[08:43] a house full of green :) [08:43] http://ci.ubuntu.com/smokeng/trusty/touch/mako/166:20140206:20140115.1/6473/ [08:43] good work!! [08:44] cjwatson: rsalveti: tvoss: thanks! [08:44] asac, yup :) [08:45] asac, with that, happy Thursday [08:45] beautiful [08:45] you too! [08:45] ogra_: ^^ thanks! [08:47] and better: we got ten more silo ppas set up :) === psivaa changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: psivaa | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: - [09:15] 09:43 < asac> a house full of green :) [09:15] 09:43 < asac> http://ci.ubuntu.com/smokeng/trusty/touch/mako/166:20140206:20140115.1/6473/ [09:15] 09:43 < asac> good work!! [09:15] didrocks: sil2100: ^^ [09:15] :) [09:15] thanks [09:15] \o/ [09:15] morning :) [09:15] asac: thanks, first things we checked in the morning [09:15] 09:47 < asac> and better: we got ten more silo ppas set up :) [09:15] (even before getting up from bed :p) [09:15] asac: morning! Yes, we saw the test results during breakfast [09:15] lol [09:15] asac: yep! [09:15] saw the email [09:15] asac: and it's an image with Mir! [09:16] indeed! [09:16] thats the best [09:16] no more unity8 cfrash [09:16] crash* [09:16] there is just one failure on maguro we need to investigate [09:16] (the messaging-app one) [09:16] asac: we are waiting for the dogfooders though [09:16] right [09:34] https://plus.google.com/hangouts/_/calendar/Y2Fub25pY2FsLmNvbV91cTRvNmQyMWJvNmJ0bm1mcW9xZWtsNTdnOEBncm91cC5jYWxlbmRhci5nb29nbGUuY29t.us2orfbhb8ssqjui2u15tajj3s [09:50] davmor2: talk to msm about your N10 [09:50] thx [09:59] asac: FYI, autopilot has some wrong packaging change [09:59] so we reject this landing [09:59] didrocks: what was the change? [10:00] asac: dependencies and they removed [!powerpc] [10:00] didrocks: can you keep stuff in the silo so they just need to respin? [10:00] i think we had them long enough in there to justify not throwing them out [10:00] asac: yeah, they need to push one more MP or one more commit [10:00] right [10:00] that's what I mean by rejecting [10:00] so yeah. if reject means: cant publish, then thats fine [10:00] yeah [10:00] (ratehr than go back to starting point) [10:00] hehe [10:00] but it means that they are not following the MP guidelines [10:01] didrocks: tell them they should get their MPs that change packages signed off by core-devs [10:01] during the MP review [10:01] this will prevent such late coming surprises [10:01] yeah [10:01] (which is in line with our MR excellence thingy that asks folks to get more reviews from all that need to do that) [10:04] didrocks: s/tell/recommend and give hint/ :) [10:05] asac: yep [10:05] didrocks: send them a nice mail, explaining them that and give them hint how they can iterate (e.g. update MPs and hit build) and CC jfunk [10:06] * asac recommends that :) [10:06] asac: yeah, we're handling that, maybe as well there is another reason, fetching the history [10:06] ack [10:15] sil2100, I'm confused: https://code.launchpad.net/~thomas-voss/dbus-cpp/force_gcc_4.7/+merge/204909 builds fine locally with dpkg-buildpackage, debuild and bzr bd [10:27] cyphermox_, around? [10:28] tvoss: hi! No symbols mismatch? [10:28] sil2100, hang on, still investigating [10:31] tvoss, could be a builder issue: dpkg-architecture: warning: Couldn't determine gcc system type, falling back to default (native compilation) [10:31] sh: 1: i386-linux-gnu-gcc-4.7: not found [10:33] sergiusens, where do you see that? [10:33] tvoss, in the build logs for the failure [10:35] tvoss, in the i386 build here: https://code.launchpad.net/~thomas-voss/dbus-cpp/force_gcc_4.7/+merge/204909/comments/479295 [10:35] sergiusens, hmmm, interesting, don't see it on the amd64 build though [10:35] sergiusens, ack, seeing it now, thx [10:35] sil2100, with http://pastebin.ubuntu.com/6884470/ I get missing symbols [10:36] sil2100, that is, that patch applied on top of my branch [10:36] dpkg-shlibdeps: