=== chihchun_afk is now known as chihchun
[05:36] <robru> Mirv: I don't know if it makes sense for you, but dual silos can now have manual sources, so if you're doing the same qt packages in vivid and wily, the train can handle that in a single silo now
[05:37] <Mirv> robru: yeah I was thinking about it, it sounds really good. so far I've needed two silos for all OTA Qt fixes!
[05:37] <Mirv> robru: it's not the same packages, but same patch being applied. wily has 5.4.2, vivid 5.4.1. but I guess since I upload the manually train is fine with that?
[05:38] <robru> Mirv: yeah train won't enforce the versions matching or anything. it'll just watch the builds & do the publishing
[05:38] <Mirv> robru: the train rocks!
[05:38] <robru> Mirv: thanks. long way to go still ;-)
[05:39] <robru> Mirv: I'll be around for a bit (just playing video games). ping me if anything explodes
[05:39] <Mirv> robru: ok, have fun
[05:39] <robru> Thanks
=== chihchun is now known as chihchun_afk
=== marcusto_ is now known as marcustomlinson
=== chihchun_afk is now known as chihchun
[07:58] <Saviq> Ursinha, we don't need anything special for our qml tests these days. we run them all in xvfb so we only basically need CPU
[08:05] <seb128> hey there
[08:05] <seb128> seems like there is an issue with the uitk/rc-proposed that makes some store apps not start
[08:05] <seb128> bug #1503901
[08:05] <ubot5`> bug 1503901 in ubuntu-ui-toolkit (Ubuntu) "Authenticator app won't open with latest SDK" [Undecided,New] https://launchpad.net/bugs/1503901
[08:05] <seb128> should probably be an ota7 blocker
[08:07] <seb128> Mirv, ^
[08:07] <Mirv> seb128: targeting so
[08:07] <seb128> Mirv, thanks
[08:12] <jibel> seb128, thanks for the bug, continue testing OTA7. ;)
[08:12] <seb128> jibel, yw!
[08:13] <sil2100> Blockers make me sad
[08:13] <jibel> seb128, did you reproduce it?
[08:13] <jibel> it works fine here
[08:15] <seb128> no, just tried and it works (but I don't have an active config in that app so maybe the clear state is not enough to hit the issue), but mzanetti did apparently, he is discussing in on #sdk
[08:16] <mzanetti> yes, found the issue
[08:16] <mzanetti> http://paste.ubuntu.com/12712813/
[08:16] <mzanetti> here's a standalone example to repro it
[08:23] <jibel> mzanetti, it is a problem in the sdk or the app?
[08:24] <mzanetti> well, the app doesn't have the prettiest code ever in there, but nevertheless it should not crash the SDK
[08:24] <mzanetti> as this was working until now I guess it should be fixed in the sdk
[08:24] <jibel> okay
[08:25] <seb128> the code does nothing buggy/illegal
[08:25] <jibel> bzoltan_, can you look at bug 1503901
[08:25] <ubot5`> bug 1503901 in Canonical System Image "Authenticator app won't open with latest SDK" [Critical,New] https://launchpad.net/bugs/1503901
[08:25] <jibel> bzoltan_, it'll block OTA7 as it breaks existing apps
[08:27] <jibel> mzanetti, can you update the bug report with your findings please.
[08:27] <mzanetti> done
[08:27] <jibel> thanks
[08:56] <bzoltan_> Mirv:  how happy the CI train is with git repos?
[08:57] <anpok> cihelp: the krillins used by our mir-mediumtests-runner-touch are those flashed with rc-proposed images?
[08:59] <psivaa> anpok: yes,
[08:59] <psivaa> rc-proposed/ubuntu/krillin/version-299 was the latest
[08:59] <sil2100> bzoltan_: not very happy I suppose
[08:59] <sil2100> bzoltan_: it's still hard coded to bzr from what I know
=== chihchun is now known as chihchun_afk
[09:00]  * psivaa -> errands
[09:01] <anpok> psivaa: thx..
