[05:03] hi [05:09] Morning hikiko [05:09] Morning duflu :-) [06:04] Mirv: i'm trying to compile the Qt5.6 package from landing ppa 011 for the raspberry pi with videocore eglfs support. i got it all to compile in a ppa but then all the tests fail because the ppa builder does not have videocore graphics hardware. is there some way i can work around this? (like disable the tests) [06:05] here's the log https://launchpadlibrarian.net/274404982/buildlog_ubuntu-xenial-armhf.qtbase-opensource-src_5.6.1+dfsg-3ubuntu1~xenialoverlay1~4ppa1_BUILDING.txt.gz [06:06] and here's the ppa: https://launchpad.net/~a-j-buxton/+archive/ubuntu/qt5-raspi-eglfs [06:20] ali1234: edit debian/rules so that override_dh_auto_test: section is empty (that is, an empty line after that and then override_dh_auto_clean: starts already) [06:22] Good morning [06:27] Mirv: thanks, i hope it works this time :) [06:34] morning [06:34] Hey pitti, andyrock. [07:01] Laney, are you adding the origin_hostname support for snap to gnome-software? [08:03] morning [08:03] robert_ancell: not doing anything on gnome-software right now [08:03] Laney, ok [08:03] robert_ancell: I would look at PK 1.0 and slimming down the apt plugin first of all [08:03] unless you or someone else does first [08:03] but not for a week or two [08:05] I'm busy on snap / paid support at the moment [08:05] nod [08:10] hey Laney [08:12] pitti! [08:12] how are you? [08:14] Laney: quite fine, thanks! went for a run yesterday evening still, felt much better after that and a shower [08:14] excellent [08:14] did you have to sit on a train for ages while it was stuck? [08:15] Laney: no, just in Cologne, I went to the lounge [08:16] I have a frequent traveller card for the Bahn and thus can use them for free [08:16] so, more time for hacking :-) [08:17] ooh [08:17] Laney: hmm: [laney] port session-migration: TODO → this is done, isn't it? [08:17] (at least in git) [08:17] or is there something missing for that? [08:19] pitti: I didn't upload s-m itself yet [08:19] (shortly) [08:20] Laney: right, I meant that functinoally it is working in the s-g-s git [08:20] sure [08:20] thanks [08:20] if that's DONE, then DONE [08:20] I set it to (in git): INPROGRESS, like the others [08:21] ok, thanks [08:22] dear libvirt, please enable your units [08:23] * Laney screams at its rules file [08:24] hey Laney pitti [08:24] & robert_ancell [08:24] seb128, Guten Morgen [08:24] tag seb128 [08:24] happy friday to you [08:25] bonjour seb128 ! [08:25] hey robert_ancell [08:25] o/ [08:26] * seb128 was out of bread and yogurt so went outside to grab a snack and coffee [08:26] robert_ancell, did you investigate g-s/snap/store icon issue? [08:26] seb128, no [08:26] to figure out if the bug is on the g-s side or the store one [08:26] k [08:27] Actually, I had a quick look before I left for here. My thought was the store side [08:27] Or the snapd side at least [08:27] which was my feeling as well [08:27] We pretty much just do whatever snapd tells us [08:27] but the store guys dimissed it [08:27] e.g bug #1588385 [08:27] bug 1588385 in Snappy "icon urls no longer listed" [High,Invalid] https://launchpad.net/bugs/1588385 [08:28] so if we don't bounce it back their way it's not going to be look at I think [08:28] seb128, we talked about making a libsnap-glib lib. I think that will make it much easier to hack up little progs to test these things. [08:28] right [08:34] pitti, you reverted the GNOME_SESSION_ID workitem from done to inprogess, was that wanted? [08:48] Morning [08:51] hey Trevinho, how are you? [08:53] seb128: hi seb128, allright... Had a free day yesterday, some local sightseeing. There's always things near your place that you don't expect it exists ;-) [08:54] seb128: I saw, fixed again, thanks [08:54] pitti, yw! [08:55] Trevinho, ah, nice, didn't take friday off as well to just do a long and relaxing