=== salem_ is now known as _salem === chihchun_afk is now known as chihchun === _salem is now known as salem_ === salem_ is now known as _salem === _salem is now known as salem_ === salem_ is now known as _salem [07:07] morning trainguards! any idea why silo 27 failed on dependencies https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-027/+build/7767209 on ppc and arm but built fine on others? https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-027/+packages [07:10] moreover, the status of the silo in citrain says all has been built fine, but ppa status disagrees? [07:26] pstolowski: unity8 hasn't ever built for arm64 powerpc ppc64el, therefore train thinks it's ok. like it says, dependencies are missing. [07:26] pstolowski: note armhf built fine, just arm64 didn't [07:28] Mirv, hmm, so all is fine with that silo? [07:47] pstolowski: looks so [07:47] Mirv, yeah, thanks for explaining! [07:48] So... so hot! [07:49] Mirv, davmor2, ogra_, popey: I would opt for skipping todays morning meeting, not much to discuss [07:49] +1 [07:49] And I would have to put something on me for the hangout [07:52] ogra_: ok then [07:52] sil2100: please don't share more [07:52] (although, likewise) [07:56] GO TEAM PYJAMAS! [07:56] ;) [07:57] bathrobe rockers please :) [07:58] sil2100, hmm now the armhf build has dependency issues too [07:58] failing on obexd-server [08:00] huh? [08:03] ogra_: anyway, looking into that [08:03] Thanks for the heads up! [08:04] UH ! [08:04] so the ubuntu-system-settings landing sits on top of wily packages ? [08:05] (at least that is what the changelog suggests) [08:06] https://launchpadlibrarian.net/213772726/ubuntu-system-settings_0.3%2B15.04.20150806-0ubuntu1_source.changes [08:06] ubuntu-system-settings (0.3+15.10.20150804-0ubuntu1) wily; urgency=medium [08:06] ogra_: yeah, dual landings work like that [08:06] We don't re-write history, just the top most changelog entry [08:06] well, do you also not re-write dependencies ? [08:07] pulling in the wily bluetooth deps might indeed explain this [08:07] (assuming they are different) [08:08] No, it's the lander that needs to make sure the deps are sane on both platforms [08:09] The recent u-s-s upload didn't have any packaging changes [08:09] But maybe the one before that [08:09] Need to dig into the changes for that [08:09] sil2100: you can just turn off the camera like ogra_ does ;) [08:10] https://launchpadlibrarian.net/213772795/ubuntu-system-settings_0.3%2B15.04.20150413-0ubuntu1_0.3%2B15.04.20150806-0ubuntu1.diff.gz [08:10] - qtdeclarative5-ubuntu-settings-components (>> 0.2), [08:10] + qtdeclarative5-ubuntu-settings-components (>> 0.6), [08:10] thats the only possibly relevant dependency change [08:10] Launchpad doesn't work for me right now [08:10] ugh [08:11] ogra_: the overlay has 0.6+15.04.20150716-0ubuntu1 of that package [08:12] yeah, and there doesnt seem to be anything that could break it like that [08:12] its not that [08:13] https://lists.launchpad.net/landing-team-changes/msg00368.html <- maybe this landing was responsible? [08:13] As it's trying to install ubuntu-system-settings-online-accounts [08:13] but it most likely is system-settings related ... i386 has the same error (on top of the old dep issue) [08:14] * No-change test rebuild for g++5 ABI transition [08:14] huh ?? [08:14] oh [08:14] the wily changelog confused me again [08:14] Don't worry about the changelog [08:15] hm, no dep changes there through [08:15] *though [08:15] sil2100, if i wouldnt know better, i'd say the seed change from wily landed in vivid [08:15] but thats not possible [08:16] Nothing like that landed: https://lists.launchpad.net/landing-team-changes/ [08:18] Ok, I'll try digging into that on my device [08:18] no, but we also have security and updates enabled during build [08:18] Yes, but nothing landed there too re: the seeds [08:18] https://launchpad.net/ubuntu/+source/ubuntu-touch-meta [08:19] right [08:24] I think I'm getting close to the problem [08:28] hah, ok, see the problem... [08:28] ogra_: so the problem is in libtimezonemap1 (from