[02:05] === trainguards: IMAGE 28 building (started: 20141121 02:05) === === chihchun_afk is now known as chihchun [03:30] === trainguards: IMAGE 28 DONE (finished: 20141121 03:30) === [03:30] === changelog: http://people.canonical.com/~ogra/touch-image-stats/28.changes === === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun [07:03] Mirv, hey, is it possible to do a binary copy of oxide-qt 1.3.5-0ubuntu0.14.10.1~rtm from https://launchpad.net/~phablet-team/+archive/ubuntu/ppa to RTM silo 3 ? [07:04] Mirv, robru tried to do it last night for me, but he says it failed because it’s considered cross-series, but I’m pretty sure we did that for 1.3.4 and it worked [07:16] oSoMoN: yes, it's possible [07:16] just needs correct parameters on the command line, LP web UI does not work [07:16] Mirv, right, that’s what I thought, but robru wasn’t sure how to do it, so he said better to wait till you or sil2100 can do it [07:17] Mirv, could you do it for me, please? [07:17] Mirv: yeah please share the command, i only know the web ui [07:23] oSoMoN: done [07:23] robru: ./copy-package -b --from=~phablet-team/ubuntu/ppa --from-suite=utopic --to=~ci-train-ppa-service/ubuntu-rtm/landing-003 --to-suite=14.09 oxide-qt [07:23] Mirv, thanks! [07:24] took a while, I needed to convert also this to the new style syntax so that I don't make Colin grumpy again :) [07:24] the new syntaxt is much better [07:25] Mirv: Hmmmmmmm thanks, seems simple enough [07:26] Weird that that would work in cli but not in the web ui [08:55] cihelp, I need somehelp with autopilot. I get "Start directory is not importable" and it won't say *why* when files are there, and I didn't change the .py files http://paste.ubuntu.com/9143520/ [08:57] jibel, ogra_: how's the testing going? [08:57] * sil2100 didn't see a status e-mail in his inbox [08:58] kalikiana: cihelp is for CI related issues, i.e. ones you can't reproduce locally ;) I have no idea about what is wrong here but I would try by getting rid of: "09:46:50.312 WARNING emulators:26 - The ubuntuuitoolkit.emulators module is deprecated. Import the autopilot helpers from the top-level ubuntuuitoolkit module." [09:00] 'top-level' and import errors.... sound like either your PYTHONPATH is wrong or something related (i.e. you're trying to import something that can't be imported from any starting point in PYTHONPATH) [09:01] vila: I used cihelp as leo is in the wrong timezone for this point in time and it's kinda urgent :-] [09:01] kalikiana: or you run into a collision where two modules have the same name at different places starting from different PYTHONPATH entries ;) [09:01] kalikiana: ack, but elopio is in QA not CI ;) [09:01] kalikiana: thomi may be ? [09:02] vila: the paths etc. are all untouched, normal device, vivid, previously ran other ap tests perfectly - but this particular test gives the error, others still run [09:02] kalikiana: aprt from the general python issues, I probably know less than you about autopilot ;) [09:03] kalikiana: but still in your paste: raise ImportError('Start directory is not importable: %r' % start_dir) [09:03] vila: alright, I'll ask qa then, thanks anyway [09:04] yes it's there, except the files exist, as the paste shows :-( [09:04] kalikiana: ooooh, you're trying to run a single test ! Do you still encounter the issue if you try to run the whole suite ? [09:05] kalikiana: they may exist but unittest doesn't know how to import them, it probably try to start with '... Oh wait, it's even clearer, it tries to import 'ubuntuuitoolkit.tests.test_launcher.LauncherQtTestTestCase.test_can_run_qt_test_case' [09:05] kalikiana: that's a test method in a test class, far from being a module [09:05] no, but it takes ages to find out if it would then run the test… and I *know* that the test name is correct [09:05] it's been run