[02:10] === trainguard: IMAGE 39 building (started: 20140520 02:10) === [03:30] === trainguard: IMAGE 39 DONE (finished: 20140520 03:30) === [03:30] === changelog: http://people.canonical.com/~ogra/touch-image-stats/39.changes === === wgrant changed the topic of #ubuntu-ci-eng to: Launchpad offline 06:00 - 06:30 UTC | Ubuntu CI Engineering Team | Vanguard: cihelp | CI Train support - US: robru, cyphermox, rsalveti - EU: sil2100, Mirv, didrocks | CITrain support no answer: use mup bot after 30 minutes, but choose right timezone | Known issues: - === wgrant changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cihelp | CI Train support - US: robru, cyphermox, rsalveti - EU: sil2100, Mirv, didrocks | CITrain support no answer: use mup bot after 30 minutes, but choose right timezone | Known issues: - [06:34] Morning all [06:37] mornio [06:37] so weird to see a davmor2 at this time! :) [06:45] davmor2: do we have a bug for the mis-shapen previews in the video lens? [06:45] I can't remember what i did and didnt file yesterday [06:58] sil2100: morning. bug 1321138 [06:58] bug 1321138 in media-hub (Ubuntu) "Video playback choppy then freezes #37 mako" [Undecided,New] https://launchpad.net/bugs/1321138 [06:58] popey: hey, the sprint is starting at 8:30? [06:59] it started 30 mins ago, yes [07:00] oh, I know for who (asac), this will be hard next week :p [07:13] sil2100: bug 1321146 [07:13] bug 1321146 in unity8 (Ubuntu) "Video scope has malformed thumbnails #37 mako" [Undecided,New] https://launchpad.net/bugs/1321146 [07:32] popey: morning! [07:32] uh [07:33] Not looking good [07:33] yeah, i suspect some of these issues will magically disappear with a reboot [07:33] root@ubuntu-phablet:~# uptime 09:33:42 up 15:20, 1 user, load average: 0.21, 0.35, 0.5 [07:33] barry, really? let me check [07:42] popey: the video playback one worries me the most - can you try reconfirming it after a reboot? [07:43] doing now [07:47] sil2100: bug 1321161 [07:47] bug 1321161 in unity8 (Ubuntu) "Unlocking welcome screen with two fingers doesn't unlock properly" [Undecided,New] https://launchpad.net/bugs/1321161 [07:50] popey: now this sounds funny ;) === psivaa changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: psivaa | CI Train support - US: robru, cyphermox, rsalveti - EU: sil2100, Mirv, didrocks | CITrain support no answer: use mup bot after 30 minutes, but choose right timezone | Known issues: - [08:30] ogra_, psivaa, Mirv: firefox crashed... [08:30] bad firefox [08:31] naughty fox [08:50] popey: look over here [08:50] balloons, can you send me a link to your autopkgtest notes from this morning? [08:51] balloons, I took some if it helps: https://docs.google.com/a/canonical.com/document/d/1gfOwR5ih1o6vb7Vgj4TjTN5TIWPKPxRjfVEg-8WAlxk/edit# [08:55] fginther, yes I can share my doc also [08:56] davmor2: look over here :þ [09:02] fginther, https://docs.google.com/a/canonical.com/document/d/1AKkKHMOUBAHmRUKJRgsPZaZia4kfoH0kN5-kuxtgXZE/edit [09:02] fginther, we should probably combine notes and do a proper write-up [09:03] balloons, agreed, we can sync up later today [09:06] sil2100: so the rerun completed with more clock app failures and the same number(4) of galler app failures [09:06] ouch [09:06] Mirv: ^ [09:06] psivaa: thanks! [09:06] the clock app has 8 failures in the run. and Mirv knows about it [09:06] :) [09:06] davmor2: it was white while running ap tests ☹ === vrruiz_ is now known as rvr [09:18] sil2100: yeah, that's ouch. renato has a meeting of sort (well, now he sits back there) and we'll try to have a meet with nik90_ after lunch or so about the clock app [09:19] Mirv: thanks, yeah, nik90_ should know more [09:19] Would be nice to get the number down, not up ;p [09:19] it's weird though, is it something maybe that qtpim is slower and that's why it's failing, or what's it about? since it doesn't seem qtpim itself is failing in any way. [09:19] sil2100: I thought a rerun would bring it down! :) [09:20] ;p [09:23] Mirv, sil2100: hey was just looking at your conversation regarding clock app :) [09:23] so I did have some amount of failing tests with clock myself too, but a rerun brought it down to the usual suspect of 1 failing test [09:24] nik90_: hey! yes we highlighted you :) me and renato were thinking about meeting you to get your wise thoughts on why the tests might be failing. [09:24] Mirv: do you have the link to the jenkins clock app test failures results? [09:24] nik90_: http://ci.ubuntu.com/smokeng/utopic/touch/mako/39:20140520:20140513.2/8112/ubuntu_clock_app/ [09:25] it was 4 but bumped to 8 with rerun. and it was 4 for me as well locally (I think) but a rerun brought it down to 1. however, it seems more flaky than it was before. [09:26] Mirv: me and elopio are currently determining the failure, [09:27] Mirv: but to be honest, it is definitely not in clock app but upstream project :) since we havent pushed anything to clock for almost 2 weeks now due to the AP failures [09:27] nik90_: yes so what did get updated was qtpim [09:27] Mirv: we are also noticing failures in the uitk emulator functions that clock uses [09:27] nik90_: so the question is how's that affecting the tests, especially since it seems sometimes it's just the "normal" 1 failing test and sometimes more. has qtpim become more slower and there should be more delays, or what? [09:28] nik90_: yeah, it's a bit weird all. let me and renato join you shortly (it seems renato might have time in 15 mins or so) [09:28] Mirv: ok [10:01] barry, I can confirm the exit bug, the other one I'm not able to reproduce [10:01] barry, I'm looking into it [10:04] didrocks, can I get a silo for this MR? https://code.launchpad.net/~ogra/media-hub/do-not-start-on-desktop/+merge/220218 [10:05] balloons, the calculator app tests passed on mako: http://s-jenkins.ubuntu-ci:8080/job/generic-click-autopilot-runner-mako/172/ [10:06] jhodapp, s/didrocks/sil2100/ [10:06] :) [10:06] ogra_, didrocks not around? [10:06] fginther, we're trying to add the pep8/pyflakes hooks for all core apps this week; here's the meta-bug: https://bugs.launchpad.net/ubuntu-weather-app/+bug/1317198 [10:06] Ubuntu bug 1317198 in Ubuntu Clock App "Python code should conform to pep8 and pyflakes" [Medium,In progress] [10:06] jhodapp: sil2100 is in charge [10:07] jhodapp, nope, but teams change ;) [10:07] fginther, right now dropping-letters and weather are ready [10:07] balloons, ack, I can get that started [10:07] didrocks, ogra_: ah ok, we need to update the subject line for this room then [10:07] yeah, was waiting for Malta for that [10:07] fginther, excellent. Since they pass let's turn it on there as well. I'll ask the guys to try a merge then for calc [10:08] ok [10:08] let's try this again :) [10:08] balloons, I'll let you know when it's ready [10:08] sil2100, can I get a silo for this MR? https://code.launchpad.net/~ogra/media-hub/do-not-start-on-desktop/+merge/220218 [10:10] fginther, excellent. Expect popey or myself to ping you with new apps to add the hooks to as we land them :-) [10:10] sil2100: Mirv: I would like to ask for a Silo for the line 27 [10:13] bzoltan: landing-017 [10:13] balloons, can you please review? https://code.launchpad.net/~fginther/cupstream2distro-config/coreapps-pep8-dropping-letters-and-weather/+merge/220222 [10:13] Mirv: thanks [10:14] fginther, do you want to do them all as one at once? [10:14] for instance, sudoku is ready now too :-) [10:15] balloons, I'll just update the MP for that. I just don't want this to linger too long and risk the branch regressing [10:15] fginther, ok, perfect [10:15] fginther, we have an MP to test calc when ready ;-) [10:16] fginther, https://code.launchpad.net/~rpadovani/ubuntu-calculator-app/setDeprecatedToolbar/+merge/220223 [10:20] balloons, can you review this one now? https://code.launchpad.net/~fginther/cupstream2distro-config/calculator-touch-testing-autolanding/+merge/220225 [10:21] fginther, yes. And I'll approve the pep8 merge