=== chihchun_afk is now known as chihchun
[09:03] <zbenjamin> sil2100: sigh :(
[09:04] <zbenjamin> sil2100: any chance this will change? Like now? :D
=== chihchun is now known as chihchun_afk
[09:11] <sil2100> zbenjamin: it's in the plans, yes
=== chihchun_afk is now known as chihchun
[09:14] <Mirv> bzoltan_: same answer as sil2100
[09:15] <bzoltan_> Mirv:  thanks
=== sil2100 changed the topic of #ubuntu-ci-eng to: Train trouble? ping trainguards | CI problems? ping cihelp | Train: http://bit.ly/1hGZsfS | QA Signoffs: http://bit.ly/1qMAKYd | Known Issues: OTA-7 #1 candidate testing in progress, landing gates open
[09:30] <zbenjamin> sil2100: that would be awesome
[09:34] <seb128> jibel, oSoMoN_, shouldn't the settings from silo 49 land for ota7 if the confined webbrowser landed? (it has a fix needed to make the non click .desktop be used)
[09:35] <oSoMoN_> seb128, it would be nice to have indeed, but not essential I would say
[09:35] <seb128> oSoMoN_, well, the silo is safe and waiting for qa since monday and fix an annoying bug
[09:35] <seb128> unsure why it didn't get verified/published
[09:36] <oSoMoN_> seb128, not sure either, but it’s tagged for next milestone, that would explain why QA didn’t validate it
[09:36] <seb128> where are things "tagged"?
[09:37] <oSoMoN_> seb128, in the trello board that tracks QA’s work, and I believe the tags are assigned based on the LP milestone of the corresponding bug reports
[09:37] <seb128> where is that trello again?
[09:37] <oSoMoN_> jibel, can you confirm the above?
[09:37] <mzanetti> jibel, do you know if this is going to be included for OTA-7 https://bugs.launchpad.net/ubuntu/+source/network-manager/+bug/1480877
[09:37] <ubot5`> Ubuntu bug 1480877 in location-service (Ubuntu) "Access points' "PropertiesChanged" dbus signals freeze UI on mobile devices" [Undecided,Incomplete]
[09:37] <oSoMoN_> trello.com/b/AE3swczu/silo-testing
[09:37] <seb128> thanks
[09:58] <jibel> mzanetti, the fix for network-manager is in OTA7, not sure if there is also a fix need for location-service and indiator-network
[09:59] <jibel> needed*
[10:01] <jibel> seb128, it didn't land because none of the fixes were targeted for this OTA
[10:02] <mzanetti> jibel, ah ok, great... just asking because it rendered my phone completely useless on the conference I've been the last two days
[10:02] <mzanetti> but I wasn't up to date with the very latest image either
=== oSoMoN_ is now known as oSoMoN
[10:23] <Ursinha> Saviq: thank you
[10:24] <Saviq> Ursinha, basically, the only need we have there is that Jenkins runs our autopkgtests, which now run all our QML tests
[10:26] <Ursinha> Saviq: right, we were trying to understand if hardware was required for the slaves, if not it's easier
=== chihchun is now known as chihchun_afk
[11:25] <bzoltan_> sil2100: Mirv: the bugfix branch for #1503901 is already in silo. I have fresh archive tests, so i expect the silo to be QA ready tomorrow monring.
[11:26] <bzoltan_> mup: are you lazy or what?
[11:26] <bzoltan_> https://bugs.launchpad.net/ubuntu/+source/ubuntu-ui-toolkit/+bug/1503901
[11:26] <ubot5`> Ubuntu bug 1503901 in ubuntu-ui-toolkit (Ubuntu) "Authenticator app won't open with latest SDK" [Critical,In progress]
[11:27] <jibel> bzoltan_, not before tomorrow morning?
[11:27] <jibel> bzoltan_, we already started regression testing and an update of uitk invalidate almost everything
[11:27] <bzoltan_> jibel: The build is ready in half an hour
[11:28] <jibel> k
[11:29] <bzoltan_> jibel:  it is a bug what was reported 2am my time. It was tagged up just 2 hours ago. I have the fix in silo already.
=== alan_g is now known as alan_g|lunch
[11:31] <bzoltan_> jibel: popey: I could use a good list of click packages what come with even the simplest AP test to extend the UITK test plan.
