[00:34] <robert_ancell> attente, awake?
[00:34] <attente> robert_ancell: yep! how's it going?
[00:35] <robert_ancell> attente, good. I've been hacking around with g-s 3.22.
[00:35] <robert_ancell> I've made a wip/ubuntu-zesty branch with out changes on top of that. I *think* I've got everything across, but if you could keep an eye out in case something got dropped.
[00:36] <robert_ancell> I figure it's good enough to release to zesty and we can fix if we find anything
[00:36] <attente> robert_ancell: that's awesome, thanks! i can take a look at it tomorrow
[06:07] <hikiko> hello
[06:11] <siva_machina> hello
[06:53] <pitti> Good morning
[06:55] <didrocks> good morning pitti
[06:57] <pitti> bonjour didrocks, ça va ?
[06:58] <didrocks> pitti: ça va bien, et toi ?
[07:13] <pitti> didrocks: je vais bien aussi, merci ! The Bucharest sprint was busy, and I had a "household" swap day yesterday, so now off to attacking the backlog :)
[07:14] <didrocks> pitti: oh, I can imagine the backlog! Good luck with this :)
[07:59] <desrt> happy tuesday!
[08:05] <pitti> hey desrt, wie gehts?
[08:06] <desrt> pretty good :D
[08:06] <desrt> last day of the madness
[08:27] <pitti> desrt: oh, which madness?
[08:31] <desrt> US election day
[08:34] <flexiondotorg> desrt, pitti didrocks Morning
[08:35] <didrocks> hey flexiondotorg
[08:35] <didrocks> morning desrt
[08:35] <desrt> hi didrocks, flexiondotorg
[08:41] <hikiko> hey
[08:41] <hikiko> I have a super weird+annoying problem with u7 sessions in Y
[08:41] <hikiko> and I can't figure out what caused it
[08:41] <hikiko> I start xfce with startx
[08:42] <hikiko> and u7 from lightdm
[08:42] <hikiko> and when I leave the u7 tty for any reason
[08:43] <hikiko> compiz stops and then I see the login screen and if I login a new session starts
[08:43] <hikiko> I attached compiz to gdb
[08:43] <hikiko> and it doesn't crash, it exits normally
[08:43] <hikiko> which means that something else in the system is responsible for that
[08:44] <hikiko> and I only have this problem on Y since I upgraded :/
[08:44] <hikiko> do you have any idea what I should check?
[08:44] <hikiko> my settings are the default
[08:44] <hikiko> gsettings
[08:48] <flexiondotorg> Morning hikiko
[08:48] <hikiko> hi flexiondotorg :)
[08:52] <Sweet5hark> moin
[08:57] <willcooke> morning all
[08:57] <hikiko> hi wi
[08:57] <hikiko> hi willcooke :p
[08:58]  * hikiko goes to debug to see where the exit signal comes from
[08:58] <hikiko> bbiab
[09:03] <Laney> hey hey
[09:03] <willcooke> morning Laney
[09:04] <Laney> hey willcooke
[09:04] <seb128> good morning willcooke Laney hikiko
[09:04] <Laney> how was center parcs?
[09:04] <willcooke> Laney, great thanks!  The kids all loved it
[09:05] <Laney> nice!
[09:06]  * Laney wants to go in a pool with wave machine
[09:06] <seb128> :-)
[09:06] <Laney> hey seb128, hoe gaat het?
[09:06] <seb128> willcooke, did you enjoy it as well?
[09:06] <willcooke> Laney, have you been to the one in Sherwood?
[09:06] <Laney> willcooke: nope
[09:06] <willcooke> seb128, yeah it was good, but I need another holiday :)
[09:07] <willcooke> Laney, it's worth it I think as a grown-ups mid-week break
[09:07] <seb128> Laney, goed, dankje! jou?
[09:07] <Laney> we usually end up going on holidays you need a holiday from
[09:07] <Laney> like staying in youth hostels in december
[09:07] <willcooke> ha
[09:08] <Laney> seb128: ik ben geweldig!
[09:08] <Sweet5hark> whoa, this zigbee worm is scary: spreading over the air (NOT over existing network topography, but directly from device to device), triggering citywide epileptic attacks while all wifi will be jammed. And all that just with one vulnerable lightbulb.
[09:08] <Laney> ik klom gisteravond
[09:09] <seb128> leuk!
[09:09]  * seb128 googles geweldig, that's not something people on daily basis ;p
[09:09]  * Sweet5hark has babelfish hiccups.
[09:10] <Laney> en fietste naar huis in de kou
[09:10] <Laney> hahahaha
[09:10]  * Laney is secretly fluent in Dutch don't you know
[09:10] <Laney> hi Sweet5hark!
[09:11] <Sweet5hark> hrhr
[09:11] <flexiondotorg> Morning Laney, seb128, Sweet5hark, willcooke
[09:11] <willcooke> morning flexiondotorg
[09:11] <seb128> :-)
[09:11] <seb128> hey flexiondotorg, how are you?
[09:12] <flexiondotorg> Mentally fine. Have managed to surface an old rugby injury so my back is not so great :-(
[09:13] <seb128> :-/
[09:15] <davmor2> Morning all
[09:15] <Sweet5hark> flexiondotorg: ouch, does a hot-water bottle or infrared help?
[09:16] <willcooke> howdy davmor2
[09:17] <flexiondotorg> Sweet5hark, Heated pad.
[09:17] <flexiondotorg> And walking.
[09:18]  * Laney nods davmor2 
[09:22] <willcooke> Brexit is gettin' real: http://www.bbc.co.uk/news/uk-37904703
[09:26] <seb128> that's a positive impact
[09:27] <seb128> you end up eating less chocolate/sugar with smaller portions ;-)
[09:27] <willcooke> :)
[09:33] <Sweet5hark> willcooke: "eat less chocolate, we cant afford those 350 bazillion to the NHS"?