error: couldn't find library libdbus-cpp.so.1 needed by debian/dbus-cpp-dev-examples/usr/lib/arm-linux-gnueabihf/dbus-cpp/examples/upower/upower (ELF format: 'elf32-littlearm-hfabi'; RPATH: '') [10:39] tvoss: strange, you sure it's using 4.7? Since you saw that in the silo the same modifications (without the name change) had symbols mismatch [10:41] sil2100, I'm sure, checked it twice [10:42] sil2100, looking at the symbols that are missing, it strikes me that they all refer to libstdc++ types or ctors/dtors/operators. And I would expect those symbols to change with compiler versions [10:42] sil2100, I cannot easily diff the symbols as some are reported as missing, which confuses the diff command line on the wiki [10:43] tvoss: ok, this is strange, but maybe something in trunk is causing the symbols not mismatch anymore? [10:44] sil2100, not sure how that would cause the issues I'm seeing, I did the same steps that CI did [10:45] sil2100, so I would suggest that I revert the chanes to the binary package version I did in that branch, and then see how I can fix the symbol issue. I would appreciate some help with that [10:45] asac: will do [10:47] asac: all new silos taken into account now! [10:48] didrocks: last two images have worked, g+ and facebook links are magically working again I don't know who fixed that :) everything looks pretty stable need to test the calls and texts and stuff but on the whole pretty reliable so far :) [10:48] davmor2: maybe a Miry-thingy? :p [10:48] davmor2: oh please check with omer, he's dogfooding as well [10:48] so just gather data [10:48] and +1 or -1 :) [10:49] didrocks: didn't make the call sorry, emergency at home will try tomorrow once we know what is happening today. [10:49] davmor2: thanks! just ensure you don't dup work [10:49] didrocks: no worries [10:53] sil2100, still around? :) [10:54] tvoss: yes, been moving from one room to another ;) [10:54] tvoss: let me test the branch here though - will you change the package name in the end, or revert it to libdbus-cpp1? [10:55] I mean, is the change to libdbus-cpp1.1 going to stick? [10:58] psivaa: Can you check the work that plars talked about in bug #1275786 happened? [10:58] bug 1275786 in Ubuntu CI Services "Please add ubuntu-system-settings-autopilot to touch smoke testing" [Undecided,In progress] https://launchpad.net/bugs/1275786 [10:58] We don't see it in the jobs on q-jenkins or the dashboard [10:58] Laney: sure, 1 sec [11:11] sil2100, I thought we should keep it at 1.1 [11:12] ogra_, didrocks: autopilot in release pocket, we kicking a new image? [11:12] Laney: as per the comment on the bug, this is still in the testing branch. [11:12] Laney: i could still not find that in the dev jenkins that we use internally. [11:12] psivaa: Oh, that doesn't mean 'testing' like 'the test results'? [11:13] mmm [11:13] Laney: no. [11:13] I see [11:13] but it's not there anyway, eh? [11:14] Laney: i could not find it. may be plars is running somewhere that i am not looking at. [11:14] secret testing dungeon [11:15] Laney: heh, i'll wait for him to get the latest. should be online in a couple of hrs [11:15] k [11:18] tvoss: if yes, then I can try helping out with symbols then - it's a pain, and I think we might divide it for clarity into more than one symbols file [11:18] tvoss: but in such a way that we don't duplicate symbols [11:19] didrocks: is maguro now dropped it seems to think that the only upgrade path is to 161 [11:19] sil2100, appreciate your help, but we should start automating symbol updates asap, they are a huge time sink [11:19] didrocks: meh nevermind [11:19] sil2100: done btw [11:19] sil2100, pushed [11:19] didrocks: \o/ [11:20] didrocks: I'd changed the mode to stable [11:25] cyphermox_, http://162.213.34.102/job/landing-001-2-publish/ just a few lines in two differen tpackages there [11:28] sil2100, ah, sorry, was afk for a bit [11:28] but i see didier already