[11:32] <bzoltan_> popey: I would wget the click packages, install on the device and run the AP tests if they have such
[11:32] <Mirv> bzoltan_: thanks!
[11:43] <sil2100> bzoltan_: thanks!
=== _salem is now known as salem_
[11:58] <seb128> jibel, ok, should we land it still? or declare it a minor bug with the new feature and delay to next ota update?
=== alan_g|lunch is now known as alan_g
[12:00] <rvr> popey: Are you in charge of reminders-app (Notes)?
[12:03] <popey> rvr: wassup?
[12:03] <rvr> popey: There are some missing strings in the app
[12:04] <rvr> popey: A language update would be cool
[12:05] <rvr> popey: For example, in Spanish, "close" button is presented in English. Also, the Evernote prompt at first start also has a phrase still in English.
[12:05] <rvr> popey: I think both are now translated in Launchpad
[12:06] <popey> rvr: Ok, I'll put a call out for language update, thanks rvr
[12:10] <jibel> seb128, bug 1501428? I don't think it's minor as it exposes wrong or missing information in security settings. We should try to land it and include it iun next respin but I'll defer to Pat for the final decision
[12:10] <ubot5`> bug 1501428 in ubuntu-system-settings (Ubuntu) "TrustStoreModel fails to expose application name for non-click app" [High,In progress] https://launchpad.net/bugs/1501428
[12:12] <seb128> jibel, right, good ... just to know, what should be done next time to flag the silo as "should be consider for the coming ota"?
[12:12] <seb128> because I failed at that
[12:12] <seb128> IRC naging doesn't seem to be an answer either
[12:12] <seb128> (well it works, but we shouldn't rely on that)
[12:13] <jibel> seb128, add a canonical-devices-system-image task. It'll be reviewed by the project team and targeted for an ota
[12:13] <seb128> k
[12:13] <seb128> thanks
[12:14] <jibel> seb128, if it's not reviewed on time and you think it's important pinging people helps
[12:14] <jibel> seb128, the MR is not reviewed and top approved
[12:14] <jibel> https://code.launchpad.net/~ken-vandine/ubuntu-system-settings/lp1501428/+merge/273066
[12:16] <seb128> jibel, that's not in the landing/silo?
[12:16] <seb128> jibel, https://requests.ci-train.ubuntu.com/#/ticket/452
[12:17] <seb128> jibel, we wanted to land the safe/simpler version of the fix which is in the silo for that ota and review/land Ken's change for next one
[12:18] <jibel> seb128, right, sorry, I was checking the ones attached to the bug
=== alan_g is now known as alan_g|lunch
[13:33] <jibel> pmcgowan, what do you think about landing silo 49 in OTA7? it contains a fix for bug 1501428 ie the item is blank for the webbrowser in security settings.
[13:33] <ubot5`> bug 1501428 in ubuntu-system-settings (Ubuntu) "TrustStoreModel fails to expose application name for non-click app" [High,In progress] https://launchpad.net/bugs/1501428
[13:33] <jibel> pmcgowan, other fixes in the silo are minor and low risk
[13:34] <pmcgowan> looking
[13:35] <jibel> we blocked ota6 for a similar issue with the camera
[13:38] <pmcgowan> jibel, yes looks good to me
[13:38] <pmcgowan> fall out of previous landing
[13:38] <pmcgowan> I will tag the other bugs too
[13:41] <mzanetti> pmcgowan, I think we want to fix this too: https://bugs.launchpad.net/ubuntu/+source/unity-system-compositor/+bug/1502145
[13:41] <ubot5`> Ubuntu bug 1502145 in unity-system-compositor (Ubuntu) "rc-proposed r140, krillin: screen does not blank after timeout expires" [Undecided,Confirmed]
[13:45] <pmcgowan> mzanetti, but there is no fix there :(
[13:45] <pmcgowan> is that a recent regression I assume?
=== alan_g|lunch is now known as alan_g
[14:07] <pmcgowan> mzanetti, ?
[14:09] <mzanetti> pmcgowan, no, no fix yet, but the issue is quite critical IMHO
[14:10] <pmcgowan> mzanetti, and it can be reproduced?