[09:34] <willcooke> :DD
[09:34] <xnox> Why choose the lesser evil?
[09:35] <davmor2> Sweet5hark: no you have it all wrong they want you to eat more chocolate and die of obesity then they don't have to treat you at all
[09:36] <flocculant> willcooke: just wait till we can only get english chocolate
[09:36] <davmor2> flocculant: Why English chocolate is good :P
[09:37] <didrocks> (hem)
[09:37] <flocculant> yea - but I don't think any is made by english companies anymore - there'll be none at all :(
[09:37] <Sweet5hark> davmor2: that wont happen, you guys will blackmail the continent to bring chocolate otherwise the continent wont get any english tea!
[09:38] <flocculant> Sweet5hark: ha ha
[09:38] <davmor2> hahaha
[09:46] <hikiko> xserver crashes when I switch TTY on Y
[09:46] <hikiko> receives sigabrt
[09:46] <hikiko> that's the problem
[09:46] <hikiko> xf86VTEnter is the function
[09:46] <hikiko> anyone who has experienced the prob in Y?
[09:47] <hikiko> https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-intel/+bug/30395
[09:47] <ubot5`> Ubuntu bug 30395 in xserver-xorg-video-intel (Ubuntu) "X-Server Crashes when switch back from terminal tty" [Medium,Fix released]
[09:48] <hikiko> it seems that we have a bug report for something similar
[09:48] <hikiko> but... 2006 :p
[09:50] <hikiko> https://www.mail-archive.com/search?l=desktop-packages@lists.launchpad.net&q=subject:%22\[Desktop\-packages\]+\[Bug+1569351\]+Re\%3A+Xorg+crashed+with+SIGABRT+in%09OsAbort\%28\%29%22&o=newest&f=1
[09:57] <hikiko> question:
[09:58] <hikiko> can I downgrade the xserver?
[09:59] <seb128> sure
[09:59] <seb128> tjaalton might be able to help you with the bug
[09:59] <hikiko> :D
[09:59] <seb128> you could also try the modesettings drivers instead of the intel one
[10:01] <hikiko> the 2nd bug report shows that nvidia have the problem too what are the modesettings drivers?
[10:02] <seb128> they replace the intel ones
[10:03] <seb128> but if you are on nvidia that's probably not a solution
[10:03] <seb128> did you report your bug?
[10:03] <hikiko> I think that since the gdb shows that it's xorg that crashes the downgrade is better
[10:03] <seb128> the one you mentioned is from 2006
[10:03] <hikiko> no because there's a bug report already
[10:03] <hikiko> no no
[10:03] <hikiko> the 2nd link
[10:03] <hikiko>  https://www.mail-archive.com/search?l=desktop-packages@lists.launchpad.net&q=subject:%22\[Desktop\-packages\]+\[Bug+1569351\]+Re\%3A+Xorg+crashed+with+SIGABRT+in%09OsAbort\%28\%29%22&o=newest&f=1
[10:04] <hikiko> that's 2016
[10:04] <seb128> you have the same backtrace
[10:04] <seb128> downgrading is a workaround
[10:04] <seb128> better to get the bug investigated and fixed if we can
[10:05] <hikiko> yep I am going to report it
[10:05] <hikiko> seb128, and maybe, well do we have an xorg developer here? should I try to debug xorg?
[10:06] <seb128> tjaalton is looking after xorg, he can maybe hint you about where/how to report the issue and if it's a known one
[10:07] <hikiko> ok I'll ping him and do a bug report in the meantime, thank you seb128 :)
[10:07] <seb128> yw!
[10:15] <pitti> Laney: workers died because apparently vivid images got removed from glance :(
[10:15] <pitti> i. e. that needs an RT (that already happened before)
[10:15] <pitti> or alternatively, some discussion whether we can drop vivid at last
[10:21] <seb128> pitti, why do we need vivid images?
[10:22] <seb128> the distro serie is still supported for touch
[10:22] <seb128> but it doesn't mean we need to do extra work it?
[10:22] <pitti> seb128: well, the train requests tests for landings to it
[10:22] <pitti> and we need to be able to run those
[10:23] <seb128> ah, right
[10:23] <Laney> hey pitti
[10:23] <seb128> fun
[10:25] <Laney> pitti: where do you see that?
[10:25] <pitti> Laney: I just ran the maintenance cron job and restarted them, but previously all workers died like that
[10:26]  * flexiondotorg nips out for a quick walk
[10:26] <Laney> pitti: I don't see a queue, did those requests get dropped?
[10:26] <Laney> failed I suppose
[10:27] <pitti> Laney: no, as I just restarted them, they are currently pending in the "wait 5 mins and then retry" x3 loop
[10:27] <pitti> after that, the worker will die again and punt them back to the queue
[10:27] <pitti> i. e. in about 10 mins
[10:27] <pitti> Laney: I restarted them to process the pending zesty ones
[10:28] <Laney> they don't show in running when this happens?
[10:32] <Laney> pitti: anyways, do you have a link to the previous RT?
[10:33] <pitti> Laney: no, when the worker does sleep(300) there is no logtail, as the actual autopkgtest run has finished
[10:33] <pitti> Laney: that's a potential improvement indeed
[10:33] <Laney> the same worker has the queue item still
[10:33]  * pitti searches email archive for that RT
[10:34] <Laney> that explains it
[10:34] <Laney> I forgot the reason for this wait 5 minutes × 3 thing instead of putting things back in the queue
[10:34] <pitti> Laney: https://rt.admin.canonical.com/Ticket/Display.html?id=90342
[10:35] <pitti> Laney: it avoids having to add a counter to the AMQP message (you can't modify it really)
[10:35] <pitti> Laney: so you can give the cloud 5 minutes to take a breath (and maybe let another test finish), then you have a better chance of having enough quota
[10:36] <pitti> but if it fails repeatedly, let the worker die so the others have more breathing room
[10:36] <pitti> or a test doesn't get retried infinitely often in a tight loop
[10:38] <Laney> pitti: ok, I can see that it keeps tests out of the rotation for a bit
[10:39] <Laney> pitti: so why did the old adt image get deleted?