kicked one === MacSlow is now known as MacSlow|lunch [12:18] sil2100: http://paste.ubuntu.com/6884858/ [12:19] didrocks: thanks :) === alan_g is now known as alan_g|lunch [12:43] I love green, awesome [12:43] with the new mir and so on [12:51] rsalveti: indeed, this really made our day in the morning [12:52] rsalveti, yeah, now we just need to make bfiller fix that darn dislaer-app crash [12:54] hahah [12:54] didrocks: http://xkcd.com/1296/ [13:01] lol === cjohnston changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cjohnston | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: - [13:08] didrocks: do you know if the ci train ppa builds are also respecting components? [13:09] didrocks: like, giving errors when a package from main starts depending on some others from universe? [13:09] probably not yet [13:11] when do we expect qt 5 packages build on ppc again? [13:11] rsalveti: no, I remember we enabled the components option in the daily-release ppa for a while but there were some issues with it. We tried to remember with Laney but failed [13:11] something to clearly look at [13:12] right [13:12] xnox: ahah :) [13:12] xnox: there is one commit in unity7 where njpatel put "I made a boo boo" [13:12] lol [13:13] thostr_: when we'll migrate on 5.2 [13:14] didrocks: so, what do we do until then? ignore all ppc builds? === MacSlow|lunch is now known as MacSlow [13:15] thostr_: well, the system is looking at the previous publication of your component [13:16] so, if you already have it built, yeah, you need to do that painful transition until 5.2 is there [13:16] but if the previous one didn't built one powerpc, the system knows it and ignore the dep-wait [13:30] not just powerpc, also arm64 and ppc64el in many cases [13:45] cjwatson: yeah, we are just talking about the jenkins job first, which is using the ppas which don't build arm64 and ppc64el [13:45] but I'm looking at all arch for proposed->release, don't worry :) === alan_g|lunch is now known as alan_g [13:49] right, though once we're less hw-constrained those PPAs should be building arm64 and ppc64el too === Ursinha-afk is now known as Ursinha [14:11] psivaa: Laney: ah, my bad. I added it, but didn't create the job for it. I'll do that at the end of the current image run to avoid disruption, and then run it alone so that it gets added to the daily results [14:13] plars: sweet === Ursinha-afk is now known as Ursinha [14:19] mhr3_, I have unity-scopes-shell building in https://launchpad.net/~ci-train-ppa-service/+archive/landing-003/ , will be ready for testing soon. [14:20] robru, awesome [14:22] ogra_: ping [14:23] sil2100, answered in the PM [14:24] ogra_: thanks ;) [14:40] didrocks, ogra_, psivaa, davmor2: you know how the dogfooding is going of image 166? [14:40] Can we promote that? [14:40] om265er was doing that [14:41] sil2100: done as far as I was aware didrocks knew about it this morning/early afternoon [14:41] davmor2: you did both mako and maguro? [14:42] sil2100: I did none om26er did them I'm testing qt 5.2 [14:42] it's really broken [14:42] sniff [14:42] whats broken with it ? [14:42] davmor2: oh, you mean Omer finished testing [14:42] ogra_, didrocks: are we promoting the image then? [14:43] * ogra_ hasnt heard back from omar ... so up to didrocks [14:43] kgunn: can we try to land latest mir already? :-) [14:43] ogra_: the dialer, installing click packages, the weather app, the indicators are in a different order (not sure if that is a big issue) etc etc [14:43] * rsalveti wants the hwcomposer 1.2 support [14:44] kgunn: also, were you able to get someone to start working on enabling run-time backend selection in mir? [14:44] davmor2, bah, thats bad news [14:45] and now I can't file a bloody bug grrrrrrrrrrr manually it is then https://docs.google.com/a/canonical.com/spreadsheet/ccc?key=0AjuCdq68GSyVdGI4dGllUUxyZGxhc0tZWFhqNnJaaFE#gid=0 [14:45] ogra_: ^ [14:45] rsalveti: sorry...had to run my kid to school....whole hour...its snowing here, crazy [14:45] ogra_, didrocks: so, Omer finished testing, it looks green for me [14:45] rsalveti: alf is going to look at it today for scope [14:45] kgunn: haha, no worries [14:45] meanwhile I'm melting here [14:46] w/o looking he thot couple to a few days [14:46] * kgunn wishes he were melting [14:46] kgunn: great [14:46] * ogra_ has nice spring weather ... 