[14:11] <mzanetti> pmcgowan, my krilling here seems to turn itself on all the time and never turn off again
[14:11] <pmcgowan> just krillin or do we know
[14:11] <pmcgowan> I am not seeing on mx4
[14:12] <mzanetti> I'll try repro with something else
[14:18] <ChrisTownsend> trainguards: Hi, we no longer need https://requests.ci-train.ubuntu.com/#/ticket/401 since the change in that silo was merged in a different request yesterday.  Do I just click "Abandon" to get rid of it?
[14:19] <sil2100> ChrisTownsend: yes, should be enough :)
[14:19] <ChrisTownsend> sil2100: Ok, thanks
[14:19] <sil2100> jibel, davmor2, rvr: is ubuntu/landing-023 in the queue to land?
[14:19] <sil2100> Since it's UITK so I suppose it's good to land this ASAP, right?
[14:20] <jibel> sil2100, not marked ready for QA but we will take it
[14:20] <sil2100> Ah, right, it's QA Required still
[14:20] <sil2100> bzoltan_: piing
[14:22] <bzoltan_> sil2100:  tell me
[14:22] <sil2100> bzoltan_: how's silo 23 going? Ready for QA sign-off soon?
[14:23] <bzoltan_> sil2100: Hmm... do you guys want me to run a real UITK Test Plan? That takes ~24h. The archive test to compare the silo tests with I have started 10 hours ago and in few hours it is ready... teh base
[14:24] <bzoltan_> sil2100:  but the build is there, I ca do sanity check on it... but that is not a proper UITK test plan
[14:24] <sil2100> jibel: ^ what does QA think? How much testing would you like on that one?
[14:25] <bzoltan_> sil2100: jibel: the change compate to the released version is subzero
[14:25] <bzoltan_> sil2100: jibel: it is a cherry pick landiing with few lines of code change
[14:27] <jibel> bzoltan_, what is the risk of potential regression introduced by this change?
[14:31] <bzoltan_> jibel: less then any landing before... check the diff, it is super minimal.
[14:32] <jibel> bzoltan_, I saw the diff but what is the impact on the rest of the toolkit? to target testing around this area instead of re-running things that are not impacted at all
[14:32] <bzoltan_> jibel: I do the test plan anyway tonight .. so I will tell in the morning if I find something. But I would suggest to proceed with this version
[15:20] <greyback> lol at citrain page with JS disabled
[15:27] <anpok> hehe
[15:32] <mzanetti> pmcgowan, so far no luck reproducing on mako.. still happening on my krillin all the time
[15:35] <pmcgowan> mzanetti, how do you trigger it?
[15:36] <mzanetti> pmcgowan, I don't do anything special... I just upgraded to the latest rc-proposed last night and it happens
[15:38] <pmcgowan> jibel, ^
[15:39] <mzanetti> faenil reported this on monday already... I just didn't upgrade earlier because I was travelling
[15:47] <Saviq> sil2100, hey, do you know what's the status of dbgsym packages in PPAs? I read somewhere LP supports publishing those now, but has to be enabled? could we enable in silos and overlay?
[15:48] <sil2100> Saviq: the overlay has those enabled IIRC
[15:48] <sil2100> Not sure about silos tho
[15:50] <Saviq> sil2100, hmm I wonder, wouldn't it mean train has to copy them actively from silos?
[15:50] <Saviq> sil2100, like, overlay can only have them if it builds itself, when copied, they have to be there in silos in the first place?
[15:52] <Saviq> sil2100, oh, silos do have them, but maybe not in the archive? apt can't see them fwiw
[15:53] <sil2100> Saviq: http://ppa.launchpad.net/ci-train-ppa-service/stable-phone-overlay/ubuntu/dists/vivid/main/debug/ <- the overlay seems to have them too
[15:54] <Saviq> sil2100, ok, but they're not added to Packages?
[15:54] <Saviq> i.e. I can't simply use apt to install them?
[15:54] <Saviq> or should I add main/debug?