on another machine [09:05] kalikiana: indeed, the test name is correct, but that still doesn't make it a correct *module* name [09:06] oh, I'll try if I can't list the modules somewhere… I know there was a command [09:06] kalikiana: that same invocation with that single test on the command line has run on a different machine ? [09:06] yes [09:07] sil2100, me neither (at least about the testing ... there were unnsurprisingly 400 other mails in mine :P (like every day)) [09:07] kalikiana: does 'phablet-test-run ubuntuuitoolkit.tests.test_launcher -v' works ? [09:07] even autopilot3 list agrees the test is there [09:08] kalikiana: I'm not arguing the test name is not valid ;) [09:08] kalikiana: I'm reading your paste and interpret it as: look, I cannot import a test method, it's not a python module [09:08] test_launcher → same exception [09:09] kalikiana: same traceback ? Some value in the exception ? [09:10] same as before, no actual error message besides failure to import [09:10] kalikiana: can you just paste it ? [09:11] kalikiana: especially the ImportError: Start directory is not importable: part [09:11] vila: http://paste.ubuntu.com/9143709/ [09:11] as far as I see exactly the same [09:12] well, no, now it complains that the *module* can't be imported [09:13] which at least makes a little more sense [09:13] python's unitest is notoriously bad at reporting import errors: it doesn't report *why* the import failed [09:16] kalikiana: I don't know how autopilot handle partial loading of a test suite :-/ But if you know how to run the full test suite and that pass, the issue is there: running a subset doesn't work for 'ubuntuuitoolkit.tests.test_launcher'. [09:17] kalikiana: I would try ''ubuntuuitoolkit.tests' and then 'ubuntuuitoolkit' if that makes sense [09:18] kalikiana: oh ! And that's python3.4 ! Gee, I'm out of my league :-/ Still stuck with 2.7 in all projects I'm involved :-( [09:18] kalikiana: on the other hand, it's unlikely that this particular piece of code has changed a lot (test discovery) [09:19] jibel, sil2100 https://bugs.launchpad.net/ubuntu-rtm/+source/cgmanager/+bug/1394919 [09:19] Launchpad bug 1394919 in cgmanager (Ubuntu RTM) "constant crash in trying to collect info for recoverable error" [Undecided,New] [09:19] i think that is caused by the "UI freezes" fix [09:20] kalikiana: the way it work[ed] is that unittest starts with 'ubuntuuitoolkit.tests.test_launcher' and attempts to import the higher level modules (ubuntuuitoolkit.tests first, is it there ? No ? Try 'ubuntuuitoolkit', it it there ? No ? Highest level ? Yes -> fails) [09:21] kalikiana: if ''ubuntuuitoolkit.tests' has been imported but doesn't have a 'test_launcher' symbol defined -> fail (which could be why it says: AttributeError: 'module' object has no attribute 'test_launcher') [09:23] sil2100, I didn't send any status yet, I must talk to Europeans first [09:25] * kalikiana just discovered a "feature", ap runs even after the script was stopped - it would be useful if there was a way to actually control it :-P [09:25] jibel: ACK [09:27] https://bugs.launchpad.net/ubuntu-rtm/+source/unity-scopes-shell/+bug/1394922 [09:27] Launchpad bug 1394922 in unity-scopes-shell (Ubuntu RTM) "scoperunner produces a lot of crash files in rtm 166" [Undecided,New] [09:28] hm, this doesn't look too good [09:28] neither does the cgmanager one [09:30] tehr scoperunner one seems to be the same we see in smoke testing [09:38] sil2100, hey, any idea about this failure https://launchpadlibrarian.net/190816573/buildlog_ubuntu-vivid-i386.unity8_8.01%2B15.04.20141121-0ubuntu1_FAILEDTOBUILD.txt.gz ? [09:52] jibel: just give us a sign on your verdict and we'll all forward it to victorp [09:58] Saviq: looking [10:15] vila: FYI seems that I'm hitting a phablet-test-run bug, ap itself isn't to blame if