10°C and sunshine ... (while it should be -10 and snowing) [14:46] 10C is fine [14:46] yep [14:47] just enough to leave the terrace door open while working [14:47] totally unusual though [14:47] kgunn: are you first going to release mir 1.5 before we try another landing? [14:47] or can we already flush trunk? [14:48] rsalveti: just looking...at spreadsheet...i can punch merge now right ? [14:48] afaik trunk is already 2/3 weeks ahead of 1.4 [14:48] kgunn: already did that yesterday [14:48] kgunn: we're all good [14:48] rsalveti: oh well...not really... unless you mean trunk == lp:mir/devel :) [14:48] we're weird [14:48] on mir team [14:48] kgunn: yeah, devel [14:49] we need to flush things faster for mwc and 4.4 support [14:49] rsalveti: oh, no...the good news is, while pipe was clogged i updated [14:49] ++ [14:49] rsalveti: oh yeah...i missed that one damn commit by a day [14:49] rsalveti: it went in on devel y'day [14:49] right [14:50] will queue up another mir quickly...just need to talk to guys to see if anything else will come thru we should "wait" on [14:50] wait == 2 days maybe [14:50] 2 days is fine [14:52] mhr3_, looks like the build is ready. let me know how your testing goes and then I can publish it if successful [15:02] robru, i pushed one more rev to the mped branch, and trying to build again doesn't work, what to do? [15:07] mhr3_, how are you trying to build? I think I need to build it myself... [15:07] mhr3_, just want to confirm, same MPs right? just a new commit, no new MPs? [15:07] robru, i just clicked the build button again [15:07] right [15:08] robru, i knew you need to do reconfiguration if i made a new branch, so thought doing this would be easier [15:08] apparently i was wrong :P [15:09] mhr3_, no, I still think this is easier... [15:09] mhr3_, but something goofy is going on. CSS isn't loading here so the page looks all impossible to read [15:09] did i break jenkins? [15:10] yey me! [15:11] mhr3_, I dunno, restarted firefox and it seems fine. my system is messed lately. ok, I reran the build with 'ignore_step' selected and it seems to be working. I'm not sure why it didn't notice the new commit before [15:12] oh well.. adding to train weirdness [15:12] robru, still can't believe you called it train... trains and nice and cosy :P [15:13] mhr3_, haha, i didn't come up with the name. despite a few warts, this system is *wayyyy* slicker than what we had before. Now we get packages built in PPA directly from MPs. makes it way easier to install the stuff on the phablet for testing. way easier to automate testing as well === alan_g is now known as alan_g|tea === greyback is now known as greyback|away === alan_g|tea is now known as alan_g [15:32] mhr3_, ok it looks like the new commit got built. please test! [15:33] ogra_: is there a nice way to get the crash report off the phone in a sensible format? Currently apport-cli -c /path/to/crash keeps telling me its not an official app so can't report grrrrr [15:34] mhr3_, actually can you check that the new commit made it in? suddenly I have doubts... [15:34] davmor2, hmm, no idea, ask pitti [15:34] robru, it's in the diff, so yea [15:34] mhr3_, ok great. still getting the hang of this train thing ;-) [15:38] ogra_: thanks will do :) [15:40] ogra_: so! [15:40] sil2100, still waiting for a word from your boss [15:40] ogra_: didrocks said a +1 on promoting 166! [15:40] * ogra_ doesnt want to be cursed at in french if he released without approval [15:40] ogra_: he's busy in a meeting but I poked