=== vibhav is now known as Guest48005 === Guest48005 is now known as vibhav === jono is now known as Guest19637 [08:03] hey sil2100, how are you? [08:08] didrocks: morning! Rather fine, and you? ;) [08:09] sil2100: I'm ok, thanks! [08:09] sil2100: I have two questions for you :) [08:10] 1. do you mind having a look at the indicators tests? one of the tests search is failing on the 3 configs [08:10] (we also have 2 additional failed tests on intel, would be great to have a grasp) [08:10] 2. did you make any progress on the appmenu-qt/Qt5 side? [08:11] didrocks: 1. I'll take a look at it [08:11] didrocks: 2. there is some progress - I ported appmenu-qt to Qt5 but am still working on the appmenu plugin-framework side of Qt5, but I'm going in the right direction I think [08:12] sil2100: 1. & 2. -> thanks :) [08:13] sil2100: did you get support from agateau? [08:13] sil2100: we'll need a FFe I guess for the appmenu part [08:27] didrocks: I wrote him an e-mail, but he didn't reply yet - but renato gave me some pointers, as well as agateau's blog posts about the inclusion of appmenu in Qt4 helped out [08:29] sil2100: great ;) I'm not sure if it's the holidays for him, let me try to grab him [08:29] yeah, he seems away [08:29] anyway, thanks sil2100! keep me posted for the indicator tests. If it's a false positive, we can force a daily already [08:48] didrocks: I've been looking at the failures, and the search tests seem to be a false positive - a really strange one, since it fails the same way on all machines [08:49] It fails to show the application lens on the first super+a [08:49] Not sure why! [08:49] Cannot reproduce it here, hm [08:49] sil2100: interesting… would be good to workaround/fix it as it will block all releases in the next days [08:49] Also, no previously executed tests seem to have done anything to break this [08:49] didrocks: could you re-start the indicator build test maybe? [08:49] sil2100: it's the one that we added yesterday [08:50] sil2100: well, it failed on all 3 configs, I doubt restarting will change anything, isn't it? [08:50] No, it's not the one added - it's one of the existing ones [08:50] sil2100: test_search was not executed before for indicators [08:50] Ah [08:50] Ok, hm [08:51] sil2100: look at the unity ones [08:51] failing the same [08:51] I just looked and indeed saw the same thing [08:51] Ok, then this requires more investigation then [08:51] ! [08:52] yep :) [09:20] didrocks: ok, it seems to be a real regression - the first Super+a press does not work [09:21] sil2100: oh, it rings a bell to me [09:21] sil2100: maybe we need a separate test for that? [09:21] didrocks: it would be hard to do, since with autopilot we're never sure what test will be executed first ;) [09:22] I can try looking into unity to check what could be the source for this regression [09:22] sil2100: oh, I just mean a separate test [09:22] sil2100: so that we'll see this one failing as well [09:22] sil2100: on the source check -> yes please ;) === mmrazik is now known as mmrazik|lunch === mmrazik|lunch is now known as mmrazik [12:08] sil2100: hey, did you get any luck? === MacSlow is now known as MacSlow|lunch === _salem is now known as salem_ [12:41] didrocks: building now something testable [12:41] sweet! [13:02] btw dash search was faster in 12.10 than it is in raring [13:03] om26er: something to tell to Satoris I guess [13:03] the rest didn't change [13:04] didrocks, well the dash itself is feeling faster on 12.10 to me, dunno why [13:05] om26er: the blur algorithm change? [13:06] didrocks, that maybe related but wasn't that supposed to fasten the dash ;) [13:06] it was, but you never know :) === dandrader is now known as dandrader|afk === dandrader|afk is now known as dandrader === MacSlow|lunch is now known as MacSlow [13:43] didrocks, can you look at https://code.launchpad.net/~ken-vandine/dee-qt/prepare_for_raring/+merge/151617 [13:44] i really want to get that landed in raring, the folks working on the core apps will need it [13:56] didrocks, I would like to bump Unity to version 7 (but leave the API version at 6) so we can do a sensible upstream release before FF today -- what are your thoughts? === mmrazik is now known as mmrazik|afk [14:52] bregma: hey! [14:52] bregma: I'm really wondering about the benefit/cost [14:52] didrocks, ho! [14:52] bregma: changing the version to 7 without changing the API version should work [14:52] meaning the path isn't changed for the assets [14:52] but it needs checking [14:53] bregma: do you think the risk/cost benefit worthes it? [14:53] kenvandine: hey! the changes looks ok. I'm wondering though if we