when you add sudoku [10:23] balloons, the pep8 MP has been updated [10:24] fginther, both approved [10:27] balloons, thanks, I'll have them deployed by the end of lunch [10:31] jhodapp: in a meeting, one moment [10:31] sil2100, thanks === MacSlow_ is now known as MacSlow|lunch [11:30] ogra_: ping! Do you have a moment for a packaging ACK? Looks safe - cross-compile enhancements and some 'new' symbols (no symbols removed) -> https://ci-train.ubuntu.com/job/landing-011-2-publish/35/artifact/packaging_changes_unity-scopes-api_0.4.6+14.10.20140519-0ubuntu1.diff [11:32] sil2100, well, with my limited cross building knowledge it looks ok ... but you should probably ask someone more familiar with cross stuff additionally [11:33] ogra_: I checked it against documentation online and it makes sense, but let me pass it forward as well :) [11:33] ogra_: thanks! [11:33] np [11:38] python3:any would be slightly more usual than python3:native, but either should be fine [11:38] :native is pretty new and there are rough edges in its support, but I think it works fine on our builders (I guess it must since you got this far) [11:51] ogra_: another +1 needed, a no-brainer :) https://ci-train.ubuntu.com/job/landing-017-2-publish/lastSuccessfulBuild/artifact/packaging_changes_qtcreator-plugin-ubuntu_3.0.1+14.10.20140520-0ubuntu1.diff <- new file installed ;p [11:52] sil2100, yeah, ack [11:52] ogra_: thanks! [11:52] jhodapp: hi! [11:53] jhodapp: so, you want a new landing for that merge? Are you CI-Trained (i.e. can you be a lander?) [11:53] sil2100, thats a no brainer too (and already pre-tested on all devices) [11:53] ogra_: ok, let me add that then [11:54] Ah! I see it's your branch ;p [11:54] heh [11:56] sil2100, yes [11:56] very very simple landing :) [11:56] ogra_: silo assigned o/ Silo 19, waiting for the spreadsheet to refresh [11:56] sil2100, i will also need a silo for line 24 at some point ... but i might hog it for more than a day to run all tests etc (thats a tricky one that most likely will revel other probs during testing) [11:56] so only if we are not short on silos [11:59] ogra_: I'm building the package in that silo [11:59] i.e. the media-hub one [11:59] sil2100, yeah, i was referring to my landing, not jims ;) [11:59] ogra_: I know ;) [11:59] k === cjohnston changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cjohnston | CI Train support - US: robru, cyphermox, rsalveti - EU: sil2100, Mirv, didrocks | CITrain support no answer: use mup bot after 30 minutes, but choose right timezone | Known issues: - === alan_g is now known as alan_g|lunch [12:32] hey there, cjwatson would like to land a new click, I added an entry on the spreadsheet, would be great if someone could get us a silo! [12:33] sil2100, ^^ [12:33] o/ [12:33] stgraber, you will need somepone from the store people to approve it afterwards iirc [12:33] Just got the ping from the bot :) [12:34] ogra_: could be, I don't know but hope cjwatson does :) I usually just push stuff on the spreadsheet, click buttons and when cjwatson says it's good to push to the archive, I click the other button (or he does, whatever) [12:34] stgraber, and i think sergiusens to actually update the click_list (though i think ubuntu-release owns that so you could perhaps do it yourself) [12:35] stgraber, yeah, just saying ... clicks need special treatment [12:36] ogra_: stgraber lp:click-sync for the image [12:36] * ogra_ is out for a bit [12:36] Ok, we're low on silos, but let me assign one anyway - we should have 2 freed pretty soon anyway [12:36] there are no silos for click though; that's a missing train feature; fginther et.al. are working on it for airline I think [12:37] right, but most clicks have deb equivalents anyway [12:38] ogra_: most won't after the unity8 switch [12:38] Only core apps [12:38] ogra_: no, click itself, not a click package [12:38] the store is not involved [12:39] (this is why I always say "click packages", not "clicks") [12:40] stgraber: silo 20 for