=== marcusto_ is now known as marcustomlinson === marcusto_ is now known as marcustomlinson [09:31] ping trainguards [09:31] anpok_: pong [09:31] AlbertA resolved the boottest issues last night [09:31] in silo004 [09:32] do we have to deo the our test process? [09:32] *redo [09:33] anpok_: what packages had to be changed to get those working? [09:33] Since normally we recommend re-testing, but it also depends on what changes and where had to be made [09:33] General recommendation is: please re-test, since even the smallest change or even a rebuild can cause potential trouble [09:34] sil2100: we reverted changes inside mir - that affected libmirprotobuf0 [09:35] ok [10:28] sil2100: < davmor2> karni: you might want to ping sill2100 to get a unique number and mark the other section as yes so we get a ticket in trello [10:28] sil2100: is this something you can help with:)? [10:29] sil2100: I've added Telegram row in line 53 of the CI Train sheet /Tarballs and Clicks/. The one in row 50 should be marked as failed, if it's something you can do. [10:39] karni: on it! [10:39] :) [10:39] sil2100: cheers! :) [10:40] wow, thats quite a list [10:44] ogra_: what is [10:55] davmor2, that silo4 list [10:55] (of packages) [10:56] ogra_: yes an it increases all the time :) [10:56] elopio: hi, quick question: in an autopilot test, is there a reason why calling self.skipTest() will not actually skip the test ? is there anything extra i should do to make it work ? and is it documented anywhere ? [10:56] davmor2, yeah, the insatiable anpok_ :) [10:57] hmm [10:57] i just noticed that qtmir-gles has a different version number.. [10:58] in other words.. those are not the droids you are looking for [11:48] sil2100: hi, do you know where the branch for ^ is by any chance? [11:50] Laney, https://code.launchpad.net/~indicator-applet-developers/indicator-datetime/trunk.15.10 ? [11:51] doesn't have 0710.1 [11:51] Laney, https://code.launchpad.net/~indicator-applet-developers/indicator-datetime/trunk.15.04 in fact it seems [11:51] http://bazaar.launchpad.net/~indicator-applet-developers/indicator-datetime/trunk.15.04/revision/419 [11:52] k, what happened there? [11:53] dual landing I guess?* [11:54] unsure why they branched if they dual land though [11:54] charles or tedg might know [11:57] it also has a very unhelpful changelog [11:58] trainguards: can you please help me find out how to see the test output from https://launchpadlibrarian.net/212289175/buildlog_ubuntu-vivid-amd64.online-accounts-api_0.1%2B15.04.20150722-0ubuntu1_BUILDING.txt.gz ? [11:58] I got some failed tests, and I'd like to see what's failing [12:04] mardy, they can't, is "make tests" not outputing the errors on stdout/err? [12:04] seb128: no, it's producing a file [12:05] seb128: it's how cmake works, I'm afraid; how do you deal with it in u-s-s? [12:05] mardy, to debug those I usually makes debian/rules cat the log [12:05] cmake is buggy in that regard imho [12:06] seb128: I agree; ok, I'll try working around that in debian/rules [12:06] seb128: thanks [12:07] yw [12:09] mardy, I think something like that should work [12:09] override_dh_auto_test: [12:10] dh_auto_test || cat dir/somelog.log [12:10] mardy, non tested, but I did hacks like that before with success [12:13] seb128: cool, I'll try === _salem is now known as salem_ [12:20] hey sil2100, can you please publish silo 19 for me? [12:23] @trainguards: had to reconnect a few minutes ago.. is silo-004 ok now? [12:35] jhodapp: on it, was eating lunch [12:36] sil2100, np, thanks! [12:36] anpok: let me try publishing [12:37] jhodapp: https://code.launchpad.net/~phablet-team/qtubuntu-media-signals/fix-black-video-issues/+merge/262363 still needs approval [12:37] sil2100, alright, one min [12:38] anpok: grrrr... platform-api needs a rebuild ;/ [12:38] Version of platform-api at dest changed from 3.0.0+15.10.20150706-0ubuntu1 to 3.0.0+15.10.20150707-0ubuntu1 