him personally [15:41] ok [15:46] ..."Waiting for other process to release the global lock" ... [15:46] * ogra_ twiddles thumbs [15:51] mhr3_, ok just ping me when you're satisfied with the testing and I can publish [15:51] robru, i'm satisfied [15:51] robru, all the code it touches doesn't run in the default image so it's completely safe [15:51] mhr3_, also it turns out that technically you are supposed to click build. for future reference, you have to check 'IGNORE_STEP' to make it rebuild new commits. [15:51] mhr3_, ok great [15:51] === Image 166 Promoted === [15:52] that's why noone noticed that i'm fixing a possible crash :) [15:55] mhr3_, ok I published it. once you see it get past -proposed and into the archive, you can go ahead and click 'merge and clean' [15:57] \o/ === cjohnston changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: doanac | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: - === gatox is now known as gatox_lunch [16:33] thostr_, ok, upstart-app-launch is ready for building in silo 5. [16:33] robru: thanks [16:34] thostr_, you're welcome! ping me when you're satisfied with the testing and then I can publish it for you [16:34] I'll... will take a while though [16:42] thostr_, ahhh it looks like we all EOD in an hour ;-) so don't rush, we'll publish it tomorrow morning. [16:42] thostr_, after this sprint is over we'll have better geographic dispersion for 24h landing-availability [16:42] robru: that will be very handy then [16:43] thostr_, next week ;-) [16:43] seb128, looks like settings finished building if you want to start testing [16:44] mhr3_, scopes-shell is in the release pocket! please 'merge and clean' that landing [16:44] robru, done, thx [16:49] robru, I've tests running for 10 minutes or so [16:50] seb128, haha, ok. [16:51] didrocks, We're getting an odd failure: http://162.213.34.102/job/landing-005-1-build/30/console [16:51] didrocks, Could that be due to the override version? [16:53] I'm getting an exception of packagemanager in silo1... could anybody check? [16:59] tedg: meeting, I'll have a look in few sec [17:00] Mirv: https://launchpad.net/~mathieu-tl/+archive/ppa [17:00] didrocks, Who's in the meeting? Should I ping them so they need to leave too? ;-) [17:01] tedg: we are coming :) [17:01] ahah :) [17:02] cyphermox_: merci [17:02] tedg: ah, this is expected [17:02] tedg: robru did rerun it with the correct parameter [17:02] to tell "don't care about what's in the archive" [17:02] we can explain you more later on [17:02] plars: joining? [17:03] didrocks: yes, one sec [17:03] didrocks, Okay, is there something we should be putting in the landing sheet to signal that? [17:03] tedg: no, you can do that on your own [17:03] tedg: it's the "ignore destination version" (or whatever it's called) [17:04] didrocks, I can't :-) thostr_ ^ [17:04] yeah, thomas has power! [17:04] tedg: let me try [17:04] thostr_, I think that robru already did [17:04] thostr_, There's a build 31 running [17:05] tedg: ok [17:05] tedg: if not, just reping [17:05] Will do, was more mentioning for the future. [17:05] updated wpa + nm, restarted network-manager [17:10] Mirv: ok === greyback|away is now known as greyback === gatox_lunch is now known as gatox [17:48] ogra_: why didn't you share Didier's post yet?! [17:50] sil2100, sorry, was catfeeding [17:50] :) === alan_g is now known as alan_g|EOD === doanac changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cihelp | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: - === plars changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: plars | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: - === vila_ is now known as vila [18:59] pmcgowan, ogra_: I know the bug reports for the QT5.2 are really short for all of those I can pad them out with steps and so on I've also been grabbing logs and crash reports as I've gone. Mirv has let me know that the multimedia is still down