[03:28] <thomi> Could someome please take a look at the failures happening here when they get a chance? https://jenkins.qa.ubuntu.com/job/autopilot-trusty-amd64-ci/2/console
[06:35]  * Mirv the critical bug filer
[06:44] <didrocks> :)
[06:44] <didrocks> Mirv: karma!
[06:45] <Mirv> that indeed
[06:46] <Mirv> ok the stack status / critical bugs start now to be up-to-date reflecting the real situation of stacks. I'm still running apps tests, and waiting for info if it's known that some unity8 tests fail on desktop.
[06:47] <Mirv> some config updates, two new critical bugs this morning for test failures
[06:48] <Mirv> oh, three bugs, that is
[06:48] <Mirv> forgot about indicators
[06:55] <didrocks> nice to get that cleaned, thanks for tracking Mirv :)
[07:12] <Mirv> yw, now unity8 bugged too, and apps sharing the sdk bug
[07:27] <vila> didrocks, Mirv : Let me start by some clarification: thanks a lot for your help yesterday, I couldn't have reached the point where bug #1244324 were filed without you.
[07:27] <didrocks> glamor-egl? I'm not even aware about that one
[07:28] <Mirv> interesting, it's now the default on some radeons?
[07:28] <didrocks>  Glamor is a library for accelerating 2D graphics using GL functions.
[07:28] <Mirv> I thought it was mainly only the newest ones like Radeon 7000
[07:28] <didrocks> hum…
[07:28] <vila> didrocks, Mirv: My fresh understanding is that it's the only option mlankhorst needed to reproduce locally even on a different hardware
[07:29] <Mirv> since AMD hasn't worked on normal 2d acceleration support anymore on those newest ones
[07:29] <didrocks> vila: do you know if all ATI hw are using it or we just got "lucky" :)
[07:29] <didrocks> Mirv: so, it's a layer on top of gles?
[07:29] <vila> didrocks: no and mlankhorst is EOW
[07:29] <Mirv> didrocks: it bypasses the need to code EXA acceleration when new cards come in
[07:29] <vila> Mirv: but yeah, I *think* that's to provide 3D
[07:29] <didrocks> interesting
[07:30] <Mirv> vila: no, it's to provide 2D acceleration with cards that don't have it otherwise
[07:30] <Mirv> vila: which Radeon the machine was running with again?
[07:30] <didrocks> Mirv: so, it should be for underperforming cards, right?
[07:30] <vila> 7750 (as in qa-radeon-7750 ;)
[07:30] <Mirv> "GLAMOR provides 2D acceleration on the Radeon HD 7000 "Southern Islands" GPUs and newer Radeon GPUs"
[07:30] <Mirv> vila: oh, right, it's one of the newest ones, right
[07:30] <vila> Mirv: where do you get that ?
[07:30] <Mirv> vila: http://www.phoronix.com/scan.php?page=news_item&px=MTQ2MzA
[07:31] <vila> Mirv: here we go ! *Now* the knowledge spreads :)
[07:31] <Mirv> didrocks: no, for new cards that AMD/community doesn't have resources (or otherwise deems better) to build a separate 2D accel support
[07:31] <didrocks> Mirv: ah, ok, starting to make sense :)
[07:32] <Mirv> so it turns out that glamor is not yet bug free
[07:33] <didrocks> unsurprinsingly ;)
[07:33] <vila> Mirv: and you googled (^Wduckduckgo'ed) that page with which keywords ? ;)
[07:33] <Mirv> vila: glamor default
[07:34] <vila> Mirv: thanks, will put that in the incident report when I get there \o/
[07:35] <vila> Mirv: indeed ranked 4 in https://duckduckgo.com/?q=glamor%20default&kl=us-en&kp=-1 well done !
[07:41] <sil2100> didrocks: hi! Did you have a discussion about the AP switch to 1.3 with thomi in the end? Since when he started his day he was unaware of anything
[07:42] <sil2100> Like, with the possibility of switching back lp:autopilot to /1.3
[07:44] <didrocks> sil2100: it was on IRC, he was pinged for it
[07:44] <didrocks> sil2100: hey! btw ;)
[07:44] <didrocks> sil2100: so I thought he got the info
[07:45] <didrocks> sil2100: did you saw my email?
[07:45] <sil2100> didrocks: yes yes, that's also why I'm asking - since we might want to chat with thomi about the possibility of doing that switch again
[07:46] <sil2100> With all the proper explaination to him 'why' ;)
[07:46] <didrocks> sil2100: I'm clearly not happy that thomi turned lp:autopilot to point to 1.4 with having a backward compatbility issue (that's not sustainable) and that they did that without converting the app AP tests
[07:46] <didrocks> sil2100: that's what we explained on IRC, if people stay on IRC to get pings but don't scrollback…
[07:46] <sil2100> My understanding in the past was that it was being worked on
[07:46] <sil2100> But it seemed that it didn't...
[07:46] <didrocks> sil2100: yeah, but not turning lp:autopilot to the incompatible version *beforehand*
[07:47] <didrocks> I wonder if management won't force autopilot to be backward compatible anyway
[07:47] <didrocks> like the sdk, we shipped a version to the community
[07:47] <didrocks> it's time to ensure our API is stable
[07:47] <didrocks> or at least having good practices, like a deprecation period
[07:47] <sil2100> didrocks: true, I guess with crucial components like this I think we need to provide a more-or-less stable API even between versions
[07:47] <sil2100> Right
[07:48] <didrocks> seems we have neither of them :(
[07:48] <didrocks> sil2100: anyway, otherwise, how went your "hunt for getting all stack yellow"?
[07:48] <ogra_> hrm
[07:48] <didrocks> I think AP 1.4 was part of it?
[07:48]  * ogra_ sees no landing sheet entry for mterry's seed change 
[07:48] <didrocks> sil2100: btw, you were pinged on this 1.3 switch yesterday as well, but it seems you were not there for 2 hours (and during the meeting)
[07:49] <ogra_> nor do i see it being discussed in any backlog
[07:49] <didrocks> ogra_: ah, I was wondering if he did that directly with you
[07:49] <ogra_> nope
[07:49] <sil2100> didrocks: yes, along with psivaa we were trying to find the problems of the AP tests and this was the main culprit of most issues we were encountering back then
[07:49] <didrocks> sil2100: so, all should be yellow now?
[07:49] <ogra_> and it sounds a little harmful to drop qtaudioengine
[07:50] <didrocks> ogra_: yeah, I thought you were involved, so knew about it ;)
[07:50] <ogra_> no, not at all
[07:50] <sil2100> didrocks: sorry about that, been AFK after a late lunch and missed the time :/
[07:50] <didrocks> ogra_: it's from sdk-libs, so it's still installed by the touch meta package?
[07:50] <ogra_> yep
[07:50] <sil2100> didrocks: we have normal AP issues now mostly
[07:50] <didrocks> sil2100: ok, so you are tracking that with upstream to get those fixed?
[07:51] <didrocks> ogra_: so not *that* urgent, we can have a discussion with mterry if you want
[07:51] <ogra_> didrocks, yeah, i just dont feel like rolling an image before we did that
[07:52] <didrocks> ogra_: sorry, I probably didn't get it then. sdk-libs is just when you want to install the sdk to develop, right?
[07:52] <sil2100> didrocks: yes, we're basically waiting for the right people to pop up
[07:52] <didrocks> or ubuntu-touch -> deps on sdk-libs -> which deps on the audio stuff?
[07:53] <didrocks> sil2100: ok, do you see landings we can proceed meanwhile? (from the landing ask)
[07:53] <didrocks> sil2100: feel free to add them (just not target an image yet, we'll discuss that in the meeting)
[07:57] <ogra_> didrocks, no, thats sdk-libs-dev
[07:57] <didrocks> oh ok
[07:57] <ogra_> sdk-libs actually carries our API
[07:58] <didrocks> ok, so we can't really redo without knowing the impact
[07:58] <didrocks> nothing in the seed?
[07:58]  * didrocks bzr checkout lp:~ubuntu-core-dev/ubuntu-seeds/ubuntu-touch.trusty
[07:59] <didrocks>   Drop QtAudioEngine plugin from sdk-libs, it is currently unused and will pull a lot of stuff into main
[07:59] <didrocks> - * qtdeclarative5-qtaudioengine-plugin
[07:59] <didrocks> Mirv: any opinion? ^
[07:59] <didrocks> do we use another audio system for qml?