[10:39] <FJKong_> Laney: hello
[10:39] <Laney> hi
[10:39] <pitti> Laney: good question, I don't know -- maybe someone thinking "oh, vivid images, these can go"
[10:41] <FJKong_> one question: source code number is 3.5,old package number is 1:3.5-1, if new upstram source code version number is 3.5.1 package number is 1:3.5.1-0 is this ok?
[10:42] <FJKong_> 1:3.5.1-0 is newer than 1:3.5-1?
[10:42] <Laney> FJKong_: If in Ubuntu: 1:3.5.1-0ubuntu1, if Debian: 1:3.5.1-1
[10:42] <FJKong_> thanks
[10:42] <Laney> FJKong_: Do `dpkg --compare-versions v1 gt v2 && echo yes'
[10:43] <Laney> have a look at "man dpkg" and search for "compare-versions" to find out more about that
[10:43] <FJKong_> ok
[10:43] <FJKong_> cool
[10:47] <Laney> pitti: ok, filed
[10:48] <pitti> Laney: cheers
[10:48] <Laney> pitti: so what drives the daily image updating?
[10:48] <Laney> I see build-adt-image-all-clouds but not how that is invoked
[10:49] <pitti> RELEASE=${RELEASE:-$(distro-info --devel)}
[10:49] <pitti> Laney: correct, I only do this for the devel series; I use the daily official images for stables
[10:50] <pitti> Laney: that is slower for every test (as the setup script has much more stuff to do), but simpler as we don't need to generate custom adt images for all releases
[10:50] <Laney> pitti: ah right
[10:51] <pitti> Laney: i. e. create-nova-image-new-release is (a) an optimization over daily cloud images, and (b) needed for opening a new release while we don't have official cloud images for it yet
[10:51] <pitti> Laney: for (a) that essentially just pre-applies setup-testbed to the image, so that for the actual test it's a fast no-op
[10:52] <pitti> Laney: and for (b) you call it manually on the previous stable image (like the latest yakkety image for z), and it'll then dist-upgrade to zesty
[10:52] <Laney> I should practice that at some point
[10:53] <seb128> Laney, pitti, unsure if you saw bug #1633162 but it's another issue due to the systemd user session
[10:53] <ubot5`> bug 1633162 in nautilus (Ubuntu) "Files application throws an error 'Oops! Something went wrong' when clicked on 'Recents'" [High,Confirmed] https://launchpad.net/bugs/1633162
[10:53] <pitti> Laney: you can practice by taking a trusty daily and creating a wily image (and then delete it again)
[10:54] <pitti> Laney: wily is still marked as supported for reasons that are beyond me, so it's on archive.u.c.
[10:55] <pitti> seb128: I saw it fly by, yes; slowly working through the backlog from last week, I'll look at bug 1637758 next
[10:55] <ubot5`> bug 1637758 in lightdm (Ubuntu) "Revision 2409 introduces regression" [Undecided,New] https://launchpad.net/bugs/1637758
[10:56] <Laney> pitti: ack, I'll try that
[10:57] <Laney> pitti: ok, I see the regex stuff for selecting an image now so I understand how this works I think
[10:58] <Laney> seb128: yeah
[11:04] <Laney> pitti: I subscribed you to a bug regarding the lightdm change too, if you saw that - could be the same root cause (multiple running u-s-ds)
[11:04] <Laney> ah yes I think it is - that one talks about xsettings
[11:05] <pitti> Laney: yeah, I forgot to stop the session
[11:05]  * Laney has duped it now
[11:05] <pitti>            State: closing
[11:05] <pitti>             Unit: session-c1.scope
[11:05] <pitti>                   ├─2720 /usr/lib/at-spi2-core/at-spi2-registryd --use-gnome-session
[11:05] <pitti>                   ├─2822 nm-applet
[11:06] <pitti>                   └─2824 /usr/lib/unity-settings-daemon/unity-settings-daemon
[11:06] <pitti> is what I get on my system
[11:06] <Laney> same (see https://bugs.launchpad.net/ubuntu/+source/lightdm/+bug/1639658/comments/3)
[11:06] <ubot5`> Ubuntu bug 1637758 in lightdm (Ubuntu) "duplicate for #1639658 Revision 2409 introduces regression" [Undecided,Confirmed]
[11:06] <pitti> but this is z only, so just "urgent", not "panic"
[11:06] <Laney> ya
[11:06] <Laney> we'd have noticed on Y already, the menus thing is pretty obvious :)
[11:37] <andyrock> morning
[11:38] <seb128> hey andyrock! how are you?
[11:38] <andyrock> terrible...
[11:39] <seb128> :-(
[11:39]  * andyrock is reviewing Trevinho's branches
[11:39] <andyrock> ahaha
[11:39] <seb128> lol
[11:39] <Trevinho> bastard :-D
[11:39] <seb128> did he rewrote half of unity over night again?
[11:39] <Trevinho> hi seb128
[11:39] <Trevinho> seb128: it's ooold stuff
[11:39] <seb128> hey Trevinho! how are you?
[11:39] <Trevinho> seb128: ok, good... I've spent yesterday evening in snappying some stuff
[11:40] <Trevinho> seb128: remmina, actually... although I had to use this workaround to maket it properly load from an absolute path.... https://github.com/3v1n0/Remmina-snap/blob/master/snapcraft.yaml#L72 (+ the run script)
[11:41] <seb128> Trevinho, that's quite hackish
[11:41] <Trevinho> that come up since last saturday I was at the Italian LOCO team meeting... And so there was some interest in snappying...
[11:42] <seb128> couldn't you just use /snap/<name>/current?
