=== rsalveti_ is now known as rsalveti === mhall119_ is now known as mhall119 [03:04] === trainguard: IMAGE 269 building (started: 20140331 03:05) === [04:31] that's a longer than usual image build I think [04:31] imgbot: stunt [04:31] * imgbot rolls on its back and purrs [04:44] === trainguard: IMAGE 269 DONE (finished: 20140331 04:45) === [04:45] === changelog: http://people.canonical.com/~ogra/touch-image-stats/269.changes === [07:40] I guess there's a good chance for two-in-a-row 100% pass, 0 crashes image results, nice! [07:42] Mirv: we are still at 50% of tests, let's cross fingers :) [07:43] Mirv: I'm a little bit suspicious of the 0 crashers though [07:43] in any flavor, nothing… === psivaa_ is now known as psivaa [07:44] psivaa: good morning, and welcome back! FYI ^ (probably something to look at, it's not that I don't trust how good our quality is, but still… that's weird :p) [07:45] didrocks: good morning. thx :). ack, will take a look at it in a bit [07:45] thanks psivaa :) [07:46] yw :) [07:47] didrocks: yeah, too good to be true :) 267 did see a lot of changes, though, but let's see what psivaa will find out [07:48] there were lxc, telephony-service updates [07:48] ouai, but I see no reason why system-service would be fixed for instance [07:51] imgbot, status 269 [07:51] I have no idea what PRIVMSG means [07:51] bah [07:51] ogra_: stop harassing the bot! :) [07:51] it worked yesterday :( [07:51] imgbot: stunt stunt [07:52] double stunt doesn't work [07:52] ogra_: making a bot working on Sunday? [07:52] see, then, it's going on strike! [07:52] imgbot, stop [07:52] AAAAARRRGH !!! (dying) [07:53] at lest that works :P [07:54] * didrocks files a report to the police [07:54] imgbot, status 269 [07:54] I have no idea what PRIVMSG means [07:54] imgbot, status 269 [07:54] bah, silly ... i typoed the command :P [07:55] hmm, except that it doesnt seem to work [07:57] imgbot, status 269 [07:57] imgbot, status 269 [07:57] I have no idea what PRIVMSG means [07:58] ogra_: time to rewrite it in go! it doesn't like your extra spaces [07:58] I have no idea what PRIVMSG means [08:01] imgbot, status 269 [08:02] imgbot, status 269 mako [08:03] imgbot, status 269 [08:05] imgbot, status 269 [08:05] ogra_, that bot doesn't want to talk to you [08:10] didrocks: so, it does look like there were no crashes in the last two images and even in the current one, there hasn't been any yet [08:10] imgbot, status 269 [08:10] Image 269 for mako has not finished the tests, status is: Running [08:11] psivaa: waow, that's seem like a dream, are we sure, of the results, on all devices? can we have something preventing to write to /var/crash? [08:11] now that was a silly typo :P [08:11] imgbot, status 268 [08:11] Image 268 test results on mako - Total: 667 Pass: 667 Crashes: 0 Rate: 100% [08:11] imgbot, status 268 flo [08:11] ogra_: but you have tests for it, right? :p [08:12] Image 268 test results on flo - Total: 667 Pass: 660 Crashes: 0 Rate: 98.7% [08:12] didrocks, indeed, there is "set -e" at top of the two shellscripts the bot consiste of :P [08:13] ogra_: rohhh :p [08:13] anyway, you gan get the current image test status now [08:13] next i'll make it announce when the tests start and finis [08:13] h [08:23] imgbot, status bug 1 [08:23] bug 1 in Ubuntu Malaysia LoCo Team "Microsoft has a majority market share" [Critical,In progress] https://launchpad.net/bugs/1 [08:23] imgbot, status 270 makop [08:23] I have no idea what makop means [08:23] imgbot, status 270 mako [08:24] * seb128 stops annoying the bot (and the channel) [08:24] Image 270 for mako has not started testing yet [08:24] seb128, no bugs, it only knows touch images [08:25] well ... [08:25] not only [08:25] imgbot, stunt [08:25] * imgbot rolls on its back and purrs [08:25] see :) [08:25] imgbot, get a cookie [08:30] didrocks: writing to /var/crash does not seem to be an issue, i am able to write there as phablet as well as root [08:30] hum… [08:31] psivaa: coming to the meeting? [08:31] imgbot, status 268 [08:32] Image 268 test results on mako - Total: 667 Pass: 667 Crashes: 0 Rate: 100% [08:32] didrocks: i am in a ho.. must be a wrong one [08:32] psivaa: https://plus.google.com/hangouts/_/canonical.com/landing-meeting [08:32] psivaa, ^^^ can you check why there are no crashers at all ? [08:32] ogra_: can you read the backlog? :p [08:33] :P [08:39] didrocks, cron runs 1h earlier now [08:39] ogra_: \o/ [08:39] thanks [08:51] didrocks, silo for 59 pls? [08:53] mhr3: can you ping the whole EU team please? We try to avoid names :) [08:54] mhr3: however, running it for you this time :p [08:57] popey: I think it might be related to this but I can't find it currently let me dig a little more https://bugs.launchpad.net/ubuntu/+source/unity-scope-click/+bug/1279481 [08:57] Ubuntu bug 1279481 in unity-scope-click (Ubuntu) "scope uses app title to uniquify list of installed applications" [Critical,In progress] [08:59] popey: so, let's do this slowly, I'm not sure about the process.. there's http://s-jenkins.ubuntu-ci:8080/job/gallery-app-click-from-branch/lastSuccessfulBuild/artifact/out/com.ubuntu.gallery_2.9.1.941_armhf.click available [08:59] davmor2: thats been critical for 2 weeks ☹ [08:59] Mirv: looks good [08:59] popey: indeed [08:59] popey: so has it been tested so is the next step upload or are there other steps? [09:00] Mirv: so you have the credentials to login and update the gallery app by pressing "new version" at https://myapps.developer.ubuntu.com/dev/click-apps/507/ [09:00] Mirv: i would not assume it had been tested [09:00] "someone" should run the AP tests for that click on a device [09:01] popey: no, I've credentials to the 'click-toolbelt' tool used by sergio [09:01] ah, i know nothing about that [09:01] popey: how could I convince "someone" to run the AP tests? :D [09:01] I'll do it. [09:01] that was easy! [09:01] thanks [09:01] np [09:02] the notes I have use CLICK_UPDOWN_UPLOAD_URL=https://upload.apps.ubuntu.com/ env variable when calling the upload tool [09:04] Mirv: ok, will get back to you if it passes. [09:04] popey: and won't say anything if it doesn't pass? :) [09:04] typically what's happened before is sergio or balloons have uploaded to the store, I download and then test and find it's broken [09:04] exactly [09:04] and then bounce it back [09:05] i'd rather we test on device before upload to save you waiting for me to test, and me bouncing it back and waiting for reply [09:08] didrocks, right, will randomize my order, atm it's sil and then you :) [09:09] mhr3: just ping the 3 of us [09:09] mhr3: sil is sick today [09:10] mhr3: o/ [09:19] Mirv: http://paste.ubuntu.com/7183950/ I can't setup the test environment on the phone ☹ [09:19] some JSON error [09:20] eh.. I did run phablet-click-test-setup without parameters successfully in the morning [09:20] oh, ubuntuuitoolkit.. [09:21] oh, actually that too worked, I often have removed that and unity8 from /usr/bin/phablet-click-test-setup manually [09:21] since they don't work if one is testing a landing PPA with updated unity8/uitk [09:21] but I hadn't edited the file after last update [09:27] Mirv: do i need to wait for you to do something? [09:30] imgbot, status 269 [09:30] Image 269 test results on mako - Total: 667 Pass: 667 Crashes: 0 Rate: 100% [09:30] popey: for phablet-click-test-setup to work? no. I'm just suggesting that if there is a temporary problem with downloading ubuntuuitoolkit tests to the device, comment out the two lines 65-66 in phablet-click-test-setup (and add ')') to workaround the problem [09:31] another 100% one :) [09:35] Mirv: it still barfs.. http://paste.ubuntu.com/7184021/ [09:40] popey: well remove the unity8 too, but still weird [09:40] ok [09:43] popey: http://pastebin.ubuntu.com/7184049/ [09:44] kk [09:45] Mirv: still fails for me [09:45] http://paste.ubuntu.com/7184055/ [09:51] popey: something is borken on your side probably then :( [09:51] bah [09:51] image 269, what you running? [09:51] on mako [09:52] #269 here too [09:52] and mako [09:52] RW or RO? [09:52] right, RW [09:53] mine is RO [09:53] that JSON sounds like happening on host though [09:53] * popey dist-upgrades === cjwatson_ is now known as cjwatson [09:59] Mirv: phablet-click-test-setup worked for me on Thursday/Friday even with a PPA with updated uitk [09:59] not anymore :( [09:59] right, I'm stuck then, I can't test anything now ☹ [10:04] right, found it.. [10:04] adb shell click list --user=phablet --manifest [10:04] that is executed [10:04] Mirv, can you pls publish 004 ? [10:04] http://paste.ubuntu.com/7184114/ [10:05] so i need to remove an app.. [10:06] popey, ugh [10:07] "my_cgroup" sounds realted to the new cgroup manager ? [10:07] t1mp: ok I believe it has never worked for me, since as you see from the error log it tries to download the version of UITK that is in the PPA, and that shouldn't work since it tries to get it from archives. if you do the setup _before_ adding the UITK PPA (or refreshing apt info), then it works [10:07] mhr3: sure [10:09] that's weird, I'm quite sure I did it after [10:09] Morning all [10:09] mirv: and zoltan's scripts for testing everything has it after [10:09] ogra_: no idea, but sounds plausible. [10:09] i wonder what click does with cgroups though [10:10] cjohnston, is there any direct involvement ^^ or is that just transient ? [10:10] (see popey's last paste) [10:11] (or probably just a confusingly named variable ?) [10:11] Mirv: it is confusing. the script does get stuff from the archive so I don't know how it worked for me before [10:12] bug 1300130 [10:12] bug 1300130 in click (Ubuntu) "UnboundLocalError: local variable 'my_cgroup' referenced before assignment" [Undecided,New] https://launchpad.net/bugs/1300130 [10:13] didrocks: I'd need a packaging ack for thumbnailer that adds new dependencies which all seem to be correctly in main: http://162.213.34.102/job/landing-004-2-publish/lastSuccessfulBuild/artifact/packaging_changes_thumbnailer_1.1+14.04.20140331-0ubuntu1.diff [10:13] also the QML plugin packaging looks good [10:14] t1mp: that is confusing. if you do apt-add-repository + apt-get update on device, it simply shouldn't work. hmm, the only thing I can think of that it uses host also, and I tend to not have the new UITK installed on host while you probably have? [10:15] oh, no, that doesn't make sense [10:15] well, I'm not really expert in click test setup other than workarounding the problems I've hit [10:17] Mirv: +1 [10:21] Mirv: so to test click packages with a new UITK, I should purge the PPA (or flash a new image), then run phablet-click-test-setup, and then add the PPA? [10:25] thanks didier [10:26] t1mp: no, no, or that's not what I do. what I do is edit the click-test-setup to not download the UITK or unity8 since you don't need to have those downloaded for running click tests. [10:27] btw I filed a bug #phablet-tools in February [10:27] bug #1280279 even [10:27] bug 1280279 in phablet-tools (Ubuntu) "phablet-click-test-setup tries to download unity8 and ui-toolkit from main archives, failing when those are being tested" [Undecided,New] https://launchpad.net/bugs/1280279 [10:31] mhr3: thumbnailer published, after checking the packaging changes [10:31] Mirv, ty [10:32] Mirv, what does it mean that it's in unapproved? [10:33] trusty is frozen - all uploads require manual release team approval [10:33] as it happens I'm already looking at it [10:34] this is an awful changelog. what do I have to do to make it less bad in future? [10:34] "New minor release", no more details [10:34] that's reviewer-hostile [10:36] cjwatson: you will notice it's similar to lxc "New upstream bugfix release." btw. But yeah, agreed on the changelog [10:36] (even if lxc then details the packaging changes) [10:36] projects that have a real separate upstream existence generally have the upstream changes available somewhere [10:37] even if I have to go look them up separately [10:37] true [10:37] https://linuxcontainers.org/news/ [10:37] mhr3: ^ [10:37] right should be at least on the NEWS file IMHO though for easier reviewer [10:38] yep, it doesn't have to be in debian/changelog but it should be described somewhere [10:38] the changelog issue seems quite recurrent in our upstream projects [10:38] especially for those having a separate release branch… [10:40] anyway, having paged through the diff, accepted [10:40] Mirv: did you succeed in publishing the gallery app in store? [10:40] ogra_: if you want my attention you need to spell my nick right [10:41] cjwatson, lol, sorry [10:41] ogra_: click doesn't do anything with cgroups [10:41] yeah, would have surprised me [10:41] well, it's clear in the traceback, that's from apport [10:42] didrocks: we don't have test results yet of it, so pending that [10:42] Mirv: ok [10:43] hi [10:43] i've got silo 001 tested, now ready for publishing [10:43] (and being recycled) [10:44] also i have a bug fix on line 60 for when a silo is available [10:44] Mirv or didrocks ^^ (as I don't see sil2100 today) [10:44] popey: just remove the app with the broken manifest for now and get the author to fix it; I'll fix click list to just log such things on stderr rather than crashing [10:45] Mirv: ok, thanks. I'll remove the uitk lines from the script then [10:45] cjwatson: ok [10:47] Mirv: sooooo I want to do this qtdeclarative upload, how do I go about getting "all AP tests" run or whatever you said needed to happen? [10:51] Laney: for me it's currently like http://pastebin.ubuntu.com/7184268/ - in RW mode, after updating to the PPA with updated qtdeclarative. the four commands at the top need to be executed separately (first two on device, last two from host) [10:51] then I run ./run_all.sh &> results/mybestestqtdeclarative.txt [10:52] ok [10:52] and then I wonder about whatever flakiness there is and rerun individual test suites until certain enough there are no regressions [10:52] I made a request in line 61 for it [10:52] dbarth: just a second [10:55] ogra_: you know when you type cj you get cjohnston not cjwatson right as I play catch up on the irc backlog ;) [10:55] dbarth: landing-007 [10:55] Laney: landing-008 [10:56] ta [10:56] davmor2, wrong ... xchat has a patch that makes it pick the last cj i have talked to ... i just hadnt talked to one of them since my xchat was restarted on sat. [10:57] ogra_: haha [10:58] Mirv: cool [10:58] dbarth: and 001 just published [10:59] ah, was just going to ask ;) [10:59] perfect, thanks! [11:00] no problem [11:01] Mirv: gallery passed! Feel free to push to store at your leisure and I'll get it approved [11:02] popey: cool! let's see what explodes. [11:05] stuff is happening [11:06] popey: ok, it ended with "Application uploaded successfully." and "https://myapps.developer.ubuntu.com/dev/click-apps/507/" [11:07] popey: can you check how it looks like? I don't have access to the myapps page. [11:09] Mirv: yes, i see it submitted [11:09] approved Mirv [11:10] \o/ I'm approved [11:10] popey: did gallery remove videos though? [11:11] davmor2: not part of my test [11:11] popey: no indeed but the whole reason of the update was to fix that right? === MacSlow is now known as MacSlow|lunch [11:12] davmor2: I didnt look at the diff [11:13] thanks Mirv, popey! [11:17] didrocks: ogra_: ev: raised bug #1300156 for the crashes not being reported [11:17] bug 1300156 in Apport "Unhandled exception in apport/report.py in touch images" [Undecided,New] https://launchpad.net/bugs/1300156 [11:18] psivaa: see #ubuntu-touch where I pinged you :) [11:18] psivaa: all fixed and an upload is in unapproved [11:19] didrocks: ohh, i dont see the ping. but glad that got fixed. [11:20] I duped you bug to bug #1296026 [11:20] bug 1296026 in apport (Ubuntu) "apport crashes at reading cgroup" [Undecided,Fix committed] https://launchpad.net/bugs/1296026 [11:30] Mirv, yes, silo pls :) [11:30] mhr3: ;) [11:31] mhr3: landing-004. I assumed you wouldn't have put it to Ready Yes if it wouldn't be ;) [11:31] Mirv, thanks [11:31] ogra_, do you know any packaging expert I can get some help from? [11:50] thank psivaa === josepht_ changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: josepht_ | 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 === MacSlow|lunch is now known as MacSlow === josepht_ is now known as josepht [12:37] ok, going for some exercise [12:39] ogra_: do you mind kicking an image once apport is in the release pocket? (not yet) [12:40] didrocks, ok [12:41] thx! [12:45] you