[03:27] <pitti> Good morning
[03:30] <thumper> hi pitti
[03:31] <thumper> pitti: happy 10 years
[03:31] <pitti> hey thumper; thanks!
[03:31] <thumper> pitti: I'm nearly at 8
[03:31] <pitti> thumper: crazy how fast that piles up, isn't it?
[03:31] <thumper> uh ha
[03:32] <thumper> my daughters don't really remember me working away from home
[03:32] <thumper> eldest was 5 when I started
[03:45] <thumper> pitti: what are you doing up at 5:30am anyway?
[03:45] <pitti> thumper: getting up and start working :)
[03:46] <pitti> thumper: my wife needs to get up at 4:50, then it depends on how much I can still sleep
[03:46] <pitti> sometimes not any more, sometimes until 6:30
[03:46] <thumper> ugh... too early for me
[03:47] <pitti> took some time for me to get used to it, too
[03:47] <sarnold> careful pitti, at this rate, you'll be getting up before you go to bed..
[03:47] <pitti> heh; but I've been doing this for ~ 3 years now ;)
[03:48] <sarnold> two years back, when I joined, it seemed early, but this seems -super- early, hehe ;)
[03:49] <pitti> sarnold: yeah, certainly not the natural and recommended proper care for a hacker
[03:50] <sarnold> pitti: well, certainly not -this- hacker.. :)
[07:07] <seb128> good morning desktopers
[07:08] <didrocks> morning seb128
[07:08] <seb128> lut didrocks
[07:22] <seb128> robert_ancell, hey, how are you? what are you doing still online? ;-)
[07:22] <robert_ancell> seb128, bug fixing..
[07:22] <seb128> lightdm?
[07:36] <robert_ancell> seb128, yeah
[07:36] <robert_ancell> bug 1364725
[07:37] <ubot5> bug 1364725 in lightdm (Ubuntu Trusty) "logind session ID not used due to race condition" [High,In progress] https://launchpad.net/bugs/1364725
[07:40] <seb128> yeah, saw that one in my emails
[07:40] <seb128> robert_ancell, btw Laney is on vac, so don't worry if you don't get much feedback on the gnome-desktop transition this week
[07:40] <seb128> I'm going to try to have a look, but it's not at the top of my list
[07:41] <seb128> would be nice if the Ubuntu GNOME team was helping reviewing/testing since they are the ones that need that upload to land
[07:43] <robert_ancell> :)
[08:10] <darkxst> seb128, robert_ancell, I am planning to take a look at the gnome-desktop branches, but been quite busy this week, will get to it in the next day or 2
[08:11] <seb128> thanks
[08:11] <robert_ancell> darkxst, ta
[08:59] <tkamppeter_> will
[08:59] <tkamppeter_> sorry
=== tkamppeter_ is now known as tkamppeter
=== MacSlow is now known as MacSlow|lunch
=== oCrazyLemn is now known as CrazyLemon
[11:23] <ali1234> https://bugs.launchpad.net/ubuntu/+source/thunar/+bug/1310788
[11:23] <ubot5> Ubuntu bug 1310788 in unity-gtk-module (Ubuntu) "Thunar exists with segfault when accessing the context menu a second time" [Undecided,Confirmed]
[11:23] <ali1234> this appears to be another crash caused by unity peeking into application menus
[11:24] <larsu> ali1234: unity shouldn't look into context menus
[11:24] <ali1234> yeah, i know it *shouldn't*
[11:24] <ali1234> but apparently it is
[11:24] <ali1234> https://bugs.launchpad.net/ubuntu/+source/thunar/+bug/1363975 seems to be the same crash
[11:24] <ubot5> Ubuntu bug 1310788 in unity-gtk-module (Ubuntu) "duplicate for #1363975 Thunar exists with segfault when accessing the context menu a second time" [Undecided,Confirmed]
[11:24] <ali1234> e.u.c has a backtrace, it is definitely dying inside unity libs
[11:25] <larsu> ali1234: interesting. I'll have a look, thanks for bringing it up
[11:25] <ali1234> thanks
=== MacSlow|lunch is now known as MacSlow
[11:32] <larsu> ali1234: does this happen on trusty for you? I can't seem to reproduce it here
[11:33] <ali1234> no, but then i don't have unity installed
[11:33] <ali1234> it certainly happens for someone somewhere at least once a day so it can't be that hard to reproduce
[11:36] <larsu> sure, I was just trying to find out if this might have been fixed already and just needs a backport
[11:36] <larsu> (also, we do have quite some users. Just because it happens to someone somewhere doesn't mean it's easy to reproduce)
[11:37] <ali1234> thunar doesn't :)
[11:48] <ali1234> i can't reproduce it on a system with only unity and thunar installed
[11:49] <larsu> what else do you need?
