[01:01] <fos> how do I find out how much data is on the hard drive
[01:05] <rpadovani> !support | fos
[01:06] <fos> thanks
[09:13] <tsdgeos> Cimi: what's delaying the review of https://code.launchpad.net/~aacid/unity8/photoscopeimprovements ?
[09:14] <tsdgeos> dednick: do you remember why not top approve of https://code.launchpad.net/~aacid/unity8/list_on_bottom_swipe/+merge/235266 ?
[09:21] <Saviq> dandrader, hey, about shell rotation... we need a notion of native orientation
[09:22] <Saviq> dandrader, and that's what the dash, the dialer need and apps need to be able to declare, that they run only in native orientation of the device
[09:23] <dandrader> Saviq, we have it already
[09:23] <dandrader> Saviq, check OrientatedShell.qml properties
[09:23] <dandrader> OrientedShell.qml
[09:23] <Saviq> dandrader, that native orientation we need to be able to override, too (like we want the N7 to be landscape, even though hardware wants portrait)
[09:23] <Saviq> dandrader, but is that a supported choice for the .desktop file?
[09:25] <dandrader> Saviq, mzanetti, any thoughts on that? I'm having a very hard time convincing uitk guys: https://code.launchpad.net/~dandrader/ubuntu-ui-toolkit/dontReparentPopup/+merge/242313Sa
[09:25] <dandrader> Saviq, mzanetti and https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1394511
[09:26] <Saviq> dandrader, (where) have they voiced their concerns?
[09:26] <dandrader> Saviq, #sdk. a loooong discussion there
[09:27] <dandrader> Saviq, started about 1.5 hours ago
[09:30] <dandrader> Saviq, that's primary orientation
[09:30] <dandrader> Saviq, not native orientation
[09:30] <dandrader> Saviq, N7's native orientation is portrait, but its primary orientation is landscape
[09:31] <dandrader> Saviq, as per OrientedShell's definition of native and primary orientations
[09:31] <dandrader> Saviq, which differs a bit from QScreen's one
[09:32] <dandrader> Saviq, but overall this is still a WIP area as I just started tackling tablet vs. phone in shellRotation
[09:33] <dandrader> Saviq, so my idea is that dash  would have "X-Ubuntu-Supported--Orienteations=primary" in its desktop file
[09:33] <dandrader> jeez, so many typos...
[09:34] <Saviq> dandrader|afk, yes, something of the sort
[09:38] <Saviq> mzanetti, snap-after-drag and rounded-bfb conflict, could you rebase one on the other please
[09:38] <mzanetti> Saviq: ack
[09:42] <mzanetti> Saviq: https://code.launchpad.net/~mzanetti/unity8/rounded-bfb/+merge/242322
[09:43] <Saviq> mzanetti, tx
[09:50] <tsdgeos> mzanetti: with https://code.launchpad.net/~mzanetti/unity8/reversible-spread/+merge/241274 there's no test for endPhase 2
[09:51] <mzanetti> tsdgeos: yeah, because that doesn't exist any more
[09:51] <mzanetti> it doesn't go to phase2 any more as long as dragging
[09:51] <mzanetti> but it snaps to phase 2 after releasing, there's still a test for that
[09:52] <tsdgeos> mzanetti: stop dragging then?
[09:52] <tsdgeos> is there?
[09:52] <mzanetti> yep
[09:52] <tsdgeos> ah right
[09:52] <tsdgeos> oki
[09:52] <mzanetti> very closely to the left edge though... so there's only 2 grid units or so that you could drag further in theory
[09:52] <mzanetti> but that would cross the line of the phase2 animation and look wierd when snapping
[09:53] <mzanetti> so I decided to not allow that
[10:17] <Saviq> mzanetti, you need to merge trunk into launcher-snap and merge into rounded-bfb, I'm getting criss-crossed otherwise
[10:17] <Saviq> *remerge into
[10:18] <mzanetti> Saviq: on it
[10:19] <mzanetti> Saviq: should be better now
[10:20] <Saviq> mzanetti, tx
[10:26] <dednick> tsdgeos: cant remember. but it's had a few changes since i approved.
[10:26] <tsdgeos> dednick: could you re-review?
[10:26] <dednick> tsdgeos: ya. doing now
[10:35] <tsdgeos> i open a wikipedia result preview and the dash is at 40% cpu :S
[10:35] <tsdgeos> can you guys reproduce?
[10:39] <Saviq> tsdgeos, any one in particular?
[10:39] <tsdgeos> Saviq: i'm opening max verstappen
[10:40] <tsdgeos> but i don't think it matters
[10:40] <Saviq> tsdgeos, yes, confirmed
[10:40] <Saviq> tsdgeos, but it settled down
[10:40] <tsdgeos> not for me
[10:40] <tsdgeos> only when the screen locked
[10:40] <tsdgeos> it did settle
[10:41] <tsdgeos> let's see if i can get it in the pc
[10:41] <mzanetti> tsdgeos: well, it gets sigstopped
[10:41] <tsdgeos> mzanetti: sure but then i came back and was fine
[10:41] <mzanetti> hmm. ok
[10:41] <tsdgeos> which is weird
[10:41] <mzanetti> yeah
[10:41] <Saviq> mzanetti, it doesn't (get sigstopped)
[10:41] <Saviq> mzanetti, dash is not lifecycled
[10:42] <mzanetti> Saviq: it does get sigstopped, just not sigtermed
[10:42] <Saviq> mzanetti, believe me, it's not sigstopped
[10:42] <greyback_> it's not sigstopped
[10:42] <greyback_> we removed that a month ago or so
[10:42] <mzanetti> did you guys revert my commit then?