do have common files with dee-qt in precise for the precise -> next LTS upgrade? [14:53] kenvandine: also, while we are at it, what do you think spending some time building the daily release and doing all that in one shot? [14:53] i wouldn't think so [14:54] kenvandine: do you mind checking at packages.ubuntu.com? [14:54] I think the benefit of being able to distinguish between the release in 12.10 and the release in raring/going forward is worth it from a support point of view [14:54] kenvandine: just one binary package, should be quick :) [14:54] sure [14:54] and an upstream release is good for downstreams [14:54] and we can close milestoned bugs [14:54] (which downstreams? ;)) [14:54] bregma: do you mind trying bumping debian/changelog and the upstream version [14:54] the imaginary ones I would like to see sprout like mushrooms after a rain [14:54] building the package [14:54] and installing it [14:54] bregma: heh [14:55] bregma: just to check that lenses still starts and so on? [14:55] I've been working on that, queued up after another test I'm running [14:56] bregma: if it works on a fresh session, please go ahead :) [14:56] unfortunately I'll have to upgrade my build machine to raring to do it proper , which means I may miss FF [14:56] bregma: well, bumping the version for bumping the version is not a "feature" [14:56] OK [14:56] bregma: so as long we don't require rebuilding the world, I'm ok with it :) [14:56] just put it in a sensible way in debian/changelog [14:56] all I want is a new package version, no other changes, so if it Just Works I will propose it [14:57] bregma: excellent! [14:57] bregma: I really hope that we'll get a clearer view soon [14:57] on if/not we release raring [14:57] yes, indeed [14:57] if we release raring, we'll want to branch, too, before release goes out [14:58] another reason to bump the package version [14:58] bregma: yeah, let's see how it goes [14:58] didrocks: would you mind if I did a test-build in ps-indicators-autopilot-release-testing with a custom list of tests to perform? [14:58] didrocks: or is there another job for that purpose? [14:59] bregma: I was expecting to go on "daily release to a ppa" when we will close to next LTS ;) [14:59] bregma: not that soon! [14:59] sil2100: oh no worry, please be my guest :) [14:59] didrocks: since I tried using autopilot-run-custom-branch, but it doesn't work somehow, probably veebers has some WIP regarding that [14:59] didrocks: thanks! [14:59] sil2100: but we do have everything built on the unity stack [14:59] sil2100: if you want to check the global result [15:00] I know, I just want to start one suite [15:00] ok [15:00] didrocks, no file conflicts [15:00] combination of soname change and multiarch makes it nice and clean :) [15:00] kenvandine: ahah, I would have bet so! great ;) [15:01] kenvandine: daily release? I think it's maybe time for folks/friends? ;) [15:01] folks? [15:01] kenvandine: or you do want to do the first shot today [15:01] I meant friends :) [15:01] i'll propose a branch enabling friends today [15:02] kenvandine: and we work on that together tomorrow? [15:02] kenvandine: so that we define stacks [15:02] seb128, don't forget my NEW review :) [15:02] kenvandine, oh, yeah [15:02] didrocks, i won't be in tomorrow :) [15:02] it's 2 packages for now [15:02] libfriends and friends [15:02] small stack :) [15:03] i just added an example to qml-friends that i'll use as the basis for an autopilot test [15:03] kenvandine, no COPYING/license in the tarball? [15:04] kenvandine: but we need dee in daily release, isn't it? [15:04] dee-qt* [15:04] kenvandine: ok, let's wait a little bit for friends that you do your autopilot test [15:06] kenvandine, accepted, please added a COPYING though [15:06] seb128, will do... [15:06] kenvandine, thanks ;-) [15:07] didrocks, i feel good about libfriends and friends even without qml-friends autopilot tests [15:08] libfriends test suite actually runs friends in test mode [15:08] it is a good integration test [15:08] qml-friends needs tests before i want to add that to the dailies [15:08] seb128, added :) [15:08] good [15:09] kenvandine: but, those 2 needs to be installed somewhere to work together, isn't it? [15:10] yes [15:10] humm [15:11] kenvandine: ok, so we need a jenkins job taking latest daily, and doing that, isn't it? [15:11] installing both [15:11] libfriends pulls in friends as a build dep to test [15:11] yeah [15:11] ah [15:11] so that might be good enough [15:11] si, it can be ran while libfriends is building? [15:11] yeah [15:11] friends doesn't depend on libfriends [15:11] so you already do run integration