[02:43] Trevinho, i would think you would override-prime [02:43] then manually prime it and add the magic after prime [02:43] assuming override-prime exists :) [06:31] good morning [06:33] mornin! [06:34] Morning didrocks, mirrorinthewall[ [06:37] hey mirrorinthewall[, duflu [06:42] something random I was thinking, in ubuntu software center would it be a useful/easy feature to add for people to be able to right click > "add to favorites"? If so where would I submit this [06:45] mirrorinthewall[: I think this is an upstream request for gnome software: https://gitlab.gnome.org/GNOME/gnome-software/issues [06:57] good morning desktoppers, happy Wednesday! [07:01] Morning oSoMoN [07:04] hey duflu [07:13] sil2100, morning! Does Poland have a single timezone? === pstolowski|afk is now known as pstolowski [07:16] salut oSoMoN :) [07:16] duflu: IIRC, it's the same than german/french one [07:16] Yeah, just checking. [07:17] (and italy/spain ofc, for not forgetting anyone around ;)) [07:17] Only the UK is different... for a change [07:18] yes! ;) [07:19] duflu: morning! Yes ;) [07:19] So koza not being around right now indicates he may not be around all day [07:25] salut didrocks [07:27] screen blanking goes off after 10s but it's set to 15min. Any idea how to fix that without logging off? [07:28] I guess changing the timeout doesn't reset it, correct? [07:28] no [07:29] I think L_aney is the one knowing more about those loginctl commands to know what owns the lock and reset… [07:38] morning all 0/ [07:38] hey c-lobrano [07:39] hi didrocks :) how are you? [07:42] I'm good, thanks! Yourself? [07:45] good morning desktopers [07:46] Morning jibel, c-lobrano, seb128 [07:46] seb128, are we meeting today? koza is not around [07:46] hey seb128 [07:47] didrocks: I'm fine, thanks. Keep replying on transparency effects :D [07:47] c-lobrano: I see, I see ;) [07:49] hey duflu, I would vote for skipping unless there is something you want to discuss [07:49] re didrocks [07:49] hey c-lobrano [07:49] seb128: 0/ [07:49] seb128, nothing... I am waiting to finish work on bionic regressions at least before switching to cosmic, which in turn is blocking more bluez work [07:50] duflu, k, let's skip then [07:52] morning [07:53] hey willcooke [07:54] Morning willcooke. No meeting it seems. [07:55] duflu, looks that way. I "saw" koza last Thursday and he said he's mega busy with a customer project [07:55] so I guess he's bunkered down [07:56] hey willcooke, since koza is not there and neither duflu or I had topics to discuss we just decided to skip, ok with you? or was there anything you wanted to discuss for bluetooth? [07:56] nothing pressing [07:56] fine by me [07:57] what are the non pressing things? ;) [07:57] Talk about getting the BlueZ snap building from the same LP project/code as the deb [07:57] trying to drop/upstream any patches we carry [07:58] and get a single build which can be used as a deb or a snap for all devices (Iot & Desktop) [07:58] -/- [07:58] Why don't Apple earpods (??) mic work [07:58] *doesn't [07:58] -/- [07:58] EOF [08:00] Morning seb128, do you have a gut feeling idea offhand what bug #1772238 might be about? [08:00] bug 1772238 in Ubuntu Budgie ".desktop file strings shown untranslated in Budgie" [High,Confirmed] https://launchpad.net/bugs/1772238 [08:02] yo [08:07] willcooke, I see, who is building the snap today is that koza and do you know what source they are using? [08:07] willcooke, for the earpods I guess it's a matter of somebody getting some and debugging [08:07] hey GunnarHj Laney, how are you? [08:07] chrisccoulson, hi, and another try, please push your packaging branches! [08:08] hey desktopers [08:08] seb128, yeah koza. I'm pretty sure he's already using the same source as the deb, but wanted to be sure. [08:08] k, so yeah, we need him to be around to discuss that one [08:09] hey ricotz [08:09] ricotz, what are those packaging branches you keep mentioning about? [08:09] seb128, hi [08:09] firefox and thunderbird [08:10] there are changes uploaded that he forgot to push to the vcs? === ecloud_ is now known as ecloud_wfh [08:10] yes, that is why I am trying to make him push