[08:00] <didrocks> like maybe   * qtdeclarative5-qtmultimedia-plugin
[08:03] <ogra_> we use pulse
[08:03] <ogra_> but not sure through which channel
[08:03] <ogra_> for Ot apps this might still be needed to talk to pulse
[08:07]  * didrocks smells we use it through qtmultimedia for qml apps
[08:12] <Mirv> didrocks: yes so zoltan + me approved removing it for now, there is other support but game makers may want positional audio so it should be a goal to add it back (+ openal & co. to main) at some point
[08:13] <didrocks> Mirv: ok, but droping for now is fine, nothing is using it?
[08:13] <didrocks> ogra_: ^
[08:13] <ogra_> Mirv, and this has been tested with apps from the store on the phones ?
[08:14] <ogra_> (not sure how many we ship by default that would make use of this)
[08:14] <Mirv> didrocks: ogra_: a few weeks ago it was checked that no packages were found that would declare a dependency on it (I grepped the whole file system), but I did not have and every store app installed
[08:15] <ogra_> Mirv, all apps using the sdk should just depend on sdk-libs ... which in turn will hide the dep
[08:15] <didrocks> the issue is that on saucy we dep on it
[08:15] <ogra_> my concern is that we are changing the API
[08:15] <Mirv> ogra_: yeah, that's why I used some grep of the whole filesystem to find mentions of it
[08:15] <ogra_> without *any* tests or even discussion
[08:15] <didrocks> so we are potentially telling people "please use it"
[08:15] <didrocks> and now we are dropping the functionality from v2 of the sdk
[08:16] <ogra_> Mirv, wasnt the agreement that we just drop the harmful pieces from qtdeclarative5-qtmultimedia-plugin ?
[08:16] <Mirv> actually the root of this was to get cordova to main for saucy, which is why the temporary drop was wanted. they're still driving it, though.
[08:16] <Mirv> ogra_: this is it? dropping the audioengine package from qtmultimedia
[08:16] <sil2100> thomi: ping
[08:17] <Mirv> if I could decide I'd keep it and get the dependencies to main for trusty.
[08:17] <ogra_> Mirv, as i understood the audioengine plugin weas still wanted just without the harmful lib as dep
[08:17] <Mirv> ogra_: ok, well openal is a strict dependency to the audioengine, just not to the normal audio/multimedia support
[08:18] <Mirv> it's not like "the" Qt audio engine, it's a 3d positional audio engine
[08:18] <ogra_> Mirv, so what was the patch that i saw that just removed openal from this package in saucy ?
[08:18] <ogra_> without completely removing it
[08:19]  * ogra_ tries to find the bug again ... i know there was a finer grained change than bluntly removing the whole plugin
[08:19] <ogra_> (wasnt that even written by you  ?!?)
[08:19] <Mirv> ogra_: I haven't seen such. it doesn't build the qtaudioengine if you drop the dependency.
[08:23] <ogra_> ah
[08:23] <ogra_> https://code.launchpad.net/~timo-jyrinki/ubuntu/trusty/qtmultimedia-opensource-src/drop_qtaudioengine
[08:23] <ogra_> that was the one i remembered
[08:23] <Mirv> bug #1206268 is the main bug
[08:23] <ogra_> and indeed you are right, no code changes
[08:23] <ogra_> but again, we have a hard defined ABI that lives in sdk-libs
[08:23] <ogra_> and this API was just changed without discussion or warning
[08:24] <ogra_> i agree that it is unlikely that someone uses positional audio in touch games
[08:24] <ogra_> nontheless we guarantee API stability and such changes need a broad discussion imho
[08:24] <Mirv> it wasn't really changed, of course, because it has been blocked. I've just stated that from my part it's ok.
[08:25] <ogra_> right
[08:25] <Mirv> I think the real issue here is that there has been no-one driving that MIR really
[08:25] <Mirv> I've helped when I've been asked to
[08:25] <ogra_> well
[08:25] <Mirv> like, no "one" driving
[08:25] <ogra_> i saw someone driving it ... and messing it up several times now
[08:26] <ogra_> which makes me cautious about changes caused by it
[08:26] <ogra_> lets keep that for the meeting, i dont really know what to do here
[08:27] <ogra_> while it is technically right to drop it, the way it was dropped was definitely as wroing as it could be in the light that we promote API stability
[08:27] <Mirv> I believe a MIR like that should be primarily driven by the team behind it, ie webapps team? but then in practice they've asked robru to do it.
[08:27]  * ogra_ gets coffee before the meeting, brb
[08:27] <Mirv> and then there have been zoltan, mterry, me, xnox striking at it too
[08:31] <didrocks> sil2100: Mirv: ogra_: coming?
[08:31] <Mirv> yep
[08:32] <sil2100> Coming
[08:35] <asac> omg my noternet starts again
[08:36] <asac> cyphermox: i really need this threshold fix
[08:36] <asac> cyphermox: wpa is on a 35/70 quality AP while i have one with 70 right next to it
[08:37] <vila> asac: you broke
[08:37] <ogra_> asac, youre not moving anymore OMG !
[08:38] <asac> will be back in 5 minutes
[08:38] <ogra_> double vision !
[08:38] <asac> i should be bak
[08:38] <ogra_> yep
[08:51] <sil2100> Mirv: can you take a look here? https://code.launchpad.net/~sil2100/cupstream2distro-config/trusty_fixes_projects/+merge/192602
[08:51] <sil2100> :)
[08:51] <sil2100> Mirv: since we decided to use this for now
[08:56] <Mirv> sil2100: I had already disabled u-s-c a few hours ago
[08:57] <Mirv> approved otherwise
[09:06] <sil2100> Mirv: yes, I fixed that ;) Thanks
[09:08] <sil2100> I need to get my phone updated to trusty I guess
[09:08] <sil2100> ogra_: how can I get trusty on my phone?
[09:10] <sil2100> AH, ok, channel trusty
[09:13] <xnox> Mirv: your patches look fine =) but web-browser app has been attempted to be seeded too early (before qtmultimedia was done).
[09:14] <didrocks> sil2100: trusty-proposed please :)
[09:14] <sil2100> Using that actually ;)
[09:14] <didrocks> you want the latest of latest!
[09:14] <didrocks> the best of the best!
[09:14] <didrocks> :)
[09:14] <ogra_> sil2100, adb shell system-image-cli -c trusty -b 0
[09:14] <didrocks> ogra_: that do work now?
[09:14] <ogra_> or even trusty-proposed :)
[09:14] <didrocks> did you get why it wasn't working for you yesterday?
[09:15] <ogra_> didrocks, -b 0 ...
[09:15]  * didrocks uses man
[09:15] <xnox> Mirv: ogra_: re: api/abi stability our 13.10 sdk is unchanged ;-)
[09:15] <ogra_> it only does the upüdate if i tell it it is on revision 0 of the new channel
[09:15] <ogra_> xnox, how do you mean ?
[09:15] <didrocks> ogra_: oh ok
[09:16] <didrocks> so we still expect to have a higher version
[09:16] <xnox> ogra_: well we are not going retrospecivily drop qtaudioengine in 13.10, it's a change in 14.04 sdk and up. And well we can easily scan current clicks in the store about it, and add review warning if somebody does use it.
[09:16] <ogra_> didrocks, it automatically uses the latest image in the new channel
[09:17] <ogra_> xnox, we are also not dropping anything from the touch seeds (and the API) that wasnt even publically discussed
[09:18] <ogra_> and had no entry on the landing plan ... and had not a single AP test etc
[09:18] <ogra_> xnox, i'm in the middle of reverting it now and will send a mail to -devel and -phone
[09:18] <didrocks> ogra_: why do we need -b0 then, it's not because we have image 100 on the other channel and 5 < 100?
[09:19] <ogra_> didrocks, yeah, i think thats the reason, barry would know more
[09:20] <didrocks> ok :)
[09:21] <vila> didrocks: waiting for your ping ;)
[09:21] <didrocks> vila: yeah, in 3 minutes :)
[09:21]  * ogra_ wonders why msm is in the next meeting
[09:22]  * ogra_ gets more coffee
[09:22] <vila> ogra_: she organized it may be ;)
[09:22] <ogra_> ah
[09:22] <vila> didrocks: cool !