w.e? ;-) [08:56] Naaa... I've to use my holidays for some seaside too :-P [08:56] :-) === greyback_ is now known as greyback === hikiko is now known as hikiko|ln [10:36] all software should be banned from having directories that start with 'de' and sort before 'debian' [10:36] * Laney writes to ftpmaster [11:07] Laney: like "debconf"? :-) [11:07] or debhelper [11:08] exactly [11:13] Laney: are you still working on the two gnome-keyring-* jobs? [11:13] Laney: I'm currently looking at openssh, and these two jobs need to coordinate (gnome-keyring-ssh must come Before=ssh-agent.service) [11:14] so that g-k wins (if enabled) [11:14] pitti: No, I finished, and I did that already :-) [11:14] * pitti hugs Laney [11:16] Laney: you still want to wait with the shutdown patch until upstream reviews it, I take it? [11:16] (we have this pkill workaround for now) [11:16] pitti: yeah, your hack is okay for the meantime [11:17] * Laney is going to switch to working on some of the last 3.20 bits now [11:17] Laney: +ExecStartPre=/usr/bin/test -z "${SSH_AUTH_SOCK}" [11:18] Laney: this is a bit of a wart, as the unit would fail if this command fails [11:18] same for the grep [11:18] I think this is better moved into the /bin/sh [11:19] Laney: I'll take care of it, so you do the 3.20 bits [11:21] pitti: I know, there's isn't a Condition type thing for this [11:22] if it actually matters then of course you can move it [11:24] something like http://paste.ubuntu.com/20440267/, testing now [11:24] related to this [11:25] Jul 22 12:24:58 yakkety-systemd-session systemd[2493]: [/usr/lib/systemd/user/gpg-agent.service:12] Invalid escape sequences in line, correcting: "/bin/sh -ec ' out=$$(gpgconf --list-options gpg [11:25] Jul 22 12:24:58 yakkety-systemd-session systemd[2493]: [/usr/lib/systemd/user/gpg-agent.service:17] Invalid escape sequences in line, correcting: "/bin/sh -ec ' out=$$(gpgconf --list-options gpg [11:25] systemd is complaining about something in these multi line scripts [11:25] oh thanks, will look [11:26] in addition to them being ugly [11:27] print \$10 [11:27] this presumably [11:27] * pitti grabs some lunch and then does these three [11:28] hmm -- no pancakes or Broedje Vermeer today, just plain bread *sob* [11:28] seb128: ^ can you telegram some? [11:28] pitti: is a failed unit a problem other than cosmetic? [11:37] Laney: mostly cosmetic, unless some other unit Requires= it [11:37] ok [11:38] I liked having a way to specify preconditions [11:38] pitti, need to invent pancake-over-ip [11:38] ├─gnome-keyring-ssh.service [11:38] │ └─2364 /usr/bin/gnome-keyring-daemon --start --components ssh [11:38] hmm, this isn't the intended daemon [11:39] Jul 22 13:37:54 pid1 gnome-keyring-daemon[2112]: ** Message: couldn't access control socket: /run/user/1000/keyring/control: No such file or directory [11:39] same as with the upstart unit, this smells fishy; anyway, post-lunch problem [11:39] that's supposed to be made by the pam thing [12:19] hi all [12:21] anyone knows what could interfere with the background transparency of the popup from switcher module? [12:21] in one case "custom build" i have access to the 'background color' option, in a .deb package from same project i don't [12:22] i mean, i have access to it but it doesn't change anything [12:37] ksamak, what switcher? the compiz/unity alt-tab? Trevinho andyrock hikiko might be able to help you with that [12:43] hikiko, hi. lool installed -proposed in a VM to test the low-gfx mode and his unity desktop doesn't work anymore. The launcher doesn't appear, the only thing that displays is the screen background. Any idea what might have happened? And how can he get back to a working desktop? unity --reset doesn't help... [12:44] hikiko: it's likely that the breakage is unrelated but rather a set of settings that my system can't recover from because of hard crashes after trying the proposed