[10:42] <mzanetti> ah
[10:42] <Saviq> mzanetti, yes, the scopes middleware can't deal with that (it was losing data)
[10:43] <greyback_> when we changed over to using UAL for lifecycle
[10:43] <mzanetti> sure. the same as with other apps...
[10:43] <Saviq> mzanetti, "other apps" don't talk to services via sockets / zmq
[10:44] <tsdgeos> Saviq: i wonder what's causing that, do we have a spinner hidden somehwere eating the cpu?
[10:44]  * tsdgeos tries to reproduce on the desktop
[10:44] <mzanetti> well, it still breaks network connections
[10:44] <mzanetti> and such things
[10:44] <Saviq> mzanetti, and you really don't need to restate your disdain over the lifecycle every time you get a chance, it's getting tiring
[10:44] <Saviq> tsdgeos, it could be the bottom activity indicator
[10:45] <Saviq> tsdgeos, it's not limited to wikipedia, reddit has the same issue (I'm thinking images)
[10:48] <dednick> tsdgeos: need to up unity8 required version of unity-api?
[10:50] <tsdgeos> dednick: let me see
[10:52] <tsdgeos> Saviq: may be the activity indicator (that is not even shown), i'm adding soem debug
[10:53] <tsdgeos> dednick: yeah we did that and then got lost in a merge since someone else "took" 7.93 for unity-api :D
[10:54] <tsdgeos> i'll up it to 7.94
[10:55] <tsdgeos> dednick: pushed
[10:56] <dednick> tsdgeos: ta. just testing changes now
[11:00] <dednick> tsdgeos: do you need to update translations, or is that done separately?
[11:00] <tsdgeos> Saviq: did we decide on anything? ↑
[11:00] <tsdgeos> update on the same MR makes for conflicts, update after makes for we forget :D
[11:01] <Saviq> tsdgeos, same MP
[11:01] <tsdgeos> ok :)
[11:01] <Saviq> tsdgeos, we don't change it that often
[11:01] <sil2100> dednick: ping
[11:01] <dednick> sil2100: yo
[11:01] <Saviq> tsdgeos, that is, until we get hook support in train, right sil2100? ;)
[11:02] <tsdgeos> dednick: pushed the pot_file then
[11:02] <tsdgeos> dednick: thanks for bringing it up :)
[11:02] <facundobatista> Hola
[11:04] <dednick> tsdgeos: cool. changes look fine. just running qmltests.
[11:08] <tsdgeos> what
[11:09] <tsdgeos> the indicator is now grey?
[11:09] <tsdgeos> Saviq: i can barely see it :/
[11:13] <tsdgeos> wait it's orange in krillin and gray in mako?¿
[11:14] <tsdgeos> ah, i guess someone "broke" Theme.palette.selected.foreground
[11:27] <Saviq> tsdgeos, yes, UITK
[11:28] <tsdgeos> Saviq: shall we file a bug so we don't forget?
[11:28] <Saviq> tsdgeos, we need to fix it locally
[11:28] <Saviq> tsdgeos, lemme MP
[11:29] <tsdgeos> have you guys run QSG_VISUALIZE=overdraw on the phone?
[11:30] <Saviq> I expect Gerry did
[11:30] <tsdgeos> i'm getting lots of
[11:30] <tsdgeos> QOpenGLShader::link: "L0010 Uniform 'tweak' differ on precision"
[11:30] <tsdgeos> and "nothing"
[11:30] <mzanetti> tsdgeos: yeah, works fine for me
[11:31] <mzanetti> tried last time about 3 weeks ago
[11:31] <tsdgeos> :/
[11:31] <tsdgeos> mzanetti: mako? krillin?
[11:31] <tsdgeos> how did you run it?
[11:31] <mzanetti> don't remember which device it was. I've set it in the upstart session. lemme try
[11:32] <Saviq> tsdgeos, 'restart unity8 QSG_VISUALIZE=overdraw'
[11:33] <mzanetti> doesn't seem to work here atm either
[11:34] <mzanetti> causes some artifacts, but not the thing we want
[11:34] <sil2100> Saviq: ;)
[11:34] <Saviq> sil2100, shall we get that into the CI team backlog?
[11:35] <tsdgeos> Saviq: does that work for you? i get regular unity8 :S
[11:35] <tsdgeos> let's restart
[11:35] <sil2100> dednick: sorry, got distracted... so, someone from QA stumbled upon an issue with indicators yesterday - we tried confirming this bug, but actually everything seems to be working fine for everyone else, but I guessed you might take a look at that:
[11:36] <Saviq> tsdgeos, yup, works here
[11:36] <mzanetti> Saviq: mako or krillin?
[11:36] <sil2100> dednick: https://bugs.launchpad.net/ubuntu/+source/ubuntu-system-settings/+bug/1394392
[11:36] <Saviq> mzanetti, krillin
[11:36] <mzanetti> hmm.. not working here on krillin
[11:36] <sil2100> dednick: might this be related to the recent sync-with-settings changes?
[11:37] <Saviq> restart unity8 QSG_VISUALIZE=overdraw
[11:37] <Saviq> mzanetti, tsdgeos ↑ that's all I did (it's vivid, though)
[11:37] <mzanetti> working on mako here
[11:38] <tsdgeos> oh
[11:38] <dednick> sil2100: that more looks like a bluetooth issue. I'll need to take a look
[11:38] <tsdgeos> so i'll change phone :D
[11:38] <mzanetti> yeah. I think I have vivid on mako and rtm on krillin
[11:38] <dednick> sil2100: as in system settings telling bluetooth to enable, but it doesnt.