[09:22] <ogra_> well, she is listed as participant :)
[09:22] <didrocks> ogra_: no coffee for you! you have 8 minutes to promote the image :p
[09:23] <didrocks> omg I pinged ogra_ too late
[09:23] <didrocks> he has already left for coffee
[09:23] <didrocks> we're screwed :)
[09:23] <Mirv> noooooo
[09:23] <didrocks> heh
[09:23] <Mirv> didrocks: two packaging acks, to get trusty landings started. those also consist of the whole diff of what's in the changes. http://10.97.0.1:8080/view/cu2d/view/Head/view/WebCreds/job/cu2d-webcred-head-3.0publish/lastSuccessfulBuild/artifact/packaging_changes_signon-ui_0.15+14.04.20131024.2-0ubuntu1.diff
[09:24] <Mirv> + http://10.97.0.1:8080/view/cu2d/view/Head/view/Friends/job/cu2d-friends-head-3.0publish/lastSuccessfulBuild/artifact/packaging_changes_libfriends_0.1.2+14.04.20131024.2-0ubuntu1.diff
[09:24] <didrocks> waow, my 2 first packaging ack for this cycle, we should celebrate :)
[09:24] <Mirv> yep :)
[09:24] <didrocks> Mirv: btw, I think it's time for you to move on the upload access side
[09:24] <didrocks> let me check that the packaging fixes are fine first :)
[09:25] <didrocks> Mirv: -IT_PROG_INTLTOOL([0.40])
[09:25] <didrocks> +IT_PROG_INTLTOOL(0.50.0)
[09:25] <Mirv> didrocks: yeah I already started btw https://wiki.ubuntu.com/TimoJyrinki/DeveloperApplication-PPU as well during saucy but I've been too exhausted to hunt for approvals
[09:25] <didrocks> it could have bumped the build-dep in debian/control on intltool
[09:25] <didrocks> Mirv: ah, great! let's discuss that next week
[09:25] <xnox> ogra_: reverting what? nothing was uploaded as far as I can tell, or changed.
[09:25] <Mirv> well I asked seb and he promised but never did ;)
[09:25] <alan_g> does anyone know what causes this? https://jenkins.qa.ubuntu.com/job/mir-android-trusty-i386-build/2/console (it doesn't seem to have anything to do with the code changes)
[09:26] <didrocks> Mirv: can you remind me about this?
[09:26] <didrocks> Mirv: not a blocker for intltool, but nice if you can then merge a fix for it
[09:26] <didrocks> Mirv: oh, signon builds on all archs? nice. +1 then :)
[09:26] <vila> Mirv: Can I haz a button after: '/!\ Remember that this diff only represents packaging changes and build tools diff, not the whole content diff!' :-p
[09:26] <Mirv> didrocks: remind next week? yes.
[09:26] <didrocks> please publish both
[09:26] <Mirv> didrocks: the libfriends only specified intltool in general as a dependency, but I can add a version
[09:27] <didrocks> Mirv: would be nice!
[09:27] <xnox> ogra_: and well *crap* happens, and we do probably want webrowser-app on the desktop / converged story. It is unfortunate that it pulled in crap in-advertadly, it would have been spotted earlier if our sdk was fully in main.
[09:27] <Mirv> ok
[09:29] <Mirv> didrocks: I don't think the signon-ui really builds on all archs (needs qtdeclarative), but the policy has been to remove those tri-arch defines. also I guess with Qt 5.2 and removal of V8 JavaScript engine they might start to work as well.
[09:30] <didrocks> Mirv: if you are sure it will move from PROPOSED, I'm fine :)
[09:32] <Mirv> didrocks: it should, there has been now powerpc/arm64 releases in trusty, and the powerpc one in saucy was removed in August
[09:32] <didrocks> Mirv: ok, let's go then!
[09:35] <vila> cjwatson: how should I ask to get https://cloud-images.ubuntu.com/query/trusty/server/daily-dl.current.txt non 404'ing ?
[09:35] <vila> meh s/how/who./
[09:37] <vila> cjwatson: alternatively, ~same question to have trusty in http://cdimage.ubuntu.com/ubuntu-server/daily/current/ ?
[09:47] <cjwatson> vila: me and I'll be working on it today
[09:47] <cjwatson> vila: only for the latter though, can't do the cloud images
[09:47] <vila> cjwatson: great ! thanks !
[09:47] <cjwatson> try smoser or utlemming for those
[09:48] <vila> cjwatson: right, US tzs for both so will have to wait
[09:48] <vila> cjwatson: perfect answer, thanks
[09:49] <cjwatson> didrocks: we're getting back to the point where if things don't migrate I might actually notice ...
[09:51]  * vila &
[09:59] <cjwatson> ubuntu-keyboard needs to be rebuilt against libpinyin4
[10:17] <didrocks> cjwatson: yeah, I don't want you to have to suffer from us though
[10:17] <cjwatson> I won't
[10:18] <didrocks> cjwatson: we have an ubuntu-keyboard in progress, so that can maybe catch it
[10:18] <cjwatson> touch's contribution to stuff stuck in -proposed is a drop in the ocean
[10:18] <didrocks> yeah, I imagine :) just don't want to add *more* to you ;)
[10:18] <ogra_> heh
[10:22] <didrocks> sil2100: ubuntu-keyboard going fine? (see ^)
[10:23] <ogra_> didrocks, asac, popey, calls and sms tested on maguro for image #5, looks fine, but we need someone to mail avengers while popey is out
[10:23] <didrocks> ogra_: can you do it? Do you know what's popey's procedure?
[10:23] <ogra_> didrocks, no
[10:23] <ogra_> thats why i mentioned it :)
[10:24] <didrocks> maybe asac will know :)
[10:24] <ogra_> avengers is a members only ML
[10:25]  * ogra_ is afk for a bit
[10:29] <asac> didrocks: why dont you become part of the list and do the announces?
[10:29] <didrocks> asac: I'm all in for that, but I even don't know what the avengers team is
[10:29] <didrocks> https://launchpad.net/~avengers doesn't seem to be that one
[10:29] <didrocks> neither https://launchpad.net/~avenger
[10:32] <asac> didrocks: https://launchpad.net/~ubuntu-avengers
[10:32] <asac> :-P
[10:32]  * asac has a good awesome bar :)
[10:33] <didrocks> asac: it's an invite only list, I need then Jono or Julien to add me
[10:48] <didrocks> asac: so, meanwhile, can you send that email so that we can promote the image?
[10:50]  * didrocks hopes sil2100 published before the rerun of all stacks ubuntu-keyboard
[10:56] <didrocks> sil2100: did you get an early lunch this time? :)
[11:12] <asac> didrocks: ok sending
[11:12] <asac> didrocks: was the image tested on mako at all?
[11:13] <asac> (with popey being gone)
[11:13] <didrocks> asac: yeah, popey did the testing this morning early
[11:13] <didrocks> (mentionned during the meeting)
[11:13] <didrocks> ogra_: please promote!
[11:16] <didrocks> Mirv: dialer-app -> fine for publishing?
[11:17] <didrocks> (same question for content-hub)
[11:18] <asac> didrocks: done
[11:18] <didrocks> thanks asac ;)
[11:19] <asac> do we have a 2 factor app for ubuntu phone in store?
[11:19] <asac> factor auth app
[11:19] <asac> 2
[11:19] <didrocks> yeah, u<something>
[11:19] <asac> really?
[11:19] <asac> cool
[11:19] <asac> heh. seems the click scope again is gone (Guess crashed)
[11:19]  * asac reboots phone
[11:20] <didrocks> asac: http://notyetthere.org/?p=351
[11:21] <asac> cool
[11:21]  * asac tries system update
[11:21] <asac> first
[11:39] <Mirv> didrocks: so ken tested them? for dialer-app, there was a desktop test failure for which I filed a bug in the morning. content-hub, which is also marked as ready to land, is fine from cu2d perspective and I could releaseit.
