=== Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha [08:05] happy friday desktopers! [08:06] hey, happy friday! [08:07] Laney, to you too! [08:07] Laney, we got your grey&rainy weather, I'm not thanking you for that :p [08:08] hah [08:08] that's what you get for showing off earlier in the week :P [08:08] seems so :/ [08:08] oh, another settings update landing in saucy [08:09] that and the new intel driver [08:09] things start looking good for release ;-) [08:09] ship it === sil2100_ is now known as sil2100 [08:38] good morning people [08:38] how are we all doing [08:40] czajkowski, hey, happy friday! we are doing well indeed ;-) [08:41] excellent glad to hear it [08:41] any plans for the weekend ? [08:42] celebrating didrocks getting married [08:43] which involves spending 6 hours in a train to go there tomorrow and then same time to come back on sunday [08:43] that should do for the w.e ;-) [08:43] czajkowski, you? [08:45] that is quite some train [08:45] Morning! Ah, so that's the reason for Didier's holidaying, huh? ;) [08:48] seb128: dinner with my mum my bf family over the weekend [08:48] and hanging out, I've had a busy week [08:50] sil2100, good morning, indeed [08:51] Laney, yeah, the line from here to Lyon is a slow one :/ [08:51] oh well, didrocks doesn't get married every week [08:52] and those trains are nice, they have a nice seats, are quiet, have power plugs ... they even have a "bar" wagon with food and drinks [08:52] I guess I'm going to take my laptop and play some good starcraft or something [08:53] nice [08:53] i'm sure it'll be great! [08:53] ;-) [08:54] pitti vuntz and some others are going to be there, I'm sure it's going to be fun ;-) [08:54] (well not in the train, but for the weeding) [09:11] oh noes [09:13] something is weird [09:13] http://paste.ubuntu.com/6191464/ [09:13] oh [09:13] is not disabled [09:13] weird [09:15] Laney, what is weird? [09:15] I misread that message [09:15] oh ok [09:15] anyway I restarted my laptop and now shortcut keys don't work [09:16] is the plugin enable? [09:16] (not unity) [09:16] yes [09:16] shortcut = multimedia keys? [09:16] yeah, and e.g. ctrl-alt-t for terminal [09:16] volume [09:17] weird, ... do you have anything in the g-s-d log? [09:20] oh, didn't they stop doing media keys? [09:20] it might have been so long since i rebooted that i was on 3.6 before [09:20] they did, but darkxst reverted that [09:20] for unity only [09:20] right [09:21] this is panel/xmonad [09:21] gnome-shell handles those keys [09:21] oh [09:21] :( [09:21] yeah, you are probably screwed then :/ [09:21] darkxst: is that supposed to be broken? [09:21] I also get some weird notification bubbles instead of osd now [09:21] we should change the "for unity" to be "for any non-gnome-shell session" [09:22] +1, I tried to use gnome-panel and got similar issues [09:23] hmm, it watches for unity on dbus and turns media keys on then [09:23] right [09:23] it should watch for gnome-shell and turn it off [09:24] mitya57: any idea what the notification problem is? [09:24] Laney, what notifications do you get? [09:24] I haven't yet debugged it, and on a different machine now [09:24] been so long since I've seen them but it might be notification-daemon ones [09:24] with an 'x' in the top right [09:25] yes, for me those were notification-daemon ones [09:25] Laney, my bet is that it's due to robru changes in http://launchpadlibrarian.net/124264231/notify-osd_0.9.34-0ubuntu5_0.9.35daily12.11.28-0ubuntu1.diff.gz [09:25] ... with 'x' and part of the message somewhere outside the screen :) [09:25] +++ notify-osd-0.9.35daily12.11.28/data/org.freedesktop.Notifications.service.in 2012-11-28 06:30:35.000000000 +0000 [09:25] @@ -1,3 +1,3 @@ [09:25] [D-BUS Service] [09:25] Name=org.freedesktop.Notifications [09:25] -Exec=/bin/sh -c 'if [ ! -x /usr/lib/notification-daemon/notification-daemon ] || [ "$GDMSESSION" = guest-restricted ] || [ "$GDMSESSION" = gnome-classic-guest-restricted ] [ "$GDMSESSION" = default -a "$(basename `readlink /etc/alternatives/x-session-manager`)" = gnome-session ] || [ "$GDMSESSION" = ubuntu ] || [ "$GDMSESSION" = ubuntu-2d ]; then exec /usr/lib/notify-osd/notify-osd; else exec /usr/lib/notification-daemon/notification-daemon; [09:25] seb128: Error: "@" is not a valid command. [09:25] fi' [09:25] +Exec=@LIBEXECDIR@/notify-osd [09:25] Laney, ^ he dropped our diff when putting the packaging inline upstream [09:26] ah [09:26] I saw that a bit back, but since it was not creating apparent issues we decided to not bother (we being didrocks and I, we discussed it on this channel) [09:26] Laney, seb128 media keys are not supposed to be broken [09:26] darkxst, what is supposed to handle them in a xmonad session? [09:27] darkxst, it seems you made the revert specific to unity [09:27] darkxst: I don't think panel handles them [09:27] so you get broken there [09:27] oh, what is xmonad even? [09:27] it's just a window manager [09:27] would be the same for metacity i guess [09:28] I can re-do the patch for GNOME instead of Unity [09:29] Like invert it? [09:29] I don't know if anything else handles the keys atm, but if not that sounds like a good idea [09:29] yeh like check for a GNOME session [09:29] instead [09:29] * Laney nods [09:29] let me know if you want testing [09:30] rather than checking for unity and enable, check for gnome-shell and disable in that case only? [09:30] seb128, yah [09:30] might still break flashback, but thats generally just broken anyway [09:31] * Laney tries putting the osd dbus service file back as it was [09:31] darkxst: why? [09:31] (why will it break flashback and why it is broken anyway) [09:33] mitya57, flashback is mostly more like Unity, but it uses the GNOME desktop session name (I believe) [09:34] darkxst: can you check for some gnome-shell D-Bus interface instead of checking the session name? [09:35] yes but its not reliable, because unity clones some of the gnome-shell d-bus interfaces ;( [09:37] anyway there are lots of patches, across a bunch of packages that check for XDG_CURRENT_DESKTOP [09:37] and in most cases the results will be wrong for flashback [09:38] we can rename the flashback sessions to ubuntu-* then [09:44] ok the notification-daemon thing isn't because of the dbus service file [09:44] it's because it's now started via xdg autostart [09:45] You may then want to add something like this: https://mail.gnome.org/archives/gnome-flashback-list/2013-September/msg00069.html [09:45] AutostartCondition=GNOME3 unless-session [09:45] gnome [09:45] Laney, that somewhat makes sense I guess, it makes easier to decide on what desktop it should be activated, compared to dbus activation which has no OnlyShowIn === vrruiz_ is now known as rvr [09:47] I'd have thought that we always want notify-osd (if installed) on Ubuntu [09:48] "we"? [09:48] KDE and gnome-shell probably don't [09:48] not sure about xfce [09:48] don't install it then [09:48] indeed they do not [09:49] then you should get the correct one dbus activated for you [09:50] oh, I see [09:50] notification-daemon isn't dbus activated any more [09:51] there are people who install "vanilla" Ubuntu and then install and use gnome-shell [09:51] they shouldn't get notify-osd [09:52] how do you tell? [09:54] * mitya57 thinks gnome-shell notifications are better integrated and more consistent with other parts of the shell [09:56] ok, so [09:56] seb128: can I ask you for some packaging change ACKs? [09:56] what's DESKTOP_SESSION under gnome shell? [09:56] sil2100, sure [09:58] Laney, I think the consensus, last time we discussed it on #debian-gnome was that dbus activation was wrong and that the notification agents should all use autostarts rather, so they can target their environment with ShowOnlyIn [09:58] Laney, "gnome" [09:59] seb128: https://jenkins.qa.ubuntu.com/view/cu2d/view/Saucy/view/Indicators/job/cu2d-indicators-saucy-3.0publish/56/artifact/packaging_changes_indicator-sound_12.10.2+13.10.20131004-0ubuntu1.diff <- this one seems sane [09:59] Laney, e.g the notification are part of the desktop experience [09:59] seb128: https://jenkins.qa.ubuntu.com/view/cu2d/view/Saucy/view/Indicators/job/cu2d-indicators-saucy-3.0publish/56/artifact/packaging_changes_indicator-bluetooth_0.0.6+13.10.20131004-0ubuntu1.diff <- not sure about the rationale of this one, but on desktop it makes sense I guess [09:59] sil2100, +1 (I acked that mr for upstream ;-) [09:59] sil2100, it was bringing the phone settings on some other images, +1 as well [10:00] darkxst, Laney: on my Debian machine it was "default" or "gnome" depending on what was selected in gdm, though I'm not sure if it's the case in Ubuntu [10:03] I guess if the goal is to have every "gnome" environment having n-daemon then I have to change it locally [10:04] It's not an explicit goal though; something changed here to make it start for me now when it didn't before [10:04] Laney, no, I think the goal is to have gnome-shell/kde/(xfce?) uses their own notification system [10:05] Laney, we should use notify-osd is gnome-panel sessions imhp [10:05] imho [10:05] that's our default implementation, if a desktop doesn't specifically specify they want something else [10:07] glatzor, hey [10:10] glatzor, have you seen https://bugs.launchpad.net/ubuntu/+source/apturl/+bug/1200775 ? is that an aptdaemon issue/something that changed there which makes port in other code needed? [10:10] Launchpad bug 1200775 in apturl (Ubuntu) "apturl-gtk crashed with AttributeError in __init__(): 'InstallBackendAptdaemon' object has no attribute 'connect'" [High,Confirmed] [10:14] M3e^tkj8 [10:14] doh [10:16] time for a new password ;-) [10:17] seems so, silly lock screen! [10:27] glatzor, unping, that's an update-manager/pitti issue [10:27] darkxst: re patches in bug 1189309: do you know why the first one was uploaded instead of the second one? [10:27] Launchpad bug 1189309 in network-manager-applet (Ubuntu) "nm-applet crashed with SIGSEGV in gtk_status_icon_set_visible()" [Medium,Fix released] https://launchpad.net/bugs/1189309 [10:40] glatzor, just for the record, not a pitti issue, he just happened to have uploaded the version, found the right people now and reassigned ;-) [10:46] I upgraded today and have a black screen (with mouse pointer) when I try to log in. compiz is running, unity-panel-service is running. Any guesses where I could start debugging? [10:48] Hmm, and gnome-control-center-unity has a greater version in raring-updates than in saucy; don't know if that's directly related but it doesn't seem good [10:49] cjwatson, that's specific to g-c-c, should have an impact on your session [10:49] YM shouldn't? [10:49] cjwatson, can you share your ~/.cache/upstart/gnome-session.log? [10:49] cjwatson, yes, sorry [10:49] cjwatson, that's just a .so for g-c-c, that's not your issue [10:49] but I'm going to make sure we fix the lower version issue [10:49] oh, here, now it's started, but it took forever [10:50] http://paste.ubuntu.com/6191729/ [10:50] it took at least a couple of minutes to start with no visible feedback [10:50] those WARN 2013-10-04 11:44:19 unity.glib.dbus.server GLibDBusServer.cpp:586 Can't register object 'com.canonical.Unity.Launcher' yet as we don't have a connection, waiting for it... [10:50] are weird [10:50] Trevinho, bregma: ^ did you see that before? [10:51] cjwatson, I wonder if the dbus changes from xnox could create an issue... [10:51] seb128: I don't have his change yet [10:52] oh, not his latest one, but I do have a previous one [10:52] dbus 1.6.12-0ubuntu6 [10:52] does seem a plausible possibility [10:52] cjwatson, is that reproducible? e.g if you start a guest session? [10:53] seb128: well, dbus --fork & expect fork, should only make "started dbus" event be emitted later. so things that e.g. "start on starting dbus" will have invalid dbus_session_bus environment variable, which only starts working after "started