you [12:41] sil2100: building already, thanks! [12:46] sil2100: so I hit a couple of issues that I need to report but neither should block the image. On the whole everything is looking good. The network off on update flash being the most serious thing [12:46] davmor2: excellent, thanks! [12:46] cyphermox_: could you give me a sign once you're around? [12:47] s/update flash/ "update/flash" === MacSlow|lunch is now known as MacSlow [12:49] sil2100, could we add https://code.launchpad.net/~mir-team/unity-mir/development-branch/+merge/220249 to silo 18? [12:49] sil2100, would save us some time and a silo [12:49] ah I got logged out... interesting [12:50] sil2100, ok, can I get a reconf of silo 18 please? [12:50] Saviq: I guess we could, unity-mir is in 2 silos right now, but both are 'for testing' [12:50] Saviq: or maybe you know if they are ready with the greeter split? [12:50] sil2100, we're close [12:51] sil2100, but this one will be quicker [12:51] Ok [12:51] Reconfiguring [13:02] cjwatson, oh ... heh, ok [13:03] sil2100, silo20 is ready to land [13:03] ergh [13:03] silo19, sorry [13:05] ogra_: ok, let me publish [13:05] thx [13:06] * sil2100 jumps out to the vet [13:09] Mirv: around? [13:09] mandel: hi. thanks. yeah, the canceled bug has been seen in at least two contexts [13:09] Mirv: i wanted to get back to that silo offer of yours ;) [13:16] barry, weird, do you have more info about when it has happened? [13:16] barry, all my scripts seem to work ok [13:16] mandel: the system-image client test suite evokes it almost immediately [13:17] dbarth: working on it! :) [13:17] barry, that is using tox, correct? [13:17] barry, I aslo think that the bug regarding the emulator can be ignored.. [13:17] barry, and, I have been upgrading my phone with udm and have had no issues.. weird. [13:17] mandel: tox is just the runner, but yes. if you branch lp:ubuntu-system-image and then run tox you will see it [13:17] (i'm running it again now) [13:19] barry, ok, can you also take a look at the following: https://code.launchpad.net/~mandel/ubuntu-download-manager/exit-not-working/+merge/220211 === alan_g|lunch is now known as alan_g [13:19] barry, udm now uses cmake, so mkdir build; cd build; cmake ..; make -j4 [13:19] barry, or just build a package :) [13:20] mandel: the exit problem actually wasn't a terrible show stopper. i worked around it by sigterm'ing the process. the canceled bug is much more serious [13:20] barry, yes, but I wonder.. why does it only happen in the tests, I mean, no one has complained about the system not being able to update [13:20] barry, and we have had severa images in devel [13:20] mandel: but let me try building from that branch. i'm a big fan of fixing the easy ones first :) === Ursinha changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: Ursinha | CI Train support - US: robru, cyphermox, rsalveti - EU: sil2100, Mirv, didrocks | CITrain support no answer: use mup bot after 30 minutes, but choose right timezone | Known issues: - [13:21] barry, in the mean time, I'll do some tests in the real hardware [13:21] mandel: the only thing i can think of is that the test runs udm in a slightly different mode, i.e. -stoppable [13:21] barry, could be [13:22] mandel: one thing i think we very definitely want to do is add some gating automated integration tests. i.e. no new udm or si can land unless the integration between the two (and preferably *also* system settings) has passed. we've talked about this for a long while now, but i think we really need to get this done for 14.10 [13:34] sil2100: hey [13:41] mandel: http://paste.ubuntu.com/7493007/ [13:41] line 5131 [13:44] ogra_: so what was the magic that was needed to be done to get unity8 migrating to release pocket? [13:44] I've seen it mentioned many times but I still don't remember what it actually is about [13:45] the magic is called "patience" [13:45] should be fully automatic now thanks to cjwatson [13:45] oh it should be, ok. unity-scope-click