[11:39] <sil2100> dednick: as mentioned, we had 2 people try to confirm this and the result was negative, so it might have been some singular case
[11:39] <dednick> sil2100: right. i just saw that.
[11:39] <dednick> sil2100: i've had troubles enabling bluetooth on my desktop after disabling, but not on phone.
[11:48] <dednick> sil2100: https://bugs.launchpad.net/ubuntu/+source/ubuntu-system-settings/+bug/1394392/comments/2
[11:51] <sil2100> dednick: thanks! Will have to poke Chris when he's up about this
[11:52] <facundobatista> does anybody know where the data that is sent in the User Agent comes from? (when client doing requests to the server remote scopes)
[11:56] <Saviq> facundobatista, pstolowski would know, but pete-woods might, too
[11:57] <pete-woods> Saviq, facundobatista: I don't know off the top of my head, but will dig through the code
[11:57] <facundobatista> Saviq, thanks! Pawel surely knows, but as I didn't find him, didn't know whom to ask :)
[11:58] <facundobatista> pete-woods, thanks
[11:58] <pete-woods> facundobatista:
[11:58] <pete-woods> http://paste.ubuntu.com/9121256/
[11:59] <Saviq> tsdgeos, https://code.launchpad.net/~saviq/unity8/dash-fix-colors/+merge/242336
[11:59] <pete-woods> http://paste.ubuntu.com/9121275/
[11:59] <pete-woods> facundobatista: ^ more complete
[12:02] <facundobatista> pete-woods, it should be building a "partner" parameter in the UA... is this already in the code?
[12:02] <Cimi> tsdgeos, someone to test it?
[12:03] <tsdgeos> Cimi: well it has your name in there, abstain if you're not going to ;)
[12:03] <tsdgeos> Saviq: looks good, let me review more
[12:04] <Cimi> tsdgeos, code approved then
[12:04] <Cimi> tsdgeos, as said, I had issues when I tested it
[12:04] <tsdgeos> Cimi: which issues?
[12:04] <Cimi> on my krillin
[12:05] <Cimi> tsdgeos, activity indicator staying on screen for several seconds
[12:05] <Cimi> tsdgeos, when refreshing
[12:10] <Saviq> Cimi, that's happening on stock rtm as well, problem in events scope
[12:10] <Cimi> Saviq, just on the scopes though
[12:10] <Cimi> sorry photo scope
[12:10] <Cimi> but anyway, we can merge then
[12:11] <Saviq> Cimi, still, the indicator is controlled by the backend
[12:30] <tsdgeos> dednick: \o/
[12:31] <dednick> tsdgeos: :)
[12:31] <tsdgeos> now we just need someone to review https://code.launchpad.net/~stolowski/unity-scopes-shell/feeds
[12:31] <tsdgeos> where's pstolowski hiding :D
[12:31] <artmello> ls
[12:32] <tsdgeos> pete-woods: do you guys have someone to review https://code.launchpad.net/~stolowski/unity-scopes-shell/feeds/+merge/239396 ?
[12:34] <tsdgeos> mzanetti: about https://code.launchpad.net/~mzanetti/unity8/reversible-spread/+merge/241274 ? no care about other stages?
[12:34] <mzanetti> tsdgeos: the tablet stage is changed too
[12:35] <mzanetti> line 74 in the diff
[12:37] <tsdgeos> man i can't read ^_^
[12:37] <mzanetti> no worries. first time I just scrolled over it I though: "huh? did I really forget to commit it? I'm sure I changed it" :D
[12:37] <mzanetti> took me 2 attempts to see it too
[12:39] <tsdgeos> mzanetti: how do i test it thoough, there's no tryTableStage :'(
[12:39] <mzanetti> tsdgeos: you can ./run.sh it and make the window fullscreen. it should switch to tablet mode
[12:39] <tsdgeos> that true
[12:40] <pete-woods> tsdgeos: it's on my list of stuff to review
[12:40] <tsdgeos> pete-woods: oki :)
[12:40] <Saviq> mzanetti, can you check, in the spread, swipe an app up and down, crossing the "center" position, does it jump around for you?
[12:41] <mzanetti> Saviq: trunk or in that no-jumping branch?
[12:41] <Saviq> mzanetti, silo 30
[12:41] <Saviq> mzanetti, but check on whatever
[12:41] <Saviq> mzanetti, but yeah, sounds like your no-jump branch made it jump when it crosses the center
[12:42]  * Saviq reboots to check if it persists
[12:43] <Saviq> yup
[12:44] <Saviq> mzanetti, also, did vesar see the close threshold increase?
[12:44] <Saviq> crash
[12:44] <mzanetti> Saviq: vesar is on parental leave unfortunately
[12:44] <Saviq> mzanetti, right, was wondering if he's back yet
[12:44] <mzanetti> but the only reason we didn't increase it before was the jumping
[12:45] <mzanetti> Saviq: afaik he'll be back on Mon
[12:45] <mzanetti> Saviq: when he comes back we might increase it even more
[12:45] <Saviq> mzanetti, kk
[12:46] <Saviq> atm stuff's crashing for me, not sure if it's happening in vivid-proposed yet
[12:46] <Saviq> and I'm not getting cores, not enough memory for some reason
[12:47] <mzanetti> Saviq: dang. it jumps indeed
[12:47] <mzanetti> Saviq: will fix
[12:47] <Saviq> k
[12:48] <Saviq> damn
[12:49]  * Saviq reflashes to see if it crashes in stock vivid-proposed, suspects Qt 5.3.2
[12:54] <Saviq> tsdgeos, in lp:~aacid/unity8/networkingstatus shouldn't onlineStatusChanged go away completely?
