=== johnlage is now known as JOHNLAGE === JOHNLAGE is now known as johnlage [03:56] I installed Android Studio according to the instructions on this page: https://wiki.ubuntu.com/ubuntu-make [03:56] and supposedly that also installs the Android SDK [03:56] what path does it install the SDK to? [03:57] I ask because I am getting this error: Error: ANDROID_HOME is not set and "android" command not in your PATH. You must fulfill at least one of these conditions. === duflu_ is now known as duflu [06:40] Good morning [07:14] good morning [07:27] morning [07:28] hi didrocks! [07:28] hi pitti! [07:28] hey larsu ;) [07:28] cold going to an end? [07:29] hopefully. It's definitely better than yesterday [07:29] thanks for asking :) [07:29] how are you? [07:30] I'm great thanks! [07:30] starting to get crazy without doing any exercise and so no work break [07:31] I'll try to do some cycling today during the break, seems the weather will permit that [07:32] :) [08:06] Laney: :/ https://bugzilla.gnome.org/show_bug.cgi?id=739567#c2 [08:06] Gnome bug 739567 in general "Comment in sol.c is needlessly offensive" [Normal,Unconfirmed] [08:06] head -> desk === qengho_ is now known as CardinalFang [08:22] good morning desktopers [08:24] hey seb128 [08:26] lut didrocks :-) [08:26] seb128: do you mind reviewing jayatana in the NEW queue? I just pushed it and as I've done quite a lot of packaging cleanup, don't want to ack it myself :) [08:26] didrocks, sure can do [08:26] thanks! [08:47] * willcooke -> still sick [08:48] bbl [08:48] larsu, ^ gonna miss our 1:1 sorry [09:04] hey hey [09:04] larsu: grim [09:05] Laney: sorry, didn't want to start your morning like that. Good morning! :) [09:07] morning Laney! [09:07] hey Laney [09:10] happy friday folks ;-) [09:10] Laney: I didn't know we had this --enable-distro thing in gnome-terminal [09:10] that's profoundly ugly [09:11] I think he did it after getting told off [09:14] but why add all this complexity? [09:14] I really wonder why he cares so much [09:15] avoiding a complete climbdown? [09:40] didrocks, is systemd supposed to write a syslog or dmesg? [09:40] I've neither of those being written and journalctl is empty [09:43] seb128: yeah, it's supposed to forward to /var/log/syslog [09:43] seb128: sudo journalctl is empty? [09:44] didrocks, oh, no, you need to sudo it... [09:44] yeah, it's a bug on the list [09:44] so that people on the admin group can as well without sudo [09:45] it should still forward to syslog [09:45] right, that got me confused [09:45] empty journal, no hint that you need to use sudo [09:45] it doesn't [09:45] but I rm-ed syslog (to have fresh logs) [09:45] hum, I doubt you created /var/log/journald, right? [09:45] ah [09:45] so maybe it skips it if the file doesn't exist? [09:45] probably it fallbacks to permanent journald storage then [09:45] no, but I rmed /var/log/*log* [09:46] well rather *.log *.1 syslog dmesg [09:46] seb128: quite easy to know, if you sudo journalctl [09:46] well, that is working, thanks [09:46] do you only have the current boot? [09:46] or older logs [09:46] yes [09:46] only current [09:46] ah, so it disables forwarding and not activate permanent storage, interesting [09:47] did you rm and reboot? [09:47] or didn't reboot yet [09:50] reboot [09:50] so, not an opened fd [09:51] worth logging a bug IMHO [09:51] can be a syslog thing TBH [09:53] bug saying what? [09:53] "deleting syslog should activate journal mode"? [09:54] seb128: I would rather expect: deleting logs in /var/log/ shouldn't stop logging further boot [09:55] k [09:55] (via syslog) [09:55] well, boots are logged [09:55] or at least the current one is [09:55] since it's in journalctl [09:56] seb128: persistent logging then? [09:56] or "logging on disk" [09:57] didrocks, let me check, maybe it has old logs [09:57] can be, didn't check yet [09:57] but sudo systemctl --list-boots has only one entry [09:57] I guess that would record previous boot if it had those [09:58] I reboot 3 times since [09:58] the --boot option helps says "ID or, if unspecified, the current boot" [09:58] so I guess --list-boots should list all boots [09:58] since there is only 1 there is an issue with persistency [09:58] seb128: if you don't have permanent journald logging, it will only list last boot [09:59] so even with the syslog forwarding working, you won't get that [09:59] (I guess you s/systemctl/journalctl) [10:00] yeah, sorry, journalctl ;-) [10:00] basically, it's: [10:00] grrr, autofingers, I keep wanting to