dbus" [10:53] * xnox ponders if we should be exporting dbus_session_bus only at post-start [10:53] guest session starts quickly; let me try rebooting [10:53] well, it sort of starts - I don't have indicators [10:54] seb128: that changed fixed unity on the phone.... [10:54] cjwatson, that's a known issue and fixed in lightdm trunk, alt-f2 /usr/lib/unity/unity-panel-service to get them [10:54] not convinced this is running with the same session type though [10:55] I had those messages with previous dbus FWIW [10:55] or start the guest from the greeter rather than from the indicator-session [10:55] Laney, but no hang? [10:55] no [10:55] dist-upgrading now [10:55] new xorg, intel drivers too I suppose [10:56] k, maybe those messages are a red-hearing [10:56] herring [10:56] * Laney restarts [10:58] sigh, this time it worked fine [11:07] OK, two more reboots and it's still fine, so this doesn't seem reproducible at all. Sorry to bother you [11:07] * cjwatson will clean up the cause of his accidentally slightly mixed raring/saucy system though [11:11] seb128, I often see that warning during startup, I don't think it means there's a problem [11:14] bregma, right, seemed like it could create delay, but it seems they are happening on systems which don't have delays [11:14] cjwatson, yeah, I don't know, maybe something hit a segfault and apport blocked stuff while processing the issue... === MacSlow is now known as MacSlow|lunch [12:12] sil2100: https://code.launchpad.net/~timo-jyrinki/cupstream2distro-config/sdk_add_libthumbnailer0/+merge/189287 [12:13] Mirv: done! [12:14] sil2100: thanks! === MacSlow|lunch is now known as MacSlow [12:49] seb128: yes, I saw that warning and it's just fine... [12:49] seb128: the fact is that we ask for the same name in multiple places, the first which arrives takes it [12:50] seb128: and we need to keep that way since some component are optional [12:51] seb128: ah, I misread... that's related to object registration... That's fine as well, basically we delay the obj registration on bus until we don't have a valid connection... [13:03] Trevinho, right, thanks [13:08] seb128: I've a question for you... Have you read around reports of not-working scroll areas (such as gedit or evolution mail composer) with mouse scroll wheel (they work with touchpad!)? [13:08] Trevinho, no I didn't, and I use all day long and didn't notice any issue there [13:09] seb128: so I'm unlucky.... THat's really weird... No scroll with mouse, but it works with touchpad... and it's the only place [13:09] seb128: couldn't be due to the usage of smooth scrolling not supported by my device? [13:09] Trevinho, did you try in gtk3-demo? could be in gtk3 textview widgets only? [13:10] seb128: not yet [13:11] attente, what do we need to release the indicator-stack? Is it a manual button push or is there a problem with the autolanding for the stack, like tests failing? [13:11] mterry, you should ask sil2100 and Mirv about landing [13:12] sil2100, Mirv: ^ :) [13:12] mterry: we just did the landing of indicators, does anything else need landing than just the indicators themselves? [13:12] mterry, https://launchpad.net/ubuntu/saucy/+queue?queue_state=1&queue_text=indicator [13:12] sil2100, oh, I guess they haven't hit saucy yet [13:12] mterry, I guess you want to nag -release then [13:13] mterry: yes, they're in the unapproved queue actually [13:21] seb128: so, it doesn't work in any text widget... but also neither in the gtk3-demo app itself: so I can't scroll on the sidebar, in the source code or in the Info tab [13:22] seb128: ah, no sorry the sidebar works [13:22] (not when unfocused, though as it used to be) [13:22] Trevinho, is that only under unity? [13:22] seb128: let me try to switch [13:31] sil2100, mterry: could you please review https://code.launchpad.net/~seb128/gnome-control-center-unity/swap-color-use-gdkcolor/+merge/189309 , it's an easy one [13:32] sil2100, we are going to need g-c-c-unity published as well since