[12:55] <tsdgeos> Saviq: yes it should
[12:57] <tsdgeos> Saviq: pushd
[12:57] <Saviq> tsdgeos, thanks
[13:03] <mzanetti> Saviq: fixed. should I just push to the same branch?
[13:04] <Saviq> mzanetti, yeah
[13:04] <Saviq> mzanetti, do you have a bit of time to try something out for me?
[13:04] <mzanetti> Saviq: done
[13:04] <mzanetti> Saviq: can do, yeah
[13:04] <Saviq> mzanetti, get latest devel-proposed on mako
[13:04] <Saviq> mzanetti, citrain device-upgrade 30
[13:04] <Saviq> mzanetti, and try to get it to crash by launching xbmcremote and closing it repeatedly
[13:05] <Saviq> try swiping the launcher in over the xbmcr splash screen
[13:05] <mzanetti> ack
[13:05] <Saviq> I'm getting a crash rather reliably, but can't get core
[13:05] <mzanetti> flashing
[13:05]  * Saviq tries with gdb in the mean time
[13:14] <Saviq> mzanetti, I managed to get a trace
[13:14] <mzanetti> cool
[13:14] <mzanetti> Saviq: any pointers where it's crashing?
[13:14] <Saviq> http://pastebin.ubuntu.com/9122539/
[13:14] <Saviq> dandrader, ↑
[13:15] <Saviq> not sure how the silo unearthed this TH
[13:16] <Saviq> TBH
[13:20] <Saviq> /food
[13:21] <dandrader> Saviq, so no idea of the use case that caused this?
[13:23] <dandrader> Saviq, is there a bug with this stack trace? it's worth investigating
[13:33] <Saviq> dandrader, actually, yes
[13:33] <Saviq> dandrader, I just launch an app and start playing with the launcher, at some point I get the crash
[13:34] <Saviq> dandrader, I was only able to repro on vivid + silo 30 though, no idea what unearthed this issue
[13:36] <dandrader> Saviq, and what's in silo 30? (I don't know how to find silo URLs)
[13:37] <Saviq> dandrader, you go to http://people.canonical.com/~platform/citrain_dashboard/
[13:37] <Saviq> dandrader, silo 30 is http://people.canonical.com/~platform/citrain_dashboard/#?distro=ubuntu&q=landing-030
[13:51] <dandrader> Saviq, define "playing with the launcher": just scrolling its icons up & down?
[13:52] <Saviq> dandrader, no, swipe it in, tap the app for launcher to go away
[13:52] <Saviq> dandrader, rinse'n'repeat
[13:53] <dandrader> ok
[13:55] <dandrader> Saviq,  "citrain device-upgrade 30" <- is that a terminal command?
[13:56] <Saviq> dandrader, yes
[13:56] <Saviq> dandrader, from phablet-tools-citrain
[13:56] <Saviq> dandrader, and you need to pass your sudo password as the last argument
[14:06] <tsdgeos> mzanetti: there's a huge flickering on doing the "alt+tab" swipe on the spread
[14:06] <tsdgeos> not caused by your patch
[14:06] <tsdgeos> but i don't remeber it being there
[14:07] <tsdgeos> mzanetti: when switching between apps, not with the dash
[14:07] <tsdgeos> can you reproduce?
[14:07] <mzanetti> uh
[14:08] <mzanetti> tsdgeos: in trunk?
[14:08] <tsdgeos> mzanetti: and in rtm
[14:08] <mzanetti> indeed
[14:08] <mzanetti> feck
[14:08] <tsdgeos> no idea where this is coming from
[14:08] <tsdgeos> want a bug?
[14:08] <mzanetti> me neither atm
[14:09] <mzanetti> meh
[14:10] <mzanetti> not happening in ./run.sh
[14:12] <mzanetti> tsdgeos: might be the antialiasing thing
[14:12] <tsdgeos> mzanetti: but that's not on rtm yet
[14:12] <mzanetti> hmm, indeed
[14:15] <mzanetti> tsdgeos: seems it is
[14:15] <mzanetti> https://code.launchpad.net/~unity-team/unity8/rtm-14.09
[14:15] <Saviq> mzanetti, it's not
[14:16] <Saviq> mzanetti, it's in silo http://people.canonical.com/~platform/citrain_dashboard/#?distro=ubuntu-rtm&q=landing-004
[14:16] <sil2100> Trevinho, andyrock: hey guys! You around?
[14:16] <mzanetti> Saviq: ah ok... it seems commit 1933 does say it in the log though
[14:16] <mzanetti> err, 1399
[14:16] <tsdgeos> what's the correct syntax for this?
[14:16] <tsdgeos> ubuntu-device-flash touch --revision=150  --channel=ubuntu-touch/ubuntu-rtm/14.09-proposed
[14:17] <Saviq> mzanetti, the bug links are from the branch
[14:17] <sil2100> hmmm, on second thought, actuall it's on my side
[14:17] <Saviq> mzanetti, which does, now, contain the antialiasing
[14:17] <sil2100> Trevinho, andyrock: nvm!