I run it by hand on the device; http://paste.ubuntu.com/9144667/ https://bugs.launchpad.net/ubuntu/+source/phablet-tools/+bug/1394932 [10:15] Launchpad bug 1394932 in phablet-tools (Ubuntu) "ImportError: Start directory is not importable on valid test folder" [Critical,New] [10:16] could somebody review https://code.launchpad.net/~nik90/ubuntu-seeds/add-qml-connectivity/+merge/237442 ? [10:16] does that make sense? [10:16] I guess it needs to be done for vidid and not utopic in any case? [10:16] ogra_, ^ [10:17] kalikiana: \o/ Glad you're unblocked, sorry I couldn't help better... [10:28] sil2100, looks like I can reproduce that in a amd64 chroot locally (not in a native i386 chroot) [10:33] popey, rhuddie will test the music-app [10:36] popey, I'll start music-app shortly, so please let me know where to get latest version [10:39] hah [10:40] Mirv, https://launchpad.net/ubuntu/+source/connectivity-api [10:40] Mirv, this needs a rebuild against new Qt I believe [10:42] oh and we got a circular dep.. [10:42] Mirv, http://paste.ubuntu.com/9145379/ [10:43] which is the same as https://launchpadlibrarian.net/190816573/buildlog_ubuntu-vivid-i386.unity8_8.01%2B15.04.20141121-0ubuntu1_FAILEDTOBUILD.txt.gz [10:43] Saviq: it does not depend on the abi virtual packages, it shouldn't need a rebuild [10:44] Mirv, right, I think the problem might be the circular dep then [10:44] yeah, circular deps are always bad anyhow [10:44] unity8 → libconnectivity-qt1-dev → libconnectivity-qt1 → indicator-network → unity8 [10:44] ouch [10:44] break it! :) [10:44] but how would that build on amd64... [10:44] yeah, I don't think indicator-network should dep on unity8... [10:45] but then that's what it uses to ask for passwords [10:46] seb128, needs a bug and approval from the mgmt if it is for rtm ... [10:46] for vivid it can indeed just go in [10:46] (if the packages are there and work :P ) [10:48] rhuddie: its linked in the trello card [10:49] popey, great. just wanted to check that was correct one. Thanks. [10:53] jibel white or black smoke? [10:54] grey [10:55] victorp, there is this crash ogra_ found [10:55] bug 1394919 [10:55] bug 1394919 in cgmanager (Ubuntu RTM) "constant crash in trying to collect info for recoverable error" [Undecided,New] https://launchpad.net/bugs/1394919 [10:55] huh! [10:55] likely a regression [10:56] victorp, bug 1394919 and bug 1394922 ... and Chipaca just reported a system-settings hang in bug #1394944 [10:56] bug 1394922 in unity-scopes-shell (Ubuntu RTM) "scoperunner produces a lot of crash files in rtm 166" [Undecided,New] https://launchpad.net/bugs/1394922 [10:56] bug 1394944 in ubuntu-system-settings (Ubuntu) "blank screen in battery → screen brightness" [Undecided,New] https://launchpad.net/bugs/1394944 [10:57] i have also seen a good bunch of webapp crashes but then noticed that i had not cleaned the whole QML cache ... will keep an eye on that if anything happens again [10:57] Mirv, sil2100, not sure what I'm doing here, the FTBFS only occurs here when trying to build i386 from amd64 chroot, and comes down to indicator-network depending on python3-xdg:i386, which doesn't exist [10:57] the cgmanager stuff seems serious though ... i had three session crashes since upgrading to 166 [10:58] ogra_, is this all stuff from 166? [10:58] yes [11:00] Wellark, hey, who's maintaining connectivity-api? [11:00] ogra_ btw, I cant reproduce the screenbrightness issue [11:01] (not that that means it is not a bug) [11:01] victorp, me neither, i asked for more info, it doesnt dsay rtm or krillin [11:01] Wellark, also, could we make indicator-network not depend on unity8? otherwise we're getting a circular build-dep since we're trying to use libconnectivity-qt1 [11:02] victorp: in any case, bugs #1394919 and #1394922 look like pretty serious issues [11:02] bug 