[11:42] <Trevinho> seb128: indeeeeed, but no
[11:42] <seb128> great
[11:42] <seb128> why?
[11:42] <Trevinho> seb128: if I'd use that prefix, then cmake would try to install in /snap/<name>/current//snap/<name>/current
[11:42] <Trevinho> and... upstream code is not wrong in that sense...
[11:42] <seb128> right
[11:43] <Trevinho> so unless we don't have proper mount bindings, I don't see other solution than this :-(
[11:43] <seb128> then you can use an organize rule to move /snap/<name>/current/usr to /usr
[11:43] <Trevinho> being /tmp the only absolute path we've
[11:43] <seb128> which gives you back things under /snap/<name>/current
[11:43] <Trevinho> oh
[11:44] <Trevinho> seb128: do you have an exapmle?
[11:44] <seb128> Trevinho, https://github.com/8none1/gedit310/blob/master/snapcraft.yaml#L19
[11:44] <seb128> that's autotools not cmake
[11:44] <seb128> but the principle is the same
[11:44] <Trevinho> nice... but yeah
[11:45] <seb128> be careful you should do that in a part that has the build only
[11:45] <seb128> if you have stage packages in the same part you might end moving their content as well
[11:45] <seb128> which you don't want to do
[11:45] <Trevinho> I see
[11:47] <Trevinho> seb128: I've got this problem with appindicator though https://bugs.launchpad.net/snappy/+bug/1639988
[11:47] <ubot5`> Ubuntu bug 1639988 in Snappy "Snaps using libappindicator and unity7 plug can't show app-indicators" [Undecided,New]
[11:48] <Trevinho> see the test snap in the bottom, you can compile it in seconds
[11:48] <seb128> weird
[11:48] <seb128> indicators should work
[11:49] <seb128> I'm about to go for lunch but can try after that
[11:50] <Trevinho> seb128: sure, well... It used to, but I get permission problems
[14:17] <xnox> seb128, shall we have a UOS session about "3rd debate on building Ubuntu desktop i386 isos"?
[14:17] <tedg> mpt: Is it "a URL" or "an URL" ?
[14:18] <seb128> xnox, do you think a discussion in a session is likely to useful to get to a conclusion?
[14:19] <xnox> seb128, i really really do hope so. Last time we got to (a) i386 is canonical supported architecture (b) Xorg stack on i386 is continued to be supported
[14:19] <xnox> but we didn't get to timings
[14:19] <xnox> thus I'm hoping to achieve consensus that yes we may stop building ubuntu-zesty-desktop-i386.iso
[14:19] <seb128> k, register one then I guess
[14:19]  * ogra_ grins about tedg ... an american asking a kiwi about english grammar 
[14:20]  * seb128 is writing to you from an i386 ubuntu install
[14:20] <xnox> tedg, english a & an are based on how things are pronounced, not written. Thus it's a URL
[14:20] <xnox> and e.g. an honest answer (because h is silent)
[14:21] <xnox> http://www.quickanddirtytips.com/education/grammar/a-versus-an -> hehe "a unicorn" yet "an umbrella"
[14:21] <tedg> xnox: Yeah, I found that reference, but I wasn't sure on the case of "U" in "URL" because it is closer to "you" than "un"
[14:22]  * xnox believes that you is a consonant sound
[14:22]  * tedg isn't sure of anything anymore
[14:22] <xnox> http://itknowledgeexchange.techtarget.com/writing-for-business/which-is-correct-a-url-or-an-url/
[14:23] <xnox> Although the U at the first of URL is a vowel, the abbreviation is pronounced “you-are-ell.” Because the letter Y is a consonant in this case, use a rather than an.
[14:23] <tedg> But I was figuring mpt had done a fair amount of work with Mozilla, he'd know for sure :-)
[14:24] <xnox> tedg, somehow i suspect that mpt would say use "Location" =) hence the Ctrl+L is the shortcut for "type the URL in"
[14:24] <tedg> Heh, that's cheating! ;-)
[14:25] <tedg> xnox: Oh, also while you're here... I have on my TODO to follow up with you and make sure you talked to pitti about backporting the graphical-session target to xenial.
[14:25] <tedg> Is that all good?
[14:25] <tedg> (or at least on the TODO list)
[14:27] <xnox> tedg, well that's backwards. Before we backport things (a) are all migrations done in zesty (b) do we need any remaining bobs in yakkety (c) what do we need in xenial (i'm guess all of it?!)
[14:28] <xnox> tedg, cause e.g. we will not be switching xenial folks on classic to systemd graphical session.
[14:28] <tedg> xnox: We need it to get into ubuntu-core so that we can use it to based the unity8-snap on.
[14:28] <tedg> xnox: So it's not a matter of migrating xenial folks as much as laying a foundation.
[14:29] <xnox> right but you said you need system systemd and system pam to start graphical target for you..... yet mix in user session units from the unity8 snap; which snapd does not support at the moment.
[14:30] <tedg> Sure, working on that too. But trying to make everything land.
[14:31] <tedg> We need snapd to do user session units, I have a meeting on that tomorrow :-)
[14:31] <tedg> I don't think that getting the graphical-session target into xenial needs to block on that though.
[14:34] <xnox> tedg, right.
[14:34] <xnox> tedg, note, it's not just user .service files; but user .target; user .socket; etc.
[14:35] <tedg> xnox: Man, now you want ALL the units! ;-)
[14:35] <pitti> tedg: oh, you literally just want the two *.target unit files, not the entire set of fixes and ports to *use* them?
[14:35] <pitti> tedg: I can cram that into the next SRU, sure (but not sure how much that actually helps without the supporting infra)
[14:35] <tedg> pitti: I'm not entirely certain there. We want to be able to make a snap that can be part of the graphic-session target.
[14:36] <xnox> pitti, well i think the polling method for starting graphical-target needs to go in too.
[14:36] <pitti> like systemctl start --wait?
