[05:55] GOdo morning [05:55] Laney: argh, again? I already restarted cyclops-024 twice yesterday [09:42] seb128: calendar is a bit MEH click on today nothing happens click on add an calendar for a second time and it crashes and the corners still aren't right which I thought was fixed else where [09:44] davmor2, hey [09:45] davmor2, hey, we are only starting on bugfixes after ff, those are worth reporting ... today doesn't do much indeed, but it brings you to the current month if you are on another one [09:46] davmor2, what do you call "add an calendar"? [09:47] dpm, pitti, how do we get translation included in gnome langpacks for new components? e.g gnome-calendar? [09:48] seb128: so I added my own work calender and then added a secondary one for the team and at that point it didn't show the second so I readded it and the app crashed [09:48] morning seb128. I think the first step is to approve the templates. Let me have a look inLP [09:48] davmor2, did you add from a file? from online account? [09:49] dpm, oh, right [09:49] dpm, https://translations.launchpad.net/ubuntu/xenial/+source/gnome-calendar/+imports [09:49] I can approve it [09:49] dpm, thanks! [09:49] ok, cool [09:50] I knew about that but forgot the details [09:50] seb128: add from web the online accounts part it missing all together, also the cog that says open online accounts does nothing [09:50] I wonder if there are other in that case [09:50] seb128, I think from there then translations will be imported/exported and the langpacks will pick them up automatically [09:50] davmor2, for me it opens unity-control-center on the online accounts panel [09:50] dpm, does it need to have a priority bumped? [09:50] yeah, we need to have a look at the xenial queue, which is what I trying to do [09:50] dpm, at least to be in the "to translate" top list? [09:50] seb128, yeah [09:51] dpm, https://translations.launchpad.net/ubuntu/xenial/+source/gnome-calendar/+pots/gnome-calendar/+admin right? what score do we use usually? [09:52] seb128, that page is a bit old, but can give some guidance: https://wiki.ubuntu.com/Translations/TemplatesPriority [09:53] seb128, the other option is to look at where the gnome templates are in LP, choose one where you think the priority should be similar, and assign gnome-calendar about the same [09:53] I'm scanning through https://translations.launchpad.net/ubuntu/xenial/+imports?field.filter_status=NEEDS_REVIEW&field.filter_extension=pot to see if I spot anything interesting to approve [09:53] dpm, thanks, nautilus is 7390 and gedit 7360 [09:53] seb128: this one looks to be the crasher https://bugs.launchpad.net/ubuntu/+source/gnome-calendar/+bug/1545259 [09:53] Error: launchpad bug 1545259 not found [09:54] davmor2, having steps added would be nice! [09:54] seb128, I notice yesterday that gnome-center is untranslated, probably also waiting for the template to be approved [09:54] dpm, g-c-c is in universe now [09:54] dpm, I put 7300 for gnome-calendar [09:55] seb128: yeap adding the one I did not going to guarantee they are the cause it might just of decided to die [09:56] seb128: also it is using an alarming amount of cpu according to top so I'll have a play with it today and just file bugs for everything [09:57] seb128, non, I meant gnome-software, not g-c-c -> it seems gnome-software's translations can't be imported because it doesn't generate a .pot file? https://translations.launchpad.net/ubuntu/xenial/+source/gnome-software/+imports?field.filter_status=all&field.filter_extension=all [09:57] dpm, let me look at that [09:57] cool, thanks [09:57] davmor2, cpu use, indeed, wth [09:59] davmor2, ok, you convinced me to look at bit to calendar and upstream some bugs today, thanks ;-) [10:00] seb128: to be honest today is the first day I decided to try and set it up and it's failing a lot all over the place :) [10:01] davmor2, yeah, we just need a good round of shaking bugs out [10:02] seb128: I'll prod it with a stick I should probably do the same with usb creator and gnome-software there wasn't any other new stuff was there? [10:03] davmor2, gnome-software you don't really need to do, I did the stick poking this week and we already have a good list of known issues [10:03] davmor2, no new software, gedit got a new major version update (3.10 to 3.18) though [10:04] seb128: yeah what I'm thinking though is I'll book a day with jibel just to hit the new software on the desktop that is new and critical and hit it all hard that day [10:10] seb128: so things like Calendar is new and useful but you can have calendar in evolution and thunderbird and .... so not so critical, gnome-software is the only gui for installing software (that we will support) so has to