tests during build? :) [15:12] yes [15:12] the unit tests in libfriends... i think those are good integration tests for friends [15:12] \o/ [15:12] since it actually runs friends and really calls it's functions [15:12] excellent [15:12] yeah, so they are not unit tests :) [15:12] we were very thorough with friends :) [15:12] but I only care about integration tests TBH [15:12] kenvandine: do you have upstream merger? [15:13] well, the are unit tests for libfriends... a side effect is integration testing of friends [15:13] no [15:13] ok, let's get that fixed! [15:13] what do i need to do to add that to the merger [15:13] kenvandine: pinging mmrazik|afk ;) [15:13] kenvandine: do you want to add the stack definition? [15:13] yeah [15:13] creating the friends stack, I think you deserve it :) [15:14] kenvandine: https://launchpad.net/cupstream2distro-config/trunk [15:14] you do have the stacks/head/ dir [15:14] create a friends.cfg [15:14] similar to the unity one for instance [15:25] didrocks, can i model it after the webapp one? [15:25] kenvandine: yes, an easier one maybe :) [15:26] right :) [15:26] kenvandine: the additional info will be added by mmrazik|afk for upstream merging (we are transitionning to having one single file for all those infos) [15:26] kenvandine: please call the file "friends", not "friens-head" [15:26] ok [15:26] kenvandine: as I added the head/ directory [15:28] https://code.launchpad.net/~ken-vandine/cupstream2distro-config/friends-stack/+merge/152199 [15:28] didrocks, ^^ [15:29] mmrazik|afk, when you're around, can you help me get friends added to the merger? [15:29] kenvandine: approving [15:29] kenvandine: you can push to trunk [15:29] kenvandine: so that we win some time :) [15:29] great [15:29] kenvandine: then, we need to create from your template the jenkins jobs [15:30] in the same project [15:30] daily-release directory [15:30] you have the cu2d-update-stack command [15:30] it should be used like this: [15:30] ./cu2d-update-stack -U <../path/to/stack/file> [15:31] kenvandine: as mmrazik|afk and fginther did some change to the tool, you can have bad surprises, tell me if it exit with non 0 :) [15:31] that will create the jenkins job? [15:31] kenvandine: you need to be connected to the vpn and having your credential setup [15:31] yep :) [15:31] kenvandine: in the future, if you change something in the template, like add/remove a project, it will update it with the same command [15:33] (it's also setting up bzr to bind to our needs) [15:33] should it be the .1 or .6 IP in my cred file? [15:34] .1 [15:34] nevermind [15:34] i looked at the wrong browser tab :) [15:34] * didrocks sees someone who never configured his cred file :p [15:34] kenvandine: the file should be ~/.cu2d.cred [15:38] humm... said it couldn't find my credentials [15:39] ~/.cu2d.cred ? [15:39] kenvandine: ^ [15:39] yeah [15:40] oh... i don't remember where i got the token in there... but it is the same as the one in the email sent to all of us [15:40] so i guess that isn't mine :) [15:40] kenvandine, token is from your jenkins user page [15:41] kenvandine: no, its the same [15:41] kenvandine: you do have an account on jenkins, right? [15:41] yes [15:41] i am logged in [15:41] kenvandine: can you try -C ? [15:41] 2013-03-07 10:36:40,690 ERROR Credentials not found. Aborting! [15:41] so it didn't find it... but it's right [15:41] ok [15:42] still says not found [15:43] kenvandine: do you mind pasting me the cred file content you have? [15:43] ok [15:43] let me try also [15:43] reconfigured webapp, working here [15:43] so it's not fginther breaking everything :p [15:44] whew! [15:45] mterry: hey hey hey! [15:46] mterry: now that cyphermox published the indicators, I think you can publish unity :) [15:46] we are at 199 daily landing right *now* [15:46] you can have the 200th upload! [15:47] cyphermox, ah. the test failures were no big deal? [15:47] didrocks, wake me when it's 1000th [15:47] :) [15:47] mterry: nah [15:47] :) [15:50] cyphermox, didrocks : done. Today is FF, right? I'm still unclear if we are doing 13.04 or not, but if we are, we should communicate that to the unity team [15:50] indeed [15:51] didrocks, also... I think as an archive admin, you can demote ubuntuone-couch if you like. I was just reminded of that fact in #ubuntuone [15:51] mterry: yeah, sent an email this morning to the team leads [15:51] so far it is unclear whether we will release 13.04 formally so carrying on with FF as usual [15:51] mterry: sorry, I should have CC you guys [15:51] cyphermox: kenvandine robru ^ [15:51] didrocks, frankly, it can probably be dropped from the