we didn't have any landing for it since we changed the naming scheme and the raring-updates version is newer than the saucy one, which is going to create upgrade issues [13:33] seb128: looking [13:34] seb128: not in a guest session... maybe i've messed up mine :/ [13:34] seb128: ok, add it to the Landing Asks, I'll pick it up and release anyway [13:34] Trevinho, could be a setting causing problems :/ [13:34] sil2100, thanks, do we need landing ask for non touch code? [13:35] sil2100, that's a gnome-control-center fix, e.g desktop only [13:35] only if it touches touch [13:35] ogra_, it doesn't, it's a gnome-control-center panel :p [13:35] for other stuff the release team ius responsible as usual [13:42] tkamppeter_, bregma: your login issue is https://bugs.launchpad.net/ubuntu/+source/lightdm/+bug/1036830 and should be fixed with https://code.launchpad.net/~robert-ancell/lightdm/handle-corrupt-xauthority/+merge/188256 [13:42] Launchpad bug 1036830 in lightdm (Ubuntu) "Can't log in unless remove .Xauthority or use gdm" [Undecided,Confirmed] [13:42] huzzah [13:43] seb128: I wonder ;) Not really sure personally if we use the Landing Asks only for things related to touch or to overall now, I'm always confused [13:43] sil2100, ogra said only touch [13:43] seb128: then let's skip that indeed - is everything merged in already that needs release? [13:44] sil2100, no, I need https://code.launchpad.net/~seb128/gnome-control-center-unity/swap-color-use-gdkcolor/+merge/189309 review/approved by you or mterry [13:45] I'll install it and test now that it's built [13:46] sil2100, ok, I already that here, basically open the appearance capplet, go to color, pick a horizontal or vertical 2 color option, pick 2 colors, see that it does the effect on the background, click the swap button, see that it works === nessita_ is now known as nessita [13:47] mterry, hey, ^ if you want to do an easy review for me (that's the same gdkcolor fix I did some months ago, but in another place of the code I forgot to change back then) [13:47] oh, you just did [13:47] mterry, thanks ;-) [13:47] sil2100, mterry code approved, so we should be good ;-) [13:48] sil2100, and I see you approved as well [13:48] rock on ;-) [13:49] Approved! [13:49] Let's get it merged and released ;p [14:06] mterry, I'd like to branch and release upstream unity-greeter 13.10.3 and then bump trunk to 14.01 ... any objections? [14:07] seb128: about that scrolling issue, i've also tried to log on gnome flashback session...and.... if I run it after logging out from the unity one then it doesn't work, but if I select it as my first session after boot it does work [14:08] seb128: so, do you have any idea what could be set wrong in that unity session (not pribably much unity related)? [14:08] as the unity guest session work as expected [14:17] sil2100, and the change is merged in trunk, so should we go with the new build [14:17] Trevinho, no idea, maybe ibus? [14:17] that would be weird though [14:18] Trevinho, log out, go to a vt, check what processes might still be running for your user? [14:19] ok, I'll check it... since it must be saved on something that is volatile.. [14:20] seb128: ok, let me spin it on === tkamppeter_ is now known as tkamppeter [14:24] bregma, I just pushed 13.10.3 to saucy, so yeah, bump trunk [14:26] seb128, thanks, is this fix already in Saucy? [14:27] tkamppeter, no it's not [14:28] seb128, will it get in soon (next lightdm?)