them [08:11] GunnarHj, that bug description is not very useful, it doesn't even mention the locale used [08:11] ricotz, just get the debdiff from launchpad and commit it? ;) [08:12] seb128, I am close to doing that, but I am hoping for that his commits are structured for easier cherry-picking [08:12] seb128: Tested with German and Swedish, so it seems to be a general, not locale dependent issue. (The strings are translated.) [08:13] GunnarHj, can you give the output of "locale" and "env"? [08:14] GunnarHj, also is that only gnome-control-center or the desktop files from other applications as well (in the menu/dash or whatever they have that lists those) [08:14] willcooke, did you/someone try using the earpods in HFP/HSP mode? IIRC the default A2DP mode does not (by design) support simultaneous output and input [08:15] seb128: Will check that out and get back. [08:15] GunnarHj, I'm downloading an iso, let's see [08:15] ... it's a common problem with headsets [08:15] hey seb128 [08:15] I'm doing good, what about you? [08:15] Morning Laney [08:16] hi duflu [08:16] how's it going? [08:16] I'm good thanks, though I've a start of cold I think [08:16] Laney, going well. You? [08:19] spiffing [08:19] seb128: not another one :< [08:19] that's kids for you it seems :/ [08:20] hey hey Laney [08:20] jolly good [08:20] I say [08:21] duflu, I dont have any, but will pass that on and find out [08:21] willcooke, there was a bug with links to the BT spec pointing out the limitation. Can't find it today :/ [08:21] Apple tends to invent its own specs to work around such things [08:22] :)( [09:00] seb128: Added another comment to the bug report. [09:19] morning [09:20] hey Trevinho [09:20] you good? [09:20] hey Trevinho, how are you? [09:20] Hi Trevinho [09:21] hi duflu, seb128 and Laney [09:21] yes all good [09:21] you guys? [09:21] cloudy today [09:21] it's because you are leaving [09:21] hey Trevinho [09:22] it's grey/rainy here as well this morning [09:22] hi didrocks [09:23] eh, London is also getting gray... As it feels I'm leaving [09:23] GunnarHj, I saw, I downloaded the iso and I'm going to poke directly in a bit [09:23] :-D [09:24] Leaving.... London. ? [09:25] an eminently sensible decision [09:25] You leave us? Well I'll leave you first [09:26] But also return periodically [09:26] night duflu [09:26] willcooke, no, I was speaking as the EU [09:27] ah [09:28] Actually, speaking as Trevinho. [09:28] Enough confusion [09:29] duflu: ah, no, no... it was "london feels that I'm leaving, thus it cries" :-P [09:32] right. I'm confused. coffee time. (thanks duflu ;-)) [09:43] there we go. morning :-) [10:03] OK then. Night willcooke :) [10:03] And all [12:22] hey all [12:42] hey andyrock, how are you? [12:44] hey andyrock [12:52] seb128: all good you? [12:53] andyrock, I'm good thanks [13:01] fossfreedom: Since you are going to fix with budgie-desktop-environment, how about adding an override of [13:01] where "NotShowIn=GNOME;Unity;" is dropped? That would make IBus Preferences show up in the menu, which would be a step in addressing (or working around) bug #1755831. [13:01] bug 1755831 in Ubuntu Budgie "ibus not working properly" [Wishlist,Confirmed] https://launchpad.net/bugs/1755831 [13:03] fossfreedom: Hmm.. I was talking about /usr/share/applications/ibus-setup.desktop [13:05] GunnarHj, looks like fossfreedom figured out the issue, I was thinking about that on the way back from lunchand that was my guess that they probably had a buggy override :) [13:06] seb128: Right, I saw his entries in the bug report. Great! Hence my ping about another Budgie issue (about using IBus). [13:13] Laney: do you know if we can condition user system service unit based on a file in user's $HOME? I didn't find examples of ConditionPathExists in /usr/lib/systemd/user/ [13:14] I guess I can be ConditionPathExists="%h/foo/bar" [13:14] just wondering if you have experience [13:15] also: systemctl status xdg-desktop-portal-gtk.service [13:15] Unit xdg-desktop-portal-gtk.service could not be found. [13:15] where the service is in /usr/lib/systemd/user/ [13:16] unsure how this whole systemd user session is working [13:16] ah, you have to add --user [13:17] didrocks, https://www.freedesktop.org/software/systemd/man/systemd.unit.html#Specifiers %h [13:17] %h User home directory [13:17] xnox: yeah, that's what I found and asked for ConditionPathExists="%h/foo/bar" [13:17] if you confirm it works that way, great! :) thanks [13:17] didrocks, it should, no idea why you have "" around it. [13:18] didrocks, i believe you must not use "" [13:18] strings, we did some protection in the path for spaces [13:18] but in that case, it doesn't really matter [13:18] didrocks, it must be systemd-escaped [13:18] oh right, that… [13:19] $ systemd-escape -p '/somepath with space' [13:19] somepath\x20with\x20space [13:20] great ;) [13:20] thx [13:22] xnox: when I look at /usr/lib/systemd/user/pulseaudio.socket, I see that there is a WantedBy=sockets.target [13:22] yes [13:22] I was expecting to see somewhere on the file system sockets.target.wantedby + a symlink to the socket [13:22] as for system systemd unit [13:22] you'd think so wouldn't you =) [13:22] magic [13:22] all user units are auto-enabled for all users by default [13:22] I see /var/lib/gdm3/.config/systemd/user/sockets.target.wants/pulseaudio.socket [13:23] ahhh [13:23] its an audio socket, probably it doesnt work by looking bt by listening instead :P [13:23] ok, so all user units are auto enabled… Interesting, that system vs user have a different policy [13:23] that is semi-true about auto-enabling things. One needs not to run --enable most of the time, for most of the user units. [13:23] xnox: I think you remember my long thread on distro policy and how to cope with those ;) [13:23] ah [13:24] didrocks, yeah, there are weird rules as to what is auto-enabled, and what is not =) [13:24] hum, I guess I should just try then? [13:24] yeah [13:24] "nice" :p [13:24] * didrocks has a sense of dejà vu, twice in a less than an hour [13:24] ok, I'll give it a try and see ;) [13:25] (thanks again) [13:32] willcooke: ping [13:33] I am in "asking to join" status for the hangout [13:35] jbicha, huh, I moved it to tomorrow, you should have got a new invite [13:35] sec [13:36] oh, I didn't read carefully enough and didn't click "add to my calendar" [13:36] :) [13:36] jbicha, I've invited your other google acc to the new meeting so that it will resend the invite [13:36] sorry for the confusion jbicha [13:37] ok, thanks :) [13:38] my big point is that if we follow https://wiki.ubuntu.com/StableReleaseUpdates#Procedure [13:38] once a bug is fixed in cosmic, we have to target it to bionic, we can't use the rls-bb-incoming tag since that isn't useful for Fix Released bugs [13:39] but we can assign people to bugs === jhernandez is now known as jhernandez_ [14:08] xnox: so, WantedBy=default.target for user session doesn't start the unit on startup, we can't like for system units running enable at package install time ofc, any idea what's the standard way? [14:09] I think that's where Laney knows ;) [14:09] what do you mean? [14:09] I'm going to try shipping /usr/lib/systemd/user/default.target.wants/ symlink [14:09] Add the symlink [14:09] there's no dh_systemd stuff for user units [14:10] Laney: it seems xnox was telling that systemd was treating user units as enabled by default [14:10] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=890509 [14:10] Debian bug 890509 in debhelper,init-system-helpers "dh_installsystemd/init-system-helpers: support user units" [Wishlist,Open] [14:10] I don't know about this rule [14:11] ok, I'll just ship the symlink manually then! [14:12] thanks Laney [14:13] ++ [14:20] Laney: seb128: FYI, if you logout quickly after login in, this is what keeps the session active for a little while (I would say ~30s): http://paste.ubuntu.com/p/whsYY9tbPK/ [14:21] didrocks, i don't think default.target is the one you want. [14:21] didrocks, what is this unit for? should it run on graphical desktop start / login; or on console/ssh logins too? [14:22] xnox: both are fine [14:23] it's sending a previously collected report [14:24] didrocks, on every login?! [14:24] xnox: there is a conditional path [14:24] so man 7 systemd.special gives example of a nautilus.service