[11:47] <didrocks> Mirv: yes, please release it :)
[11:47] <didrocks> for dialer-app 6> ok
[11:47] <didrocks> Mirv: can you anotate dialer-app test failure on the landing ask?
[11:47] <didrocks> Mirv: did you ping upstream as well?
[11:47] <didrocks> (in case they are not on top of their bugs?)
[11:51] <Mirv> didrocks: ok, ok, and no ping yet, I suggested on the call that sil2100 would ping about the AP failures
[11:51] <didrocks> sil2100: have you? ^
[11:54] <ogra_> asac, "Autthenticator" is the 2fa app
[11:54] <Mirv> didrocks: http://10.97.0.1:8080/view/cu2d/view/Head/view/Services/job/cu2d-services-head-3.0publish/lastSuccessfulBuild/artifact/packaging_changes_content-hub_0.0+14.04.20131024.2-0ubuntu1.diff - seems ok to me, indeed adds the QML plugin docs to the doc package
[11:54] <didrocks> Mirv: +1 ;)
[11:57] <asac> ogra_: wow... even qrcode works :)
[11:57] <vila> asac, didrocks : 2fa on touch is called Authenticator, darn ogra_ beated me to it ;)
[11:57]  * asac super impressed
[11:58] <ogra_> asac, mzanetti's work :)
[11:58] <vila> asac: be aware that it seems you can't configure both the g one and authenticator on sso
[11:58] <asac> vila: sure?
[11:58] <vila> or rather, I did add authenticator but sso seems to recognize the codes from that g one
[11:59] <ogra_> asac, didrocks, popey, so i released 5/20131024 ... http://people.canonical.com/~ogra/touch-image-stats/20131024.changes is the matching changelog
[11:59] <asac> well. it has now both devices
[11:59] <vila> asac: only light tests, no time to dig at the time
[11:59]  * asac hopes
[11:59] <asac> will find out soon enough... at least yubi and android worked well
[11:59] <asac> together
[11:59] <ogra_> yeah, shouldnt cause issues
[12:00] <didrocks> ogra_: niceness!
[12:00] <ogra_> didrocks, well, someone needs to mail :)
[12:00] <didrocks> ogra_: asac did
[12:00] <sil2100> Early ;p
[12:00] <ogra_> bah, i just had the launcher crash
[12:01]  * ogra_ has never seen that 
[12:01] <ogra_> all swipe gestures work, just not the left to right one that brings you back to the shell (or reveals the launcher)
[12:02]  * didrocks really goes for a run
[12:03] <vila> ogra_: not at all or only not enough to revel the shell ?
[12:03] <vila> *reveal
[12:03] <ogra_> not at all
[12:03] <ogra_> i have seen the "snap back" thing before that you seem to refer to
[12:03] <vila> never seen then
[12:04] <vila> right, any workaround for that ? (I generally just reboot)
[12:04] <ogra_> well, i rebooted already ...
[12:04] <ogra_> apart from filing a bug and nagging Mir and unity people ?
[12:04] <ogra_> no :)
[12:04] <didrocks> in fact, I can't go for a run, no music :/
[12:04] <ogra_> oh ?
[12:04]  * didrocks plugs it and put that on the "later on" plate (so after the meeting)
[12:05] <didrocks> ogra_: sucky battery, I charged it for 30 minutes and it went off in 10…
[12:05] <ogra_> ah, just wanted to say, music playback works fine here :)
[12:05] <didrocks> ahah, no, I don't use my nexus4 while running :p
[12:05] <ogra_> why not ?
[12:05] <didrocks> too heavy
[12:06] <ogra_> ah
[12:06] <didrocks> (and risky)
[12:06] <didrocks> if I feel down as I already did… ;)
[12:06]  * vila hands his unused ipod to didrocks  ;)
[12:06] <didrocks> or a car strikes me while riding a bike…
[12:06] <didrocks> no more didrocks, but at least, the company has the neus 4
[12:06] <didrocks> nexus 4
[12:07] <didrocks> so didn't loose everything :p
[12:07] <mzanetti> vila: first release of authenticator had a bug which made it fail to store the counter. Should be fixed now and it should be fine with having Google authenticator and Ubuntu authenticator registered at the same time
[12:07] <cjwatson> heh
[12:07] <cjwatson> I use my Nexus 4 while cycling quite happily
[12:07] <cjwatson> (for navigation)
[12:07] <vila> mzanetti: omg, that would be awesome !
[12:08] <asac> mzanetti: its awesome! thanks for this app
[12:08] <didrocks> cjwatson: do you have something to hang it? (or you just cycle with one hand?)
[12:08] <cjwatson> didrocks: upper jacket pocket
[12:08] <cjwatson> I do want to get a handlebar mount for it at some point; in the meantime as long as it's close enough I can hear voice directions
[12:08] <didrocks> ah ok, so you don't use the screen, just the voice directions
[12:08] <cjwatson> I'm not suicidal, I use both hands :)
[12:09] <didrocks> heh :)
[12:09]  * ogra_ uses it that way in the car actually
[12:09] <mzanetti> asac: you're welcome
[12:09] <ogra_> voice directions via earphone
[12:09] <ogra_> no distracting screen
[12:09] <didrocks> well, even with both hands, in Lyon, just doing the 5 kms that separate me from the park is a little bit suicidal sometimes…
[12:09] <cjwatson> yeah, occasionally I have to stop and look at the screen, but on a bike that's not a big deal
[12:09] <ogra_> haha, move to a safe place then :)
[12:10] <cjwatson> would be more troublesome if I drove
[12:10] <didrocks> ogra_: well, I like to not needing owning a car TBH
[12:10] <didrocks> and just rent one when needed :)
[12:10] <vila> mzanetti: urgh, weird case here, at some point auth wsa removed from the phone, I just re-installed it BUT it remembers the previous settings, how can I reset that ?
[12:11] <asac> psivaa: plars: can we give http://reports.qa.ubuntu.com/smokeng/trusty/touch_custom/ some love too?
[12:11] <mzanetti> vila: you can just swipe the account away to delete it
[12:11] <asac> psivaa: plars: like retrying and complaining in meetings if stuff is bad there?
[12:11] <asac> thanks
[12:12] <vila> mzanetti: a.w.e.s.o.m.e ;)
[12:12] <mzanetti> :)
[12:13] <ogra_> which reminds me to invest another hour to package more webapps today :)
[12:13] <mzanetti> ogra_: yeah! I already finished the blocks game
[12:14] <ogra_> oh my !
[12:14] <mzanetti> ;D
[12:14] <vila> mzanetti: woohoo it works !
[12:14] <ogra_> the site thats from has a bunch more, no worries :)
[12:14] <mzanetti> :D
[12:14] <vila> now I can get rid of my SIM-less android phone \o/
[12:14] <mzanetti> that was my intention ^^
[12:14]  * ogra_ just discovered the mobile mode of dict.cc yesterday evening :)
[12:14] <vila> ogra_: and thanks for imdb !
[12:15] <ogra_> just need to find an icon
[12:15] <vila> ogra_: when will I be able to watch the trailers ? ;-D
[12:15] <ogra_> which is usually the part taking the most time when packaging a webapp
[12:15] <ogra_> vila, ask jhodapp
[12:15] <ogra_> he is working on fixing the web streaming mode
[12:15] <ogra_> you can watch them in trusty btw
[12:16] <ogra_> like the first 100 frames or so ... then it becomes a slideshow ...
[12:16] <ogra_> ... and then turns into a podcast :P
[12:16] <vila> ogra_: you mean with image #5 or on the des... haa,  :)
[12:16] <mzanetti> lol
[12:16] <ogra_> from image #4 on it should work
[12:16] <ogra_> just the rendering backend isnt finished yet
[12:17] <didrocks> ogra_: "progressive degradation"
[12:17] <ogra_> nah its all improving :P
[12:17] <ogra_> saves your bandwith ... doesnt do that wasteful video stuff once you gotr an impression iof the pics
[12:17] <ogra_> and falls back to the true info via audio ...
[12:18] <ogra_> we should just call it a feature :P
[12:19] <sil2100> didrocks, Mirv: yes, I was pinging people ragarding the issues - I guess the dialer-app and the 'apps' AP issues are being worked on right now, let me check the status
[12:31] <Mirv> sil2100: so osomon? I'd have pinged him too in the morning but he was not up yet. but I assigned the bug to him.