? [14:32] mterry, am I correct in believing unity-greeter does not have any CI or autolanding? [14:36] bregma, it does have autolanding, but yesterday I got in the bad habit of uploading some small fixes without it, if that's what you're looking at [14:36] naughty boy [14:43] tkamppeter, next week most likely, but you can probably build/try trunk if you want to confirm it fixes your issue [15:06] bregma, is christownsend off today? [15:06] seb128 yes [15:06] bregma, hum, ok [15:08] bregma, I was looking at https://bugs.launchpad.net/ubuntu/+source/gsettings-desktop-schemas/+bug/1201405 but I can't reproduce the issue [15:08] Launchpad bug 1201405 in gsettings-desktop-schemas (Ubuntu) "[regression] Wall: keyboard bindings for 'Move up & Move down' are being set to "super+prior & super+next'" [High,In progress] [15:09] bregma, do you know if that's randomly happening? [15:09] bregma, I tried with a daily iso in a vm or with a guest session on my work machine,ctrl-alt-up/down works [15:09] seb128, we've never seen it happen [15:10] ok, the description made it seems like it would happy for everyone [15:25] seb128: g-c-c-u published! [15:25] seb128, the patch fixes the problem. Thank you very much! [15:25] Why do we patch some extra methods into org.gnome.SessionManager? [15:37] * Laney starts a T-cycle things to do list [15:37] they seem unnecessary now [15:50] sil2100, great, thanks! [15:50] tkamppeter, thanks for confirming! [15:50] Laney, what methods? [15:50] RequestShutdown and RequestReboot [15:50] Laney, T is nice, but we are not done yet with bug fixing for S :p [15:50] I know [15:50] Laney, I think chrisccoulson added those back in the days, not sure why now [15:50] that's why it's a to do list and not things i'm doing now [15:50] :P [15:51] I think at the time g-session didn't have anything to do that [15:51] https://launchpad.net/ubuntu/+source/gnome-session/2.26.0svn20090408-0ubuntu2 [15:51] yeah, saw it [15:52] man we had a lot of arches back then [15:52] blast from the past === ember_ is now known as ember [15:52] it still exists! [15:52] i was younger than 30 when i did that. true story! [15:53] Laney, chrisccoulson: https://lists.ubuntu.com/archives/ubuntu-devel/2009-April/028102.html [15:53] that has the explanations [15:55] man [15:55] you ubuntu archaeologist [15:55] lol [15:56] the good old times, when chrisccoulson was still a desktopers [15:56] hah :) [15:57] seb128, what? i didn't make any changes to notify-osd [15:57] robru, you dropped our Ubuntu diff when you inlined the packaging [15:58] seb128, when was that? that must have been forever ago, i don't even remember it [15:58] robru, yeah, it's almost a year ago [15:59] nov 2012 [15:59] seb128, hah, wow. so what problem is happening now? [15:59] turned out to not be to do with that [15:59] robru, the problem was not due to that [15:59] ah, ok ;-) [15:59] robru, I just mentioned that in case [16:00] seb128, so anyways, powersaving works fine on a guest account, and also in the VT! what dotfile can possibly be in my account that breaks powersaving?? [16:00] robru, I noticed the change earlier this year, and discussed it with didrocks, but we decided it was not creating issues to not have it so we didn't add it back [16:00] robru, I would recommend to dump the gsettings config of the g-s-d power plugin and compare the values [16:00] maybe some gsettings key in there [16:01] robru, e.g "gsettings list-recursively org.gnome.settings-daemon.plugins.power > keys" [16:01] robru, do that for your user and guest and diff the files [16:01] seb128, how do I do that? I looked at it with dconf-editor and it claimed there was no schema available (but i checked; of course the schema is installed, otherwise everything would crash without it) [16:03] seb128, http://paste.ubuntu.com/6192824/ can this be it?? [16:04] robru, shouldn't [16:04] robru, that's the only diff? [16:04] seb128, yeah! [16:05] robru, you might want to just dump the whole db (e.g drop the org.gnome.settings-daemon.plugins.power) argument [16:05] robru, thank diff and grep for anything that contains power dpms, suspend etc [16:05] thank->then [16:05] seb128, ok. in a meeting now, will try that shortly [16:05] seb128, thanks [16:06] robru, yw [16:35] robru: hello man! [16:36] robru: could you take a look at this branch here? Make sure it makes sense, since there are REALLY minor nit-picks [16:36] robru: https://code.launchpad.net/~sil2100/ubuntu-settings-components/small_packaging_mods/+merge/189368 [16:36] robru: since I think you did the packaging, and it's good ;) [16:36] sil2100, yeah, sure [16:36] I go now, have a nice weekend [16:37] sil2100, ok, you as well [16:38] sil2100, thanks, you as well! [16:38] sil2100, wait, still around? [16:38] sil2100, why did you change the boilerplate "if you arent' a memeber..." text? that is the same in all daily released packages... [16:39] robru: actually, the 'trends' changed ;) [16:39] what is ubuntu-settings-components? [16:39] robru: it's a nitpick, but we try to get things this way from now on, check: https://wiki.ubuntu.com/DailyRelease/InlinePackaging (as it got modified) [16:39] * Laney perks up [16:39] sil2100, so we have to change that for all of them? actually I like this new boilerplate, since you can copy & paste without having to change the team name [16:40] robru: indeed! I think we don't have to, we might do that 'by chance' if you make some other modifications, but I would say it doesn't make sense to change that on purpose [16:40] robru: I'm using this for new components for now ;) [16:40] See you guys! [16:40] sil2100, ok ok, good night! [16:42] Laney, I don't actually know! ;-) [16:42] haha [16:43] Laney, I just package it ;-) [16:45] * Laney eyes GunnarHj [16:45] you're supposed to wait for FFes to be accepted before uploading :P [16:49] Laney: any update on g-s-d shortcuts stuff? [16:50] mitya57: darkxst is going to fix the patch [16:51] Laney: ah, right, and what about notify-osd then? [16:51] don't know what to do [16:52] I uninstalled it [16:52] :p [16:52] why don't you like what I suggested (AutostartCondition=GNOME3 unless-session gnome)? [16:53] that's what it has [16:53] it just means that it autostarts on all non-shell sessions [16:53] i.e. no way to get osd on those if you have n-daemon installed [16:54] ah, and your session is not managed by gnome-session? [16:54] it is [16:54] but it's called gnome-xmonad [16:54] so you get autostarted [16:55] so notification-daemon has that AutostartCondition? [16:55] sure does [16:55] it's probably wrong [16:56] what else would you have? [16:56] I think on GNOME sessions we want either notify-osd or nothing [16:56] seb128, grepping through gsettings for power/suspend/dpms, i just don't see anything like that. [16:56] no notifications? [16:56] that would be a bit bad [16:56] nothing=managed by gnome-shell [16:57] we're talking about non-shell gnome sessions [16:57] all I can think of is a wrapper to exit if the dbus service file exists [16:58] then you get osd if it's installed and can't force daemon [16:58] don't know if that matters [16:59] seb128, wait, in ...plugins.power, I have three keys that don't seem to be part of the schema. idle-dim-battery, sleep-display-ac, and sleep-display-battery. could those have caused this? [17:01] Laney: we can make notification-daemon recommend notify-osd, this way notify-osd will be present on all environments that depend on notification-daemon [17:02] notify-osd is anyway a successor for notification-daemon, and the latter is kind of abandoned [17:02] it's a part of gnome flashback [17:03] i don't think we could do anything as aggressive that [17:03] as that [17:03] robru, could be, but weird if they are not in the schemas, that shouldn't be possible [17:04] last commit 2012-09-03 [17:04] talk to the ubuntu-gnome guys [17:04] I don't know any person that will prefer n-d to notify-osd [17:04] it's their thing really [17:04] darkxst: ^ [17:04] seb128, dunno, do you have those keys? dconf-editor says 'no schema' when I click on them, but other keys in that same plugin have a schema. my guess would be that the schema changed, but i still have those keys lying around from before [17:04] Laney, they don't use it, gnome-shell has its own built in notifications [17:05] the flashback session does [17:05] I assume they care about that at least a bit [17:05] robru, oh, you didn't say you were using dconf(-editor), those are leftover keys indeed [17:05]