just want the crashes back :( [12:57] Mirv, 004 rdy to publish === alan_g|lunch is now known as alan_g [12:59] mhr3: looks good, needs packaging ack, maybe ogra? http://162.213.34.102/job/landing-004-2-publish/41/artifact/packaging_changes_unity-scopes-shell_0.4.0+14.04.20140331-0ubuntu1.diff [13:00] also, EODish, usually filled by sil2100 until robru/cypher_mox are about [13:01] for once someone is getting rid of dependencies instead of adding them :) [13:01] Mirv, i assume that has been built already ? [13:01] (i.e. the makefile changes have been tested) [13:02] ogra_: yep, it's in the landing PPA where mhr3_ has tested if from [13:02] Mirv, then ack [13:02] thank you [13:19] didrocks: have we not hit traincon0 again now? [13:19] balloons: mirv pushed gallery 941 to the store and I approved it [13:19] fyi [13:19] === trainguard: IMAGE 270 building (started: 20140331 13:20) === [13:22] popey, cool.. nice to have Mirv pushing his own stuff :-) [13:22] heh [13:23] didrocks: 269 has mostly worked for me, apps are opening more slowly might need popey to confirm that running side by side with promoted, I've had one crash this morning but that might of been before I updated, the system still feels a little flaky though can't put my finger on why though [13:23] balloons: any news on music app? I can't run ap tests on it because of the UnicodeDecodeError ? [13:23] UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 33484: ordinal not in range(128) [13:23] that [13:24] popey: have you seen that some of the apps are showing odd characters in the descriptions I'm assuming that is down to unicode too [13:25] davmor2: such as? [13:25] bug 1300230 [13:25] bug 1300230 in Ubuntu Music App "UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 33484: ordinal not in range(128)" [Undecided,New] https://launchpad.net/bugs/1300230 [13:25] popey: black diamond ? where german/chinese characters should be [13:25] balloons: ^ [13:26] davmor2: no, i mean which app? [13:26] popey, well where we stand is it affects you, and seems to be a locale thing [13:26] popey: have a look at Tradera for one [13:27] popey, Sergio gave the most helpful response. It's odd we are just now seeing this [13:27] popey: look like a chinese app next to lastpass [13:28] davmor2, sounds like unicode issues as you say.. maybe that will get more traction :-) [13:28] looks fine here davmor2 [13:28] http://popey.com/~alan/phablet/device-2014-03-31-142807.png [13:28] popey, let's try setting your locale to utf8 and have you run them [13:28] popey: Rad.io the what's is what?s [13:28] popey: no in the description of the app not running it [13:29] fginther, is https://wiki.ubuntu.com/CI/MpTouchDeviceTesting coming to core apps? [13:29] ooh [13:29] http://popey.com/~alan/phablet/device-2014-03-31-142912.png [13:29] yup [13:29] \o/ [13:29] didrocks: ^ found abug [13:29] davmor2 stikes again! [13:30] I blame ogra_ for it of course he is free to pass on the blame though :) [13:30] looks fine on the last stable image #250 [13:31] balloons, not right away, there are some pieces missing from the core apps jenkins to do the full thing. I do want to ask you about doing an automatic test after building the core apps clicks on s-jenkins [13:31] popey: thanks for that so it's been introduced recently then [13:32] balloons, I think the next step forward for the core apps is to automate that test after build [13:36] fginther, so you want to run tests on every click that you build on s-jenkins? [13:37] balloons, yes, that was the basic idea. [13:38] sounds like a nice first step.. do you need anything to be able to do that? [13:39] balloons: i dont understand why the AP tests for gallery work fine but music dont [13:39] on the same device [13:40] popey, let me blow your mind a bit more.. I don't believe anything has changed testwise since last time (and it works for me, heh) [13:40] bah [13:40] popey, I'll wager the old click will fail the same as the new one [13:40] you need the old click package? [13:40] no i have an archive [13:40] but I want to test this click not the old one [13:40] am trying on my stable phone [13:41] mmmm.. there you go.. [13:41] what? [13:41] yes, I just want to see if you can narrow done anything [13:41] I would test the new click on stable and see what happens [13:41] kk [13:41] doing now [13:41] popey: confirmation please https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1300236 [13:41] well, prepping [13:41] Ubuntu bug 1300236 in unity8 (Ubuntu) "Click Scope is shopping ? instead of characters" [Undecided,New] [13:41] but truly, I'm curious if the old click gives you the same issue [13:41] popey: I added your image there too [13:42] davmor2: done [13:42] balloons: k, will test that too [13:49] didrocks: one bug of the day https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1300236 [13:49] Ubuntu bug 1300236 in unity8 (Ubuntu) "Click Scope is showing ? instead of characters" [High,Confirmed] [13:53] popey: you don't have to long press when the app is in available only when the app is in installed and my steps were for in available :P :D [13:53] balloons, regarding the testing - I would like to coordinate these changes so that no-one is surprised. Should I just work with you? [13:53] meh [13:55] doanac, I'm curious, why do you need to run system settle before unlocking the device? wouldn't it be just as effective after? [13:55] balloons: yeah, fails on 250 image too [13:55] fginther: if it took too long to settle