[12:31] <Mirv> sil2100: the dialer-app is still assigned to bfiller, did you find someone to look at that during European day?
[12:33] <sil2100> Mirv: yes, I poked oSoMoN about that - dialer-app I posted to renato_, who forwarded it to tiagosh
[12:33] <sil2100> Who is said to be working on that
[12:34] <sil2100> Mirv: osomon is at lunch now, but he confirmed that he's looking at the issue
[12:37] <Mirv> sil2100: cool, I didn't know about tiagosh
[12:44] <psivaa> asac: sure, will take a look
[12:50] <sil2100> ogra_: do you have a moment? Could you take a look at http://10.97.0.1:8080/view/cu2d/view/Head/view/Services/job/cu2d-services-head-3.0publish/lastSuccessfulBuild/artifact/packaging_changes_ubuntu-keyboard_0.99.trunk.phablet2+14.04.20131025-0ubuntu1.diff ?
[12:50] <sil2100> I'm a bit wondering about the (<< 1.4) part
[12:51] <ogra_> well, if it would load i could look at it :P
[12:51] <sil2100> !
[12:51] <cjwatson> sil2100: the Depends change is safe
[12:51] <sil2100> One moment ;)
[12:51] <sil2100> https://jenkins.qa.ubuntu.com/view/cu2d/view/Head/view/Services/job/cu2d-services-head-3.0publish/lastSuccessfulBuild/artifact/packaging_changes_ubuntu-keyboard_0.99.trunk.phablet2+14.04.20131025-0ubuntu1.diff
[12:52] <ogra_> ah, a lot better
[12:52] <cjwatson> it's a bit pointless, but it's safe
[12:52] <ogra_> yeah, wont break a thing
[12:52] <sil2100> cjwatson: that's what I was wondering, why we even need that right now
[12:52] <sil2100> But ok, then I publish, thanks guys
[12:52] <cjwatson> the package is going to have to be updated later anyway; but it's not too unreasonable to document the current restriction in the metadata
[12:53] <ogra_> is there any .install file that might need to be adjusted to the removals ?
[12:53] <ogra_> if there isnt, i'd say its fine
[12:55] <fginther> morning
[12:56]  * ogra_ scratches head ... i have a webapp click package that seems to not work, even though nothing seems to be wrong ... 
[12:56]  * ogra_ wonders if that could be caused by rolling it as root inside a chroot 
[12:58] <didrocks> ogra_: any issue on the perms? as apparmor just kill the app…
[12:58] <ogra_> didrocks, thats what i'm wondewring, is it forbidden to build click packages (or their contents) as root
[12:58] <ogra_> oh !
[12:58] <ogra_> WOW !
[12:59] <ogra_> root@ubuntu-phablet:/# grep DEN /var/log/kern.log |wc -l
[12:59] <ogra_> 80
[12:59] <ogra_> thats a lot denials
[12:59] <didrocks> here you go :)
[12:59] <ogra_> didrocks, no, my webapp isnt in there
[12:59] <didrocks> interesting, so we have that many other pieces denied by apparmor? we should definitively make a list and work on it
[13:00] <ogra_> seems many webapps would like to access /run/shm
[13:00] <didrocks> hum, weird…
[13:00] <didrocks> I would think it would be more the container acting wrongly, wouldn't it?
[13:01] <ogra_> and xnox' G+ app seems to like to fiddle with the pulse config in all places you can imagine
[13:01] <mzanetti> ogra_: ah, btw. what I wanted to ask. for those games, are you just running webbrowser-app with the given url or are you creating a qml file with the UbuntuWebView?
[13:01] <ogra_> mzanetti, just the former
[13:02] <ogra_> they are all plain HTML5 games that you can as well just play in the browser
[13:02] <mzanetti> ogra_: mhm... because I noticed they can be flicked up/down which is not nice during the gameplay.
[13:02] <mzanetti> ogra_: the UbuntuWebView would give you the possiblity to disable that
[13:02] <ogra_> mzanetti, i tried that but that seems to not work unless i actually use the SDK
[13:03] <ogra_> it seems ot be missing bits and pieces if i just try to put a qml wrapper in place and build a click package
[13:03] <xnox> ogra_: honestly, not by intend =) "with pulse config" ?! what did I do wrong? =))))
[13:03] <ogra_> mzanetti, http://daker.me/2013/10/package-your-webapp-for-ubuntu-touch.html
[13:03] <ogra_> mzanetti, thats what i tried to follow
[13:03] <mzanetti> yeah... not working?
[13:03] <ogra_> xnox, i'm pretty sure thats more a jdstrand thing
[13:04] <ogra_> mzanetti, nope
[13:04] <jdstrand> ogra_: can you paste those denials?
[13:04] <mzanetti> hmm... can you tell me more? maybe I'm able to help you
[13:04] <ogra_> jdstrand, http://paste.ubuntu.com/6300513/
[13:04] <ogra_> mzanetti, i'll have to dig up the click for my slashdot app again ... will take a bit since i poked around a lot in there (and borke it)
[13:05] <mzanetti> ogra_: ok sure. just drop me a mail with details when ready and I'll have a look
[13:05] <ogra_> mzanetti, oh, looking at these denials it seems the slashdot test app is in the list too
[13:05] <jdstrand> ogra_: you need the audio policy group
[13:05] <xnox> ogra_: right if the webapps policy is too restrictive, than jdstrand should add stuff to it.
[13:05] <jdstrand> ogra_: for webapps, I recommend audio, networking, video and location
[13:05] <ogra_> jdstrand, well, xnox does :)
[13:06] <jdstrand> xnox: ^
[13:06] <jdstrand> xnox: fyi, https://wiki.ubuntu.com/SecurityTeam/Specifications/WebAppsConfinement
[13:06] <ogra_> jdstrand, i usually only use video where appropriate ... but have usually audio and networking
[13:06] <xnox> jdstrand: i think i only use webapps policy from the easy wrapper, or shall I declare those additional ones?
[13:06] <jdstrand> xnox: what easy wrapper?
[13:06] <ogra_> mzanetti, Oct 24 12:21:18 ubuntu-phablet kernel: [14309.845991] type=1400 audit(1382610078.837:199): apparmor="DENIED" operation="exec" parent=1977 profile="com.ubuntu.developer.ogra.slashdot_slashdot_0.1" name="/usr/bin/qtchooser" pid=17494 comm="exec-line-exec" requested_mask="x" denied_mask="x" fsuid=32011 ouid=0
[13:06] <ogra_> so thats not the code but some missing policy when using the qml wrapper
[13:07] <xnox> jdstrand: let me look.
[13:07] <mzanetti> ogra_: mhm... right. at that point it's like a native app to the policy I guess
[13:07] <jdstrand> fyi, qtchooser exec is allowed in the ubuntu-sdk and ubuntu-webapp templates
[13:08] <ogra_> root@anubis:/root/apps/slashdot-app/package# cat slashdot.json
[13:08] <ogra_> {
[13:08] <ogra_> "template": "ubuntu-webapp",
[13:08] <ogra_> "policy_groups": [
[13:08] <ogra_> "networking"
[13:08] <ogra_> ],
[13:08] <ogra_> "policy_version": 1.0
[13:08] <ogra_> }
[13:08] <ogra_> well ...
[13:09] <ogra_> that would mean it shoudl work :=
[13:09] <ogra_> :)
[13:09] <xnox> jdstrand: right, so I have template "ubuntu-webapp", but indeed only networking in the policy_groups. I guess I should add more.
[13:09] <xnox> jdstrand: i thought template "ubuntu-webapps" would do that for me.
[13:09] <ogra_> xnox, video and audio at least
[13:10] <xnox> ogra_: can you file a bug against https://launchpad.net/click-webapps ?
[13:10] <xnox> ogra_: that's where the source code is for my webapps & bug tracking.
[13:10] <xnox> ogra_: such that I don't forget. And the one to do URLPatterns.
[13:11] <jdstrand> xnox: no-- the template just allows the browser to run. arguably, networking should be included in it by default, but, meh. the others aren't necessarily needed for all webapps, so those definitely make sense to not be included
[13:12] <jdstrand> xnox: but, I think sane defaults are all of them
[13:12] <xnox> jdstrand: ack. Now I understand.