[14:17] <Saviq> tsdgeos, --revision before touch
[14:17] <Trevinho> sil2100: ah ok :)
[14:17] <tsdgeos> meh, i was pretty sure i had trid that
[14:17] <sil2100> Trevinho: I'm working on fixing the changelog-bug issue you guys reported in citrain :)
[14:18] <Trevinho> sil2100: cool
[14:23] <tsdgeos> so revision 150 doesn't ahve that problem
[14:26] <sil2100> Trevinho: ok, so I have a really simple fix for that - would you mind if the 'Fixes: bug, bug, bug' string in the commit message would be changed to 'Fixes: #bug, #bug, #bug'? (i.e. # appended to the bug number)
[14:27] <Trevinho> sil2100: no, I think it's nicer
[14:27] <Trevinho> sil2100: adding LP: also?
[14:27] <sil2100> Well, it's not necessary, I would be a bit worried that it would 'bloat' the commit message
[14:28] <sil2100> Since it would mean the bigger list would be like this: "Fixes: LP: #123456, LP: #234567, LP: #345678"
[14:29] <sil2100> As you see, there are multiple ':' then in the commit-message, which looks a bit dirrty
[14:29] <Trevinho> sil2100: irc seems to like it :D
[14:29] <Trevinho> as you prefer btw, it's not a big deal
[14:29] <Saviq> tsdgeos, are you seeing this on rtm?
[14:30] <tsdgeos> Saviq: yes, that's why i'm bisecting
[14:30] <Saviq> tsdgeos, ugh :?
[14:30] <tsdgeos> afaik we have not changed much/at all in rtm
[14:30] <Saviq> https://launchpad.net/ubuntu-rtm/+source/unity8
[14:31] <Saviq> tsdgeos, thanks, let me know if I need to chase someone
[14:31] <tsdgeos> Saviq: i'll try to find out in which image it was introduced
[14:31] <tsdgeos> 157 is fine too
[14:32] <Saviq> tsdgeos, FWIW I was seeing that for a long time, but with the dash only
[14:32] <tsdgeos> it's a good thing this thing caches the iamges
[14:32] <Saviq> and only sometimes
[14:32] <tsdgeos> flashing is faster :)
[14:33] <tsdgeos> i fear i'll flash 166 (latest) again and it will be gone
[15:04] <tsdgeos> Saviq: it's ubuntu-app-launch
[15:04] <Saviq> huh
[15:04] <tsdgeos> yeah
[15:04] <Saviq> hmm
[15:05] <tsdgeos> well i went to 165
[15:05] <tsdgeos> works
[15:05] <tsdgeos> 166 fails
[15:05] <tsdgeos> went back to 165
[15:05] <tsdgeos> updated updated gir1.2-ubuntu-app-launch-2 libubuntu-app-launch2 ubuntu-app-launch ubuntu-app-launch-tools
[15:05] <tsdgeos> ad now fails
[15:05] <tsdgeos> let me revert them
[15:06] <Saviq> tedg, greyback, any idea how ual could cause apps flickering when focusing them?
[15:07] <tsdgeos> any idea where i can find the 0.4+14.10.20141031~rtm-0ubuntu1 packages of ual?
[15:07] <greyback> Saviq: flickering? As in visually?
[15:07] <Saviq> greyback, yes
[15:07] <greyback> do their contents flicker, like on a focus/unfocus/focus flicker?
[15:07] <Saviq> greyback, yes, goes to black
[15:07] <tedg> Trying to think what changed there…
[15:07] <tedg> Let me check
[15:08] <Saviq> tsdgeos, https://launchpad.net/ubuntu-rtm/14.09/+source/ubuntu-app-launch
[15:08] <Saviq> tsdgeos, https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu-rtm/landing-010/+build/6525799
[15:08] <Saviq> to be exact
[15:08] <tsdgeos> tx
[15:08] <tedg> So the change we made was to pause on the cgmanager connection.
[15:08] <greyback> Saviq: app suspend/resume timinging changed slightly maybe
[15:09] <dandrader> Saviq, reproduced the crash and think what the cause is
[15:09] <Saviq> greyback, that's what I was thinking, but would that cause a flicker?
[15:09] <tedg> So there could be other events happening that would cause a fliker while we're waiting?
[15:09] <Saviq> greyback, shouldn't we hold the last frame we had?
[15:10] <greyback> Saviq: if app was suspended before it got unfocus event, and then focused and then resumed, the unfocus & focus events might be stuck in a queue until the app resumed
[15:10] <greyback> then it reacts on unfocus, then focus
[15:10] <greyback> guess though
[15:11] <Saviq> greyback, and we stop rendering on unfocus (I mean apps do) due to the visibility event or whatnot?
[15:11] <greyback> Saviq: there are separate focus and visibility events
[15:12] <Saviq> greyback, yeah, unfocus shouldn't cause any real difference, visibility stops rendering though
[15:12] <greyback> Saviq: right, except if UI is designed to change on focus/unfocus
[15:12] <Saviq> greyback, but it's not like either should send up a blank frame?
[15:12] <Saviq> greyback, yeah, but that's different
[15:12] <greyback> Saviq: blank frame? No
[15:12] <greyback> ok, that wasn't clear to me
[15:12] <Saviq> greyback, exactly, so where are we getting the black frame from :/
[15:13] <greyback> Saviq: only thought is Mir
[15:13] <tsdgeos> totally
[15:13] <tsdgeos> reverting those packages makes the flickering away
[15:13] <tsdgeos> let's try with 166 and downgrading those now
[15:14] <greyback> tsdgeos: that's bizarre. Reverting ual fixed the black frame flicker?
[15:14] <tsdgeos> correct
[15:17] <greyback> freaky
[15:18] <greyback> I dunno how but perhaps we're pulling a buffer from Mir which wasn't rendered to by the client yet. Should not be possible though
[15:21] <dandrader> Saviq, dang it. now I cannot reproduce it anymore....