1394919 in cgmanager (Ubuntu RTM) "constant crash in trying to collect info for recoverable error" [Undecided,New] https://launchpad.net/bugs/1394919 [11:02] bug 1394922 in unity-scopes-shell (Ubuntu RTM) "scoperunner produces a lot of crash files in rtm 166" [Undecided,New] https://launchpad.net/bugs/1394922 [11:02] ogra_, so far worring bugs, but none of them topblockers [11:03] jibel: any other serious issues reported during regression testing? [11:03] Saviq: hmmm [11:03] victorp, you think the session crashing isnt one ? [11:03] well, what is the impact to the user? [11:03] seems like the systems manages not to let the crashes impact UX [11:04] sil2100, if you give me 30 minutes I'll finish the list of issues we found :) [11:04] jibel, that sounds like a good idea :) [11:04] Saviq: sorry, was in a meeting - this sounds a bit strange, I think connectivity-api was managed by Wellark and pete-woods [11:05] sil2100: don't finger me on that one :p [11:05] jibel: waiting with anticipation ;) [11:05] victorp, apps that were running dont work anymore after a session crash [11:05] it's Antti's baby [11:05] until you reboot [11:05] * jibel kills OSD [11:05] pete-woods: ok, sorry ;p Just remember seeing you in the initial commits ;) [11:05] or kill them explicitly [11:05] ogra_, ah! yes that is nasty [11:05] which one is that? [11:06] 919 or 922? [11:06] victorp, the session does the lifecycle mgmt ... if it dies the apps stay suspended ... dangling dead [11:06] thats 919 [11:06] sil2100, so, if I understand correctly, anything that would like to link against libconnectivity-qt1 will fail on i386 if the builder's amd64 [11:07] ogra_, ah, now I have seen your comment in the bug.. [11:07] victorp, i havent heard from others about such an issue though ... (and i used the phone a lot last night to keep it busy) [11:07] could we put the impact on the bug please [11:07] victorp, we usually do afrer we have confirmation ... [11:07] ogra_, apps stop working? nope I have not seen it either [11:08] victorp, kill unity8 from the commandline with some apps open [11:08] ogra_, oks, well let see if it gets confirmed, then yes it looks like a blocker [11:08] then try to start the apps that were open [11:08] they wont start [11:08] (after the session respawned i mean) [11:09] we are missing a watchdog for this that cleans up running apps when the session dies [11:09] (there is work on forcing a reboot when the session crashes ... but thats still in early stages) [11:09] Wellark, where RU? [11:09] when you're needed :P [11:10] victorp, hmm,, llooks like the brightness issue was actually 166 and krillin ... [11:11] victorp, system-sessings is mostly pure QML ... given the QML cache issues it could well be caused by the not cleaned up cache [11:11] ogra_, I cannot reproduce the brightness issue either, but I completely wiped the phone. [11:12] ogra_, or it could be the dbus high load that he mentions, but seems like noone can repro [11:12] we should take all issues in apps with a grain of salt on 166 ... especially from people that OTAed to 166 [11:12] jibel, I ota'd.. but GM will be a fresh install anyway :) [11:12] there is confirmation from a community person [11:13] ogra_, right, but not wide spread to make it a blocked imho [11:13] plus as you said might be precompiling [11:13] right [11:13] we need to be careful about judging apps in general on this image [11:15] victorp, oh, and seems rickspencer ran into a 100% CPU dbus bug yesterday ... most likely the one we have open already ... [11:15] (we duplicated it during landing meeting) [11:16] yeah, I run into one too [11:16] they need to be fixed for ota-1, that is for sure [11:16] Yeah, but the 100% CPU issue in dbus is investigated all the time, not sure if we'll be able to get a fix quick [11:16] It seems to be a tricky issue [11:16] sil2100, well, i guess silo16 will at least killl a part of that [11:16] but thats not for landing yet [11:17] tvoss: hey! Did you have any luck with the 100% CPU dbus investigation? [11:17] ogra_: right... [11:17] sil2100, nope, not yet [11:17] and as i said ... just a part .. [11:20] sil2100, what's even weirder... it built fine yesterday https://launchpadlibrarian.net/190747862/buildlog_ubuntu-vivid-i386.unity8_8.01%2B15.04.20141120-0ubuntu1_UPLOADING.txt.gz ¿? [11:20] on an amd64/i386 machine, too [11:21] * Saviq does not get it :| [11:21] /food [11:21] Saviq: wait, are you sure it's not transient? Ah, right, you an repro on a local chroot... Let's take a look if there have been any changes in the archive [11:22] sil2100, victorp we are all looking into the issue [11:28] tvoss: ogra_ mentioned that dbus is a lot quieter with the new upowerd, as seen in silo rtm 16 (not for landing) [11:42] Saviq: I have no idea what changed and suddenly cause this to fail, nothing got uploaded to vivid in the meantime that could have changed the situation [11:42] Saviq: I also saw that it actually passed yesterday on an older builder [11:43] But that shouldn't have any effect [11:48] sil2100, yeah, it's *weird*, it's like apt resolution stopped working suddenly === karni is now known as karni-lunch [11:48] sil2100, as the actual failure for me seems to be that the dep chain results in python3-xdg:i386, which for some reason is not fulfilled by python3-xdg:all ?? [12:34] trainguards: can oxide-qt 1.3.5-0ubuntu1 be binary-copied from https://launchpad.net/~ubuntu-mozilla-security/+archive/ubuntu/ppa to vivid silo 17 ? [12:34] Saviq: something is b0rken it seems, will still look into that after lunch [12:34] oSoMoN: sure [12:35] sil2100, thanks [12:35] But so far this looks like a ghost story ;) But I'm sure there's a logical explaination [12:36] sil2100, thanks! the PPA size might need to be increased first, as it already contains a copy of oxide (1.3.4) [12:36] I just saw that, yeah, we'll have to poke someone regarding that [12:36] Let me go to webops [12:37] actually binary copies might work regardless [12:38] yes, it'd work [12:38] Mirv: they overwrite older packages? Or not obey the size limit? [12:38] sil2100: they don't obey / check only if it's under the limit before starting the copy. so yes it'd work. [12:38] sil2100: so I've had 20GB of stuff in 2GB PPA when doing Qt landing backups ;) [12:38] Mirv: yay \o/ Ok, let me do that then [12:38] Wow [12:39] don't file a bug, for me it's a feature [12:39] haha [12:40] I'm pretty sure no one here would want this feature to go away ;) [12:40] oSoMoN: ok, copied, should be published in the PPA soonish [12:40] sil2100, awesome, thanks! === karni-lunch is now known as karni [12:41] that’s an awesome feature I would say :) [12:46] http://people.canonical.com/~jhm/barajas/master/device_krillin-20141121-7e18b86.tar.xz [12:46] http://people.canonical.com/~jhm/barajas/master/device_krillin-20141121-7e18b86.changes [12:46] http://people.canonical.com/~jhm/barajas/master/device_krillin-testresults-20141121-7e18b86.ods [12:46] Wow, another vivid tarball? [12:46] vivid device tarball for krillin. ogra_ sil2100 jibel ^ [12:46] one, small, important change [12:46] john-mcaleely, go ahead [12:46] we can get some run time before we migrate it to OTA [12:47] the gpg stuff ? [12:48] john-mcaleely: sounds nice [12:48] It's generally good to use vivid as the testing platform [12:48] ogra_, yeah, the gpg stuff [12:49] sil2100, ogra_ I'll push it now then [12:49] sil2100, it isnt ... he hates developers and wants to make our life harder by adding secutity checks !!! [12:49] :) [12:49] lol [12:50] It's all for our good [12:50] or maybe [12:50] bwahahaha [12:50] !!! [12:50] hah [12:50] ;) [12:50] sil2100, ogra_ pushed. [12:50] thanks :) [12:51] psivaa: come to think of it, hmmm [12:52] psivaa: we don't have vivid krillin smoketesing results? Since the rtm-dashboard only has utopic results - and those seem really old (from last month) [12:53] sil2100: let me take a look, not sure if that was requested to be setup [12:54] it was surely there before [12:57] sil2100: ogra_: they are running, but not being pushed to the rtm dashboard. now that http://dashboard.ubuntu-ci:8080/ is disfunctional, and vivid not being rtm, we need to find a home for that [12:58] i'll raise it in the meeting today [12:58] thanks ! [12:59] Right, thanks! === alan_g is now known as alan_g|lunch [13:06] rsalveti: hey! === jhodapp is now known as jhodapp|sick [13:25] sil2100: may i ask for a reconfig of the Silo28. I just added the gles branch what comes from the different project. [13:25] or Mirv ^ [13:26] hmm, my phone isn't showing up in adb devices... [13:26] even after a reboot [13:26] bzoltan: doing! [13:26] sil2100: thanks [13:26] bzoltan: I lost my mouse pointer and I'm in the hangout... it's a bit hard to find buttons to push :D [13:28] bzoltan: done :) [13:28] Mirv: ouch! [13:28] * sil2100 goes to lunch finally [13:39] Hey guys I see an issue with ciborium it keeps saying my card is connect and disconnect [13:39] abyway back to the holiday now [13:40] ok, can't get the phone to show up at all over adb [13:41] mako is fine, krillin not === alan_g|lunch is now known as alan_g [14:04] sil2100: Mirv: I am done with silo28. I have pasted the URL to the test logs. The silo is still creating the -gles package, but that is not the scope of the tests... Once the -gles are done, would you please push the new UITK to Vivid? [14:06] sil2100: hey [14:17] ogra_, Hi! I am testing silo 2 -- What is the regression potential for this change ? [14:18] the first startup of apps after OTA will be slower than usual to create the new cahe [14:19] 1sec max [14:28] om26er, oh, and beyond the above ... zero [14:30] ogra_, alright, good to know. [14:31] jibel, ogra_, om26er: are both the oxide and qml-cache silos being signed-off right now? [14:32] no [14:32] only qml [14:32] oxide gets its own image after promotion [14:32] one... at a time. [14:32] bzoltan: ok. and I'm ready with 002 to vivid. [14:32] Ok, now I see :) [14:37] ogra_, Well the silo seems to do whats written on the tin. Should I just approve or try it a second time ? [14:37] * om26er is clueless ;-) [14:38] jibel, popey, I finished testing the music app. No issues found. I recorded results here: https://docs.google.com/a/canonical.com/spreadsheets/d/1gFoUz9zQ38LovfLLIcHU5SHTuDDBQvHTSPVcrIxJUlc/edit#gid=0 [14:38] om26er, if the cache is empty after OTA we're fine [14:39] rhuddie: magic, thanks very much [14:39] om26er, a second run should just give you the same result ... else we have a very serious upstart probllem ;) [14:39] rhuddie: "Needed to reboot after install for app to work (othwise it just displays spinner)" - pull down on app scope should have worked. [14:40] can't believe we still have that bug [14:40] been there since day 1. [14:41] ogra_, it was all clear till your first reply. Second reply made it a little confusion, retry or retry not ? :p [14:41] popey, the app launched, but displayed the spinner instead of displaying "No music found" [14:41] rhuddie: even after pulling down to refresh click scope? [14:42] om26er, dont, unless you want to waste time :) [14:42] popey, I didn't do that, so let me check again [14:43] rhuddie: i have experienced that myself, its because the click scope still points to the old app version you had. either pull down or search for music in the scope at the top and it launches [14:43] okki, approved. [14:44] yay, thanks [14:45] rsalveti: hey, just wanted to know if there was some progress related to the pulseaudio part of bug #1391230 [14:45] bug 1391230 in pulseaudio (Ubuntu RTM) "[TOPBLOCKER] web audio, pulse runs at 6% and screen will not blank on idle" [Critical,Confirmed] https://launchpad.net/bugs/1391230 [14:45] ogra_, om26er: \o/ [14:45] Let me publish [14:45] sil2100: nops, still investigating [14:45] publishing [14:45] should know more on monday [14:46] ...and ogra_ was faster [14:46] rsalveti: thanks :) [14:46] heh, sorry :) [14:46] ogra_: I'm making a new mir release, but want to coordinate the seed change (https://code.launchpad.net/~mir-team/ubuntu-seeds/use-new-Mir-metapackages/+merge/242408) [14:46] ogra_: ;p [14:47] AlbertA2, yeah, i saw that, this is vivid only, right ? [14:47] Sorry is not enough! *runs away in tears* [14:47] ogra_: right [14:47] * ogra_ hugs sil2100 [14:47] * sil2100 hugs ogra_ back [14:48] ogra_: UITK would need core-dev ack. https://ci-train.ubuntu.com/job/ubuntu-landing-028-2-publish/4/artifact/packaging_changes_ubuntu-ui-toolkit_1.1.1338+15.04.20141121-0ubuntu1.diff [14:48] they manually fixed some changelog entries plus added new test related dependencies [14:49] Mirv, looks ok to me, i hope their bzr tree can cope with teh changelog hacking :P [14:49] ok. I hope so too :) [14:50] popey, confirmed it is working after doing as you suggested [14:50] ogra_: so how do we go about that? publish mir first then you'll make the seed change? [14:51] AlbertA2, is Mir ready to land in a silo already ? [14:51] ogra_: yes I just ok'ed. silo 009 [14:52] AlbertA2, then publish it, i'll do the seed change and upload a new meta to vivid [14:53] seed changess and their packages should always land together in the same image ... beyond that the order does not matter [14:53] (we build vivid images at 2am UTC, still plenty of time) [14:53] ogra_: ok good to know thanks [14:53] trainguards: vivid silo 009 is ready to publish [14:55] rhuddie: awesome [14:56] Mirv: so, the vivid plans to have Qt 5.4.0, right? [14:58] trainguards: oops sorry, branches TAed now...for mir [15:01] ogra_: I wonder if you'd be ok with reviewing that new Mir packaging changes too.. https://ci-train.ubuntu.com/job/ubuntu-landing-009-2-publish/51/artifact/packaging_changes_mir_0.9.0+15.04.20141120.1-0ubuntu1.diff it's quite big, but _mostly_ because lot of cmake file changes (included for possible interest, debian/ is what needs acking) and because of renamed packages [15:02] sil2100: yes, so this week I've been trying to get to the point where I have fast&ugly packages that can be used to detect bugs early [15:02] sil2100: next week there should be hopefully release candidate from upstream, instead of the beta I've now used [15:06] Mirv, phew ... not a small one ... but ack [15:07] Mirv: ok, then I'll prepare a merge with a fix for appmenu-qt5 not building but not release it for the time being [15:12] ogra_, what's the ETA for the new build? [15:13] cihelp: this has been saying "Recording test results" for at least half an hour (which suggests there's a problem somewhere) - http://s-jenkins.ubuntu-ci:8080/job/mir-vivid-amd64-ci/193/console [15:13] jibel, the package just landed, starting a build now ... ~90min to 2h [15:13] k [15:13] build triggered [15:14] alan_g, the problem in this case is that build 192 is going very slow. and jenkins can't complete build N if build N-1 is not finished [15:14] rtm -proposed migration... so nice [15:14] alan_g, I've been looking at the slow build for the last 20 minutes or so and I think I've got it unwedged [15:14] alan_g, dammit, it timed out [15:15] jibel: so QA is not doing exploratory testing anymore, right? I remember Dave doing that in the past [15:15] * alan_g is sympathetic to fginther's frustration [15:15] sil2100, we do, when there is time left between silos and images [15:16] Saviq: I had away message in canonical-irc, as you get kicked out on