stuff [13:02] seb128: nautilus-desktop-canvas-view.c is not in trunk neither in a deb patch [13:03] where do we take if from? :) [13:05] seb128: i've a fix for this https://bugs.launchpad.net/ubuntu/+source/nautilus/+bug/1564375 [13:05] Launchpad bug 1564375 in nautilus (Ubuntu) "Unity should take display scaling into account to avoid placement of desktop icons under launcher" [Medium,In progress] [13:05] but I'm not sure where to send it [13:06] i mean would be nice to modify directly the patch that re-introduce that file [13:08] seb128: ah found it [13:08] in trunk is under nautilus-desktop [13:09] right, they moved the desktop rendering to another component [13:09] sverdy, those macos users ... [13:09] it's likely that the opengl plugin or something got unloaded, those situations are annoying [13:10] it took me 15 min last time to reset the right configs, ended up start ccsm from a vt with the correct display and dbus env to re-enable things [13:10] seb128: tsss [13:10] seb128: ok i think we need two different patches for y and x (if you want that patch to be SRUed) [13:10] andyrock, do you know what's the proper way to reset the compiz config nowadays? [13:10] seb128: so actually sverdy made me turn off opengl acceleration exactly for this purpose, but I think it broke the overall setup [13:11] seb128: nope [13:11] it's not that doable [13:11] because the way we do profiles [13:12] lool, yeah, the annoying part is that when the opengl plugin fails to load it desactivate itself from the config and next start have it and things that depends on it like unity desactivated [13:12] mmm why it fails to load? [13:12] I think I ended up getting the DBUS_SESSION_BUS from gnome-session and set that to a vt and DISPLAY=:0 ccsm [13:12] seb128: so I turned accel back on and everything work [13:12] i've a branch to force unity loading [13:13] seb128: but the whole point of sverdy was to make me test this config [13:13] not sure how relevant it is though [13:13] well dunno what you changed exactly [13:13] but I guess disactivating opengl != using software rendering [13:14] yep [13:14] the new version is updated for software rendering [13:14] seb128: sverdy asked me to turn off opengl accel, install proposed, and then everything stopped working [13:14] unity fails to load without opengl plugin [13:14] try under kvm or such [13:14] I turned opengl accel on again and everything works [13:14] I think he wanted to show me slow rendering got faster [13:14] unaccelerated rendering I guess [13:14] right, kvm an iso [13:14] btw my week is over [13:15] happy we [13:15] andyrock, have a good w.e! [13:34] sorry I didn't see the highlight [13:34] lool, still here? [13:34] there was a problem in a past daily build [13:34] because some files weren't packed [13:34] but that is fixed now [13:35] it might be that if you use an outdated mirror [13:35] if the crash is because of that, next release will fix the problem [14:00] ok, I'm dropping offline for a while, tennis and then travelling (going to north of France for the w.e), I might be offline a bit during the trip otherwise I do catchup once arrived before calling it a week [14:00] have a nice w.e for those I don't see again today [14:00] bbl [14:03] hf seb128! [14:04] thanks, you too, have a nice w.e! [14:05] * Laney is preparing to break the world on monday [14:07] Laney: hm, did the gnome ssh agent actually work for you? I did a couple of fixes and now SSH_AUTH_SOCK is correctly exported; but "ssh 10.0.2.2" still asks me for the key pwd on the CLI, not with a graphical dialog [14:07] I wonder if these env vars you dropped are still required after all [14:08] same with upstart [14:11] hm, no, it doesn't export them any more, so nothing to eval [14:13] pitti: http://people.canonical.com/~laney/weird-things/ssh.png [14:14] yeah, that's what I would expect [14:14] so it works [14:14] strace