the screen would lock again wouldn't it? [13:58] doanac, that wasn't my understanding. I thought this unlock method was permanent until unity8 is restarted. Should check with mterry to know for sure [13:58] doing so now [13:59] fginther: that would solve my complaint then [14:02] balloons: fails on #250 with music app 389 too. [14:02] so is this an autopilot fail? [14:08] doanac: did you ping upstream about? Do you know starting what image this is a regression for? [14:10] didrocks: must not be in my back scroll. what regression are you curious about? [14:10] doanac: sorry, too many d here :p [14:10] davmor2: ^ [14:10] (we should kill them all) ;) [14:11] didrocks: yeah starting at di :P I'm just chatting about it on touch popey says it isn't effecting 250 so it is a recent change I'm guessing [14:12] /nick dopey [14:13] popey: \o/ [14:13] davmor2: you are going to bisect? [14:13] davmor2: starts maybe with images which contains unity8 only [14:17] didrocks: yeah I'll start that in a bit I have a juggling act on at the minute [14:18] doanac, I was wrong, the screen can lock back without the powerd commands executed [14:19] davmor2: do you know if we have a bug for the fact that when you go back from preview, the apps lens goes back to the top? [14:19] fginther: actually - that's another thing. i think our scritps grab power locks after we reboot and before we unlock the screen [14:19] popey: not that I'm aware of [14:19] so there are a couple of things we need to hook into this logic [14:19] k, I'll file [14:20] sorry popey I meant dopey for that obviously ;) [14:20] heh [14:21] doanac, yep, reboots kinda suck [14:43] didrocks: so it looks like it might be the new apps scope in which case it will of been in play from when that landed by the sound of it [14:44] === trainguard: IMAGE 270 DONE (finished: 20140331 14:45) === [14:44] === changelog: http://people.canonical.com/~ogra/touch-image-stats/270.changes === [14:44] davmor2: do you have an exact unity8 version for Saviq? [14:46] didrocks: no not yet I was talking to alecu on a u1 channel to see who would be the best person to talk to, apparently it may of been introduce when the click scope dropped the vala backend for the c++ one he is currently looking at that by the sound of it [14:46] davmor2: ok ;) [14:46] davmor2, didrocks, dobey is looking into it [14:47] it's most probably the click scope [14:47] Mirv, still here? [14:47] Saviq: indeed I searched on music for "de" and the first search found in grooveshark is looking correctly unicoded [14:48] didrocks: ^ [14:48] Mirv, silo for line #10 and #26 pls? (cc: didrocks) [14:49] mhr3: maybe try cyphermox at this time as well :) [14:49] he should be around [14:49] moo? [14:49] haha :D [14:49] didrocks, see you keep changing who should i ping all the time, how am i supposed to read your mind? :P [14:49] mhr3: Mirv finished ~1h ago from now [14:50] so US timezone :) [14:50] mhr3: topic. [14:50] knowing that robru is here quite later (west coast) [14:50] cyphermox, silo for #10 and #26 pls [14:50] yup [14:58] didrocks: shouldn't the spreadsheet update with the silo number once I assigned it? :) [14:59] mhr3: landing-001 for line #10, in case it takes a while to update the spreadsheet [14:59] cyphermox: yeah, you have to wait for 3 minutes at most (so average 1.5 min) [14:59] I'm looking at line #26 now [14:59] as there is a double rsync [14:59] cyphermox, k, ty [14:59] imgbot, status 270 [14:59] Image 270 for mako has not started testing yet [14:59] slacker ... [15:03] ogra_: why haven't you tested it and given imgbot the right info then? ;) [15:21] fginther, hey, I was thinking... we're starting to look into converting our qmluitests to an autopkgtest, was it considered to run those as part of ci jobs? or are we going for airline ci directly (will we still have single-branch auto-run ci when we get to airline at all?) [15:27] Saviq, will get back to you shortly (in a meeting) [15:27] fginther, sure [15:44] didrocks: can you help us reconfiguring silo-15? [15:44] we added a new MR to the list [15:44] (powerd) [15:44] it seems I can't reconfigure it [15:44] rsalveti: a new components that wasn't in the initiale list? [15:44] component* [15:44] didrocks: yes [15:44] rsalveti: IIRC, you had extra power though to run the other job [15:45] imgbot, status 270 [15:45] Image 270 for mako has not finished the tests, status is: Running [15:45] didrocks: that's what I was trying to understand :-) [15:45] didrocks: which job? [15:45] great [15:46] rsalveti: let me first try, then, I'll tell you if nothing is broken :) [15:46] didrocks: I only have the 'debug mode' option [15:46] at http://162.213.34.102/job/landing-015-0-reconfigure/build at least [15:46] rsalveti: ok, you are on the wrong job for reconfiguring if you add new components :) [15:46] got it [15:46] (options depending on creds isn't available on jenkins :/) [15:46] do I need to use the main job? [15:46] yeah [15:47] I did that [15:47] then, checked "ignore conflicting…" [15:47] (for that one) [15:47] and click "proceed" [15:47] didrocks: hm, don't have permission to build it [15:47] oh? I thought you had all super power [15:48] me too [15:48] let me check you are in the right teams [15:48] (done) [15:48] rsalveti: added you to ~ubuntu-unity [15:48] rsalveti: so, for that job, you need first to select the right line [15:49] fginther, you can work with me, but I will be out for a week. What's the timeline you had in mind? [15:49] and