[11:50] <ali1234> well the other possibility is xfce desktop while unity is also installed
[11:50] <ali1234> but i don't have a system like that immediately available
[11:51] <larsu> I guess this commit is supposed to fix that: http://bazaar.launchpad.net/~indicator-applet-developers/unity-gtk-module/trunk.14.10/revision/331
[11:52] <ali1234> yes, indeed
[11:52] <larsu> I can't test with a xfce system right now either
[11:53] <larsu> backporting that commit should be easy
[11:54] <ali1234> yeah. of course that doesn't really fix the problem, it just prevents the conditions that trigger it :)
[11:56] <ali1234> it's also possible the user is running thunar in unity and some other condition that we don't have is causing the crash
[11:57] <larsu> ya, backporting would be a first step to make the crash go away
[11:57] <larsu> I'll gladly debug it if you find a way to reproduce this in unity
[11:57] <ali1234> what's more, the error reports stopped on 21st august, so maybe it is already fixed
[11:58] <larsu> and I'll point attente to the problem (he's maintaining u-g-m)
[11:58] <ali1234> thanks. i'll see if we can reproduce it somehow or get more information from the reporter
[12:08] <ali1234> mterry: can we get XDG_CURRENT_DESKTOP in the crash reports please?
[12:08] <mterry> ali1234, hm?  Which crash reports?
[12:08] <ali1234> on errors.ubuntu.com
[12:09] <seb128> seems like you ping the wrong person there?
[12:10] <seb128> that would rather be a question for ev/bdmurray/pitti I would say
[12:10] <ali1234> ah, my mistake
[12:10] <seb128> we do include that info on apport report to launchpad
[12:11] <seb128> so could be a e.u.c side change
[14:07] <Saviq> hi all, having a problem with lightdm here, I'm on a dual-GPU (Optimus) system here, and lightdm just will not start X when I disable Optimus and run with only the nvidia card (proprietary driver)
[14:07] <Saviq> any ideas?
[14:08] <Saviq> I can startx and run the session just fine, it's just lightdm never even tries (no Xorg log or anything, nothing interesting in lightdm log...)
[14:27] <willcooke> Saviq, sounds like that could be a robert_ancell question.  I'll ask him and get back to you (unless someone else knows in the meantime...)
[14:28] <Saviq> willcooke, yeah, I though to bug him, but not around
[14:28] <Saviq> willcooke, thanks
[14:28] <willcooke> Saviq, it's about 2am for him :)
[14:28] <Saviq> I even reinstalled, thought I broke something after having removed nvidia + nouveau due to a hardware issue
[14:28] <Saviq> willcooke, sclacker
[14:28] <Saviq> -c
[14:28] <willcooke> Saviq, I know right!
[14:30] <willcooke> Saviq, I've mailed him
[14:30] <Saviq> willcooke, thanks
[14:44] <seb128> kenvandine, hey
[14:45] <kenvandine> hey seb128
[14:45] <seb128> kenvandine, going to do an u-s-s landing in the next days? ;-)
[14:45] <kenvandine> seb128, yes... been blocked by what's in silo 10
[14:45] <kenvandine> which they are working on landing now
[14:45] <kenvandine> i have a bunch of branches building in silo 20 now
[14:45] <kenvandine> which will need a respin after 10 lands and merges
[14:46] <seb128> k
[14:46] <mvo> seb128: hi, do you know what plans Laney has with the gobject intospection multiarch work? I would love to get it in for 14.10 and I know Laney wants to push it for experimental too
[14:46] <mvo> I'm curious about timeline etc
[14:47] <seb128> mvo, hey, he's on vac until the 11
[14:47] <mvo> oh
[14:47]  * mvo scratches head
[14:47] <seb128> mvo, but he uploaded https://packages.qa.debian.org/g/gobject-introspection/news/20140831T003420Z.html
[14:47] <seb128> he said gir needs rebuilds iirc
[14:47] <seb128> let me look at my irc logs
[14:57] <mvo> seb128: I wonder if I can simply go ahead in ubuntu, ideally prepare it all in a PPA and then mass-copy. but I would need a trusted ppa for that, right?