since packages built [12:38] This is crazy [12:38] ok [12:39] sil2100, ok that's approved now [12:39] jhodapp: republishing then [12:39] thanks [12:45] sil2100, is there still an issue? [12:52] jhodapp: no, reviewing [12:52] ok [12:55] It's good [12:55] Too bad we won't be able to dual land those anymore [12:58] sil2100: now [12:59] o/ [12:59] Aw come ooooon [13:00] anpok: now qtmir... [13:08] seb128, Not sure, there was a big rebuild for Mir, wonder if that got pulled in. [13:11] tedg, but why did the wily update got commited to the 15.04 branch? [13:11] seb128, Not sure. Someone screwed up? ;-) [13:12] tedg, that was sort of part of the ping/question [13:12] or is the indicator in dual landing? [13:12] seb128, We should ask charles to be sure as I haven't been following as closely. [13:12] in which case was the branching needed/wanted? [13:12] No, none of the indicators or dual landing. [13:12] are [13:12] shame [13:12] that spare works [13:13] Eh, kinda. If you don't care about what ends up in wily, sure. [13:13] why would bugfixes not be good for wily? [13:16] Because there are always integration issues. Bug fix works on foo version of a library but not bar. [13:17] If you're not looking at those, you basically get bit rot. [13:20] * sil2100 pokes ogra_ about the endorsement [13:20] * ogra_ whistles innocently [13:42] sil2100: ^ now rebuilding qtmir-gles [13:42] anpok: thanks! [13:50] trainguards: hi guys, could I get silo 46 reconfigured? thanks! :) [13:52] https://launchpadlibrarian.net/212294587/indicator-datetime_13.10.0%2B15.10.20150720-0ubuntu1_13.10.0%2B15.10.20150722-0ubuntu1.diff.gz and it got an empty changelog [13:58] pete-woods: on it [13:58] sil2100: thanks :) [13:58] pete-woods: actually, you can now reconfigure yourself! :) [13:59] sil2100: say what?? [13:59] I keep forgetting about that [13:59] The train is becoming more and more self-service [13:59] sil2100: I don't see any links for that on the dash.. [14:00] I take it the jenkins jobs are there [14:00] anpok: ok, this is stupid... the train doesn't check all components, just one after another - now it's qtubuntu [14:00] I just need to know which to run [14:00] pete-woods: you need to use the spreadsheet :) [14:01] pete-woods: find your landing line (it's written on the dashboard), click on any cell of the row, go to the 'Landing tools' menu and Reconfigure ;) [14:01] sil2100: I'm on the line, but too stupid to find the Landing Tools menu [14:02] pete-woods: it's next' to 'Help' [14:02] d'oh [14:02] seen it now [14:02] was looking in addons, etc [14:02] :) No worries, soon the whole spreadsheet will be gone ;p [14:02] but won't you guys miss it? [14:03] hm, a little bit, so much sweat and tears soaked into that spreadsheet [14:03] google will surely miss getting all the incident reports [14:04] ha [14:09] nerochiaro: no, it should be enough. Maybe you are skipping the test but it's failing during the setUp? [14:09] elopio: i am actually skipping the test during the setup, because the condition under which to skip the test will be detected only after the window appears [14:10] nerochiaro: then that should be enough. [14:10] do you have a trace and a link to the code? [14:17] elopio: let me try that again. i have a suspicion that i want to verify before submitting an actual bug [14:20] sil2100: any idea why 0.14.0 is already in wily? [14:20] mir-0.14.0 [14:20] I don't know, I think this silo got published once right? But blocked in -proposed [14:20] that was my understanding .. yes! [14:20] but now: [14:21] Yes, I see it's in -proposed indeed [14:21] Ah, so those package versions were caused by the earlier mir landing [14:21] ? [14:21] The rebuilds requested? [14:21] http://archive.ubuntu.com/ubuntu/pool/main/m/mir/ <- there is a 0.14.0 [14:22] seems so.. yes.. [14:22] If that's the case, let's force publishing - I'll double check that and try to force it [14:24] sil2100: thx [14:25] qtubuntu-gles now as a different version number -- but if it is true the source should be identical [14:35] mandel, so what's the plan for silo 9? i see it's been approved to be published? [14:35] mandel, are you going to do a separate silo for vivid? [14:36] kenvandine, yes, that is the idea :) [14:36] kenvandine, do to click-scope [14:36] mandel, however... that silo has an unapproved merge proposal [14:36] sil2100: sorry for the fuzz .. those packages are still just in wily-proposed [14:36] Yeah, we'll have to overwrite them [14:36] mandel, can you get that published so i can rebuild and publish my other silo? [14:37] kenvandine, vivid? [14:37] kenvandine, or the wily one? [14:37] wily [14:37] anpok: It's best to look on launchpad (https://launchpad.net/ubuntu/+source/mir) or use rmadison to see the archive state [14:37] i have a dual landing silo ready to go though :) [14:37] You can't see where packages are just by looking in the archive's pool [14:37] so i need to get your's published first [14:37] mandel, do you have a vivid silo already? [14:38] kenvandine, not yet, I had issues building the wily one, just got it fixed [14:38] kenvandine, I'm in a standup, as soon as I get out I'll create the vivid one [14:38] mandel, ok, thx [14:38] kenvandine, will ping someone to approve the mr missing [14:38] the udm branch is the one not approved [14:38] at least the only one i looked at :) [14:39] anpok: I'm now double confirming if it's ok to land [14:40] it is ok [14:40] better than ok .. [14:44] anpok: yep, confirmed [14:44] Trying to force publish [14:46] anpok: this silo has bad-luck - we have unbilt revisions here https://code.launchpad.net/~unity-system-compositor-team/unity-system-compositor/trunk/+merge/262866 [14:46] anpok: https://code.launchpad.net/~mir-team/qtubuntu/qtubuntu-gles-mir-release-0.14.0/+merge/264185 <- here as well, but we know about this one [14:47] sil2100: yip [14:47] anpok: anyway, unity-system-compositior might require a rebuild (and qtubuntu-gles in the same time might be rebuilt as well, since why not?) [14:47] anpok: or... [14:47] ok [14:47] anpok: or you could revert the changes that got pushed to that u-s-c branch [14:48] Not sure if that's wise though [14:48] hmm [14:49] It's hard to say which new commits we're missing, would have to dig inside the train [14:49] I suppose the 3 recent commits [14:49] yes [14:49] I am rebuilding those. [14:50] anpok: fingers crossed that now we'll be able to JUST PUBLISH [14:50] ;) [14:51] sil2100: Hi! So I want to add a slightly modified version of the lxc package to the vivid+overlay PPA. What do I need to do to accomplish this? Is there a Wiki or something so I don't need to pester you too much? [14:52] ChrisTownsend: hey! Yes, the best way would be to use the train as then QA can do the necessary testing - we have some documentation for that: [14:52] ChrisTownsend: https://wiki.ubuntu.com/citrain/LandingProcess [14:53] ChrisTownsend: the "Landing your change to Ubuntu" has an overall train-guide, but "Landing your change to the Stable Overlay PPA" (or "Dual-landing for stable and devel" if that fits your project) is what you would be most interested probably [14:53] Since the first section is more for people that never used the train before [14:53] ChrisTownsend: you landed things through the CI Train already, right? [14:53] sil2100: Ok, thanks! I've used the train for Unity landings before, so I'm familiar with that. [14:58] sil2100: Ok, since this is not a MP, I'll need to send you (or some other trainguard) the source package, right? My plan was just to use "apt-get source lxc" in Vivid, make my changes, update changelog, and build source package. Is this correct? [14:59] sil2100: I'm looking at the "CI Train for manual source uploads" section. [15:03] ChrisTownsend: yeah :) [15:04] ChrisTownsend: well.. not exactly apt-get source [15:04] ChrisTownsend: since the overlay already has some changes on top of it [15:05] lxc-android-config 0.227 <- that's the version in overlay [15:05] sil2100: Well, it's just the lxc package itself. [15:05] Aaaaah! [15:05] Sorry, yeah ;) [15:05] ohh ... [15:05] be very very careful with that though [15:06] ChrisTownsend: right, then apt-get source is fine, sorry, miss-read that [15:06] since our architecture kind of depends on a functioning lxc container for the android layer [15:06] ChrisTownsend: but also make sure the same changes are in wily as well (if that's possible) [15:07] ogra_: Hmm, ok. What I'm doing is adding one line in debian/lxc.preinst to add a "phablet" user in the section that creates /etc/lxc/lxc-usernet. [15:08] sil2100: I don't really want this to go in the wily archive. [15:08] sil2100: Ideally, only the overlay PPA. [15:08] ChrisTownsend, well, test it very carefully, we modify the hell out of lxc on the phone already to make the container work properly (there are various bits we had to disable like network support in containers and such) [15:09] ogra_: Oh..... [15:10] ogra_: Then this won't work anyways. Back to the drawing board... === chihchun is now known as chihchun_afk [15:13] ChrisTownsend, take a look at all the override upstart jobs in lxc-android-config ... thats our "hacks" pac kage for the phone where we keep them central [15:13] ogra_: Ok, I'll take a look. Thanks! [15:29] mandel: some of your merges need approval/review: [15:29] mandel: list here: https://ci-train.ubuntu.com/job/ubuntu-landing-009-2-publish/100/console [15:37] sil2100, don't let silo 9 land yet... mandel asked me to give it a quick test and i had problems [15:37] ACK, let's switch it to 'not-tested' then [15:37] yeah [15:38] sil2100, done [15:38] kenvandine: thanks :) [15:39] * kenvandine passes silo 9 and publishes 30 :) [15:39] mandel, sorry, silo 9 will also need a rebuild [15:40] mandel, with silo 9 it's just not downloading click updates :/ [15:52] sil2100: silo 004 looks good.. [15:53] Ok, on it again [15:57] anpok: were any packaging changes necessary to fix the boottest issues? [15:58] sil2100: we did bump an ABI for mir-graphics-drivers-android. [15:58] but it turned out not to be the cause.. [15:58] sil2100: we fixed it by reverting abi breaking changes in libmirprotobuf0 [15:59] anpok: ok, let me just get someone re-approving the mir packaging changes then [15:59] ogra_, slangasek: could anyone of you review this change? https://ci-train.ubuntu.com/job/ubuntu-landing-004-2-publish/82/artifact/mir_packaging_changes.diff [16:00] Actually more slangasek would be required here, as we need an archive admin [16:02] trainguards: I saw my soli go from "in the Proposed pocket" to "empty," and my MPs were merged. Where does that mean my package is? [16:03] s/soli/silo/, apparently I'm dyslexic when I lack coffee [16:03] kyrofa: hey, this means the package has been released to the main pocket and is now available in the target archives [16:03] kyrofa: if it was a dual landing then it means that the package is now both in the vivid overlay-ppa and the wily archive [16:03] sil2100, sweet, I was hoping that was the case, but I searched packages.ubuntu.com and saw nothing. Does it take a while for that to update? [16:04] kyrofa: what package did you publish? [16:04] unity-scope-snappy [16:04] kyrofa: it's there already ;) https://launchpad.net/ubuntu/+source/unity-scope-snappy [16:04] packages.ubuntu.com is not a super reliable source of information [16:05] Not instantly reliable I would say [16:05] sil2100, heh, good to know. Alright great, thank you! [16:08] sil2100, karni: all good to go on telegram [16:08] davmor2, karni: let's ship it! [16:08] davmor2: win \o/ [16:08] yes! [16:09] It works now finally, yes? [16:09] davmor2, you tried on OTA4/5 too? [16:09] sil2100: yes, we resolved the thumbnailer regression [16:10] jibel: meh I forgot ota4, ota5 was good