autopkgtest claims to be failing though both for unity-scopes-api and unity8 [13:45] well, then i would expect an actual failure [13:46] right, so it seems then [13:46] Mirv, read -devl [13:48] indeed [13:48] we'd need unity8 though anyhow, so this's a bit unfortunate [13:49] cyphermox_: hi! [13:50] cyphermox_: so, did you take a look at bug https://bugs.launchpad.net/ubuntu/+source/indicator-network/+bug/1320780 ? [13:50] Ubuntu bug 1320780 in network-manager (Ubuntu) "After flashing, wifi network defaults to off" [Critical,Incomplete] [13:50] cyphermox_: since we decided that we would need some proffessional opinion on what could be the problem before deciding on promotion [13:53] mandel: so i just `make`d the source and dumped the executable in /usr/bin. i think that solves the exit problem, or at least enough to reliably and consistently produce the canceled bug [13:53] cyphermox_, i was wondering if all these networking/3G issues we have recently could be related to the dropping of pm-utils and powermgmt-base ... http://people.canonical.com/~ogra/touch-image-stats/29.changes ... does NM make any use of these on the phone ? [13:54] doubtful [13:54] yeah, thought so [13:54] ogra_: do you still need those removals investigated? [13:54] its just an odd dropping ... [13:55] cjwatson, well, i'd really love to know why ... as there was no dep or seed change ... [13:55] (though they shouldnt do any harm i think ... our powermanagement should fully be done by powerd or android) [13:55] rsalveti, ^^right ? [13:56] * rsalveti reads [13:57] (we dropped pm-utils and powermgmt-base from eth images) [13:57] *the [13:58] ogra_: not used by us [13:58] if used, we have something wrong :-) [13:58] not even telepathy or ofono ? [13:59] don't think so === boiko_ is now known as boiko [13:59] well, i was wondering if udev might make use of any of the scripts they ship [13:59] and via that one of the upper layers or so [13:59] do we know why they were dropped? [13:59] rsalveti, nope [13:59] just check 28, see who is depending on them [13:59] i wasted half my saturday trying to find out about it [14:00] the dep or seed change would be reflected in the above changeset [14:00] (actually checking the packages listed in 29 should be enough) [14:01] it should be, but you checked them out without luck [14:01] indeed [14:01] so maybe an alternative way could be useful [14:01] but i doubt that xkb-data would cause it in 28 http://people.canonical.com/~ogra/touch-image-stats/28.changes [14:01] :P [14:01] ogra_: upower: [14:01] Depends: libc6 (>= 2.4), libdbus-1-3 (>= 1.0.2), libdbus-glib-1-2 (>= 0.88), libglib2.0-0 (>= 2.37.3), libgudev-1.0-0 (>= 146), libimobiledevice4 (>= 0.9.7), libplist2 (>= 1.11), libpolkit-gobject-1-0 (>= 0.99), libupower-glib1 (>= 0.9.2), libusb-1.0-0 (>= 2:1.0.8), udev, dbus, pm-utils | systemd | systemd-services [14:02] but there was no change in upower [14:02] maybe because we now have systemd installed? [14:02] oh, systemd-services ... that might be it [14:02] just systemd would be enough [14:02] but again, didnt change in that image [14:03] maybe the dependency order changed when creating the image? [14:03] * ogra_ wonders if the changeset is incomplete or some such [14:03] i also checked all uploads on -cahnges on sat. [14:03] systemd would need to be installed before upower was pulled in [14:04] i know xnox was working on dependency order changes but that was way way after image 29 [14:04] like yesterday ... [14:04] or so [14:04] maybe for systemd it was before? [14:04] priority-mismatches currently wants to pull systemd up to Priority: standard, so that could well account for it [14:04] we didnt really re-order the seeds ... but in 28 i added urflkill [14:04] oh [14:05] Though I don't see systemd alone in the livefs-build-logs diff [14:05] * ogra_ didnt think about checking priority changes [14:05] (only other packages whose names contain systemd) [14:05] -services is definitely there [14:06] well, it is