the libtimezonemap source) [08:28] ogra_: u-s-s was built against the version in -proposed [08:29] ogra_: since we build out silos with -proposed enabled, and the new libtimezonemap still didn't migrate [08:29] ah ! [08:29] So u-s-s is uninstallable with other packages... we need to migrate it [08:29] There's no hard dep, but the binary packages depend on it as that's what it was built against [08:29] indeed [08:29] 0.4.3.1 - this needs to go out, let me check excuses [08:30] on a sidenote wily i386 fails too ... [08:30] The following packages have unmet dependencies: [08:30] ubuntu-touch : Depends: qtmir-android but it is not going to be installed [08:30] Depends: qtubuntu-android but it is not going to be installed [08:30] (probably not that urgent :) ) [08:31] ogra_: yeah, didn't have time for that yet ;) Since emulator is b0rken anyway [08:31] wily ... not vivid though ... [08:32] (not sure if the emulator was broken there too, now it clearly is though :) ) [08:33] hmmm, we need to somehow resolve this, I might even consider copying this package to the overlay [08:33] Since it still needs to wait a bit in the SRU queue [08:33] Building against -proposed in the case of an already released series is not a good idea ;) [08:33] yeah [08:34] at least if your silos dont have proposed-migration too :) [08:34] seb128: would you mind me copying over the binaries of libtimezonemap from -proposed to the overlay? :) [08:34] sil2100, please do [08:34] unsure what's the issue [08:34] ogra_: true ;p ALthough we would have a slightly different -proposed pocket anyway [08:34] seb128, dep issues [08:34] the soname didn't change, why is u-s-s having an updated depends? [08:35] seb128, silos buld against proposed but their packages dont go through the migration [08:35] like no public interface changed [08:35] seb128: u-s-s built against the libtimezonemap in -proposed, not sure why it actually hard-depped the binary deps [08:36] It's certainly something worth looking into, but generally copying the package is much faster [08:36] ah [08:36] DEB_DH_MAKESHLIBS_ARGS = -V [08:36] there you go :) [08:36] Oh man, I hate that one [08:36] that package always update its shlibs to current [08:36] I once got yelled at for sigining off a package with this in the old times [08:40] Packages copied, I'll re-kick the image build in a moment [08:40] yay === ljp is now known as lpotter === chihchun is now known as chihchun_afk [10:08] sil2100, hey, we have failure on version check https://ci-train.ubuntu.com/job/ubuntu-landing-005-1-build/176/console, presumably because we're trying to land in vivid overlay only. what is the best course of action? [10:09] Let me take a look [10:09] sil2100, we have single trunk for both wily and vivid [10:09] sil2100, and i don't want to dual land atm due to the issues [10:09] Ah, yeah... [10:10] This is a bit troublesome in overall, since we really wouldn't like to garbage the devel trunk with vivid commits [10:10] Since then there would be one big inconsistency if you want to get back to releasing for wily from trunk [10:11] sil2100, hmm. would bumping to 0.3 be an option? [10:12] No no, this would be just dodging the check, but the trunk would still be inconsistent [10:12] I would recommend to do something like this: [10:13] Branch off current trunk to some 15.04 branch, change (by a direct push to the 15.04 trunk) the top most debian/changelog entry to include 15.04 in its version number instead of 15.10 and use that branch instead [10:13] Once dual landings are back, you can simply drop using the 15.04 branch completely [10:13] And do everything from trunk once again [10:14] But at least this way trunk will be consistent and not go back in time with releases [10:15] sil2100, ah, i see, thanks for explaining. will try to do that [10:17] sil2100, also need to s/wily/vivid/ in that topmost changelog entry? [10:17] pstolowski: yw! I know it's a bit of work, but we already have so much things that the train does wrong that I would at least like to keep all trunk branches 'sacred' [10:17] pstolowski: you could