[15:21] <Saviq> dandrader, yeah, it's not an easy thing to repro, I was only able to do it on vivid + silo 30, even though silo 30 had nothing to do with anything related
[15:22] <dandrader> Saviq, yeah, I have vivid with silo 30 as well
[15:23] <Saviq> dandrader, what I usually did was launch xbmcremote (don't think app matters), and while it was launching I'd do "swipe launcher in" and "tap on spread" in quick succession
[15:23] <Saviq> s/spread/splash/
[15:23] <Saviq> dandrader, then, right swipe to spread, close the app, start from scratch
[15:23] <Saviq> dandrader, I'd usually crash it within 30s
[15:23] <dandrader> ok
[15:27] <tedg> So the only thing I could think of there is that we're ending up processing an idle event that you don't expect.
[15:27] <tedg> I think I could make it a custom context so no other events would process.
[15:27] <tsdgeos> tedg: greyback: i'm going to open a bug since it's very reproducible for me, you guys can't?
[15:27] <tedg> That might fix the issue.
[15:28] <tedg> tsdgeos, I'll read the bug, I'm not quite sure what I'm looking for yet.
[15:29] <greyback> tsdgeos: if you want, but will only investigate if tedg thinks it's qtmir's issue
[15:29] <tedg> Ah, on just quick right swipe?
[15:29] <Saviq> tedg, start two apps
[15:29] <tedg> If I do the long drag it doesn't seem to happen.
[15:29] <Saviq> tedg, short right swipe to toggle between apps
[15:29] <tsdgeos> tedg: yes quick swipe
[15:29] <Saviq> tedg, you get black
[15:30] <tedg> Ah, I see. I never do short swipe :-)
[15:33] <tsdgeos> tedg: https://bugs.launchpad.net/ubuntu/+source/ubuntu-app-launch/+bug/1394622
[15:45] <Cimi> mterry, can we reuse the same qsortfilterproxymodel from unity?
[15:46] <mterry> Cimi, I assume you're looking at wizard-import?  We do use the same in wizard-plugin.  wizard-import really is trying to make as little changes as possible, just to make reviewing the actual changes I do make in wizard-plugin reviewable
[15:48] <Cimi> mterry, mmm ok
[15:49] <mterry> Cimi, if I didn't do that, it would be impossible to see what I actually did.  wizard-import is real dumb -- I don't even think the wizard works right if you install it
[16:11] <tsdgeos> Saviq: i think the high cpu usage was basically the busy indicator
[16:11] <tsdgeos> just that as it was white
[16:11] <tsdgeos> i wasn't seeing it
[16:11] <Saviq> tsdgeos, ok, will keep an eye out
[16:12]  * Saviq takes an eye out and puts it on a table
[16:13] <tsdgeos> be careful it doesn't fall off
[16:13] <tsdgeos> the table i mean
[17:02] <michael_ptasz> is anyone on?
[17:05] <mzanetti> Saviq: can you think of any explanation why this would happen? https://bugs.launchpad.net/ubuntu/+source/ubuntu-app-launch/+bug/1394622
[17:05]  * mzanetti totally lost how UAL could influence this
[17:07] <greyback> mzanetti: a crazy guess: the call to focusApplication might now be blocking on the UAL resume call, and so take slightly  longer to switch the app to the front
[17:07] <mzanetti> hmm... a good one
[17:08] <mzanetti> but would a blocking call actually render black frames?
[17:08] <mzanetti> in which case the issue sounds like a bigger one
[17:12] <greyback> mzanetti: it's not a black frame IMO, it's like both foreground & background app are invisible, so the black is the Stage background
[17:14] <mzanetti> oh...
[18:19] <Saviq> greyback, yeah, that could make sense, but that would still mean we're swapping for a blank frame for some reason, and IIUC we should never do that?
[18:19] <Saviq> and apps should not have alpha, btw, so should never be transparent
[18:19] <Saviq> and for what else reason should we have both apps visible: false
[18:19] <Saviq> ?
[18:20] <Saviq> when it's not visible: false to start with (unless the one we pull in from the right edge is a copy?)
[18:21] <greyback> Saviq: apps *can* have alpha ATM, but that's not the issue.
[18:22] <greyback> I suspect it's just a glitch in our animations
[18:23] <greyback> it's not a blank frame from the application, more it set invisible for a frame somehow
[18:26] <Saviq> greyback, mhm
[18:27] <greyback> is guess tho
[18:27] <greyback> I've not tried to dig
[18:29] <Saviq> kk
[20:46] <balloons> bregma, who did the work on unity8-lxc container?
[20:48] <ChrisTownsend> balloons: Me and stgraber.
[20:48] <balloons> ChrisTownsend, brillant. Can I pick your brain a bit on it then?
[20:48] <ChrisTownsend> balloons: Sure!
[20:49] <balloons> So for the unity8-lxc-setup, I would like to point the setup at an existing iso to save myself the download. In addition, it didn't *seem* like using --redownload would use zsync to save my bandwidth
[20:50] <balloons> to workaround this I manually placed the file in the expected folder.
[20:50] <ChrisTownsend> balloons: Right, that area needs some work.
[20:51] <balloons> once complete I assume I just run unity8-lxc-session yes? it's complaing then unity8-lxc doesn't exist. And indeed lxc-ls shows me nothing
[20:51] <ChrisTownsend> balloons: So yes, manually copying the ISO to the expected place will work.  Ans yes, using something like zsync or rsync if the existing iso is there would be much better.