which I think is the cause for a lot of the high bugs hence no report I'll retest when the qt multimedia fix lands hopefully tomorrow :) [19:03] davmor2, thanks [19:07] rsalveti: the QT5.2 qtmultimedia fix is that in the ppa now? So I can trial it tomorrow :) [19:09] davmor2: not yet, sergiusens is working on that [19:09] ah cool thanks [19:11] rsalveti, right, let me get to it [19:29] Hi - can anyone tell me who I should talk to to get window-mocker landed in distro? I guess it's still on the old system, since automerger is still running for it [19:31] is it known that wifi is broken on latest proposed? [19:32] nmcli d shows no wifi device === plars changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cihelp | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: - [22:21] thomi, by any chance does veebers know? [22:21] fginther: otp, one moment [22:21] he's the "lander" contact for the other components [22:25] fginther: I'm not sure what you're asking? [22:25] veebers is on holiday today, I can probably help though [22:26] thomi, on the ci-train spreadsheet, veebers is listed as the contact for the other autopilot components. I assumed he was versed on the process or at least knows who to talk to [22:26] yeah, he knows, but.. does *everything* go through ci-train now? [22:26] and if that's the case, how come automerger is still switched on for some things? [22:27] thomi, ah, you weren't asking for it to be added to ci-train. [22:27] hmm [22:27] no [22:27] I really don;t know which process to use to land that component [22:29] thomi, you should be able to get help from the daily release team (didrocks, cyphermox_, kenvandine, robru, sil2100, Mirv) [22:30] thanks, will email them [22:30] ls [23:12] am I missing something, or do annon users no longer have ability to retry jobs on s-jenkins? [23:13] i need to retrigger merge proposal, now that protobuf/mir is fixed in the archive. [23:13] since this is just another merge proposal, I should be able to self-service myself. [23:13] why can i not click on http://s-jenkins.ubuntu-ci:8080/job/dialer-app-ci/159/rebuild ? [23:14] cihelp ^ [23:16] xnox, I can click that for you, to my knowledge there has never been anonymous access to that ability [23:16] xnox, rebuilding now [23:17] fginther: which is very odd. Given that it's using up your highly valuable time, on trivial button pushing. [23:18] fginther: or is it again the case of not being able to do a per job access control? [23:18] fginther: can we do some kind of generic remote triggers for any jobs on any jenkins, and then put a sensible ACL mediation on top of it. [23:19] cause really, anybody should be able to retry merge-proposal/ci, and the person who propose the merge should be able to self-service. [23:19] xnox, yes, jenkins ACL is not specific enough to cover this. There really should be a REST api that can be triggered to accomplish the same goal [23:19] that is, we could implement a very basic API to accomplish that task [23:20] and then "just" provide the ability to rebuild given the link [23:31] fginther: hm [23:31] "Project-based Matrix Authorization Strategy? [23:31] with that, you can enable 'Enable project-based security' per job basis, where you can set up the matrix permissions the same way as you set the global ones." [23:31] i'm not sure what half of those words mean, but it sounds good. [23:32] also https://issues.jenkins-ci.org/browse/JENKINS-1504 [23:37] xnox, thanks for the pointers, this area has been a hot button lately, I'll talk to ev about getting something worked up [23:37] fginther: i guess i should spin up a jenkins to do it. [23:38] fginther: since all jenkensi are behind VPNs we should be able to allow e.g. any authenticated with openid user to push /some/ buttons [23:38] fginther: e.g. the retry one on adt, or retry one on merge proposal thingy. [23:38] xnox, gotta go, let's pick this up again [23:39] fginther: ta! see ya =)