installed [14:06] ii systemd 204-10ubuntu3 armhf system and service manager === jhodapp__ is now known as jhodapp [14:06] so yeah, that might be it [14:06] thanks cjwatson ! [14:07] alternate dependencies can be somewhat unstable [14:07] yep [14:07] * ogra_ remebers diggin into java stuff years ago for that ... in the end we had to re-order the deps [14:07] if you definitely want one particular choice then it's a good idea to seed it explicitly [14:07] that's generally enough to cause it to be selected, if done in the correct seed [14:08] according to rsalveti and cyphermox_ we dont need them, so it is actually good they are gone [14:08] it just bugged me that i couldnt find out why the were dropped [14:09] yeah, the java issue back then was about a discrepancy between apt-get and tasks ... and the ordering of the deps they do [14:09] i think thats still the case [14:09] (that they resolve them in opposite order) [14:10] Mirv: just saw you updated qtdeclarative, remember we also need to update the -gles version [14:10] will syng them again [14:11] ogra_: err, sort of, not really [14:11] ogra_: germinate generates the tasks and it tries to imitate what apt will do. there are some corner cases, but it's certainly not "opposite order" [14:11] ah [14:12] nik90_: http://imgbin.org/index.php?page=image&id=17527 [14:12] there are some depth-first vs. breadth-first differences in some places [14:12] well, i learned that the order of stuff in the seed matters too ... [14:12] didnt know that until slangasek told me :) [14:14] ogra_: could you please advice me on debian/control fields? [14:14] i can try :) [14:15] just ask away ... if i cant there are surely enough people in here that can [14:16] ogra_: I would like to introduce a soft dependency of the qtcreator-plugin-ubuntu on the qtcreator-plugin-go. I wish that after that change the users with the qtcreator-plugin-ubuntu installed get the qtcreator-plugin-go if that is available for them from Utopic or from the SDK PPA. But does not break anything if it does not exist for them. Is that Suggests: or Recommends: ? [14:17] well, recommends are like depends ... just that you can easily uninstall them without ripping out the upper layer ... [14:18] davmor2: https://bugs.launchpad.net/indicator-datetime/+bug/1321307 [14:18] Ubuntu bug 1321307 in Indicator Date and Time "Incorrect icon used to display calendar events in the indicator-datetime" [Undecided,New] [14:18] but they get always pullled in on install (unless you explicitly tell apt not to) [14:18] so if you want it that loose you want suggests ... but these will never get automatically pulled in anywhere [14:19] barry, I have not been able to reproduce it in the phone.. weird [14:19] they are just "an info" if the user actually looks at the package ... [14:20] mandel: the canceled signal? i can reproduce it reliably in my test suite. in fact, if i disable reaction to canceled, most of my tests pass (the exception being the ones that actually test canceling). i know those canceling tests aren't influencing the results because i can kill all processes, and run a failing test in isolation and it sees the bogus canceled signal [14:20] we dont really have a field reflecting "pull this in only sometimes ... when the weather is sunny and at least 24°C ..." [14:20] mandel: are you on utopic? [14:21] barry, yes, I'm on utopic [14:21] hmm [14:21] davmor2: https://bugs.launchpad.net/ubuntu-ux/+bug/1321309 [14:21] ogra_: What happens if the recommended package does not exist? [14:21] Ubuntu bug 1321309 in Ubuntu UX "Missed calendar or alarm events need to be properly indicated" [Undecided,New] [14:22] bzoltan, the toplevel package will not install ... as i said, recommends is like a dep at install time ... it just allows you easy removal later [14:22] No, that's not true [14:23] Unsatisfied Recommends at install time won't cause an error [14:23] bzoltan: Recommends is the field you want [14:23] oh, ok [14:23] cjwatson: Super, thank