[14:36] <tedg> pitti: I imagine we'll need more than just the target files, but I don't understand how much more.
[14:36] <xnox> pitti, whatever you coded and merged upstream, yeah.
[14:36] <pitti> that's a loot more
[14:36] <pitti> it's also dbus-user-session, two handfuls of fixes for it, etc.
[14:37] <xnox> pitti, and eventually we will have these sessions in xenial: unity7 (upstart), unity8 (upstart), unity8-snap (systemd)
[14:37] <xnox> pitti, and eventually we will have these sessions in yakkety: unity7 (upstart), unity8 (systemd), unity8-snap (systemd)
[14:37] <pitti> and given that we still find subtle regression, we can't possibly switch existing xenial systems to that
[14:37] <tedg> I thought dbus-user-session was already in xenial?
[14:37] <xnox> nope.
[14:37] <pitti> the package yes, but not functional
[14:38] <xnox> tedg, your best bet at the moment is to test Unity8 snaps on yakkety classic system with snapd
[14:38] <xnox> because all the user session support is in there. and your base is the right one.
[14:39] <tedg> Sure, I guess that's fine for the moment. But I'm concerned that we don't have a path to get it so that a U8 snap could run on core 16.
[14:39] <xnox> tedg, or are you after an all-snaps system? my understanding that personal all-snaps doesn't exist yet, and should be started with work on core 16?!
[14:39] <tedg> Yes, all snaps doesn't exist yet, but it is the goal and is expected to work on core 16.
[14:39] <xnox> tedg, when you say core 16, do you mean xenial-classic with snapd? yakkety-classic with snapd? all-snaps core 16?
[14:39] <tedg> All snaps core 16.
[14:40] <tedg> Ubuntu Core Snap series 16.
[14:40] <xnox> tedg, if we can shove the dbus-user-session from zesty into your unity8 snap... we should be fine =)
[14:41] <xnox> to be honest
[14:42] <xnox> i somehow expect to use systemd from unity8 snap for the user session init; talking to system systemd from ubuntu core 16 as pid 1
[14:42] <xnox> pitti, what do you think ^
[14:42] <tedg> I don't think we can do that? I mean, we don't have any control over that part of the session AFAIK. That'd require quite a few holes.
[14:43] <xnox> horum. maybe jdstrand can help =)
[14:43] <pitti> xnox: not sure what you mean -- user systemd doesn't really talk to the system systemd
[14:43] <xnox> jdstrand, is it fine, on an all snaps system, for system PAM and system systemd; spawn pam_logind using session logind/systemd from the unity8 snap?
[14:44] <xnox> pitti, true. (i'm getting confused with like how upstart user session worked)
[14:44] <pitti> xnox: wasn't that the same? I don't think the session and system inits talked to each other directly; at most via these bridges
[14:46] <xnox> right.
[14:46] <tedg> I have a hard time imagining systemd confined in a snap. It needs access all over for cgroups and such.
[14:47] <xnox> somehow imho graphical.target; dbus-user-session; and the whole lot should be all inside the unity8 snap. Even if it means a user dbus; inside the unity8 snap; specific to unity8 and not connected to a user session dbus vissible via ssh into the all snaps system.
[14:47] <xnox> tedg, user systemd has no access to cgroups and such.
[14:47] <mpt> tedg, I use “a” because I pronounce it “Yoo Arr Ell”, which starts with a consonant sound
[14:48] <xnox> Arrrrrr
[14:48] <tedg> mpt: Makes sense to me! Thanks!
[14:48] <seb128> thanks tedg xnox mpt ;-)
[14:48] <tedg> xnox: If that's the case then every desktop has to replicate that and there's no way for snapd to do user services.
[14:49] <seb128> tedg, sorry for the noise!
[14:49] <tedg> xnox: User services would only be defined for inside the snap instead of inside the session.
[14:50] <xnox> tedg, hypothetically who do you expect to own the session; the display; and which interface to launch if two snaps are installed: unity8 and xubuntu
[14:50] <xnox> and both provide.... interface user-graphical.target?
[14:51] <xnox> or should snapd only connect one of them?
[14:51] <xnox> or if we have multiple snaps.... only connect those that you want/need? e.g. connect unity8 snap and median-scanner snap with user units; but not the xubuntu snap
[14:51] <tedg> xnox: No, it would connect each of them to lightdm:sessions
[14:52] <tedg> xnox: Then lightdm would deal with handling the multiple and starting the right one.
[14:52]  * xnox cries but lightdm has it's own systemd user session too.....
[14:52] <xnox> ..
[14:53] <xnox> i think for now we want xenial's systemd to have graphical.target unit or whatever it was and that's it. And deal with dbus-user-session et.al. later.
[14:53] <pitti> well, user systemd plus session dbus isn't going to work
[14:53] <xnox> pitti, can you please just sru in the two units into xenial and that's it, for now?
[14:54] <pitti> sure
[14:54] <xnox> pitti, right, but i have no idea how all-snaps systems currently handle multi-user, and if we can switch ubuntu core 16 to dbus-user-session; or if we can use dbus-user-session only via lightdm:session interface; or via a gadget snap or some such.
[14:55] <xnox> i should play with an all snap system in a vm and poke things experimentally.
[15:30] <willcooke> ding ding ding!
[15:30] <willcooke> #startmeeting Desktop Team Weekly Meeting - 2016-11-08
[15:30] <meetingology> Meeting started Tue Nov  8 15:30:51 2016 UTC.  The chair is willcooke. Information about MeetBot at http://wiki.ubuntu.com/meetingology.
[15:30] <meetingology> Available commands: action commands idea info link nick
[15:31] <willcooke> Roll call: andyrock, attente, desrt,  dgadomski, fjkong (out), flexiondotorg, happyaron (out), hikiko, laney, qengho, seb128, sweet5hark, themuso (out), tkamppeter, trevinho, robert_ancell (out)
[15:31] <flexiondotorg> o/
[15:31] <hikiko> hi
[15:31] <seb128> hey
[15:31] <andyrock> hi
[15:31] <Laney> meow
[15:31] <qengho> Aiee!