work and work well and is critical. So those would be the priorities for a day of hitting it :) [10:13] Trevinho: seb128 hoy. i'm currently doing the a11y integration in compiz [10:14] davmor2, makes sense! [10:14] i did an external lib, inserted it. i'm now having trouble with what looks like glib contexts [10:15] seb128: grrr calendar still has those odd top corners I thought that was fixed? [10:16] davmor2, Trevinho is working on making CSD look better, we might change calendar to use normal decorations under Unity though [10:16] ksamak, hey [10:17] seb128: ah okay when I saw it gone in other apps I thought it was fixed and this was just an odd ball [10:18] and/or Xlib calls. u know if it's possible to break some X while doing some X calls like XFetchName [10:19] davmor2, no, you still have it in a few applications places, e.g if you try to send files to a phone by bluetooth from the indicator [10:19] ksamak, you should be able to step on those from gdb yes [10:19] seb128: got some warnings from x too, saying it's got some threading problems, although i'm calling X within compiz context [10:19] seb128: thanks for the clarification on that :) [10:20] are X calls in a different thread in compiz? [10:20] do we have to rely on X to get window name? [10:20] ksamak, no idea, you probably want to talk to hikiko or andyrock or Trevinho, they work on compiz/unity [10:20] ok. [10:21] maybe i can write a mail. [10:22] ksamak: mh... Let me see better [10:23] ksamak: x calls are in main thread, in case you do stuff from glib, you can still ensure you do stuff in callbacks from the main thread by using the usual ways [10:35] Trevinho: i'm not that good with glib yet, what's the usual way to you? [10:35] in my lib i did idle_sources to insure that a11y calls are made from the right thread and glib context [10:35] but that's in c++ [10:42] ksamak: yeah you can just use g_idle_add or timeout.... [10:43] ok, same then. thx [11:50] pitti: the SRU is ready now. I'll ask AMD to test it [12:02] tseliot: thanks [12:02] pitti: thanks to you ;) === tsimonq2alt is now known as tsimonq2 [12:25] morning [12:31] hey andyrock, how are you? [12:31] Trevinho, andyrock, my unity segfaults when coming back to my session after an user switch, is that known? [12:32] nope [12:42] going for some exercice, bbl === hikiko is now known as hikiko|ln === alan_g is now known as alan_g|lunch [13:35] seb128: nope.... [13:35] seb128: got a trace? === JMulholland_ is now known as JMulholland === hikiko|ln is now known as hikiko === alan_g|lunch is now known as alan_g [14:06] so, at some point in the last several weeks, GTK-based apps stopped pulling in all their required dependencies -- not a problem on a classic desktop because of the seeding process, but it causes failure in a Libertine container unless libgl1-mesa-glx is manually installed ... just making people aware [14:06] all manner of Wayland EGL libs get pulled in, though === alexabreu is now known as alex-abreu [15:02] bregma, what error do you get when those are missing? gtk shouldn't rely on any gl [15:04] cking: What should we do about grub2 and zfs-linux upstream adding code to support grub2? rlaager proposed a distro patch to include feature from next release. Do you think it's worth the effort to update grub to include environment var and zfs to add distro patch? [15:04] I'm not so sure. [15:05] qengho, what are the opros/cons of doing this? [15:05] Mirv, what's the best qt5 metapackage to pull all of qt5 dependencies? [15:06] I was trying to have a go at snappifying the calculator app, and I got stuck at pulling qt5 [15:08] qengho, 'cos I'm not sure either :-/ [15:08] cking: the distro patch isn't as bad as usual since it's literal upstream diff. It means new updates and testing of two xenial packages. We get no ugly symlinks in /dev/ that someone somewhere might assume should be there forever. I haven't figured out the Conflicts/Depends relationship yet, but that could be ugly. [15:10] seb128, the error message is "Couldn't open libGL.so.1: libGL.so.1: cannot open shared object file: No such file or directory" [15:10] qengho, well I suspect if you can determine what the Conflicts/Depends issues are then we can make a more informed decision. As yet, I'm not clued up on that [15:10] bregma, do you know what code tries to open it? if something links against it then it should auto pick a depends [15:12] seb128, you would think, and it used to work -- I'd have to play around inside the container to figure out what's requiring GLX, and that's hard without a terminal ... which doesn't run because of missing dependencies [15:12] I shall investigate, since something definitely changed recently [15:13] if only recent