type "ctrl" [10:00] instead of ctl :p [10:00] - you always get all journalctl features for current boot [10:00] - you only get them on previous boot with journald persistent logging (not enabled by default) [10:01] - otherwise, if syslog is installed, it will fw logs to it for previous boots, and you won't have access to them via journalctl [10:01] by default, on debian/ubuntu, we enable syslog forwarding by default [10:01] we discussed in case syslog is removed to enable persistent journald logging automatically [10:02] (which is just about mkdir /var/log/journal) [10:06] cherry-picked some stuff, mir backend builds now [10:07] Laney, great! [10:07] ;-) [10:08] does anyone know what " '$nbr': '[1-6]$' means in bash? [10:09] is that "true if $nbr is a number between 1 and 6"? [10:12] haven't seen that syntax before [10:12] share the script? [10:13] Laney, /etc/init.d/console-setup [10:14] if expr "$(fgconsole 2>/dev/null || true)" : '[1-6]$' >/dev/null && \ [10:14] (! type plymouth >/dev/null || ! plymouth --ping); then [10:14] log_action_begin_msg "Setting up console font and keymap" [10:14] ah [10:14] expr, I think that's a regex match [10:15] so basically if "fgconsole" returns a number between 1 and 6? [10:15] ye [10:15] k [10:15] next question, why this check? [10:15] hint, under systemd, that number is "7" [10:15] which makes the keymap not being applied to consoles on boot [10:16] I guess I should move to -devel [10:16] * seb128 does that [10:16] 7 for me too and I'm not using systemd atm [10:17] k, so maybe I don't understand the script [10:17] ah, that's because I ran from X of courser [10:17] ah [10:17] that was my next question, how did you run it :p [10:17] 1 on vt1 [10:17] in my case I just did an "fgconsole" call in the init script and I'm looking at syslog after boot [10:17] right [10:17] it's the console you are on [10:18] but what is that supposed to be for init scripts [10:18] not manually ran [10:18] and why this check? [10:20] --force [10:20] Do not check whether we are on the console. Notice that you can be forced to hard-reboot your computer if you run setupcon [10:20] with this option and you are on a X server. [10:20] probably because of that [10:21] hum [10:22] well, that seems to be the wrong way to check if you are on an xserver [10:22] since systemd runs it on vt7 [10:22] and there is no xserver :p [10:22] seb128: s/systemd/ubuntu/ [10:22] seb128: I guess it's lightdm forcing that [10:22] systemd on ubuntu [10:22] (on fedora with systemd, it's vt1) [10:22] didrocks, doesn't happen under upstart though [10:22] so not "ubuntu" [10:22] seb128: we run on vt7, even under upstart, don't we? [10:23] hum [10:23] so why is the keymap correct under upstart? [10:23] * seb128 looks more into the issue [10:23] seb128: you mean on !x_session [10:23] no, on boot [10:23] seb128: on x, the keymap is correct, isn't it? [10:23] boot with upstart as pid1, console is azerty [10:23] use init=/bin/systemd [10:23] console is qwerty [10:24] dunno [10:24] seb128: what do you mean by "console", like, you go on emergency mode? [10:24] on X lightdm/u-s-d sets it to my user value [10:24] no [10:24] like I do ctrl-alt-f1 [10:24] yeah, so that's vt1, nothing to do with where the X server is running, right? [10:24] no, you are the one who mentioned X [10:25] that confuses me ;-) [10:25] 11:22:02 seb128 | since systemd runs it on vt7 [10:25] Xorg doesn't even run in those tests [10:25] what's the "it" then? :) [10:25] the init script [10:25] if expr "$(fgconsole 2>/dev/null || true)" : '[1-6]$' >/dev/null && \ [10:25] (! type plymouth >/dev/null || ! plymouth --ping); then [10:25] log_action_begin_msg "Setting up console font and keymap" [10:25] I put a [10:25] ah, you talked about the X server with Laney first [10:25] "fgconsole" there [10:25] ok [10:25] and look at journalctlk [10:25] are you sure it's running on vt1, this script? [10:25] and it printed "7" [10:25] no [10:25] it runs on 7 [10:25] did you try that under upstart? [10:25] systemd is spawning vts on demand [10:25] according to the journalctl log [10:26] Laney, not yet, but under upstart my keymap is azerty [10:26] so I assume the script works [10:26] it would maybe rule out this script [10:26] didn't look at the differences yet [10:26] i.e. upstart could have a job [10:26] or something else might do it there [10:26] ok, I think running it under upstart and see if it's running under vt7 or not will shade some lights [10:26] well, on what? [10:26] I still think that line [10:26] if expr "$(fgconsole 2>/dev/null || true)" : '[1-6]$' >/dev/null && \ [10:26] is buggy [10:27] whatever the outcome of upstart or other run is [10:27] that script should run on boot, next line is [10:27] if run_by_init; then [10:27] so running by init is a supported case and should work [10:27] no? [10:27] then i may happen that some other upstart script set the keyboard as well [10:28] still doesn't change the fact that console-setup should be doing it and is not? [10:28] yeah, but at least, it would rule out if this script is in use or not [10:28] (doesn't change that it's maybe buggy, of no use, and should be removed) [10:28] well, that's orthogonal [10:28] or another bug [10:28] yeah, but while we are at it, can be interesting to decipher [10:28] right, I'm in the middle of debugging [10:28] and know if we should look at some other place, or in this script [10:28] one step at the time ;-) [10:29] well, we can look at fixing that script in any case [10:29] even if it doesn't fix the keymap bug [10:29] yeah, not what I'm telling [10:30] is there some magic command for "give me grub menu on next boot"? [10:30] * seb128 tired to key pressing keys to not get the menu [10:30] keeping shift pressed? I'm sure there is some flags, but don't know [10:30] to detect faulty boot [10:30] well, half of the time that doesn't work for some reason [10:31] shift kept pressed [10:31] seb128: I saw that you have to press it during the bios [10:31] not before [10:31] at least, on my hw [10:31] too early then? [10:31] was the case for me [10:31] I think it's the same here [10:31] it's annoying :p [10:31] * seb128 reboots a third time [10:31] yeah, I'm sure there are more magic, like "it's not a state" :) [10:33] oh nice, eurostar is directly leaving from Lyon now [10:33] rather than having to change in paris? [10:33] yeah [10:34] well s/now/next May/ [10:34] (or lille, I preferred changing in Lille) [10:35] * seb128 gives up on getting a grub menu and edit /etc/default/grub [10:35] i heard there might be one direct to berlin :-o [10:35] seb128: I'm in grubby world as well [10:35] trying to have a nice patch to have an upstart boot option [10:35] but so that this only happens when we start to have systemd by default [10:36] shrug, of course I forgot to update-grub :p [10:37] :p [10:38] Laney: so, this script is running on vt7 under usptart (that's what you told)? [10:39] nah [10:39] ah [10:39] I didn't debug it [10:39] leaving that to others for now [10:39] seb128: you are rebooting under upstart or should I try? [10:39] my bet is that it's not running on vt7 under upstart [10:39] and so, it sets the locale [10:40] didrocks, I just booted under upstart [10:40] ok [10:40] it exit 0 early [10:40] oh? [10:41] seb128: /etc/init/console-setup.conf [10:41] if [ -f /lib/lsb/init-functions ]; then [10:41] . /lib/lsb/init-functions [10:41] seb128: yeah, don't look further, this is the job setting it up under upstart ^ [10:42] Hi seb128! [10:42] GunnarHj, hey [10:42] seb128: Is https://code.launchpad.net/~gunnarhj/ubuntu/vivid/gdm/config-error-dialog/+merge/239729 still on your 'list'? Also, would it help if I asked someone in the Swedish LoCo to test it, considering that I'm not able to do so myself? Or will you test anyway when reviewing? [10:42] GunnarHj, no, it was never on my list, I commented on it because it was similar to the lightdm one, but I didn't use/touch gdm for years and I don't feel like uploading it, letting that to Ubuntu GNOME team [10:43] didrocks, so seems the script bails out under upstart [10:43] seb128: Ok, subscribing ubuntu-sponsors again, then. [10:43] GunnarHj, thanks [10:43] seb128: yeah, and as we have the specific upstart job for it, that works… [10:44] seb128: I wonder if systemd isn't just supposed to have systemd-localed doing the job and something doesn't work with our config [10:44] like, it depends on /etc/locale.conf [10:44] which we don't have [10:45] didrocks, no, systemd has systemd-vconsole [10:45] http://www.freedesktop.org/software/systemd/man/vconsole.conf.html [10:45] seb128: trying to change the locale on a fedora system I have handy [10:45] ah, interesting [10:46] didrocks, but it's turned off in debian/rules because console-data/setup do that [10:46] and nobody had time to properly figure out how to replace those yet [10:46] * Disable systemd-vconsole-setup.service for now. [10:46] didrocks, see pitti's comment on https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1400393 [10:46] Launchpad