then, popup the dialogbox [15:49] (select the right line == have the cursor on the line you want to configure) [15:49] rsalveti: done for this one [15:49] didrocks: great, can build it now, awesome [15:49] rsalveti: yw! [15:50] didrocks: thanks [15:50] Saviq: should be done [15:53] balloons, I'd like to have it in place by the end of the week [15:54] fginther, sounds like someone else should have a go. I'm here tomorrow then gone. Volunteers? [15:54] balloons, who will be pushing app updates while you are out? [15:59] Saviq, I'm back. +1 on going to autopkgtest. that will plug right into the ci-airline as dep8 tests are a primary function. [16:01] Saviq, I do not expect to have single branch ci working first thing. We are focusing on replacing the current ci-train functionality in the first phase [16:01] fginther, mirv is the man on this one ;-) [16:02] cyphermox: ogra_: balloons: coming? [16:03] trying to [16:03] my sound doesnt work at all [16:03] * ogra_ only gets choppy stuff [16:04] didrocks, sorry, trying to upgrade my hangouts plugin [16:04] please ping me if there is anything urgent [16:06] plars: coming? [16:08] didrocks: oh, time change [16:08] ber [16:08] err [16:08] brt [16:10] *SH*T* [16:10] ! [16:10] so not even a new FF and HO plugin help [16:10] * ogra_ curses [16:10] fginther, okies [16:11] davmor2: https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1300302 [16:11] Ubuntu bug 1300302 in unity8 (Ubuntu) "Dead area at bottom when scrolling in scopes" [Undecided,New] [16:13] davmor2: also https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1300304 [16:13] Ubuntu bug 1300304 in unity8 (Ubuntu) "Scope loses context when you go back from preview" [Undecided,New] [16:16] davmor2: also https://bugs.launchpad.net/ubuntu/+source/webbrowser-app/+bug/1300309 [16:16] Ubuntu bug 1300309 in webbrowser-app (Ubuntu) "Links in twitter webapp are not clickable" [Undecided,New] [16:20] didrocks, lets do it on IRC [16:20] didrocks, might be easier to finish in irc? [16:21] hello [16:22] didrocks, didnt get any better [16:23] didrocks: seriously give up on the hangout you keep getting taken over by didroid!!!! [16:24] didrocks, so i was talking about bug 1290771 [16:24] bug 1290771 in Canonical Upstream To Distro "cupstream2distro should FTBFS if packages have an empty changelog" [High,New] https://launchpad.net/bugs/1290771 [16:24] davmor2: also https://bugs.launchpad.net/ubuntu/+source/webbrowser-app/+bug/1300312 [16:24] Ubuntu bug 1300312 in webbrowser-app (Ubuntu) "twitter webapp header doesn't stick" [Undecided,New] === gatox is now known as gatox_lunch [16:25] didrocks, so probably foundations can help [16:26] yeah, giving up on the hangout [16:26] since they are the ones complaining the most about it [16:26] eh, it's not exactly a top priority for us [16:26] just notifying you [16:27] davmor2: http://popey.com/~alan/phablet/device-2014-03-31-172649.png [16:27] is that the greying issue you see? [16:27] cjwatson, well, infinity made some fuss about it on the weekend and slangasek seemsed unhappy too [16:27] *seemed [16:27] popey: yeap [16:27] ogra_: sure, anyone who reviews uploads will notice it, but that's in our (essentially community) release team capacity [16:28] it's not a foundations priority [16:28] ok [16:28] we're just pointing it out :) [16:29] can someone update a landing proposal in CI rain for me? [16:29] well, adam pointed it out with a megaphone and waving red flags in both hands ... but yeah :) === josepht 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: - [16:29] on the spreadsheet, line 31 (UITK proposal) I'd like to remove the last MR (fix-swipe-delete-002) [16:29] "please stop doing this" doesn't have to be the same as "let me help you fix this" ... [16:30] ogra_: ok, so in summary… [16:30] (while I'm connected) [16:30] Laney, nope, but when i carried it forward to didier (who owns the code) he said he has no time to work on it ... which was why i suggested to ask the people that are concerned about it if they have spare cycles [16:30] ogra_: tell people to give a correct algorithm to detect that case [16:30] and I can integrate it [16:31] knowing that: [16:31] didrocks, right, now you missed a bunch of conversation .... [16:31] you can have other commits that was done before [16:31] and so generate a changelog content [16:32] didrocks, well, an entry with no chars out of space ones and "*" should be easy enough to detect [16:32] ogra_: right, sorry and all but I know at least I don't have time to help fix the changelog generator - if nobody has cycles then I guess it'll remain unfixed [16:32] seb128: you can have other commits before [16:32] didrocks, it's not a perfect solution, but seems the only buggy case we had were ending up with "* " as changelog entry [16:32] cjwatson, right, fine with me, adams unhappyness simply made me try to get the concern forward [16:32] so this commit will have no entry [16:33] didrocks, sure, but that still makes the entry with "* " buggy [16:33] what is hiding the issue [16:33] yeah, but you are hiding the fact that people edit debian/changelog without putting content [16:33] hiding? [16:33] we are speaking about rejecting/failing build for those [16:33] that would "hide" anything [16:34] wouldn't* [16:34] so I can block on that, trying to detect it [16:34] but it will clearly hide other cases [16:34] that are invalid as well [16:35] well if you reject that particular case ang change nothing else, I fail to see how