you. [14:23] "The Recommends field should list packages that would be found together with this one in all but unusual installations." ... [14:23] debian policy doesnt really answer that question :) [14:23] mandel: please try branching lp:ubuntu-system-image and see if you have the same problem. instead of running the full test suite try this: `tox --notest -r && .tox/py34/bin/python -m nose2 -vv -P test_download_winners_overwrite` [14:24] rsalveti: yes, I'm trying, I just wonder why I'm getting https://launchpadlibrarian.net/175964511/buildlog_ubuntu-utopic-amd64.qtdeclarative-opensource-src-gles_5.2.1-5ubuntu3_FAILEDTOBUILD.txt.gz [14:25] barry, ok, doing that [14:26] mandel: cool. i'm trying to see if there's any pattern [14:26] Mirv: oh, I just uploaded it directly [14:26] Mirv: but it seems to be building just fine [14:26] Mirv: https://launchpad.net/ubuntu/utopic/+source/qtdeclarative-opensource-src-gles/5.2.1-5ubuntu3 [14:26] robru, FWIW exporting ANDROID_SERIAL works just fine with citrain-push ;) [14:27] rsalveti: oh... right, that PPA had a dependency on qt5-beta2 which now has Qt 5.3.0 :S ok, thanks! I did remember it, or at least I was properly reminded via CI Train comment :) [14:27] Mirv: oh, ok then :-) [14:30] cyphermox_: hm, do you have any leads on what could be the cause of the bug #1320780 ? [14:30] bug 1320780 in network-manager (Ubuntu) "After flashing, wifi network defaults to off" [Critical,Incomplete] https://launchpad.net/bugs/1320780 [14:30] Mirv: yeah, built fine in the archive https://launchpad.net/ubuntu/+source/qtdeclarative-opensource-src-gles/5.2.1-5ubuntu3/+build/6026189 [14:31] mandel: is it possible that if an https url should 404, we'd get a canceled signal? it looks like often (i've not verified it every time) we try to download the blacklist keyring which does not exist, so we get a 404. we seem to get a canceled signal in that case [14:32] sil2100: I'm looking into it already, but I don't [14:32] something changed, clearly. stuff doesn't go stop working all of a sudden === alan_g is now known as alan_g|tea [14:32] but I have no idea what [14:34] rsalveti: excellent === jhodapp is now known as jhodapp|mtg [14:37] cyphermox_, yeah, the prob is that we had multiple images where two network related landings were kind of wonky (first urfkilll then indicator-network) and it sorted itslef only after 4-5 images [14:37] mandel: further debugging. it does seem like when we request a nonexistent blacklist keyring (an expected use case since there's no other way to tell if a blacklist exists or not), we would expect a 404 leading to an error signal, but instead we get a canceled signal and no error signal. i see the started signal, followed immediately by the canceled signal [14:37] so we cant really nail it to one specific image [14:37] unfortunate timing [14:37] ogra_: It doesn't seem to me like it's either [14:38] and frankly, that's not excuse for not being able to figure out what's broken [14:38] heh, indeed [14:38] I'm looking into it, it just takes time [14:38] but it makes it harder [14:39] simply because the actual breakage was only identified very late ... we had a bunch of images there were "expected to be broken" due to the incompletel landings === Ursinha is now known as Ursinha-afk === alan_g|tea is now known as alan_g [14:40] barry, hmmm that might be the pattern we are looking for [14:43] sergiusens, found the problem! [14:45] barry, it is weird that you could not build udm when CI was able... [14:46] barry, but you can use the deb generated by CI [14:46] look in the comments [14:47] mandel: yeah, i just cmaked it which was good enough [14:50] sil2100: ogra_: so, we can't kick a new image build before the scopes api bug is resolved and unity8 migrates to release pocket [14:50] Yeah... [14:51] well, we could ... just with no benefit for you :) [14:51] sergiusens, fixed [14:54] mandel: no thanks ;-) [14:56] ogra_: trying an OTA update right now to see if I can