[13:12] <jdstrand> xnox: so feel free to update your template
[13:15] <ogra_> bug 1244654 and bug 1244656
[13:16] <ogra_> jdstrand, so that qtchooser denial ... could that be caused by files inside the click being root owned ? http://paste.ubuntu.com/6300585/
[13:17] <jdstrand> shouldn't matter
[13:17]  * ogra_ notes that this is probably the totally wrong channel to discuss this 
[13:18] <jdstrand> ogra_: can you paste the apparmor profile for that app? cat /var/lib/apparmor/profiles/click_com.ubuntu.developer.ogra.slashdot_slashdot_0.1
[13:19] <ogra_> http://paste.ubuntu.com/6300606/
[13:27] <ogra_> oh, wow
[13:28] <plars> asac: yes, I'm talking to sfeole about it
[13:28] <ogra_> so my new webapps seem to cause entrieas in ~/.cache/upstart/application.log
[13:28] <ogra_> start: Auftrag konnte nicht gestartet werden
[13:28] <plars> asac: regarding the custom job stuff
[13:28] <plars> psivaa: ^
[13:28] <ogra_> (which means "cant run request" or some such)
[13:28] <ogra_> smells like an upstart regression
[13:29] <vila> jibel: hi
[13:29] <vila> jibel: I need to re-install qa-radeon-7750 as a jenkins/otto node, didrocks said you may have a document describing that, can I haz ? ;)
[13:29] <psivaa> plars: ack, was looking at why the diff in cutom unity8 tests. but if you are already dealing with that i'll leave it to you
[13:30] <psivaa> s/cutom/custom
[13:30] <vila> jibel: otherwise, Ill just go with a basic jenkins-slave and lp:otto in~jenkins + symlink for ~/bin and see what is missing
[13:30] <plars> psivaa: unity8 tests are pretty broken for now at least, I'm more interested in the custom test failure, but I'll check unity8 too... it didn't seem to run right
[13:31] <plars> psivaa: also there was a difference in webbrowser with the latest image. I can handle though
[13:31] <psivaa> plars: the unity8 tests in custom jobs are not even starting to run
[13:31] <jibel> vila, http://bazaar.launchpad.net/~otto-dev/otto/trunk/view/head:/doc/README
[13:32] <jibel> vila, if you don't know how to do a server install with cobbler, retoaded can help you.
[13:32] <vila> jibel: how did I miss that...
[13:33] <vila> jibel: yeah, already done (no trusty image so he did saucy + upgrade), waiting for a server image to pop up, cjwatson said he will try to get it done today
[13:33] <vila> jibel: thanks
[13:33] <jibel> vila, meanwhile you can install saucy and upgrade to trusty
[13:34] <vila> jibel: yup, already done, see above (already done yesterday evening)
[13:34] <jibel> I mean you don't really need a server image of trusty for that because this machine is not supposed to be reinstalled often
[13:35] <vila> jibel: ack, but the sooner this is automated the better
[13:36] <alan_g> fginther: do you know what caused this? https://jenkins.qa.ubuntu.com/job/mir-android-trusty-i386-build/2/console (it doesn't seem to have anything to do with the code changes)
[13:37] <fginther> alan_g, ack, looks like a missing piece of transition. I'll get that fixed
[13:38] <alan_g> fginther: thanks
[13:39] <elopio> ping sil2100: you talked with thomi about making autopilot trunk 1.4 again, right?
[13:39] <sil2100> elopio: hi!
[13:39] <elopio> hello :)
[13:39] <sil2100> elopio: well, thomi did it by himself already yesterday, since it seems he wasn't informed about the decisions that happened earlier (he missed the pings about it)
[13:39] <sil2100> elopio: so trunk is trunk now, but next week we'll be deciding on how to resolve it completely in the end
[13:40] <elopio> sil2100: yes, I see that. But otto is now taking again autopilot 1.4. I think you discussed about changing the config to take 1.3. Who can do that?
[13:41] <sil2100> elopio: otto should take 1.3 I guess
[13:42] <sil2100> elopio: yesterday I already redeployed it to point to 1.3, now it points to 1.3-trusty, which is nothing more than an alias to 1.3
[13:43] <elopio> sil2100: ah, nice! let me dig more on the failures then, becuase I thought it was just autopilot 1.4 again.
[13:46] <fginther> alan_g, the mir-android-trusty-i386-build job is working now, would you like the failed jobs to be restarted?
[13:47] <alan_g> fginther: thanks, no (I merged by hand to get it done)
[13:52] <cjwatson> Is Qt going to be bumped to 5.1 any time soon?  There are a bunch of modules waiting for it in -proposed
[13:53] <ogra_> cjwatson, well, 5.2 was just released (or is in the process to)
[13:53] <ogra_> might be that we skip 5.1 ...
[13:53] <ogra_> (topic for the sprint iirc)
[13:53] <cjwatson> well, whatever
[13:53] <cjwatson> but 5.1 is in Debian at the moment so it'd be a merge; I thought you guys liked to work stepwise
[13:54] <ogra_> rsalveti is our Qt decider i think
[13:59] <rsalveti> cjwatson: we can give it another try and see, but we got quite a few regressions in touch with 5.1
[13:59] <rsalveti> something might be fixed already as it's in debian, will take a look
[14:01] <ogra_> rsalveti, so we wont skip directly to 5.2 ?
[14:02] <rsalveti> ogra_: if we can get 5.1 to work without much effort, I'd prefer to move in smaller steps, otherwise we can just migrate to 5.2 directly
[14:02] <didrocks> jdstrand: joining the landing process feedback call?
[14:03] <ogra_> k
[14:03] <rsalveti> ogra_: just easier to find regression if we can move in smaller steps
[14:03] <ogra_> yup, agreed
[14:04] <jdstrand> didrocks: ok
[14:05] <fginther> vila, cjohnston, either of you interested in reviewing: https://code.launchpad.net/~fginther/cupstream2distro-config/core-apps-trusty/+merge/192600
[14:06] <cyphermox> asac, you on trusty? trusty already has the fix
[14:06] <cyphermox> asac: so you could take the pacakge from there too, if you want
[14:08] <cjohnston> fginther: done
[14:08] <asac> cyphermox: just a single wpa supp deb?
[14:08] <cyphermox> asac: no, that's a NM fix
[14:08] <ogra_> didrocks, phablet-tools should be ready for going in
[14:09] <cyphermox> libnm-glib4, libnm-util2, network-manager, gir1.2-networkmanager-1.0
[14:17] <vila> fginther: oops, done a while ago, forgot to say :-/
[14:20] <fginther> vila, cjohnston, I'm trying to get more exposure on these MPs, so please feel free to ask why thing are the way they are
[14:20] <fginther> vila, cjohnston and thanks
[14:21] <vila> fginther: in my case it was more about why it was needed, what triggered it and so on, context ;) But that may just be me being ignorant ;)
[14:22] <cjohnston> vila: I couldn't figure out why he was adding trusty support :-)
[14:22] <sil2100> grrr
[14:24] <vila> cjohnston: ha, so, every 6 months we... ;)
[14:24] <cjohnston> take a vacation?
[14:34] <didrocks> ogra_: \o/ sil2100 can you release it?
[14:36] <sil2100> didrocks: aye!
[14:39] <ogra_> thanks !
[14:42] <sil2100> Done
[14:52] <sil2100> hm, nice that robru wrapped and sorted the packages: long lines, but actually it's harder copy and pasting those now
[14:53] <ogra_> you mean the deps lines ?
[14:53] <sil2100> ogra_: no, I mean in cu2d-config
[14:53] <ogra_> ah
[14:54] <ogra_> i thought the debian/control lines
[14:54] <ogra_> (which i find one of the worst behaviors)
[14:54] <didrocks> ogra_: sorting deps?