[15:54] <Saviq> think that's what I was missing
[15:54] <Saviq> sry
[15:54] <Saviq> didn't know how to work with that yet
[16:01] <pmcgowan> mzanetti, sorry was otp, do you just turn the phone on and it doesn't timeout?
[16:01] <mzanetti> pmcgowan, yep
[16:02] <mzanetti> pmcgowan, also, if I receive a notification that turns it on, it won't time out either
[16:02] <pmcgowan> mzanetti, jibel could not reproduce it which may be a clue in itself
[16:03] <pmcgowan> what does upower say its state is
[16:05] <mzanetti> pmcgowan, seems u-s-c is holding it awake. alf is looking into it too
[16:06] <pmcgowan> ok
[16:06] <seb128> does anyone see why https://ci-train.ubuntu.com/job/ubuntu-landing-009-1-build/369/console is failing?
[16:07] <seb128> hum
[16:07] <seb128> ./debian/rules: Permission denied
[16:07] <seb128> wth?
[16:13] <seb128> cihelp, ^ (or is that a trainguards issue?)
[16:14] <fginther> seb128, that would be trainguard
[16:14] <seb128> fginther, thanks
[16:34] <robru> seb128: sorry was in meeting
[16:34] <seb128> no worry
[16:35] <seb128> seems like CI build doesn't like the fact that indicator-session debian/rules is not +x
[16:35] <seb128> but that was not an issue before
[16:35] <seb128> and dpkg-source and other tools usually fix that
[16:35] <seb128> unsure what changed that makes it a problem
[16:36] <robru> seb128: what changed if we started calling "rules clean" before builds
[16:36] <seb128> seems buggy
[16:36] <seb128> and why?
[16:36] <robru> seb128: i can make it chmod first but it's easier if you just set the permission for now
[16:36] <seb128> I did
[16:37] <seb128> still a bug in the system
[16:37] <robru> seb128: we did this so people could hook into that and munge their control files prior to building, eg so they can have different abi from same source package
[16:37] <seb128> I see
[16:38] <seb128> well in which case adding the chmod +x seems like a good thing to do
[16:38] <seb128> or ignoring the failing clean calls
[16:39] <robru> seb128: yeah i can fix it but will take some time to get the fix in production
[16:39] <seb128> that's ok, I did a chmod +x mp for indicator-session meanwhile
[16:39] <robru> Ok
[16:39] <seb128> robru, thanks
[16:40] <robru> You're welcome
=== alan_g is now known as alan_g|EOD
[17:36] <bzoltan_> robru: jibel: rvr: The silo23 with the tiny fix is good for QA validation
[17:42] <rvr> jgdx: ping
[17:44] <jibel> bzoltan_, good, alesage is on it
[17:45] <alesage> jibel, bzoltan_ ack
[17:45] <rvr> jgdx: "Open System Settings and call a SIM Service number containing a #, e.g. +47005#1"
[17:45] <bzoltan_> alesage: jibel:  super, thank you. Feel free to ping me, I try respond even during the night
[17:46] <rvr> jgdx: I think my SIM service numbers don't have any hash characters
[19:27] <jibel> trainguards you can publish silo 23 it is for ota7
[19:40] <robru> mterry: around for a packaging ack? https://requests.ci-train.ubuntu.com/#/silo/23
[19:40] <mterry> robru, hiyo sure
[19:40] <robru> thanks
[19:41] <robru> oh hah, it's just the trivial debian/watch
[19:41] <robru> I should fix the train to not consider that as requiring an ack...
[19:42] <robru> slangasek: can you remind me again why the -gles packages need to be separate source packages and can't just be build from the same sources?
[19:42] <mterry> robru, yeah
[19:43] <robru> mterry: need you to publish for now please
[19:43] <mterry> robru, was about to ask if I was looking at right diff  :)
[19:43] <mterry> robru, I did
[19:43] <mterry> oh whoops
[19:43] <mterry> now I did
[19:44] <robru> mterry: yeah the first time you click build it just logs you in and redirects to the form
[19:44] <robru> mterry: I'll be so happy when we finally get rid of jenkins
[19:51] <jgdx> rvr, right, it's difficult to test
[19:58] <rvr> jgdx: So, what can I do?