it would "hide" mroe [16:35] seb128: yeah, but let's say I've 3 commits [16:35] I generated debian/changelog with UNRELEASED [16:35] then, I have this commit [16:35] it would be the same situation as before, with one case of error blocked [16:35] which edits the changelog [16:35] only to change the version [16:35] so, the change will still apply [16:36] you are maybe overthinking it? [16:36] can't we just check at publishing time, don't dput if the changelog entry is buggy [16:36] didrocks, you are blocking to early ... [16:36] the changelog is built at this point [16:36] it's easier to check the result [16:36] and you won't see it [16:36] in that case [16:36] you will see the 3 other commits ocntent [16:36] but nothing else [16:37] didrocks, you only want to check the last entry of the generated source package ... if thats empty, refuse the build [16:37] which edits the changelog [16:37] only to change the version [16:37] so, the change will still apply [16:37] and you won't see it [16:37] in that case [16:37] you will see the 3 other commits ocntent [16:37] but nothing else [16:37] seb128: again, yeah, but you will hide other potential [16:37] issues [16:39] didrocks, but even cu2d creates .changes files no ? [16:39] in a hangout [16:39] didrocks, I fail to see how "rejecting one case" is hiding issue over letting it in [16:39] but anyway, that's a minor issue [16:39] grep \* /path/to.changes .... then chop off the asterisk ... if your result is an empty string, FTBFS [16:40] if somebody really wants to see it fixed they can work on a patch [16:40] this 'please run all autopilot tests' is painful [16:40] it's been like 6 hours now [16:41] davmor2: another https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1300326 [16:41] Laney, it's almost like building webkit! [16:41] Ubuntu bug 1300326 in unity8 (Ubuntu) "Shell randomly freezes with grey tint" [Undecided,New] [16:42] it'd be cool if webkit made the machine make random noises every now and again [16:42] about time we get arm64 phones [16:42] seb128: +1 [16:44] didrocks, silo bad http://162.213.34.102/job/landing-015-1-build/60/console ? [16:54] didrocks, cyphermox, rsalveti, looks like silo 015 got broken ↑ [16:54] and can't reconfigure: http://162.213.34.102/job/landing-015-0-reconfigure/26/console [16:55] looking [17:01] cyphermox, any thoughts? it looks like the ID matches in the spreadsheet, not sure why it wouldn't find it [17:01] not sure, no [17:02] I assume it changed as it was reconfigured by an admin? [17:02] cyphermox, no, the ID only gets assigned when it's new. there's a code block that checks to make sure we don't clobber IDs that are in use. [17:03] well then let's check on jenkins directly [17:03] cyphermox, the number matches in jenkins... [17:03] i mean in the files... === alan_g is now known as alan_g|EOD [17:07] robru: btw, line 25 should already ahve been assigned to silo 2... [17:07] or maybe not [17:08] cyphermox, wasn't sure about that one. silo 2 came from a different line, but many same components [17:08] yeah [17:09] cyphermox, i guess silo 2 is one of those ignoreconflicts special cases, so maybe we should assign line 25 anyway? [17:09] depends if mhr3 agrees [17:10] cyphermox, i'm going to try re-preparing silo 15 [17:10] ok [17:10] cyphermox, fine with me [17:12] cyphermox, that's really bizarre, re-prepare gave the same error. I just checked the file, the json looks fine and python parses it fine [17:12] and up to Saviq, too [17:12] Saviq, the only solution I can think of so far is to smash out the silo with a crowbar [17:13] Saviq, (i want to free your silo then reassign) [17:15] robru, please do [17:15] ok [17:15] robru, I'm worried PPAs will be unhelpful, though, as we might end up with packages of the same version as deleted ones [17:16] robru, and PPA will reject different .orig.tar.gz [17:16] Saviq, well the new silo will likely be a different one [17:16] robru, yeah, that will help :) [17:16] but also, just version bump ;-) [17:16] hmmm, silo 15 it seems is in a really bad way, I can't even free it... [17:17] :| [17:17] cyphermox, http://162.213.34.102/job/landing-015-3-merge-clean/11/console does this mean anything to you? it seems to indicate that some kind of 'global' 'config' is missing... [17:18] cyphermox, Saviq oh wait, maybe it's not the silo that's broken... maybe this is all of citrain? [17:19] yeah [17:19] hold on [17:20] cyphermox, are you poking at the server? [17:21] popey: confirmed [17:22] robru: I was, butI don't see anything meaningful [17:22] davmor2: did you say you had a bug for apps showing as available when already installed? [17:23] nvm, found it davmor2 bug 1279481 [17:23] bug 1279481 in unity-scope-click (Ubuntu) "scope uses app title to uniquify list of installed applications" [Critical,In progress] https://launchpad.net/bugs/1279481 [17:24] ah, got it [17:24] robru: the status part is there [17:24] cyphermox, I assume that cupstream2distro-config went missing? redeploy? [17:24] doubt it's relevant [17:29] robru: going to try again to free it === fginther changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: fginther | 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: - [17:32] gah [17:35] cyphermox, robru, I'm going away, so just do whatever you need to do... [17:36] sure [17:36] Saviq, ok, we're trying [17:36] robru: will reconfigure a different way [17:36] kgunn, fyi ↑ ci tracks seem to have bent... [17:36] guh [17:37] robru: I'm unable to deal with