Snap plugin. It only has one auth. [23:15] nope [23:15] The Ubuntu One plugin has the other one, and wont be visible. [23:15] they're shared [23:15] all the plugins share the same auth_array [23:16] ah [23:16] "sing out", well at least tha tmatches the optimism of the skype screenshot :0 [23:16] :) [23:17] robert_ancell: it is this way because e.g. ubuntu-reviews use ubuntuone auth that comes from a different plugin [23:17] andyrock, would it be an issue to show both accounts in that menu? [23:18] Because the U1 login could apply to the results (i.e. if we were using the Ubuntu review server for reviews). [23:18] it would be odd to have something like that: [23:18] Sign in... [23:18] Sign Out [23:18] ---- [23:18] Sign in... [23:19] Sign Out [23:19] in the same popover [23:19] we should at least agree on a way to distinguish the two [23:19] like [23:19] Snap Store [23:20] "Sign into Snap Store" [23:20] "Sign into Ubuntu One" [23:21] robert_ancell: and "Signed into SnapStore as " [23:21] ? [23:21] andyrock, I think what flagged this for me is gs_plugin_set_default_auth is adding public API to G-S, and Richard is probably going to scrutinize it. So a more generic solution is probably better. [23:21] robert_ancell: kk that works for me [23:22] my day is almost over [23:22] so I'll likely update it tomorrow [23:22] andyrock, worst case we can just distro patch out U1 for us, and we'll drop U1 in the near future [23:22] andyrock, sure, thanks!