bug 1400393 in systemd (Ubuntu) "wrong keyboard layout" [Undecided,New] [10:48] seb128: ok, seems pitti has the more context on this. I don't see the value if we have a delta there with upstream, but I think there are good reasons [10:48] didrocks, just "need to be reviewed before enabled" it seems [10:48] didrocks: I don't have much context, I didn't look into this at all yet [10:48] seb128: well, I'm afraid that there are as well "keep in sync with other init systems" consideration [10:49] I use US layout anyway, so I didn't notice [10:49] yeah, for now "compatible with sysvinit/upstart" >> "use upstream solution which works differently under systemd" [10:49] yeah, so I guess we'll have yet another compatibility layer (and a custom service) to provide [10:50] pitti: let's discuss that next week, when you have time if you wish [10:50] didrocks: or rather fix the init.d script to DTRT? [10:50] we might have our ubuntu specific bits only in the upstart job [10:51] * didrocks not really a fan to have an init script doing something that is mostly built-in [10:52] I think the init script is launched on the only available vt (vt7) [10:52] me neither -- as long as the builtin stuff DTRT, that's fine [10:52] seb128: are you going to try removing the vt check? [10:52] we should maybe look at enabled vgconsole then? [10:52] didrocks, yes [10:53] but lunch first, bbiab [11:08] I have prepared an SRU as bug 1386241, now this SRU will take longer to get verified (have to sort out some problems with printer manufacturers) and I want to get another SRU into Trusty, bug 1401835 which can get verified easily once the package is in proposed. How should I proceed? [11:08] bug 1386241 in system-config-printer (Ubuntu Trusty) "Add the full IPP Everywhere support from Utopic to Trusty" [High,Fix committed] https://launchpad.net/bugs/1386241 [11:08] bug 1401835 in system-config-printer (Ubuntu Utopic) "system-config-printer does not automatically download and install printer driver packages any more" [High,In progress] https://launchpad.net/bugs/1401835 [11:08] The conflict is in Trusty, bug 1401835 is to propose the SRU for both Utopic and Trusty. [11:19] * Laney hunts for a usb stick [11:53] blurgh [12:25] ctrl-shift-d works for inspector now [12:26] so good [12:26] thanks to ricotz === greyback__ is now known as greyback [12:44] tkamppeter, upload the easy fix first, get that moved to updates, then do another SRU with the longer to verify one [13:19] larsu: looks like gnome-tweak-tool has borderful buttons in addition to virt-manager (also a transparent background) [14:30] seb128, the long one I have uploaded already some time a go and it got approved for verification. Now it turns out that the long one needs more changes (have to work out with manufacturers) and this week I got together the short one. Is it possible to somehow reject the current version of the long one and later on, after having fixed it and the short one being completed, upload a new version of the long one to the original [14:30] bug report and get that veriofied then? [14:30] tkamppeter, sure, you can do another SRU and ask the SRU team to discard the current one, you just need to use a version > to the current one === CardinalFang is now known as qengho [14:41] Laney: noted (sorry, crazy Friday) [14:41] np [14:41] should probably use the pad [14:41] ... [14:41] ok, great, with current vivid/unity8-desktop (desktop-next), unity8 has apps running in windows mode and you can move/resize those [14:42] seb128: how about unity7? [14:42] did Trevinho's patch land? [14:42] it's early not-doing-a-lot mode (no resize handle, no alt-tab), but still cool ;-) [14:42] seb128: that's very cool though [14:42] larsu, different topic? I'm speaking about wm under unity8 [14:42] unity7 has wm since day 1 :p [14:42] it uses compiz [14:43] seb128: yes, different (but old) topic about windows with csd [14:43] Trevinho, ^ [14:43] how is that work going? [14:43] larsu: ah I was confused, I thought you were talking about the gdk mir backend... [14:43] hehe [14:43] larsu: on compiz side, not yet [14:44] so testing gtk3-demo [14:44] windows can be moved and resized, and decorations from GTK are on the left [14:44] still no border [14:44] * larsu sees this not happening this cycle [14:44] again. [14:44] seb128: using the mir-demo-shell some of these stuff is working [14:44] nice to see that it has "x_-" on the left though [14:44] larsu: no, it will :P [14:45] Trevinho, well, we are speaking unity7 