archive... [15:51] mterry: oh yeah, demoting the couch! [15:51] let's demote for now :) [15:51] * cyphermox is going to head to uni shortly (over lunch) to do one last NM test before uploadin [15:52] :) No rush on archive, as long as it's out of main [15:52] mterry: flushed from main, in universe now :) [15:52] didrocks, sweet [15:55] mterry: there is one branch that I expect larsu will want to have land today; needing a FFE and all [15:55] but it's also needing two MIRs, would you be available later to review them? [15:55] I'm going to check if they have been filed already, otherwise I'll take care of it [15:56] cyphermox, sure [15:56] cyphermox, shouldn't need an FFe if everything happens before some hour today... don't remember when [15:56] well yeah [15:57] but it's indicators, and I would really prefer if it went through the grinder [15:57] rather than uploading manually... [15:57] yeah, manual upload is so old… [15:57] hehe [15:57] old school... [15:58] and let's welcome the friends stack to daily landing! [15:58] I mean, whoever even still speaks of debdiffs anymore? ;) [15:58] yay [15:58] cyphermox: exactly! [15:59] cyphermox, well, you could start a build and all :) [15:59] yeah [15:59] just it would be tight [16:00] and the MIRs need to be done too :) [16:00] * mterry whips out his rubber stamp [16:01] j/k! j/k === dandrader is now known as dandrader|afk [16:03] didrocks, is dee-qt setup to automerge? [16:04] seb128: larsu: would have been cool to have a bug attached to your change to not end up with an empty changelog :p [16:04] kenvandine: no, nor daily landing [16:04] kenvandine: do you think we should attach that to an existing stack? [16:04] not sure where it fits [16:04] yeah, I'm wondering… [16:04] dee-qt deps on dee? [16:04] i'll need it when i add gwibber in [16:05] yes [16:05] the core apps will need it too [16:05] kenvandine: yeah, I think we'll have an infrastructure stack [16:05] i'm sure we'll have a stack for those [16:05] dee will move to it [16:05] as libunity [16:05] I guess [16:05] ok, lets just get this merged and uploaded to raring [16:05] kenvandine: let's flesh out the stack story as discussed in the session by next week [16:06] didrocks, can you approve the MP and i'll push to trunk? [16:06] and upload [16:06] kenvandine: sure, one sec [16:07] kenvandine: done :) [16:09] grrr [16:09] Unable to obtain lock held by didrocks@bazaar.launchpad.net on taotie (process #14998), acquired 342 hours, 18 minutes ago. [16:09] didrocks, can you break your lock? [16:09] kenvandine: on dee-qt? [16:09] yeah [16:09] lp:dee-qt? [16:09] waow [16:09] I did push that at some point? [16:09] crazy… ;) [16:09] kenvandine: done [16:10] ok, pushed and uploaded [16:10] i guess that'll go to sourceNEW [16:10] for the rename [16:11] hum, not sure [16:11] binNEW for sure [16:11] source new, it was in precise… [16:12] cyphermox, what's with the misc stack? === mmrazik|afk is now known as mmrazik [16:17] mterry: there was a merge that wouldn't complete [16:17] for various reasons ;) [16:17] kenvandine: I'll put the autolanding bits and pieces later today... [16:18] great [16:18] fginther: mmrazik: objection if I rename all -head file to ? [16:18] didrocks, not from me [16:18] didrocks: fine with me [16:18] let's go then :) [16:19] mmrazik, for friends, libfriends, qml-friends and dee-qt please [16:19] kenvandine: its just one cfg file, right? [16:19] mmrazik: you have the "friends cfg file" [16:19] alesage: FYI ^^^ I think we already have autolanding etc for some of those. I'll delete them [16:19] we aren't doing daily releases for qml-friends and dee-qt yet [16:19] mmrazik: but we only daily release libfriends and friends [16:19] but we want to get the merger merging them [16:20] kenvandine: ack [16:20] so I guess I need to add an option for that :) [16:20] kenvandine: there is an option for that [16:20] (i.e. option that will make the daily release machinery to ignore a particular project but still have autolanding etc) [16:20] mmrazik, ok [16:22] oh right daily_release_default [16:22] and then daily_release [16:23] mmrazik: do you mind adding false for them to the two other components than libfriends/friends only? [16:23] s/only// [16:23] didrocks: ack [16:23] thanks a lot [16:47] grrr, hw issues for jenkins test builders [17:10] kenvandine: dee-qt should be part of the friends stack? [17:24] good morning didrocks ! ;-) [17:24] hey robru! How are you? [17:24] sleepy ;-) [17:24] and you? [17:24] robru: tired :-) [17:25] robru: I'll have the list of WI setup for tomorrow I guess, I had other things with feature freeze to deal with [17:25] didrocks, we