[15:31] <desrt> *mwah
[15:32] <andyrock> willcooke: I still have to prepare the bullet list
[15:32] <andyrock> can you start with someone else? :D
[15:32] <willcooke> sure!
[15:32] <willcooke> muhahaha
[15:32] <willcooke> #topic attente
[15:33] <willcooke> Let's come back to attente too
[15:33] <attente> sorry
[15:33] <attente> i'm ready
[15:33] <willcooke> ah, cool,
[15:33] <attente> debugging some issues with libertine menus and fonts
[15:33] <attente> finished fixing fcitx-qimpanel build to allow local installation
[15:33] <attente> looked into the broken gtk menus and dpi scale issues a bit, gtk isn't reading the xsettings properly, but not sure how to fix that
[15:33] <attente> re-wrote the jhbuild plugin to use bubblewrap instead of lxc, needs tests+docs
[15:34] <attente> still need a review of content-hub-glib before landing the copy-paste stuff in gtk
[15:34] <attente> (eof)
[15:34] <willcooke> thanks attente
[15:34] <willcooke> #topic desrt
[15:34] <desrt> i figured out the security for the proxy, and implementation is underway.  will continue on that this week.
[15:34] <willcooke> woo!
[15:34] <desrt> might be interesting to see if we can use this as an alternative to the xsettings stuff to solve the issues attente is seeing
[15:34] <desrt> that's all, really
[15:34] <willcooke> thanks desrt
[15:34] <willcooke> #topic dgadomski
[15:34] <dgadomski> hey
[15:34] <dgadomski> - worked on bug #1589401, no results just yet
[15:34] <dgadomski> - returning to bug #1598183
[15:34] <dgadomski> - working on bug #1598183
[15:34] <ubot5`> bug 1589401 in NetworkManager "cannot view wifi networks after re-enabling wifi" [Medium,Confirmed] https://launchpad.net/bugs/1589401
[15:34] <ubot5`> bug 1598183 in gvfs "Operation not permitted while writing to symlinked fuse locations" [Medium,Confirmed] https://launchpad.net/bugs/1598183
[15:34] <dgadomski> eof
[15:35] <willcooke> thanks dgadomski, feel free to reach out to happyaron on #1589401 - he might be able to assit
[15:35] <willcooke> assist
[15:35] <willcooke> #topic FJKong_
[15:35] <willcooke> libproxy-0.4.13.
[15:35] <willcooke>   * New upstream release
[15:35] <willcooke>   * Remove 3 Patches
[15:35] <willcooke> klog-0.9.2.5-1
[15:35] <willcooke> make new upstream release
[15:35] <willcooke> upload to :
[15:35] <willcooke> https://mentors.debian.net/package/klog
[15:35] <willcooke> fix bug#831236
[15:35] <willcooke> * xnec2c
[15:35] <willcooke>   * New upstream release
[15:35] <willcooke>     - add icon image to fix gui-app-without-icon error
[15:36] <willcooke> #topic flexiondotorg
[15:36] <flexiondotorg> Time appropriate greetings
[15:36] <flexiondotorg> - Researched this weeks candidates for the Snap Upstream Blitz and updated their Trello board.
[15:36] <flexiondotorg> - Reviewed the ISV Snap advocacy process and documentation.
[15:36] <flexiondotorg> - Several calls learning the process for ISV Snap advocacy.
[15:36] <flexiondotorg> - Learning the snap advocacy process came round sooner than I expected and has taken longer than I thought to learn. I'm up to speed now.
[15:36] <flexiondotorg> - Reworking merge proposals for aptdaemon and update-manager (LP: #1623856) - PPA coming later for testing.
[15:36] <ubot5`> Launchpad bug 1623856 in aptdaemon (Ubuntu) "Scrolled Windows in update-manager are too small to read" [Low,In progress] https://launchpad.net/bugs/1623856
[15:36] <flexiondotorg> - Fix for deja-dup scrolled windows (LP: #1623835) - MP and PPA coming later.
[15:36] <ubot5`> Launchpad bug 1623835 in deja-dup (Ubuntu) "Updated gtk element means you need to specify a size on the details drop down or it is only 1 line high" [Low,Triaged] https://launchpad.net/bugs/1623835
[15:36] <flexiondotorg> eof
[15:36] <willcooke> thanks flexiondotorg, great stuff
[15:37] <willcooke> #topic happyaron
[15:37] <willcooke> 1. dkms: lots of bugs, upstreamed apport related updates
[15:37] <willcooke>     merge is pending because how to handle secure boot patch is not yet clear
[15:37] <willcooke> 2. NM bugs at Debian, trying to merge this week
[15:37] <willcooke> 3. sponsored fqterm uploads, synced to Ubuntu
[15:37] <willcooke> 4. batch translation review for gnome zh_CN
[15:37] <willcooke> #topic hikiko
[15:37] <hikiko> hi
[15:37] <hikiko> * 1st investigation/document for chromium on mir
[15:37] <hikiko> * compiz:
[15:37] <hikiko>   - fixed some compile errors on tests
[15:37] <hikiko>   - fixing the branches that have been reviewed by Trevinho (moving win, expo gap)
[15:37] <hikiko> * unity:
[15:37] <hikiko>   - working  on the decorations code - changes that will be useful for different tasks
[15:37] <hikiko>   - had almost fixed the switcher (alt+`) before my desktop crashed, tomorrow :p
[15:37] <hikiko> * other:
[15:37] <hikiko>   - debug an issue I had on the desktop and thought it was related to either lightdm or u7 gsettings, found out that its an xserver bug latest Y release
[15:37] <hikiko> eof
[15:37] <willcooke> thanks hikiko
[15:37] <willcooke> #topic Laney
[15:37] <Laney> • Upload livecd-rootfs to stop stripping .pyc on live images, announce to -release
[15:38] <Laney> • Some task finding, advice, review & sponsoring for FJK
[15:38] <Laney> • Turned autosync back on, babysat autopkgtest for that (managing capacity and handling emergent issues)
[15:38] <Laney> • Got some informal training in autopkgtest tasks from p_itti
[15:38] <Laney> • Poked a bit at u-s-d's xsettings manager not working in Z (1639658)
[15:38] <Laney> • Worked much more on asgen charm, implemented external storage, getting ready to import existing data so that the new version can be deployed
[15:38] <Laney> 😚
[15:38] <willcooke> :) thanks Laney
[15:38] <willcooke> #topic qengho
[15:38] <qengho> * Cr 54 still in progress. All 32-bit arches broken. We don't cross-compile. Ups
[15:38] <qengho> tream does. :(
[15:38] <qengho> That's pretty much it. All my time.