multiple channels if you use "public" away in your nick [15:17] Saviq: I will see what I can do with the deps [15:17] libconnectivity-qt is not coming from i-network source [15:17] Wellark, I'm not really sure what's teh culprit [15:17] Wellark, no, but connectivity-api depends on i-n [15:17] libconnectivity-qt1 that is [15:17] Saviq: it probably should just recommend [15:17] would that fix it? [15:18] Wellark, well, can it just recommend? [15:19] Saviq: sure. if i-network is not around (hence no connectivity dbus-service running) libconnectivity-qt1 falls back to "connected" [15:19] there is no hard coupling [15:19] other than we know that i-network provides the connectivity-service libconnectivity-qt1 tries to poke over dbus [15:20] === trainguards: RTM IMAGE 167 building (started: 20141121 15:20) === [15:20] Saviq: i-network must depend on unity8 [15:20] as it depends on the snap-decisions that unity8 provides [15:21] Wellark, yeah I know why [15:21] one way to break this circular dependency problem would be to introduce two meta packages [15:21] unity-notifications-service [15:21] and connectivity-service [15:21] as, i-network depends on the unity-notification-service, which is actually implemented by unity8 [15:22] and then again libconnectivity-qt1 dependes on connectivity-service which just happens to be impemented by i-network right now [15:22] but didrocks might not like the idea ;) [15:23] although IMO it would make the dependencies more explicit [15:25] when something depends on a well-known dbus-service, it should not matter which package implements it [15:25] from the user point of view [15:25] sil2100: if possible, I prefer conditional changes that can be landed beforehand, so that nochange rebuild is enough when landing Qt [15:25] or "client" [15:25] sil2100: can we throw in two metapackages --^ [15:29] alan_g, I ended up cleaning up a couple runaway process and restart that build that timed out. [15:29] alan_g, things look better now, but will need to figure out something to prevent this in the future [15:35] so.. next vivid image will have new qtbase, new ui-toolkit and new mir. what could possibly go wrong. [15:37] cihelp, I need a job on s-jenkins updated (the docviewer-app-click job) [15:38] balloons, updated in what way? [15:39] retoaded, it looks like it's still building with trusty, and in general might not be the current way to build clicks. It's sat unused for some time [15:39] retoaded, you can look at clock-app-click as an example [15:40] balloons, ack, will work it in amoung the morning full of meetings [15:40] retoaded, ty [15:42] fginther: thanks. [15:43] sil2100: can you republish ubuntu silo 15, MR's are approved now. also need a silo for line 60 please [15:46] trainguards, hello, may i ask for a silo for line #59? === chihchun is now known as chihchun_afk [16:01] trainguards: can vivid silo 17 be published, please? [16:04] that was quick, thanks to whoever did it :) [16:04] o/ [16:08] sil2100: Mirv: thanks for midwifing the UITK landing.. you guys just rock! [16:12] jibel: btw. what about sanity testing of vivid images? Does that still happen for every vivid image? [16:19] bzoltan: you're welcome [16:29] mterry, Hi! [16:29] om26er, hello! [16:30] === trainguards: RTM IMAGE 167 DONE (finished: 20141121 16:30) === [16:30] === changelog: http://people.canonical.com/~ogra/touch-image-stats/rtm/167.changes === [16:30] mterry, I was playing around and found that pin code lock can be bypassed easily when unity8 is start. [16:30] om26er, seems bad [16:30] jibel, sil2100 victorp ^^^^^ image 167 for you [16:30] Wow, that seemed somehow fast! [16:30] ogra_, what a service!! [16:30] :) [16:30] jibel: can we have someone sanity testing that + delta? [16:31] 70min only ! [16:31] thats a new record [16:31] But the changelog is huge [16:31]