[14:58] <seb128> mvo, that seems like a transition/something that needs a ffe to me
[14:58] <mvo> seb128: hrm, that not like it was in the old days :P
[14:58]  * mvo files one
[14:59] <seb128> mvo, indeed not!
[15:02] <seb128> mvo, oh, found back the log, http://irclogs.ubuntu.com/2014/09/02/%23ubuntu-desktop.html#t18:09
[15:03] <mvo> seb128: not that much it seems :)
[15:03] <seb128> mvo, no, but basically it's "either take the hit and do the work, or wait for things to be tested in Debian and sync"
[15:03] <seb128> you can probably help in Debian and then sync
[15:06] <mvo> seb128: ok
[15:13] <marga> Trevinho, do you think there's any data that we could provide to help with https://bugs.launchpad.net/ubuntu/+source/unity/+bug/1311316 (input field missing from Unity's lockscreen)?
[15:13] <ubot5> Ubuntu bug 1311316 in Unity "After locking screen there is no input field to type password for unlock" [High,Triaged]
[15:14] <Trevinho> marga: mh, not sure... I've a workaround in mind but it's probably something due to a race...
[15:14] <marga> Trevinho, I have quite a number of affected users that can reproduce this on a daily basis, so if there's any data to gather, I could do it.
[15:14] <marga> Yes, I agree it's due to a race.  Reports state that it's much more likely to happen when on high load
[15:15] <Trevinho> marga: ah, ok... I'm not getting that for a long time, but if you've some people affected it would be probably nice to give them a debug version through ppa, to understand more
[15:15] <marga> sure
[15:19] <didrocks> fginther: hey, did you received my email about the lost jenkins jobs?
[15:24] <fginther> didrocks, yes, the jobs were accidentally deleted. I have a backup of them, I just need to get them restored after dealing with a few other critical issues
[15:24] <didrocks> fginther: ok thanks! I hope you backed up the version which autologin/add sudoer and so on (for a reminder of what to check)
[15:24] <didrocks> fginther: I'll create the chaining jobs then
[15:25] <fginther> didrocks, yeah, in this case it was just the job configuration that was removed. the VMs should still be there
[15:25] <didrocks> great ;)
[15:25]  * fginther will also check for that
[15:25] <didrocks> thanks a lot!
[15:48] <ara> seb128, any idea why this xserver-xorg upload to trusty-proposed is not being reflected in the pending sru report? (http://people.canonical.com/~ubuntu-archive/pending-sru)
[15:48] <ara> https://launchpad.net/ubuntu/+source/xorg-server/2:1.15.1-0ubuntu2.1
[15:51] <seb128> ara, it is listed?
[15:52] <ara> seb128, no, that's my question. It is not listed in the trusty pending packages, and I don't know why :)
[15:52] <seb128> ara, well, I see it here
[15:52] <ara> seb128, sorry, I see it now, my bad
[15:52] <ara> :(
[15:52] <seb128> want a screenshot of the page?
[15:52] <seb128> ara, no worry ;-)
[15:53] <ara> seb128, thanks, and sorry :)
[15:53] <seb128> nw!
[15:55] <marga> Trevinho, I'm awfully jetlagged and so I'm not sure if I understood you correctly.  Is there a version that would help debugging already available? Are you going to make it available? Do you need anything from me?
[15:55] <Trevinho> marga: I don't have it yet, but we could do it if you have someone that can easily reproduce the issue
[15:58] <marga> I have several someones
=== Trevinho_ is now known as Trevinho
=== broder_ is now known as broder
=== davmor2_ is now known as davmor2
=== popey_ is now known as popey
=== sil2100_ is now known as sil2100
[16:53] <kenvandine> seb128_, i have that folks change ready to upload, modified from yesterday based on my chat with bigon_
[16:53] <kenvandine> seb128_, it'll introduce a NEW binary
=== seb128_ is now known as seb128
[16:54] <seb128> kenvandine, k, I'm about to be away but I can new review that a bit later
[16:54] <kenvandine> thx
[16:55] <kenvandine> i'm submitting it to debian too, but can't seem to get reportbug to work
[16:55] <kenvandine> the locking bug...