there [14:45] unity8 is a different topic I guess [14:45] wm is just starting being added to it [14:45] seb128: ah, sorry..... Yes [14:45] Trevinho: this is what you said the last two cycles... [14:46] not trying to start a huge discussion though [14:46] larsu: nope, just the last one... and did some work, but then the irreparable happened :P [14:59] seb127, how do I contact the SRU team, do they have a special IRC channel? [14:59] you subtracted 1 from seb128, how mean! [15:00] lol [15:00] tkamppeter, no, they don't, just comment on the SRU bug asking to drop the current SRU or try #ubuntu-devel [15:05] ah, dist-upgrade finished; unity8 session works [15:06] what's the key to make desktop files show up in the dash? [15:07] seb128, thanks, comment added to bug 1386241. [15:07] bug 1386241 in system-config-printer (Ubuntu Trusty) "Add the full IPP Everywhere support from Utopic to Trusty" [High,Fix committed] https://launchpad.net/bugs/1386241 [15:08] Laney, X-Ubuntu-Touch=true [15:09] ty [15:29] should adding that to gedit.desktop and running it be enough? [15:31] "running it"? [15:31] what are you trying to do? [15:31] adding the key to gedit.desktop and restarting unity8-dash should make gedit be listed in the apps icons [15:31] under unity8 [15:31] clicking on it then should start gedit [15:32] ok yeah, I just get a black screen [15:32] k, so backend is buggy [15:32] that's already the case is stock vivid [15:32] I hopped the update would fix it :/ [15:32] ya [15:33] I thought the update was buggy but the archive version is the same [15:36] you can look to .cache/upstart/legacy......log [15:36] it might have errors [15:38] http://paste.ubuntu.com/9490515/ [15:39] Laney, dunno about that sorry :/ [15:39] no error there I can see [15:39] http://paste.ubuntu.com/9490532/ 3.12 [15:39] seems basically the same [15:42] yeah [15:42] could be a change in mir that bugged it [15:42] hum [15:48] Trevinho, that libcanberra commit, how does it work? it's a build time thing and we build our libcanberra with gtk/x11 no? [15:48] seb128: no, it's a runtime check [15:49] Trevinho, shrug, ignore me, I though you added the ifdef [15:49] but that also adds the if (!GDK_IS_X11_DISPLAY(d)) call in it ;-) [15:50] seb128: also... I've branches for cogl and clutter for mir support that I'll push to master branches soon... So, it would be nice to package them... :P [15:50] Trevinho, when is that segfault happening? I didn't see it on unity8 sessions so far [15:50] Trevinho, great [15:50] seb128: but while for clutter I don't see problems in having releases, cogl people seems not to want to release soon [15:50] Trevinho, do you have any idea about current gtk nor working on unity8 in vivid? [15:50] seb128: well, running the gnoem tetris game,, [15:50] it gives a blank surface [15:51] seb128: ooops, no.. I'm latest mir + latest gdk, but not in vivid [15:51] k [15:51] (and they work) [15:51] we are not likely to get 3.16 this cycle [15:51] so we need to backport fixes to the stable version [15:52] seb128: well, a part from the gdkgl work, the rest should stay I guess [15:52] yeah [15:53] and.... gdkgl stuff got recently broken by an upstream change... that weirdly breaks us... but it looks mostly a mesa/mir issue [15:55] seb128: ah, i thought I had replied to you, but it doesn't seem the case... so the crash was happening to me when running the gnome tetris game (that uses clutter, and canberra) with mir [15:56] Trevinho, oh, makes sense, I've added the lp: # reference to the changelog and uploaded your change, thanks ;-) [15:56] seb128: np [16:01] Laney, https://bugs.launchpad.net/ubuntu/+source/cairo/+bug/1353362/comments/11 ... was that comment about the cairo merge or about the specific issue described before? [16:01] Launchpad bug 1353362 in cairo (Ubuntu) "cairo needs merge from debian" [Undecided,New] [16:02] seb128: probably related to feature freeze [16:03] seb128: hey, can you help me with https://bugs.launchpad.net/ubuntu/+source/unity-gtk-module/+bug/1307657? [16:03] didrocks, I've the feeling you skipped over your piloting shift this week, want to look at that merge ^ next week? would be good for your karma around here :-) [16:03] Launchpad bug 1307657 in unity-gtk-module (Ubuntu Trusty) "[SRU] UBUNTU_MENUPROXY should not be set in Xfce" [Undecided,In progress] [16:03] attente_, hey, help how? to land it? [16:03] yeah [16:04] is the UBUNTU_MENUPROXY env not useful anymore? [16:04] in any case, not sure I'm going to get to it today [16:04]