[15:39] <qengho> EOF
[15:39] <willcooke> thanks qengho, good luck
[15:39] <willcooke> #topic seb128
[15:39] <seb128> • debugged versions' script erroring out after the url updates done by j_bicha
[15:39] <seb128> • helped flexidotong to fix the update-manager details view having an incorrect height with the new gtk
[15:39] <seb128> • triaged recent nautilus bugs
[15:39] <seb128> • NEW queue reviews
[15:39] <seb128> • discussed/verified g-s-snap SRU
[15:39] <seb128> • reviewed/cleaned up some items from the sponsoring queue, not quite a full shift though

[15:39] <willcooke> thanks seb128!
[15:39] <willcooke> #topic Sweet5hark
[15:41] <seb128> zZZzz ;-)
[15:41] <willcooke> he was here earlier, let's come back
[15:41] <willcooke> #topic TheMuso
[15:41] <willcooke> * Uploaded an SRU for bug 1574324 for xenial and yakkety. Testing seems favourable so far, will get fresh installs set up in coming days to test myself.
[15:41] <willcooke> * Still trying to reliably reproduce bluetooth profile switching issues on xenial at least. No luck so far.
[15:41] <willcooke> * Started going through the patches we carry for Pulseaudio, some of which are to work around issues that may no longer exist. Will start to test these in the coming week.
[15:41] <willcooke> * More package merges for zesty.
[15:41] <ubot5`> bug 1574324 in pulseaudio (Ubuntu) "pulseaudio crashes when connecting to bluetooth headphones (due to ubuntu changes?)" [High,In progress] https://launchpad.net/bugs/1574324
[15:41] <willcooke> #topic tkamppeter
[15:42] <tkamppeter> - cups-filters: Backported cups-browsed-hang-on-shutdown fixes to the Xenial package for an SRU (bug 1638986).
[15:42] <tkamppeter> - cups-filters: Further work on IPP Everywhere support.
[15:42] <tkamppeter> - cups: Tested IPP Everywhere PPD generator of CUPS with non-IPP-Everywhere printers getting broken PPDs. Reported bug upstream.
[15:42] <tkamppeter> - Snappy: Created VM running Ubuntu Core 16.
[15:42] <ubot5`> bug 1638986 in cups-filters (Ubuntu Xenial) "cups/cups-browsed causing delay in shutdown (16.04)" [Undecided,In progress] https://launchpad.net/bugs/1638986
[15:42] <tkamppeter> - Bugs.
[15:42] <tkamppeter> Some picture of The Hague: https://www.dpreview.com/galleries/8150277390/albums/the-hague
[15:42] <willcooke> :D thanks tkamppeter
[15:43] <willcooke> #topic Trevinho
[15:43] <Trevinho> · Some reviews of compiz and unity branches
[15:43] <Trevinho> · Landed new unity and compiz in zeisty
[15:43] <Trevinho> · Fixed icon theme update on lockscreen (causing indicators to have a wrong icon)
[15:43] <Trevinho> · Attended the italian LoCO meeting (ubuntu-it meeting with snapping session JAM, and mir porting)
[15:43] <Trevinho> · Snapped bucklespring
[15:43] <Trevinho> · Snapped Remmina (including freerdp)
[15:43] <Trevinho> · Proposed some changes to desktop-launchers (and reported various snap-world bugs)
[15:43] <Trevinho> · Some fixes in unity8 indicators
[15:43] <Trevinho> EOF
[15:43] <willcooke> thanks Trevinho
[15:43] <willcooke> did you push your snapcraft yamls to the playpen?
[15:43] <willcooke> if not, please do!
[15:44] <willcooke> #topic robert_ancell
[15:44] <willcooke> - snapd-qt patch complete - awaiting review
[15:44] <willcooke> - gnome-software 3.22 released to zesty - patches updated
[15:44] <willcooke> #topic andyrock
[15:44] <andyrock> #1 Code Reviews
[15:44] <andyrock> #2 Updated branch lp:~azzar1/unity/properly-handle-copy-dialog
[15:44] <andyrock> #3 Fixed double lock issue lp:~azzar1/unity/fix-double-lock
[15:44] <andyrock> #4 Some debugging for lp:1635625 and workaround proposed (but Marco found another one)
[15:44] <andyrock> eof
[15:44] <willcooke> thanks andyrock!
[15:44] <willcooke> #topic Sweet5hark
[15:44] <willcooke> is he back?
[15:44] <seb128> lp 1635625
[15:45] <ubot5`> Launchpad bug 1635625 in unity (Ubuntu) "Some indicator icons are missing after unlocking the screen" [High,Fix released] https://launchpad.net/bugs/1635625
[15:45]  * Sweet5hark is back. confused by wintertime.