[20:51] <ChrisTownsend> balloons: Right, after package install, you need to run 'sudo unity8-lxc-setup'.
[20:53] <ChrisTownsend> balloons: Since the ISO is there, you may need to do 'sudo unity8-lxc-setup --rebuild-all'.
[20:53] <balloons> trying again..
[20:54] <ChrisTownsend> balloons: I've never tried it with an ISO already existing and no container set up.
[20:54] <balloons> ChrisTownsend, oh.. hmm, it's under root only
[20:55] <balloons> I see it with sudo lxc-ls only
[20:55] <ChrisTownsend> balloons: Yes, since it's a full session.
[20:55] <ChrisTownsend> balloons: Needs root privileges to be started from lightdm.
[20:56] <balloons> ChrisTownsend, so do you think running under MIR is possible using the container?
[20:56] <balloons> also, I was curious about what the best way to update would be. We can re-pull the image, or run update-lxc, or even apt-get inside the container
[20:57] <ChrisTownsend> balloons: The container is using using Mir and Unity8.  Once you have it setup, you log into it using LightDM.  Select the Unity 8 LXC session.
[20:58] <ChrisTownsend> balloons: You can either redownload the latest ISO or use 'sudo unity8-lxc-setup --update-lxc' to update the packages in the container.
[20:58] <ChrisTownsend> balloons: And yes, you can attach to the container and do apt-get inside.
[20:59] <balloons> right, just wondering what I should propagate as the best practice.
[21:01] <ChrisTownsend> balloons: I say either the update-lxc method or if someone wants the whole new image, then use --redownload.  The thing about getting a new image is if new packages are included for demo purposes or new functionality that simply updating won't pick up.
[21:01] <balloons> anyways, I'm not getting a running session ;-( http://paste.ubuntu.com/9131705/
[21:02] <ChrisTownsend> balloons: You're logging in through lightdm?
[21:02] <balloons> ChrisTownsend, right. mm.. but that would only be true of the image I suppose right? if you add it through the meta-package should be fine
[21:02] <balloons> ChrisTownsend, ohh right, my bad
[21:04] <ChrisTownsend> balloons: Well, one thing that comes to mind is there is the Gtk+ work going on and they may add Gtk apps to the ISO at some point.  Just updating the container won't pick that up.
[21:04] <ChrisTownsend> balloons: But yeah, if there is some metapackage on the ISO that adds those dependencies, then yeah, it should work.
[21:05] <balloons> so probably should get zsync going then push pulling from image :-)
[21:06] <ChrisTownsend> balloons: Yeah, I'll enter a bug report and work on that.
[21:08] <balloons> ChrisTownsend, ok, so from lightdm anyway it attempts to launch then just reloads lightdm
[21:09] <ChrisTownsend> balloons: Hmm, ok.  What version of Ubuntu is the host and are you using a real user, aka, not Guest?
[21:10] <balloons> ChrisTownsend, ahh that was the next question. Vivid, and yes a guest session
[21:10] <balloons> mir I suppose isn't happy with guest sessions?
[21:11] <ChrisTownsend> balloons: Ok, I don't think the guest session works with Unity8 yet.
[21:11] <ChrisTownsend> balloons: I create some test user for testing that stuff.
[21:12] <balloons> good to know.. I'll be updating the wiki with my notes after this as well to fill in some of these gaps
[21:12] <ChrisTownsend> balloons: And once you log in, you can't log out back to lightdm.  It's a known issue in Unity 8.
[21:13] <balloons> do we have bugs for either of these issues, just for note puroposes
[21:13] <ChrisTownsend> balloons: Cool, thanks for the Wiki help.  Hopefully what I have there is a good start, but more is better and also I have a bias on how it works and probably make unconscious assumptions.
[21:14] <ChrisTownsend> balloons: I'm not sure about the guest support, but for the logout, yes, we do.  Lemme get you the bug #.
[21:14] <ChrisTownsend> balloons: Bug #1360316 for the logout issue.
[21:14] <balloons> ChrisTownsend, certainly heh. The goal is to push some new community folks at this. Having this container is exciting
[21:15] <ChrisTownsend> balloons: It is pretty cool.  Thanks to stgraber for getting this started.
[21:17] <balloons> ohh, fast user switching did work.. that's nice
[21:17] <ChrisTownsend> balloons: I just tried a guest session on a non-container Unity 8 session and it just sits there at a black screen.  At least with the container, it kicks you back to lightdm so you can do something.
[21:22] <balloons> ChrisTownsend, ok so how do I exit?
[21:22] <balloons> as you said, no way back to lightdm easily
[21:23] <balloons> I seem to losse vt's also
[21:23] <ChrisTownsend> balloons: Um, well, uh...ssh into the machine and stop the container.
[21:23] <ChrisTownsend> balloons: Or reboot.
[21:23] <ChrisTownsend> balloons: And the VT's is a known issue.  I'll get you that bug as well.
[21:24]  * balloons wonders if he re-setup ssh on his box
[21:25] <ChrisTownsend> balloons: Bug #1383497 is for the VT issue.  It's a container thing at this point and needs to be investigated further.
[21:25] <balloons> currently I'm actually a bit stuck. using logout, restart, shutdown just returns me to unity8 and I can't login anymore :-(
[21:26] <balloons> is there a reason I can't login anymore from the unity8 lockscreen?
[21:26] <balloons> not that I set a password for the user
[21:26] <balloons> I mean, from the touch side
[21:27] <ChrisTownsend> balloons: Umm, this is a little beyond my expertise in that this is a Unity 8 issue.