shows that ssh is talking to the daemon [14:34] but it doesn't even work in plain yakkety in my VM, so maybe I'm just missing some package there [14:34] (wouldn't know what though -- fairly standard desktop install) [14:34] what is $SSH_AUTH_SOCK? [14:34] /run/user/1000/keyring/ssh, as expected [14:35] same as on my laptop [14:35] don't know then [14:35] and keyring-daemonm does wake up when I ssh [14:35] right, nevermind [14:35] :( [14:35] I got the ssh-agent working, and g-k-r gets priority when it is enabled [14:35] seahorse shows it unlocked? [14:36] Laney: no entries in "secure shell" [14:36] while I do have them on my laptop [14:36] but I'd think it would ask me anyway and auto-import it the first time [14:37] ooh - I created an ECC key, maybe keyring chokes on those [14:37] it did that for me [14:37] hmm [14:37] I just did ssh-keygen, so whatever the default is [14:38] RSA [14:38] nod [14:38] aaah! [14:39] RSA works fine indeed [14:39] sheesh [14:39] https://bugzilla.gnome.org/show_bug.cgi?id=641082 [14:39] Gnome bug 641082 in ssh-agent "Can't handle ECDSA keys" [Enhancement,New] [14:39] Laney, thank you [14:39] i also hate gtk theming now. [14:39] Laney: mystery solved, thanks and sorry for bothering [14:39] did they really change GtkProgressBar to progressbar =) #NotNice [14:39] * Laney meows at xnox [14:40] well, now it's documented API [14:40] so "shouldn't" "change" [14:41] https://developer.gnome.org/gtk3/stable/GtkProgressBar.html#id-1.3.8.6.10.6 [14:41] folks in case you missed it, call for lighting talks at FOSDEM has opened. [14:41] https://fosdem.org/2017/news/2016-07-20-call-for-participation/ [14:44] pitti, seb -> should i be at all be looking into android-property-bridge upstart event emitter at all to become systemd-ish? [14:44] or snap-based touch will not use that at all? [14:44] xnox: I figure we'll need a non-upstart mechanism for this either way [14:45] ok. [14:45] hmm [14:45] pitti, the idea i had was to generate a template unit, and then have small daemon sitting on the system side that will toggle things on & off. [14:46] and then other units could bindsto android-property-bridge@key-val or some such [14:46] xnox: right, I faintly remember; that sounds good [14:46] * xnox ponders if i should just emit things over dbus [14:46] xnox: I figure the upstart bridge is more or less that, right? (a daemon which listens to changes and translates them to upstart events) [14:46] yes. [14:47] what i did is tweak the bridge code slightly to work with pid1 as systemd =) [14:47] i should have stub code somewhere around with it. [14:48] however, i had no device to test that on =/ [14:48] does goldfish work at all? [14:48] pitti, do you have a systemd booting phone around? i though you did have that at one point.... [14:48] (pid1 systemd, with session under upstart) [14:48] I heared the ubuntu-emulator doesn't any more (but I haven't tested myself recently) [14:48] xnox: I did, but we dropped support for dual-boot, so no ubuntu phone any more [14:49] ack [14:49] xnox: but I uploaded all my changes to ubuntu proper, so any ubuntu phone ought to reasonably start with systemd (unless they made dramatic changes to the upstart jobs) [14:49] pitti, maybe sil2100 can bring a phone or two to the december cdo-core sprint =) [14:49] pitti, ack. [14:49] yes, that would be useful [15:01] I can bring two, but one is my main phone so only one would be for fun and play [15:01] ;) === JoseeAntonioR is now known as jose [15:43] man [15:43] just what is this unity-greeter problem === pesari_ is now known as pesari [17:01] BAH [17:01] i'll look more on monday [17:01] laters! [19:15] Mirv: i got the package to compile but it doesn't work :( [19:16] i had to remove an installation rule about /xcb/*/glintegrations because those didn't get built [19:16] i think maybe i need to add a different gl integration or something