[17:07] <Laney> mvo: I didn't plan on doing that for 14.10 but was rather going to do it in exp and fix things there as much as possible before next cycle
[17:07] <Laney> is it urgent for you?
[17:08] <bigon_> 18:03 < bigon> kenvandine: didn't you also forgot to move " usr/lib/*/libfolks-dummy.so.*" ?
=== bigon_ is now known as bigon
[17:08] <kenvandine> bigon, i didn't forget it, i left it since it's a lib... simpler to leave it there
[17:08] <kenvandine> it won't load without the backend plugin
[17:11] <bigon> the for libfolks-eds the lib is in the same pkg as the module
[17:11] <bigon> I guess we should follow this here non?
[17:12] <bigon> need to go now
[17:12] <kenvandine> bigon, ok, i'll do that
[17:28] <mvo> Laney: aha, ok. not really urgent, it would be nice as it would unblock ubuntu-sdk-libs:$arch but I just checked the rdepends and the amount if quite substantial, so I guess utopic is a bit too ambitious
[17:33] <mvo> Laney: I will play around a bit in a ppa and see how bad it is (or have you prepared stuff already?)
[20:06] <robert_ancell> Saviq, are you getting bug 1305006?
[20:06] <ubot5> bug 1305006 in Light Display Manager "Crashes running display server scripts for non X servers" [High,In progress] https://launchpad.net/bugs/1305006
=== desrt_ is now known as desrt
[21:03] <Saviq> robert_ancell, reading
[21:05] <Saviq> robert_ancell, I don't *think* it was crashing, let me find out again
[21:05] <robert_ancell> Saviq, do you have the lightdm.log?
[21:20] <Saviq> robert_ancell, no, no crash, this is all that's left in /var/crash; /var/log after a "initctl start lightdm" with no X ever started http://paste.ubuntu.com/8226876/
[21:20] <Saviq> robert_ancell, and I'm trying to run a standard ubuntu desktop session, no unity8 involved
[21:21] <robert_ancell> Saviq, oh interesting. What does "loginctl list-seats" give you?
[21:21] <Saviq> robert_ancell, seat0
[21:22] <robert_ancell> Because LightDM appears to think there are no seats
[21:22] <Saviq> robert_ancell, but I'm running a session through "startx" now, let me log into IRC from somewhere else
[21:22] <Saviq> brb
[21:22] <robert_ancell> And "loginctl show-seat seat0"
[21:24] <Saviq_> robert_ancell:ok, here from another pc, loginctl says seat0, 1 seats, no greeter
[21:25] <robert_ancell> I want to check if seat0 has CanGraphical=false
[21:26] <Saviq_> robert_ancell:CanGraphical=no
[21:26] <robert_ancell> That's the issue. Why does it think that?
[21:26] <Saviq_> you tell me :)
[21:27] <robert_ancell> Can you file a bug with this info an we'll try and dig into loginds logic to determine what's going on
[21:28] <robert_ancell> I suspect that disabling Optimus is making logind think there's no video card
[21:30] <Saviq_> robert_ancell:this has worked for me Fri morning, btw
[21:31] <Saviq_> except I'm on a different mobo now, different BIOS, wonder if that affects things
[21:31] <robert_ancell> Saviq_, yeah, we changed to honoring what logind specifies, so before we just ignored CanGraphical as startx will also
[21:31] <Saviq_> ah, nice coincidence... could've saved me a reinstall this morning :)
[21:33] <robert_ancell> Yeah, sorry. Was busy sleeping at 2am :P
[21:33] <Saviq_> robert_ancell:yeah, we already decided with willcooke you're a slacker
[21:33] <Saviq_> bug #1365165
[21:33] <ubot5> bug 1365165 in systemd (Ubuntu) "logind says CanGraphical=no with Optimus disabled" [Undecided,New] https://launchpad.net/bugs/1365165
[21:34] <robert_ancell> Lucky we have such little overlap he's calmed down before I next see him and he doesn't fire me
[21:34] <robert_ancell> Saviq_, ta
[21:34] <Saviq_> robert_ancell:let me know if I can get you anything else, I'll try downgrading lightdm then
[21:35] <robert_ancell> Saviq_, you should be safe with lightdm <= 1.11.7. I'll start digging into how logind does the CanGraphical determination - currently have no knowledge there
[21:35] <Saviq_> robert_ancell:good enough, thanks a bunch
[21:40] <robert_ancell> Saviq_, can you add a lspci to the bug report?