[14:55] <ogra_> didrocks, making them unpasteable
[14:55] <ogra_> by adding linebreaks after each comma
[14:55] <didrocks> ah, I like linebreaks after each comma
[14:55] <didrocks> makes diff easier to read
[14:55] <ogra_> i hate it with passion
[14:55] <didrocks> I'm just not on the trends of alphabetical order :)
[14:55] <didrocks> I think logical order (and following the build system order) is better
[14:55] <ogra_> makes it nearly impossible to quickly copy/paste to install i.e. build deps
[14:56] <Laney> you can't do that anyway
[14:56] <Laney> use mk-build-deps or something
[14:57] <ogra_> apt-get install $(echo "my paste here" | sed 's/,/ /')
[14:57] <Laney> mk-build-deps -i -r
[14:57] <ogra_> yeah yeah ... nifty tools :P
[14:57] <Laney> :)
[14:58] <cjwatson> the paste strategy fails as soon as there are versions anyway, so why cling to it ...
[14:58] <ogra_> dunno, because old habits die hard i guess :)
[15:06] <plars> can someone please enable saucy builds in the phablet-team/tools ppa?
[15:07] <sil2100> fginther: hi! How can I check if the merge job is working for a given merge already?
[15:07] <sil2100> fginther: since I have been wondering why https://code.launchpad.net/~tiagosh/dialer-app/fix-autopilot-dependency/+merge/192688 wasn't yet merged
[15:07] <sil2100> fginther: I once asked, but it popped out of my head already
[15:08] <sil2100> didrocks: when we're installing packages from the testpackages: list, we install automatically all the dependencies of those packages, right?
[15:08] <fginther> sil2100, go to the autolanding job for the branch - http://10.97.0.26:8080/job/dialer-app-autolanding/
[15:09] <didrocks> sil2100: right!
[15:09] <didrocks> no filter, nothing
[15:09] <fginther> sil2100, and check the parameters for the running (or queued jobs)
[15:09] <fginther> sil2100, in this case your MP is running
[15:09] <sil2100> Running mediumtests I see
[15:10] <sil2100> fginther: thanks!
[15:10] <fginther> sil2100, there was a backup due to all of the maguro's failing overnight
[15:10] <sil2100> fginther: are there tools to make it skip mediumtests actually? Since I don't think there's need for running those for this small merge, I tested on the daily build job that it works
[15:11] <fginther> sil2100, no there is no way to change the configuration of a running job
[15:12] <sil2100> Too bad, thanks!
[15:12] <fginther> sil2100, based on the speed of flashing phones and running tests, it should be done within an hour
[15:14] <sil2100> fginther: good to know!
[15:38] <tedg> ev, So after release, we run whoopsie but not apport... does that mean we don't get files in bug reports apport would normally attach?
[15:44] <sil2100> tedg: hi! Do you know who could look into https://bugs.launchpad.net/hud/+bug/1244704 ?
[15:47] <tedg> sil2100, Looks like an autopilot issue?  I see the HUD appear and disappear in that video: http://10.97.0.1:8080/job/autopilot-trusty-daily_release/104/label=autopilot-nvidia/artifact/results/autopilot/videos/unity.tests.test_hud.HudBehaviorTests.test_no_initial_values.ogv
[15:50] <didrocks> ogra_: sil2100: cyphermox: robru: kenvandine: if everyone is here, do you want to listen to the qbr call? we can proceed the meeting in advance if you prefer
[15:52] <sil2100> didrocks: not sure personally
[15:52] <didrocks> ok, can wait for 8 minutes then
[15:53] <sil2100> Ah, actually I can't join the call right now, I don't really have a normal phone while mobile phones have problems with connecting to the conference call line
[15:54]  * sil2100 didn't bother with a normal phone since we're using hangouts normaly anyway
[15:54] <tedg> sil2100, There's an icecast stream
[15:56] <didrocks> sil2100: can you write your summary then?
[15:56] <didrocks> sil2100: like, you are taking care of dialer-app?
[15:56] <didrocks> I saw you tracked that
[15:56] <sil2100> didrocks: dialer-app is taken care of already - just waiting for the merge to finally get merged, trying to poke people with other test failure
[15:56] <didrocks> ah, can't join the call
[15:56] <didrocks> not the hangout :)
[15:56] <didrocks> ok, making sense
[15:56] <sil2100> Right ;)
[15:56] <didrocks> don't tease me then!
[15:57] <sil2100> ;D
[15:57] <didrocks> keep that for the big meeting revelation ;)
[15:57] <sil2100> tedg: thanks for the info
[15:57] <sil2100> tedg: as for the failures, those might be AP-only related, but actually there was no change in AP 1.3, so the issue might be somewhere else as well that we suddenly ahve those failures
[15:58] <sil2100> tedg: maybe I'll poke bregma about those failures, as those are from the unity source?
[15:59] <tedg> sil2100, Yeah, it does seem odd.  The focus change one is odd as well.   But we did change that code recently-ish (two weeks ago)
[16:01] <ogra_> i'll be late for the meeting (need to catch the cat for susie, so she can get her to the vet)
[16:01] <didrocks> ogra_: °1
[16:12] <didrocks> cyphermox: 1. mir, 2. platform-api, xserver-xorg-xmir (second one is a no-change rebuild), unity-system-compositor, 3. libunity-mir
[16:13] <cyphermox> kgunn: you'll let me know when this stuff is ready?
[16:13] <cjwatson> I've got a new packagekit upstream I'd like to upload, merged from Debian; I've tested it on grouper and click handling still seems to work as I'd expect
[16:13] <cjwatson> any objections or shall I go ahead?
[16:13] <cjwatson> not especially major changes
[16:13] <cjwatson> http://paste.ubuntu.com/6301459/
[16:18] <sil2100> tedg: one more thing, did you see bug #1244522 ?
[16:19] <didrocks> plars: if you have any ringing bell -> email is fine as well :)
[16:19] <didrocks> plars: but as we'll be in the same timezone, that will be easier ;)
[16:20] <didrocks> on*
[16:20] <didrocks> well on the ***samish*** timezone :p
[16:20] <plars> didrocks: you mean next week I guess? since you'll be traveling
[16:20] <sil2100> tedg: this is also one of the blockers, I think Timo already poked someone about that but it's still better if you would check that too ;)
[16:20] <didrocks> plars: right
[16:20] <plars> didrocks: of course :)
[16:20] <sil2100> Damn, dialer-app merge takes FOREVER
[16:20] <didrocks> plars: I don't think we'll do the meeting over HO, that's why ;)
[16:20] <plars> didrocks: should make things easier for a bit :)
[16:20] <didrocks> hehe, indeed ;)
[16:20] <plars> didrocks: ah, right
[16:21] <plars> didrocks: I won't be there, but just ping on irc if there's anything urgent to make me aware of from the in-person meetings
[16:21] <didrocks> plars: will do!
[16:21] <didrocks> thanks
[16:23]  * cyphermox does to get lunch, bbl
[16:50] <sil2100> didrocks: ok, oSoMoN prepared a fix for notes-app, one for gallery-app is coming right up
[16:50] <didrocks> \o/
[16:50] <ogra_> great
[17:07] <ev> tedg: I don't quite understand what yo're asking
[17:07] <ev> you're*
[17:07] <ev> so for the desktop, when we switch off apport we still get crash reports to daisy.ubuntu.com
[17:07] <ogra_> (wasnt the first one proper texan slang ?)
[17:07] <ev> it just doesn't do the Launchpad bits
[17:07] <ev> ha!
[17:07] <ev> allow me to rephrase
[17:07] <ev> howdy partner
[17:07] <ogra_> lol
[17:08] <ev> Ubuntu Touch is another story, I reckon
[17:08] <ev> We didn't quite flip the switch for whoopsie being enabled by default on that, last I heard
[17:32] <tedg> ev, I was worried more about the log files, like hud.log, which doesn't seem to be on recent traces.
[18:01] <cyphermox> tedg: you looking at the indicator-network FTBFS?
[18:01] <cyphermox> kgunn: news for the mir stuff?
[18:04] <tedg> cyphermox, Yes, I passed it along to kenvandine
[18:04] <cyphermox> oh
[18:04] <cyphermox> heh.. I told sil I could look into it too
[18:04] <cyphermox>  ;)
[18:04] <kenvandine> cyphermox, feel free too :)
[18:04] <cyphermox> kenvandine: you already on it?
[18:04] <cyphermox> ah, alright :)
[18:04] <kenvandine> just started
[18:04] <kenvandine> either way :)
[18:04] <cyphermox> probably just the vala version no?