changes to LXC in Xenial hadn't broken things so badly..... [15:14] bregma, I'm going to have a look [15:14] should be easy to see in a pbuilder [15:14] just with ldd [15:14] no need to start code [15:42] Trevinho, a bit difficult to get a bt, I'm unsure if it segfaults or what, it's slow, reload the UI and then relock the screen, so it looks like it but not segfault in the logs nor apport report [15:42] cking: Okay, we can update grub to add env-var at any time. zfs-linux should Conflict with old, current grub when zfs removes extraneous udev symlinks. That's the only configuration that doesn't work old grub with new zfs sans symlinks. [15:42] I tried to attach gdb from vt1 [15:42] but I loose access to the vts after starting/closing a guest session :-/ [15:42] wonder wth is going on, could be an intel driver issue [15:43] qengho, ok, so this does look like a viable way forward then [15:50] cking: Right. I'll put together a debdiff for grub. Then we can change zfs at leisure. [15:50] seb128: mh, i see.. Maybe a backtrace of when it slows down might help too [15:51] attaching to th eprocess [15:51] qengho, ok, sounds like a plan. I will be out for a week or more because I have an operation next week and may be in recovery mode for a while [15:51] Trevinho, well, I think it might segfault [15:51] but I can't prove it :p [15:51] qengho, so it may take a while for me to get the ZFS fixed done [15:51] need to ssh from another box [15:51] cking: oh dear. good luck! [15:52] Feb 26 16:38:32 localhost kernel: [ 162.688698] [] dump_stack+0x41/0x59 [15:52] Feb 26 16:38:32 localhost kernel: [ 162.688703] [] warn_slowpath_common+0x87/0xc0 [15:52] Feb 26 16:38:32 localhost kernel: [ 162.688726] [] ? drm_framebuffer_reference+0x48/0x90 [drm] [15:52] qengho, thanks, ping me when I come back online and prod me to look at this. you know how one can get overloaded with email when one gets back to work after a week or two [15:52] I guess those are the video issues I get [15:52] cking: ack [15:52] thansk! [16:21] dpm: from our point of view probably ubuntu-sdk-libs. for development normally one'd use simply the SDK IDE which does not use system libraries anyway but provides the full all-included SDK [16:22] dpm: there's no actual metapackage to pull all Qt upstream modules in, the sdk-libs specifies what we include on images [16:27] thanks Mirv, I'll use ubuntu-skd-libs I think [16:44] Mirv, would you happen to know what could be causing qmlscene not to find Qt? This is again, trying to snappify the calculator app: [16:44] dpm@el-far:/snaps/bin$ ubuntu-calculator-app.ubuntu-calculator-app [16:44] qmlscene: could not find a Qt installation of 'qt5 [17:07] dpm: man qtchooser (or askubuntu) has the answer, but2 3 [17:08] yes sorry it's a common problem. one needs to have a default configuration either via a package installing it (qt5-default), environment variable or user's own configuration file [17:09] ...and the default fallback is still qt4, since packaging wise users of qt4 couldn't be aware of qtchooser but qt5 packagers and users should be aware of qtchooser [17:09] but at some point I suspect the default will change simply because qt4 is no more at some point [17:10] dpm: one can also run have the exec line qmlscene -qt5 blah.qml instead of qmlscene blah.qml [17:12] dpm: hmm actually qmlscene should be in the whitelist that it's always Qt 5 (Qt 4 didn't have qmlscene), so it might be you're also simply missing qmlscene from the installation, and the qtchooser's way of reporting errors is poor === Noskcaj_ is now known as Noskcaj === attente_ is now known as attente [17:27] seb128: is that the bt you get? [17:28] andyrock, "that"? [17:28] https://www.irccloud.com/pastebin/YfeCINYp/ [17:28] seb128: ^^^ [17:29] andyrock, that's an xorg/kernel error in syslog [17:29] not the unity one === tyhicks` is now known as tyhicks === _fortis_ is now known as _fortis === alan_g is now known as alan_g|EOW === mhall119_ is now known as mhall119 [19:09] Laney, so I've been doing some geonames work, as you know. larsu put it in git, which is a totally reasonable thing to do. But our citrain doesn't support git yet. Is there an objection if I migrate geonames to bzr and try to get it train-ified? (autolanding via citrain and all that jazz) [19:19] * mterry goes and does that anyway === dpm is now known as dpm-afk === tsimonq2alt is now known as tsimonq2 === Guest37386 is now known as fredp === fredp is now known as Guest54537 [21:56] jdstrand: i reported against isc-dhcp-server and client 4.3.3-5ubuntu7 [21:57] if you get a chance im available to try things. :D [22:12] seems like i get a DHCPRELEASE then I have to request a new lease which restarts dhclient [23:11]