[21:28] <balloons> sure, you've just been on such a roll :p
[21:29] <ChrisTownsend> lol, yeah
[21:35] <balloons> bregma, are you about?
[21:36] <balloons> ofc, I guess I could ask ChrisTownsend as well.. test his skills :-) I was curious what the roadmap, timeline, dev practices are for delivering new features. It would be nice to have some notes to accompany new stuff flowing into the image and help people keep pace with what's going on
[21:38] <ChrisTownsend> balloons: Are you talking about the Unity 8 desktop proper or features and such for just the container part?
[21:39] <balloons> ChrisTownsend, unity8 proper. I'm wondering what the best way to communicate what's happening, particularly with the focus on people using it
[21:39] <ChrisTownsend> balloons: I'd like to know the answers to those questions as well;-)
[21:39] <bregma> indeed
[21:40] <balloons> for the container, I just *KNOW* you'll be fixing my bug right away
[21:40] <ChrisTownsend> balloons: zsync?
[21:40] <balloons> ChrisTownsend, yea
[21:41]  * balloons doesn't hide his sarcasm
[21:41] <ChrisTownsend> balloons: I'm on it!
[21:41] <balloons> <#
[21:41] <balloons> <3
[21:41] <bregma> frome a vague standpoint, basic window management is coming in the next few weeks, and design will be addressing desktop-oriented issues in the next few months, but that's about as exact as we're going to get until the current phone-oriented crush is past
[21:41] <bregma> those guys have my sympathy
[21:42] <bregma> but I wish they had my software working
[21:44] <balloons> well mzanetti was showing off something more up to date than what I saw I think yesterday
[21:45]  * balloons re-reads his own sentence, is confused by his resounding command of english,  and realizes the day must be getting long
[21:46] <mzanetti> ?
[21:46] <bregma> yeah, saw that
[21:47] <mzanetti> bregma: what sofware?
[21:47] <bregma> unity8 on desktop with what looked like windows
[21:47] <balloons> hey mzanetti I was just trying to refer to your video.. ^^ exactly
[21:48] <mzanetti> yeah, I'm working on that now mostly
[21:48] <mzanetti> today I added support for a wallpaper and made the launcher showing up correctly (as opposed to the upside-down thing on the phone)
[21:49] <mzanetti> next steps is to make the mouse behave more mousy instead of touch
[21:50] <ChrisTownsend> mzanetti: I don't suppose you have that code someplace where we can try it, do you?
[21:51] <mzanetti> ChrisTownsend: https://code.launchpad.net/~mzanetti/unity8/wallpaper
[21:51] <mzanetti> should build fine with debuild. just install unity8, unity8-common and unity8-private from those packages and run it with unity8-desktop-session-mir
[21:52] <ChrisTownsend> mzanetti: Cool!  That includes the windowing stuff too?
[21:52] <mzanetti> yes
[21:52] <ChrisTownsend> mzanetti: Thank you!
[21:52] <balloons> ohh slick. mzanetti will you land that in rtm or vivid? or both?
[21:52] <mzanetti> ChrisTownsend: but keep in mind this is *really* early stages
[21:52] <ChrisTownsend> mzanetti: Understood
[21:53] <mzanetti> balloons: well, for now I'm trying to land it in vivid. not sure what will propagate through rtm, that's really not the focus for this
[21:53] <balloons> mzanetti, I ask because it seems some stuff is going to rtm first -- is that right?
[21:54] <mzanetti> balloons: no, we land into vivid and cherry-pick phone related fixes into rtm
[21:54] <mzanetti> maybe a sync of trunk -> rtm will happen at some point, but really not sure when and how
[21:54] <balloons> mzanetti, that would be my expectation. The differences then I'm seeing must simply be the default setup (scopes, etc)
[21:56] <mzanetti> balloons: probably, yes. definitely not for unity8
[21:57] <mzanetti> ChrisTownsend: ah right, you need to "gsettings set com.canonical.Unity8 usage-mode Windowed" for now, as long as we don't have all the context recognition in place
[21:58] <ChrisTownsend> mzanetti: Ok, thanks for letting me know.
[21:58] <mzanetti> have fun :)
[21:59] <ChrisTownsend> mzanetti: I will:)
[22:00] <balloons> mzanetti, I notice the keyboard doesn't in scopes
[22:00] <balloons> is that known?
[22:01] <mzanetti> balloons: more context please :)
[22:01] <mzanetti> on the phone?
[22:01] <mzanetti> or the desktop?
[22:01] <balloons> mzanetti, sorry, desktop
[22:01] <mzanetti> hmm, really?
[22:01]  * mzanetti tries
[22:02] <balloons> mzanetti, doesn't seem to work for me.. the store and searching on scopes for music or video don't work
[22:02] <mzanetti> balloons: indeed... works in system-settings but not in scopes
[22:03]  * balloons can file
[22:03] <balloons> if you'd like
[22:03] <mzanetti> balloons: ah.. it does work, but you need to focus something else and then back to scopes
[22:03] <mzanetti> balloons: probably too early to file focus bugs :)
[22:04] <mzanetti> I'm sure there's an uncountable amount of issues still
[22:06] <mzanetti> balloons: know what'd be awesome, if you could get some community guys to package up the terminal-app and dekko for amd64
[22:08] <balloons> mzanetti, that I can work on. I actually went looking for file manager and terminal first thing
[22:08] <mzanetti> yeah, not in there if not armhf :/
[22:09] <balloons> mzanetti, yep, indeed