do that, yes, not required but will add to the consistency [10:17] :) [10:18] heh, interesting view on consistency ;) [10:20] jamesh, ^ gyi [10:20] fyi even === _salem is now known as salem_ === alan_g is now known as alan_g|lunch === _morphis is now known as morphis [12:44] davmor2: hey, I suspect silo10 has slipped through the spreadsheet->bileto transition, ca you please add it to your trello board? [12:50] sil2100: ^ [12:51] greyback_: let me take a look [12:51] greyback_: ok, so it's there, but it's not set as 'Ready for QA' [12:51] greyback_: once you're done with testing you need to mark that it's ready for QA sign-off [12:52] greyback_: like in the spreadsheet, just the fields are called differently [13:01] sil2100: davmor2 oh wow, sorry, I totally missed that [13:01] I thought it said ready for QA [13:02] greyback_: no worries === jhodapp|sick is now known as jhodapp === alan_g|lunch is now known as alan_g [13:17] rvr: silo 4 has fix for gallery not starting up if you have time to test it [13:19] bfiller: Thanks! Busy with other things right now. If not me, someone else will take it for sure :) [13:19] rvr: great, thanks [13:21] sil2100, heeeeeelp [13:22] pstolowski: what's up? [13:22] sil2100, made a 'temporary' trunk, here is the MP https://code.launchpad.net/~stolowski/unity-scope-mediascanner/audio-card/+merge/267325 [13:22] sil2100, the silo still fails https://ci-train.ubuntu.com/job/ubuntu-landing-005-1-build/181/console [13:22] hm, let me seee [13:22] sil2100, i have no idea where 0.5.4+15.10.20150707-0ubuntu1 comes from [13:23] sil2100, here is the changelog of the 15.04-trunk http://bazaar.launchpad.net/~unity-team/unity-scope-mediascanner/trunk-15.04/view/head:/debian/changelog where i changed 15.10 to 15.04 [13:24] Let me look deeper what's going on [13:25] pstolowski: oh, but it's failing for unity-scopes-shell [13:25] pstolowski: the branches you showed me are for unity-scope-mediascanner - you need to do the same for unity-scopes-shell ;) [13:27] sil2100, uh, oh, indeed, ovelooked it, sorry [13:28] sil2100, but that's weird, yesterday i succesfully built unity-scopes-shell in silo 27 without needing to do that [13:28] and it was only vivid too [13:29] Interesting [13:30] btw. we need to reconf the silo to target the overlay [13:30] Since now it's targeting the main vivid archive [13:30] sil2100, oh, which one, 27? [13:30] pstolowski: 005 [13:30] I changed it, reconfiguring now [13:31] sil2100, hmm, and why not 27, it looks the same in bileto? [13:31] I don't know, you showed me links for silo 005 [13:31] So I thought you're working on 005? [13:32] sil2100, yes, i'm working on 5 now, but as i said, yesterday i built unity-scopes-shell in silo 27 (which is supposed to target vivid-overlay) [13:32] 15:22 < pstolowski> sil2100, the silo still fails https://ci-train.ubuntu.com/job/ubuntu-landing-005-1-build/181/console <- so I wanted to reconfigure this silo to target the overlay [13:32] sil2100: can dual silos still land? [13:33] I can reconf 27 too as well if it's not for overlay [13:33] greyback_: if you're able to test your silo against wily then yes ;p ;p [13:33] sil2100: right so ;) [13:34] sil2100, both 5 and 27 should target vivid-overlay, i was assuming that when 'vivid' is slected in bileto, it will actually be vivid-overlay? [13:34] pstolowski: no [13:34] pstolowski: since bileto still needs to enable people to land stuff as SRUs, right? [13:34] sil2100, hmm right [13:35] sil2100, then please reconfigure both 5 and 27 [13:35] pstolowski: so to get something landed to overlay, you select vivid and then 'Destination PPA' select the overlay (it should be available as a auto-complete option) [13:35] On it :) [13:35] * pstolowski is back to square one after 2 days of fight with silo 27 :( [13:36] No rebuild required! [13:36] At least not for the retargetting [13:36] sil2100, ah, cool [13:36] :) [13:36] sil2100, changelog versions etc not affected? [13:36] Should be good now [13:37] pstolowski: right, we're using the same versioning for overlay and vivid for the train [13:43] huh? [13:43] sil2100, i