this, let's call didrcoks [17:37] cyphermox, yeah, i'm also stumped. mup? [17:37] yah [17:38] who does it? [17:38] cyphermox, i thought you'd do it ;-) [17:38] ok [17:39] wait === gatox_lunch is now known as gatox [17:40] cyphermox, /srv/juju/vol-0000011d/var/lib/jenkins/silos/landing-015/config is blank [17:40] cyphermox, I'm gonna try copying it from the public URL [17:41] cyphermox, Saviq, kgunn: ok I fixed it [17:43] Saviq, ok, reconfigured, and I'm rebuilding for you too since you stepped out. should be ready to test when you get back [17:45] cyphermox, and I put line 25 into silo 4 [17:45] robru: hi, when you have a chance I need silos for line 62 and 63 [17:45] ok [17:46] so 15 is sitll broke for now? [17:46] cyphermox, no no, i fixed it [17:46] good [17:46] then I'll sms didrocks back to say it's good [17:47] cyphermox, somehow the config got lost from the jenkins server, but it was still live at ~platform/citrain, so I just copied the json back and then it was right as rain [17:47] good [17:48] bfiller, ok, you got silos 10 and 11 [17:48] robru: thanks [17:49] bfiller, you're welcome [18:58] lunch! === Ursinha is now known as Ursinha-afk [19:21] imgbot, status 270 [19:21] Image 270 test results on mako - Total: 667 Pass: 629 Crashes: 8 Rate: 97.0% [19:21] woah [19:24] ogra_: for some reason I keep reading that as 667 pass: 629 Crashes and thinking that kinda sucks :) [19:25] should i add some commas ?? [19:25] imgbot, status 270 [19:26] Image 270 test results on mako - Total: 667, Pass: 629, Crashes: 8, Rate: 97.0% [19:26] better ? [19:26] 8i mean the values are still pretty awful) [19:27] ogra_: \o/ [19:27] imgbot, stunt [19:27] * imgbot rolls on its back and purrs [19:27] :) [19:28] ogra_: haha [19:28] i taught it tricks on the weekend :) [19:29] ogra_: you have too much free time obviously ;) [19:30] heh [19:33] imgbot, status 270 [19:34] Image 270 test results on mako - Total: 667, Pass: 629, Crashes: 8, Rate: 97.0% [19:35] imgbot, status 270 flo [19:35] Image 270 test results on flo - Total: 667, Pass: 663, Crashes: 4, Rate: 99.0% [19:37] imgbot, stunt [19:37] * imgbot rolls on its back and purrs [19:53] robru: silo 10 ready for release [20:10] robru: cyphermox can you free up silo 5? as in cancel? [20:11] sure [20:11] sergiusens: nothing to keep? [20:11] cyphermox: nope; we will regen when needed [20:11] they are small packages [20:12] cyphermox: just leave the line in "Pending" [20:14] ok [20:18] balloons: finally got the tests running and get one failure... [20:18] balloons: http://paste.ubuntu.com/7186673/ [20:21] I finished the testing on silo-001, can someone finish it off? [20:24] popey, so what was the issue in the end? [20:24] popey, ohh the playshape thing. I didn't think andrew landed that [20:24] weird [20:25] robru: silo 9 ready for release now [20:25] robru: just a heads up, that's the one that switches the browser and webapp container to oxide [20:26] tested since friday, supports old webapps as well as the new 14.04 ones (those only are switched to oxide) [20:26] dbarth, that's exciting! [20:26] dbarth, and scary :) [20:27] a bit, but after using for a bit we think it's ready now [20:27] great [20:28] i guess the release team wants to look at this one a bit more closely before the next image, so that's why i'm mentioning [20:34] * tedg 's silo is less scary :-) [20:35] vanguard ping [20:44] tedg, silo 1 published. [20:44] well, almost. pending packaging ack [20:44] robru, Great, thanks! [20:49] robru: line 44 ready for release, as well as line 62 [20:49] thanks, will release soon [20:52] Question about ci & tests... If I push a new branch of Mir, does it run the test that are included with the source (same as running ctest) or are there separate tests that only Jenkins runs? [21:16] fginther: Do you have any info on this ^^^ ? [21:21] josharenson, jenkins does a standard binary package build using pbuilder. So if the tests are executed by debian/rules. jenkins will run them. There are some special settings for mir... [21:22] for mir, the jenkins build sets -DENABLE_MEMCHECK_OPTION=ON and -DDISABLE_GTEST_TEST_DISCOVERY=ON [21:22] and specifically for armhf: -DMIR_RUN_ACCEPTANCE_TESTS=NO and -DMIR_RUN_INTEGRATION_TESTS=NO [21:23] fginther: thanks, thats helpful. Is there an easy way to view all the commands executed for the build? [21:23] or is that really it? [21:25] josharenson, there's not a real good way to view the commands individually as jenkins uses pbuilder which does all the building [21:27] josharenson, there is a set of pbuilder hook scripts that implement the customized build options. If you are interested in these, they are contained in lp:pbuilderjenkins [21:27] josepht, the set of hooks for mir is "H05set_package_version H10enable_coverage H10disable_mir_test_runs B10gcovr_run H15enable_testing D10install_valgrind B09copy_results C09copy_results" which is viewable in the jenkins build parameter list [21:29] fginther: thanks a bunch, just trying to get as close as I can to the jenkins build locally so I don't have to push a bunch of garbage while I'm learning. [21:54] robru, thanks [21:54] Saviq, you're welcome === fginther 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: - [23:25] hey robru or cyphermox....that line 64 for platform does conflict, but with the split greeter silo that won't land for some time [23:26] i think line64 is the only legit platform-api that wants to land [23:52] robru: hey, can I get a silo for phablet-tools l65?