[18:05] <kenvandine> i am piloting... maybe this counts :)
[18:05] <kenvandine> i think it's just missing a patch
[18:05] <cyphermox> ok
[18:05] <kenvandine> i got it
[18:05] <cyphermox> ok
[18:05] <kenvandine> speaking of piloting... i forgot to check in :)
[18:07] <cyphermox> hehe
[18:37] <kenvandine> tedg, i think it is actually indicator-network that needs fixing, i pushed a branch that fixes the build
[18:37] <kenvandine> tedg, but the tests fail
[18:38] <kenvandine> vala-0.22 does have the fixes from that patch in it, but the hashtable has different types
[18:44] <dobey> fginther: hey. so i've been looking through the jlp autolander code a bit, and there's very little i see there that tarmac doesn't already do (and in some cases does it better). i think we can probably write a little bit of python to make plug-ins for the bits it doesn't do, maybe refactor a couple pieces in tarmac to work better for our needs, and switch over to using tarmac for all the autolander stuff
[18:46] <cjohnston> +1
[18:46] <cjohnston> 0000
[18:46] <fginther> dobey, very cool! the hard part is prioritizing the time to work on it :-(. Do you have a suggestion on how to proceed
[18:47] <fginther> dobey, I know vila would like to have some tests identified for our use cases first :)
[18:47] <cjohnston> fginther: the dashboard uses tarmac for autolanding...
[18:48] <dobey> fginther: not at the moment. i know i have to get a couple of the u1 projects moved over to daily releases, and the autolander isn't currently doing some of the stuff we do with tarmac, and which we need. at least, the contributor validation bit.
[18:48] <dobey> fginther: i know tarmac is very well tested. we manage hundreds of branches with it in u1 :)
[18:52] <fginther> dobey, I don't doubt that tarmac is reliable, the problem is that jenkins-launchpad-plugin is tightly integrated into our whole upstream merger CI process. It's going to take a bit of work to move and make sure our existing use cases are covered
[18:53] <cjohnston> fginther: I think it may be something to look at with that 2.0 BP
[18:53] <fginther> so we need to identify migration process
[18:53] <fginther> cjohnston, agreed
[18:54] <fginther> that would be a chance to build up smaller bits at a time
[18:56] <dobey> fginther: is there any log/documentation about why tarmac + custom plug-ins wasn't used in the first place? i know didrocks proposed some major changes to tarmac in the past that we rejected because they were too specific to the needs of just unity at the time, but i remember those being a bit different than what i'm sing in the jlp source tree now
[18:57] <fginther> dobey, I suspect that has something to do with it. A lot of process was borrow from work didrocks did. I don't have the history, but I'll poke around on a few people that are still around
[18:58] <dobey> fginther: right; just looking for something to isolate the caes that tarmac doesn't handle already :)
[19:05] <fginther> cjohnston, josepht, can you review https://code.launchpad.net/~fginther/cupstream2distro-config/mir-dev-trusty/+merge/192736? I provided a little bit of the 'why' this time.
[19:06] <cjohnston> -1 I don't think we need to support trusty :-P
[19:17] <tedg> kenvandine, Oh, good.
[19:17] <kenvandine> tedg, can you take it from here?
[19:18] <kenvandine> i got it building, but didn't look into the test failures
[19:18] <tedg> kenvandine, I wasn't really planning on upgrading to trusty...
[19:18] <tedg> Not before travelling.
[19:18] <kenvandine> pbuilder is your friend :)
[19:18] <kenvandine> tedg, want a log from the test failures?
[19:19] <tedg> kenvandine, Sure, can I say no?  :-)
[19:19] <kenvandine> http://paste.ubuntu.com/6302333/
[19:19] <kenvandine> tedg, you could, but i wouldn't listen :)
[19:20] <tedg> Hmm, those aren't vala related
[19:20] <kenvandine> nope
[19:20] <tedg> That's the non-vala code.
[19:20] <kenvandine> well, couldn't they be failing because something else changed types?
[19:22] <tedg> Eh, seems unlikely.
[19:22] <tedg> Guessing race?
[19:22] <tedg> It's not getting the notification signal
[19:24] <tedg> kenvandine, Do you have it set up?
[19:24] <kenvandine> same 5 tests fail, ran it 3 times now
[19:24] <tedg> kenvandine, Can you try moving the creation of the spy to be before teh "GetSecrets" call?
[19:26] <kenvandine> same thing
[19:30] <tedg> Hmp
[19:30] <tedg> Hmph
[19:31] <tedg> That's probably a different bug though.
[19:31] <tedg> At that point it'll probably have to be a pete-woods thing
[19:33] <kenvandine> tedg, ok, i added a link to that log in the bug and assigned it to pete-woods
[19:33] <tedg> kenvandine, Cool, thanks!
[19:34] <kenvandine> tedg, np
[19:39] <cyphermox> kgunn: mir stuff ready now? :)
[19:41] <cyphermox> kgunn: I can keep pinging every two hours or so or if you want you can let me know if there's an ETA so I don't keep bugging you for no reason
[19:43] <kgunn> cyphermox: no problem pinging
[19:44] <kgunn> we're in #ubuntu-mir
[19:44] <kgunn> mterry saw a ci failure...but i think its a flakey test, a couple of us have tested locally with no repro
[19:47] <cyphermox> ok
[19:47] <cyphermox> please let me know
[20:28] <ev> tedg: whoopsie doesn't send custom apport files to daisy.ubuntu.com, if that's what you mean
[20:28] <ev> fields*
[20:30] <tedg> ev, So it's the apport tools that attach those to the crash file?
[20:32] <ev> tedg: so if you have a HudLog field in your apport report, that wont get sent to daisy. That's added by an apport hook, which whoopsie ignores the data from: http://bazaar.launchpad.net/~daisy-pluckers/whoopsie/trunk/view/head:/src/whoopsie.c#L93
[20:32] <ev> the plan here was always to have daisy.ubuntu.com ask for additional data to be collected when it was actually needed
[20:32] <ev> rather than everyone sending up massive log files
[20:33] <ev> much like how we handle core files
[20:34] <ev> this was called server-side hooks and I never quite got to it
[20:34] <tedg> ev, Hmm, okay.  I guess I just want the log files :-)
[20:35] <ev> I hear your pain, but whoopsie can't do that just yet :)
[20:35] <tedg> ev, Would you be against adding the upstart logs to the white list in the mean time?
[20:37] <ev> tedg: alas, it wouldn't be that simple. Cassandra isn't good at blob storage. If these log files are quite big, we're not in a good place yet for their storage
[20:38] <tedg> ev, then they're very small
[20:38] <tedg> :-)
[20:38] <ev> there's a few things we could do here. Enabling compression in cassandra is one, but the move to 1.2 (where that feature is reliable) is a couple of months off by my estimate
[20:38] <ev> ha!
[20:39] <tedg> It seems to vary, I have a bunch that are 20-30K but the gnome-session one is 20M
[20:39] <ev> our code is supposed to split large values across multiple columns, but I more generally worry about the performance impact on cassandra
[20:40] <ev> and this is on the recoverable reports that we're getting lots and lots of?
[20:40] <tedg> Yeah, we could also probably just gzip them.  Couldn't search, but I dont' think that's a needed feature.
[20:40] <tedg> No, it was on crashes that it came up.
[20:41] <tedg> We've apparently got 1500 cases of dbus disappearing before HUD.  Which should theoretically be impossible.
[20:42] <tedg> I was curious if we're getting the sigterm before the crash and we're just not responding quick enough or if there was something else in the log.
[20:42] <ev> tedg: talking to IS about it right now
[20:43] <tedg> ev, Cool, thanks!
[20:45] <ev> tedg: could you wait two weeks? They feel we could handle the load now, but we're in the middle of a repair on the 'newcassandra' ring
[20:45] <ev> if you really need it now, then propose a merge to whoopsie to accept the field
[20:46] <ev> repair> http://wiki.apache.org/cassandra/Operations#Repairing_missing_or_inconsistent_data
[20:46] <tedg> ev, I don't think it's critical, we're removing the crash, so mostly hiding the error.  I was more surprised when I started digging.
[20:46] <tedg> ev, More for "next time" type of thing.
[20:46]  * ev nods