[15:45] <Sweet5hark> - building snaps on lp
[15:45] <Sweet5hark> - bumped to 5.2.3 snap and deb
[15:45] <flexiondotorg> ^ is brilliant
[15:45] <Sweet5hark> - started armhf and i386 builds (untested so far)
[15:46] <Sweet5hark> - netzpolitischer Kongress in berlin (Gruene)
[15:46] <Sweet5hark> - backporting to LTS
[15:46] <Sweet5hark> - thats it IIRC
[15:46] <Sweet5hark> EOF
[15:46] <willcooke> thanks Sweet5hark
[15:46] <willcooke> #topic AOB
[15:46] <willcooke> Anyone got news?
[15:47] <flexiondotorg> Not news, but a question.
[15:47] <flexiondotorg> I've been working on LP: #1631208
[15:47] <ubot5`> Launchpad bug 1631208 in ubuntu-themes (Ubuntu) "Tabs in Firefox dialogs are not rendered (for add-on options & Certificate Viewer)" [Low,In progress] https://launchpad.net/bugs/1631208
[15:47] <flexiondotorg> Which required a refactoring of notebooks in general. And that changes the general appearance of notebooks/tabs.
[15:48] <flexiondotorg> In order to support Firefox new visual elements are introduced.
[15:48] <flexiondotorg> Should I continue with FF support now as is, or should we consider patching FF to introduce bespoke selectors.
[15:49] <flexiondotorg> Right now it seems FF > 49 is reliant on frames for drawing boarders.
[15:49] <flexiondotorg> Which requires work around in some traditional GTK applications.
[15:49] <attente> isn't there a way to theme individual apps by name?
[15:49] <Laney> no
[15:49] <Laney> there's an upstream bug for firefox though
[15:50] <Laney> I think it'd be a good idea to explain the problem there and ask if they would accept doing that
[15:50] <attente> i could've sworn there were app-specific selectors in the theming code...
[15:50] <Laney> Only if the application has done something
[15:50] <willcooke> bear in mind that Firefox will become a snap Real Soon Now - so any changes to Ffox will need to be upstreamed rather than patched by us.
[15:50] <flexiondotorg> attente, There are app specific selectors for gedit and gnome-terminal.
[15:50] <flexiondotorg> But for frame in general, there are not.
[15:51] <attente> oh i see...
[15:51] <Laney> You can sometimes pick widgets by name
[15:51] <flexiondotorg> OK, I think I;ll split the FF notebook change out.
[15:51] <flexiondotorg> ANd just submit an MP for the gedit and gnome-terminal stuff.
[15:51] <Laney> I don't think that's a particularly great idea because they are usually implementation details and might break at any time
[15:52] <Laney> Firefox might be willing to guarantee a CSS class or something
[15:52] <flexiondotorg> To be clear, remove the FF changes to a separate branch for now.
[15:52] <flexiondotorg> And get the other issues closed out.
[15:53] <flexiondotorg> I'll then progress FF issues upstream.
[15:53] <flexiondotorg> Laney, yep. Ideally their own class(es).
[15:53] <flexiondotorg> Thanks guys.
[15:53] <willcooke> thanks flexiondotorg
[15:53] <Laney> It's kind of crap to have to have app-specific code
[15:54] <Laney> In an ideal world they would just directly look the same
[15:54] <flexiondotorg> Agreed.
[15:54] <willcooke> any more for anymore? 5... 4...
[15:54] <Laney> I guess they don't directly use GTK widgets though - so it might be the best we can get.
[15:54] <Laney> Ta
[15:54] <flexiondotorg> Laney, Correct, they do not.
[15:55] <seb128> flexiondotorg, what are those snappy upstream advocacy guidelines? something we should all read?
[15:55] <willcooke> oki, let's wrap - we can continue afterwards if needed
[15:55] <flexiondotorg> seb128, Eventually, yes.
[15:55] <willcooke> oops, sorry seb128
[15:55] <flexiondotorg> But they'll get refined through use.
[15:55] <seb128> are we being transformed in marketing team sidekicks? ;-)
[15:56] <seb128> willcooke, no worry, can be an after meeting discussing, it's just the first time I read that we have advocacy material and are supposed to follow it
[15:56] <flexiondotorg> Everyone eventually ;-)
[15:56] <willcooke> oki, closing meeting - please carry on as needed
[15:56] <willcooke> #endmeeting
[15:56] <meetingology> Meeting ended Tue Nov  8 15:56:25 2016 UTC.
[15:56] <meetingology> Minutes:        http://ubottu.com/meetingology/logs/ubuntu-desktop/2016/ubuntu-desktop.2016-11-08-15.30.moin.txt
[15:56] <flexiondotorg> seb128, It is not marketing exactly.
[15:56] <flexiondotorg> Developer to developer discussions.
[15:57] <seb128> k
[15:57] <flexiondotorg> When it become a sales exercise there is a hand off :-)
[15:57] <seb128> well I didn't know we had a specific strategy to learn
[15:59] <seb128> flexiondotorg, thanks
[15:59] <flexiondotorg> np
[16:28] <willcooke> tkamppeter, great pic of the canal.  Was that on the way to team dinner?
[16:44] <attente> did the synaptics driver break for anyone?
[16:44] <flexiondotorg> attente, What release?
[16:44] <attente> flexiondotorg: z
[16:44] <attente> i just did an update
[16:45] <flexiondotorg> I'll tell you tomorrow, re-install a test machine that has a touchpad.
[17:45] <flexiondotorg> Catch you all tomorrow desktopers
[17:45] <seb128> flexiondotorg, have a nice evening!
[17:45] <willcooke> night flexiondotorg
[18:01] <Laney> night from me too
[18:16] <seb128> night Laney
[18:17] <seb128> calling it a day as well
[18:17] <seb128> good night desktopers
[18:45] <willcooke> night all
[22:05] <attente> flexiondotorg: hey, just wanted to let you know don't worry about the touchpad testing. seems all i needed was to make a custom libinput config in /usr/share/X11/xorg.conf.d
[22:20] <flexiondotorg> OK.