haven't updated scopes-shell [13:43] on it [13:44] Ah :) [13:51] :( [13:52] sil2100, it's complaining about unity8 version now... that brings back the question why it worked in another silo yesterday where i built both unity-scopes-shell and unity8 [13:53] pstolowski: are you sure it wasn't because it was a dualo silo then? [13:53] Let me check the logs [13:53] sil2100, it was dual silo originally. then i change to vivid only and reconfigured [13:54] pstolowski: probably the build happened when it was a dual silo still or something [13:55] sil2100, no, not possible, actually you changed the silo for me yesterday, the new one was vivid from the start [13:55] Looking at the logs [13:57] pstolowski: ah... yeah, it worked for silo 27 as you changed the upstream version to 0.5.5 [13:57] pstolowski: which camouflaged the issue [13:58] Damn, this would mean you'll release a 15.04 version to trunk with that merge... eh [13:58] wonderful [13:58] Well, I suppose we could live with that [13:58] Don't want to force you rebuilding that silo, just remember it will introduce some uglyiness to your main trunk [13:59] sil2100, in that case i'd prefer to re-target to our temporary trunk [14:00] I suppose no re-test would be needed as it's the same contents [14:14] trainguards: is it a known issue that autopkgtests for qtcreator-plugin-ubuntu are failing and blocking migration of a number of packages in wily? [14:15] oSoMoN: yes [14:15] oSoMoN: it's caused by seeds [14:15] sil2100, is there anything we can do about it? This prevents building new silos for e.g. webbrowser-app, because the jenkins job complains that webbrowser-app 0.23+15.10.20150729-0ubuntu2~gcc5.1 is missing from the changelog [14:15] oSoMoN: we need new seeds updated and uploaded [14:16] oSoMoN: hm, I think it's a different case here [14:17] oSoMoN: so unblocking webbrowser-app from -proposed won't help, as this upload was anyway a direct upload [14:17] oSoMoN: what needs to be done is: [14:17] oSoMoN: first, let's double check if 0.23+15.10.20150729-0ubuntu2~gcc5.1 was a no-change rebuild for gcc [14:18] oSoMoN: if that's the case and we're not missing any real changes, two things can be done: [14:19] oSoMoN: first, the clean but longer method would be to take the missing changelog entries, adding them to the webbrowser-app trunk (by a direct bzr push to trunk) and then rebuilding the webbrowser-app silos [14:19] oSoMoN: second, faster but less cleaner, is simply using the FORCE_REBUILD flag while building, but that will simply override and ignore this changelog entry missing [14:19] Which is not super nice, but since it's a no-change rebuild, no one will cry for not having it in the changelog [14:20] sil2100, ok, I’ll take the first option and add the missing changelog entry to trunk [14:20] sil2100, thanks for the help! [14:20] oSoMoN: no problem! Any of the two is fine, but yeah ;) [14:43] sigh, so many test regressions in proposed [14:58] hmm, i wonder why so many of these packages didn't get picked up and rebuilt in the gcc5 silo ppa [15:10] jhodapp, how's your silo 21 coming along? [15:10] pmcgowan, actually silo 38 now :) [15:10] pmcgowan, it's building everything now and if it succeeds, I'll do a quick local sanity check again...if I'm happy it'll be ready for QA [15:11] great davmor2 alecu ^^ [15:12] pmcgowan: thanks [15:13] davmor2, alecu: so to test this, you'll just go through the listed test plan, which is for media-hub. We want to make sure that no existing use cases are broken. [15:14] jhodapp: I'm going to disappear in a minute but reappear later [15:14] davmor2, no worries, it's not ready to test yet anyway [15:14] so more likely monday morning then [15:15] that's perfect...I'll send you an email confirming that it's ready for you on Monday morning [15:22] cihelp: hey! Could we get qtcreator-plugin-ubuntu autopkgtests re-run? [15:22] cihelp: who should I poke regarding those? We landed a seed change that should help [15:23] http://autopkgtest.ubuntu.com/packages/q/qtcreator-plugin-ubuntu/wily/amd64/ (and i386) [15:24] sil2100, I