[21:47] <Saviq_> robert_ancell:will do
[21:52] <Saviq> robert_ancell, and yeah, downgraded lightdm works like a charm, thanks!
[21:53]  * Saviq missed his second screen
[21:55] <Saviq> robert_ancell, lspci uploaded
[21:55] <robert_ancell> Saviq, ta
[21:58] <robert_ancell> Your system is different to mine. Mine shows both an Intel and NVIDIA VGA controller on the PCI bus. Yours only seems to have NVIDIA
[21:58] <robert_ancell> But you must have an Intel controller somewhere
[21:59] <robert_ancell> RAOF, ^ any insights into Optimus and where the Intel controller would be?
[22:24] <RAOF> Saviq: When you say “disable optimus” what do you mean? Do you mean flick a bios switch to “nvidia only” or somesuch?
[22:39] <Saviq> RAOF, yes
[22:39] <Saviq> RAOF, it literally says "Enable Optimus" and a checkbox
[22:39] <Saviq> robert_ancell, ↑
[22:40] <Saviq> robert_ancell, I explicitly disable the intel one in the BIOS to be able to use external monitors properly, otherwise I end up in hybrid/prime and stuff's not so great
[22:40] <RAOF> You're using the binary nvidia drivers like all the UI heathens, aren't you?
[22:41] <Saviq> RAOF, not for long now ;)
[22:42] <Saviq> RAOF, but yeah, nouveau just doesn't cut it, and with intel alone I only get the VGA out
[22:42] <RAOF> I just wonder whether this bug is actually “logind doesn't understand that non-kms drivers can run X”.
[22:42] <Saviq> and with hybrid all hell breaks loose with multimonitor
[22:43] <RAOF> tselliot would be interested in your hybrid experiences, methinks.
[22:44] <RAOF> On the basis that he's been doing much of the “can we make hybrid work” bits.
[22:45] <robert_ancell> RAOF, yeah, that's what I was thinking
[22:45] <robert_ancell> Is that solvable?
[22:46] <RAOF> Not really, no.
[22:47] <RAOF> logind can't know if X will run without actually running X.
[22:47] <RAOF> Which makes me wonder what CanGraphical is meant to be.
[22:47] <robert_ancell> yeah
[22:47] <robert_ancell> CanGraphical means "has a graphical device associated with this seat:
[22:47] <robert_ancell> "
[22:48] <RAOF> Yeah, it's a barefaced lie.
[22:48] <robert_ancell> RAOF, is there an alternative method of telling if a binary driver device is present, i.e. can we just patch logind?
[22:48] <RAOF> Well, it *could* do that correctly, I guess. By doing pci/usb/whatever probing equivalent to what X does...
[22:49] <RAOF> Saviq: Hey, could you dump “udevadm info --export-db” output on that bug?
[22:50] <Saviq> RAOF, done
[22:51] <RAOF> robert_ancell: We could, assuming that we have udev rules that appropriately tag all possible output devices with seats, and then logind can *distinguish* all possible output devices...
[22:52] <robert_ancell> RAOF, so it might just be a missing udev rule?
[22:52] <RAOF> Plausibly.
[22:55] <RAOF> So, I'm not sure what logind checks, but Saviq's udev output shows that there's nothing with an appropriate seat tag.
[22:56] <robert_ancell> RAOF, but there is a device that could/should be tagged?
[22:56] <robert_ancell> i.e. udev at least knows about the device
[22:57] <RAOF> Not sure. There's both a drm device and P: /devices/pci0000:00/0000:00:01.0/0000:01:00.0 with DRIVER=nvidia, so at least in this case there's options.
[22:58] <RAOF> But I'm not sure if logind is (for example) looking for a framebuffer or something.