reproduce it properly [14:58] barry, I just tested the following => http://paste.ubuntu.com/7493304/ [14:58] barry, can you please run that in your system? [14:58] barry, I see an error and not a canceled signal [15:00] mandel: i want to convert that to py3 for consistency [15:00] barry, sure, I'm just editing a test script that I already have [15:01] okay, i see the error too. what about https? === Ursinha-afk is now known as Ursinha [15:02] barry, fair point, lets check that too === doanac changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: doanac | CI Train support - US: robru, cyphermox, rsalveti - EU: sil2100, Mirv, didrocks | CITrain support no answer: use mup bot after 30 minutes, but choose right timezone | Known issues: - [15:03] barry, is that for a valid cert or just testing a 404 [15:03] it's a valid self-signed cert [15:04] e.g. i tried https://www.python.org/foo [15:05] elopio: http://q-jenkins:8080/job/autopilot-release-gatekeeper/ is in quite heavy use, could there be another job that could be triggered for 5.3 AP testing? [15:06] mandel: were you able to reproduce the problem within the si test suite? [15:06] barry, sorry I got disconnected, you said something about using a valid ssl [15:07] mandel: i tried https://www.python.org/foo and that got the error (no canceled). one difference with the test suite is that i use a self-signed cert [15:07] mandel: remember there's a testing option in udm to use the self-signed cert? [15:07] barry, yes, but that should be working as it used to.. but could be a reasno [15:07] reason* [15:08] * barry is grasping at straws [15:08] barry, the tests got stuck here => http://paste.ubuntu.com/7493341/ [15:08] mandel: wtf? ;) === gatox is now known as gatox_lunch [15:09] mandel: now that you have the venvs set up, try just running: [15:09] .tox/py34/bin/python -m nose2 -vv -P test_download_winners_overwrite [15:10] barry, ack [15:12] barry, let me do a standup and I'll try that [15:12] mandel: sounds good [15:14] fginther, I'm seeing that tests are returning unstable, but are still being shown as 'passed' by jenkins [15:14] for instance, https://code.launchpad.net/~mrqtros/ubuntu-rssreader-app/two-no-article-placeholders-fix/+merge/219455 [15:16] Hey, just a heads up that content-hub update broke the 14.04 ABI used by gallery-app and needs to be reverted before we promote an image; this also means we need to revert the browser changes using the new ABI [15:16] mhr3: so I added line 33 so that as soon as the fix for scopes-api is guaranteed we can land it [15:16] sil2100, Mirv, ogra_: ^ (the content-hub thing) === seb128_ is now known as seb128 [15:17] lool, seb128: ouch [15:17] Mirv, -click is in 011 [15:17] Mirv: no need, we already have that building [15:17] mhr3: oh, right, a re-release from there [15:17] Mirv: we're using an existing silo [15:17] great then [15:18] sil2100, lool: is that what creates bug #1321304 [15:18] ? [15:18] bug 1321304 in ubuntu-system-settings (Ubuntu) "Can't set wallpaper from gallery" [Undecided,New] https://launchpad.net/bugs/1321304 [15:18] lool, seb128: let me see when that actually landed... [15:18] friday [15:19] seb128: correct [15:19] mandel: fyi: the ci debs are armhf so they don't help [15:19] Yeah, I see it now [15:19] lool, thanks for pointing it [15:19] lool, seb128: thanks for pointing this out guys [15:19] barry, not in the desktop, true [15:19] np [15:19] Who will handle the revert? [15:19] mandel: but your patch does seem to fix the .exit() problem [15:19] *sniff* i was looking forward to a promotion [15:19] sil2100, talk to kenvandine I guess? [15:20] I'm sad for folks that are back at square 1 *and* need to revert the changes, but I'm glad we catched the ABI breakage [15:20] lool: right [15:20] lool, i think we have a good solution [15:20] lool, definitely ... thanks a lot ! [15:20] kenvandine: hi! So, could you handle the revert of gallery-app/content-hub ? [15:20] quicker fix that reverting all of that [15:20]