have no idea how to rerun those. Is pitti or jibel available? ^ [15:24] jibel is off on holidays, might look for pitti in this case [15:24] fginther: thanks! [15:24] sil2100, or maybe infinity? ^ [15:29] pstolowski: jhodapp gave a status of the silo we depend on, a few lines above ^^^ [15:32] alecu, great, thanks [15:35] alecu, we have no clarity what to do with unity8 atm (in terms of branching), i discussed this with tsdgeos but we will wait for monday when other unity8 guys are back [15:39] Mirv: do you have the power to re-run autopkgtests still? [15:39] tsdgeos, also, give silo 38 a try with your music-scope changes [15:40] jhodapp: pstolowski was giving it a go i think [15:40] tsdgeos, ok great [15:40] I'm also testing it right now [15:40] i'm trying to fix/debug a crash in unity8 startup [15:40] not the scope stuff, but silo 38 [15:40] jhodapp, we planned to, but i got stuck with the build issues (may need to have separate trunks for all projects involved) [15:42] pstolowski, build issue for the music-scope silo? [15:43] jhodapp, yes (silo 5) [15:43] ok [15:45] trainguards, I am getting a signature error that I do not understand: https://ci-train.ubuntu.com/job/ubuntu-landing-057-1-build/2/console [15:46] abeato: hey! The problem is different [15:46] 2015-08-07 15:39:54,683 ERROR dch: fatal error at line 1141: [15:46] New version specified (0.108+15.04.20150807-0ubuntu1) is less than [15:46] the current version number (0.108+15.10.20150807-0ubuntu1)! Use -b to force. [15:46] You need to branch trunk for an older ubuntu series, or do a sync silo from a newer release to an older release. [15:47] sil2100, hmm, ok... when I select vivid in the citrain, that's vivid-overlay? [15:47] abeato: no, you need to also add info about the overlay in the 'Destination PPA', it should be in the autocomplete there [15:48] But reconfiguring it doesn't require a rebuild [15:48] Just change it and reconfigure your silo [15:48] ok, thanks [15:50] sil2100, so I select the overlay ppa and keep the target series to "vivid"? [15:50] Yes [15:50] ok [15:51] sil2100, I also want to sync a package from https://launchpad.net/~canonical-arm-dev/+archive/ubuntu/ppa/+packages [15:51] I have used ppa:~canonical-arm-dev/ubuntu/ppa,vivid as sync source [15:51] is that right? [15:55] sil2100, also I am nit sure about what how to reconfigure: does editing the entry+saving does it? [15:55] abeato: after you edit the entry and click save, simply click the 'Assign' button once again (as if reassigning) and proceed [15:55] * sil2100 looks if the docs are updated for that [15:56] sil2100, ok, got it [15:58] abeato: just in case it's needed: https://wiki.ubuntu.com/citrain/LandingProcess#Modifying_your_existing_landing [16:00] sil2100, great, I'll take a look [16:01] sil2100, what about the sync I want to do? should it work or I need somebody else to copy the package around (pulseaudio)? [16:01] abeato: pulseaudio hmm... you'd have to ask a trainguard or a core-dev to help with the sync [16:01] Since it's a non-train package [16:01] ok [16:11] ogra_: ugh, forgot to kick a new rc-proposed build [16:11] Kicking now [16:11] kick it then :) [16:11] like beckham ;) [16:15] davmor2, vivid doesn't boot after installing the silo...figuring that out now === chihchun_afk is now known as chihchun [16:48] robru: hi, you can go ahead and free silo 29, alreadu assigned silo 58 using the requests page [16:53] davmor2, what was fixed in the new device tarballs? [16:55] o/ [17:05] pmcgowan: user wipe mode changes mostly but I think it still needs the the one silo to land too but I'll hit that after, for now it just didn't break anything. [17:06] davmor2, thanks [17:10] boiko: thanks! [17:36] so it's just autopilot-gtk, and kde, blocking the gcc5 transition now, it seems [17:38] di dual landings build separately for each target (eg. wily and overlay), or is it a binary copy? [17:38] *do* [17:45] robru: silo 48 is mark for qa but it is a wily only build so can I throw it back to you please [17:46] davmor2: lemme see [17:46] davmor2: alright then [17:47] robru: ta [17:47] Mirv: morphis what's going on in silo 48? why was a wily silo marked as needing QA? [17:49] bregma: seems it is not exactly a copy, as each gets different version numbers generated [17:50] bregma: so different source packages, but only one ends up getting committed to the vcs upon landing [17:50] robru, can you please dput qtmultimedia from ppa:jhodapp/ubuntu/ppa to silo 38 [17:52] meh, right; devel-proposed doesn't include wily-proposed [17:54] jhodapp: sure, one sec [17:55] jhodapp: uh, "a different source with the same version is published in the destination archive" [17:56] robru, do I have to bump the pkg version each time? [17:56] jhodapp: yeah 1ubuntu4 is already in there, you'll need to upload a new 1ubuntu5 then [17:56] ok great, one min [17:56] jhodapp: yes! you can't go around having multiple different source packages with the same version number, silly. that defeats the point of having version numbers. [17:56] well sure, but I never released the last one so thought it may not matter [17:58] jhodapp: PPAs are basically the same thing as the ubuntu archive. for all intents and purposes putting it in the PPA is "releasing" it, it's just not in the blessed archive that users download from. [17:58] robru, ack [18:00] robru, ok try again [18:01] ok [18:02] jhodapp: alright it's in there, you can do a WATCH_ONLY now [18:02] awesome thanks! === jhodapp is now known as jhodapp|errand [18:19] hmm, requests seems out of sync with my silo [18:20] dobey: uh, I just hit publish, I hope you don't mean there's something missing from the PPA [18:20] dobey: what's out of sync? same MPs are listed [18:20] robru: no. the ppa is built and all [18:21] robru: it says "Building" but the packages have been built for hours [18:21] robru: and when i just set it to "publish without qa" i noticed that someone had apparently set it to "dual" and "qa required" (which were both incorrect) [18:21] dobey: dashboard should be considered authoritative there. sometimes network issues make the status update in bileto fail. [18:22] well now it's correct for the publishing state at least [18:22] weird [18:23] dobey: as for the other thing, I guess it's possible somebody else was editing your row. not sure who though [18:25] yeah. would be nice to have visible history there [18:27] dobey: that would have to wait for v2 I'm afraid ;-) === jhodapp|errand is now known as jhodapp === cwayne is now known as cwayne-afk [18:54] hmm, i guess sil didn't get anywhere with asking for the qtcreator-plugin-ubuntu autopkgtests rebuilds [19:02] michi: you forgot to check ONLY_FREE_SILO [19:18] michi: I'm also freeing silo 41 since the MP in there was released in your other silo that published. [20:01] robru, alright, got another qtmultimedia change to dput from ppa:jhodapp/ubuntu/ppa to silo 38 please (hopefully this should be the last one) [20:02] robru, actually hold off on that for now [20:07] robru, ok yes, please dput that [20:24] jhodapp: alright, copied [20:44] robru, thanks [20:45] jhodapp: you're welcome [21:41] robru: thanks. I abandoned some silos, but didn’t know that this isn’t sufficient [21:47] michi: no worries. the 'Abandon' link in requests.c.u.c is mostly for if you don't have a silo assigned already, like if you created two requests by mistake. [21:47] Ah, I see. [21:48] So, what should I do when I abandon a silo after it’s allocated? [21:48] Build again with ONLY_FREE_SILO? [21:48] michi: https://ci-train.ubuntu.com/job/ubuntu-landing-032-3-merge-clean/6/console you just about had it here, but you need to check ONLY_FREE_SILO [21:48] yeah [21:48] robru: I look at the landing process page and could find anything. [21:49] Might be good to add a sentence or two. My expectation was that abandon would do all that. [21:49] michi: sorry, I'll update that (it's a bit out of date still) [21:49] Sure, no problem. Just mentioning it :) [21:56] michi: https://wiki.ubuntu.com/citrain/LandingProcess#Abandon_a_landing ok ;-) [21:57] robru: Wow, most excellent, thank you! :) [21:57] michi: you're welcome! === salem_ is now known as _salem