have so much in common ;-) [17:25] robru: heh, isn't it? ;) [17:25] didrocks, ok, looking forward to seeing the workitems. [17:26] :-) [17:27] mmrazik, no, not sure where it belongs right now [17:27] kenvandine, didrocks: do you mind if I put it into friends.cfg? [17:27] i don't mind [17:27] mmrazik: for now no, but I'm sure it will move next week [17:27] ultimately it needs to be in like some infrastructure stack [17:27] I don't want to move the unity stack to cupstream2distro-config based config [17:27] mmrazik: is it a problem for you? [17:28] didrocks: well... we still find bugs and issues in the ci/autolanding templates etc [17:28] mmrazik: I mean, moving it from one stack file to another one [17:28] didrocks: but I'm using it for mir so it seems to stabilize :) [17:28] didrocks: I'm generating the ci/autolanding jobs per stack [17:29] didrocks: if I move dee-qt to unity stack and want autolanding I will need to regenerate autolanding jobs for the full stack [17:29] more or less [17:29] mmrazik: oh sure, TBH I think it will be in a new stack on its own [17:29] didrocks: in that case its cool [17:29] great :) [17:31] Hey, does anyone know what PPAs are currently required to build unity/phablet on raring? (the build script seems outdated) [17:54] grrrr === dandrader|afk is now known as dandrader [17:56] * sil2100 for now is out of ideas [17:57] didrocks: one possibility for the failure is that somehow, magically, the lenses_ list in Unity gets emptied and simply the first Super+a does not return the lens, but triggers re-scanning of the lenses directory [17:57] ...or something [17:58] I am unable to reproduce it in the way it happens in the test :< [17:58] didrocks, kenvandine: so besides the fact ps-jenkins can't review nor merge (no membership in super-friends) the autolanding setup is ready [17:58] sil2100: maybe you can get some help from bregma's team? [17:58] mmrazik: thanks! [18:10] robru, kenvandine: this looks like a missing build-deb to me: [18:10] https://jenkins.qa.ubuntu.com/job/friends-raring-amd64-ci/1//console [18:11] mmrazik, yes, it does look that way [18:11] mhm... but python3-setuptools is in there [18:11] it can't be... it builds in raring [18:12] could that be missing from the jenkins environment where it does ci? [18:12] kenvandine: right... this is outside the chroot. we need to install python3-setuptools on the builders [18:12] mmrazik, http://bazaar.launchpad.net/~super-friends/friends/trunk/view/head:/debian/control yeah, all that stuff it's complaining about missing is listed as a build dep, so I dunno what's gone wrong [18:12] kenvandine, robru: sorry for the buzz [18:12] mmrazik, no worries === zniavre__ is now known as zniavre [18:18] robru: the build now passes. Jenkins votes needs fixing because of the commit message. The commit message is used for the automatic merge commit during autolanding. [18:19] fixing the commit message is all you need to do. Once the branch is approved jenkins will pick it up again [18:19] mmrazik, ok, which branch? [18:19] err.sorry [18:19] robru: you will get an e-mail from launchpad :) [18:19] https://code.launchpad.net/~robru/friends/purge-accounts/+merge/151657 [18:19] oh god, that branch isn't ready to land yet! [18:20] ;-) [18:20] there are some important changes there, but it doesn't yet solve the problem that it set out to solve. [18:20] robru: if you push a new revision jenkins will notice and will run the tests again [18:20] I only mp'd it as an easy way for ken to review it. [18:20] robru: it will only merge it once the global state of the MP is Approved [18:22] mmrazik, ok, I'll be working on that one today. please don't allow any other mps to be merged just yet either, I'm a little bit notorious for mp'ing things prematurely just to get other people's input on the work. === mmrazik is now known as mmrazik|afk === mmrazik|afk is now known as mmrazik [18:22] robru: don't be stressed about the jenkins messages. As long as the global state is not approved they will only comment to provide early feedback [18:22] ok [18:22] mmrazik, thanks === mmrazik is now known as mmrazik|afk [18:40] robru, i won't approve that :) [18:45] kenvandine, I think we need some kind of "staging" trunk where I can push things, and then have the launchpad recipe build test packages into a PPA, before pushing things to trunk and to raring [18:45] why? [18:46] that kind of defeats the purpose :) [18:46] we don't need the daily ppa anymore [18:46] kenvandine, just to make it easier to test things before pushing broken crap onto all raring users ;-) [18:46] well, not for friends and libfriends [18:46]