[20:01] <slangasek> robru: you can't build -gles and non-gles as part of the same source package because of conflicting build-deps
[20:02] <robru> slangasek: why do the build deps conflict? What if we made the build deps coinstallable so that we could unify the source packages?
[20:03] <robru> slangasek: what if we made a debian/rules that installed one build dep for over build then switched to the other for the other build? ;-)
[20:03] <slangasek> robru: they conflict because GL and GLES have conflicting APIs and there's no reasonable way to make them coinstallable without doing a lot of build system work
[20:03] <slangasek> robru: then that debian/rules would be shot out of a cannon
[20:03] <robru> Hehe
[20:04] <jgdx> rvr, I'm thinking
[20:04] <robru> slangasek: what if the rules file did the builds in two separate chroots with the different build deps?
[20:05] <slangasek> robru: you don't get to do anything that requires root in a debian/rules.
[20:05] <robru> slangasek: what if we invented a user mode chroot? ;-)
[20:07] <robru> slangasek: what if, instead of the train enforcing users to manage gles twins, the train just made gles copies similarly to the way it makes vivid copies. Would reduce a lot of hassle for people i think
[20:09] <jgdx> rvr, are you able to make the device rw to test this?
[20:09] <rvr> jgdx: Yes
[20:10] <jgdx> rvr, okay, then I guess there's a way. You have to +rw and install python3-dbusmock
[20:13] <rvr> jgdx: Checking
[20:19] <jgdx> rvr, next up, stop ofono. $ sudo stop ofono ?
[20:19] <jgdx> let me know how it goes
[20:21] <rvr> jgdx: ofono stop/waiting
[20:24] <jgdx> rvr, okay, then $ sudo python3 -m dbusmock --template ofono
[20:24] <jgdx> and in a new terminal, do phablet-shell and then $ gdbus call -y -d org.ofono -o /ril_0 -m org.freedesktop.DBus.Mock.AddProperty "org.ofono.SimManager" ServiceNumbers "<{'Foobar': '+47#000'}>"
[20:24] <jgdx> should output ()
[20:24] <jgdx> open system settings and test
[20:25] <rvr> hmm
[20:28] <rvr> Done
[20:28] <rvr> ()
[20:29] <jgdx> rvr, okay, System Settings -> Phone -> Service numbers -> Foobar -> Call
[20:30] <rvr> aha!
[20:30] <rvr> Done
[20:30] <jgdx> the dialer should have +47#000
[20:30] <rvr> Yes
[20:30] <jgdx> not just +47
[20:31] <rvr> Right!
[20:33] <kgunn> robru: you'd be the bomb if you automated twins
[20:36] <robru> kgunn: I really hate the way gles are handled currently and I'd love to clean that up
[20:36] <robru> kgunn: like, it really bothers me on a deep and personal level ;-)
[20:36] <kgunn> :)
[20:36] <robru> kgunn: only because I know that a better way exists.
[20:37] <robru> kgunn: can't guarantee a timeline but I'm looking into solutions
[20:38] <kgunn> me likie
[20:38] <jgdx> rvr, what'd you get
[20:38] <rvr> jgdx: Yes, I got +47#000
[20:39] <jgdx> rvr, wheee! Thanks for you patience
[20:39] <rvr> jgdx: I'm going to do some additional checks before approving the silo
[20:40] <jgdx> okay, I'm going to bed. Night!
[20:41] <rvr> jgdx: Good night!
[20:44] <robru> kgunn: did you ever notice that the train can't generate diffs for the non-gles version of packages that have -gles twins? I just noticed today and it doesn't seem like it's a result of anything I've changed recently, but I have a hard time believing that the train always failed to diff ubuntu-ui-toolkit
[20:51] <kgunn> huh...i never noticed
[20:53] <robru> kgunn: it's due to the watch file... when it builds the gles variant, it downloads the orig.tar over the non-gles one, then renames it to the -gles one, leaving the non-gles one without an orig.tar to diff. pretty basic stuff, it's been using debianwatch files all along...
[20:53] <robru> kgunn: but yeah I never noticed until today. there must have been diffs in the past...
[21:24] <robru> wat
=== salem_ is now known as _salem