which BindsTo=graphical-session.target [14:24] but there was something else about it. [14:25] well, I don't care about a graphical-session, as said [14:26] looking at /usr/lib/systemd/user on my machine, i see things shipping symlinks in .wants and .requires directories, by hand [14:26] yes, see the discussion we had with L_aney ^ [14:26] there is no special magic for user session, it's the same rule, at least :) [14:27] there was magic.... [14:27] * xnox is still looking [14:27] are you sure you didn't get tricked by dbus activation? [14:28] didrocks, what are the files called that launch a session? [14:28] wdym? [14:28] /usr/share/xsessions/*.desktop? [14:28] why are we talking about magic? [14:28] possibly [14:29] yeah, I don't know, I guess we solved the issue, and there is no magic ;) [14:29] right, that looks broken. [14:29] or at least changed.... alot..... [14:29] didrocks: you know about path units? [14:29] depending on what you're doing you might want one of those [14:30] it used to be the case that xsessions/*.desktop file would start systemd user target for graphical session and that would start all the things. [14:30] right, that is not the case with Shell at the minute. [14:30] but looks like not to be the case anymore =/ [14:31] and there was like a shell script to interate and enable units too [14:31] but Didier has said a few times that he doesn't want this anyway so I'm not sure what the problem is [14:33] (don't think there was ever a script like that) [14:34] I feel close to having a gdm started by systemd units [14:34] Laney: yeah, I know about path units, I don't think it's a good approach for that case as we want next login to start resending a pending report, not as soon as we create the file [14:35] * Laney is trying to come up with a good unit structure [14:35] (as we create the file if we can't send it immediately due to network) [14:35] nice! [14:35] alright, that's up to you [14:36] didrocks, yeah, I already noticed "pending session closing" due pulseaudio&co, that seems worth fixing one day [14:51] Trevinho, I assigned you bug #1768610, interesting packaging bug :) [14:51] bug 1768610 in nux (Ubuntu) "leftover conffile forces GNOME is software rendering" [High,Confirmed] https://launchpad.net/bugs/1768610 [14:52] nux-tools has a Xsession.d script that does " /usr/lib/nux/unity_support_test || export LIBGL_ALWAYS_SOFTWARE=1", when the deb is removed but not purged that leads to software rendering being forced [14:52] *sigh* [14:52] conffiles suck sometime :/ [14:52] unsure what's the right fix there though [14:52] seb128, we had something similar before with like X11 snippets from upstart. Basically pick any package to force purge the connfile... E.g. gnome-shell should force purge it. [14:53] we should change the script to not export the env if the cmd is missing [14:53] seb128, there is no way in apt/dpkg to declare ConflictsPurged: [14:53] but maybe we need to clean out the conffile also [14:53] problem is one can remove the package, and the conffile is left on disk :-( [14:53] right [14:53] fixing the script in xenial and bionic would reduce the number of users hitting the issue [14:54] but not fix those who already removed the package [15:03] seb128: ouch... And that wasn't either needed anymore. I never liked such thing though [15:04] An Ubuntu desktop trigger is too much? [15:04] Hi [15:05] focus didn't switch... 🤔 [15:13] trigger what? [15:16] seb128: check if that file is there, and if the package has been purged... [16:05] Trevinho, I don't think many users hit it since I don't think we remove unity on upgrades atm [16:06] but we should fix that because at some point users are probably going to clean that out === pstolowski is now known as pstolowski|afk [17:19] night all [17:37] Hi jbicha, I have prepared the SRU of pkgbinarymangler. Can you sponsor? [17:59] GunnarHj: I thought we were going to keep it in cosmic first a bit longer? [18:34] jbicha: Well, the SRU procedure takes it time.. But you are in the middle of it, and know better than me what's a reasonable trade-off, so whenever you feel it's appropriate. === JanC_ is now known as JanC