give me a minute karni and sil2100 [16:10] I think if OTA-5 is good then it should be good [16:10] davmor2: naturally, take your time [16:10] yes [16:10] but davmor2's on the safe side to double check that [16:11] although if someone updates telegram he should see an upgrade to ota5 [16:11] sil2100: it will only take a couple of minutes after the flash [16:17] slangasek, robru: could you guys make sure to release silo 004? It needs a force publish as the previous 0.14.0 is still in -proposed [16:17] slangasek, robru: the mir packaging changes needs review by an archive admin, but then it should be fine to just force publish [16:17] robru: and a reminder from me about the compoment name ci train branch! [16:17] * sil2100 needs to jump out now [16:17] o/ [16:25] karni, jibel: video, image and text all sent on ota4 too [16:26] davmor2, \o/ [16:26] davmor2: \o/ ! [16:26] great news, thank you davmor2 [16:26] sil1200 has ofcourse disappeared now :0 [16:34] ogra_, do you happen to know if the latest wily image (152) is known to work on krillin? [16:34] fginther, https://bugs.launchpad.net/ubuntu/+source/live-build/+bug/1477051 [16:34] Ubuntu bug 1477051 in live-build (Ubuntu) "Phones on devel-proposed do not boot - /bin/sh: 1: /bin/sh: initctl: not found" [Critical,Confirmed] [16:34] infinity, is on it [16:34] ogra_, danke! [16:34] wall last nights wily buiolds are affected [16:34] *all [16:38] ogra_: There's a build going on cdimage right now with the new live-build... [16:39] yay [16:39] ogra_: Which I just realised will be broken, cause I didn't wait for it to migrate, and that build doesn't use proposed. [16:39] ogra_: So, uhm. After this, there will be ANOTHER build that fixes it. :P [16:39] heh, all fine [16:39] it is wily after all [16:39] as long as your change didnt migrate into vivid yet we're all fine [16:39] Yeahp, but I'm using wily to validate the bugfix for vivid (since the symptom and the fix are identical). [16:40] Fix is uploaded to both, so once I'm happy with the state in wily, then yay. [16:40] yeah [16:53] kenvandine: mterry: anybody got a sec for a packaging review? https://ci-train.ubuntu.com/job/ubuntu-landing-004-2-publish/82/artifact/mir_packaging_changes.diff/*view*/ [17:00] robru: that requires an archive admin review for package name changes [17:01] I'm looking at it [17:01] robru: and have all the other packaging changes in that silo already been reviewed? [17:02] slangasek: yes, they were reviewed the last time it was published. I think by mterry. [17:03] robru: ok. and why is the build-dependency on abi-compliance-checker being dropped? [17:04] slangasek: I dunno... anpok? ^^ [17:04] anpok: your mir upload in silo 004 is dropping the build-dependency on abi-compliance-checker relative to the archive, with no explanation in the changelog, no bug references, and we *just* got abi-compliance-checker promoted to main as part of an MIR to satisfy this build-depedency; nack on these packaging changes === alan_g is now known as alan_g|EOD [17:49] anyone around that can override ubuntu-system-settings promotion to release in wily? It's held because of a missing depends in autopilot, which I've proposed a fix for [17:50] kenvandine, try #ubuntu-release perhaps ? [17:50] re [17:51] slangasek: we dropped it because it was optional, and we were asked to drop ist.. [17:52] anpok: who asked you to drop it? [17:53] slangasek: we intend to add it as soon as it is in main https://bugs.launchpad.net/mir/+bug/1475982 [17:53] Ubuntu bug 1464447 in abi-compliance-checker (Ubuntu) "duplicate for #1475982 [MIR] abi-compliance-checker" [Critical,Fix released] [17:53] anpok: that's not how it works. [17:55] anpok: first, the build-dependency on abi-compliance-checker has been in the mir packaging for at least one stable release (which means the MIR should have been done /last/ cycle). Second, abi-compliance-checker is /already/ in main now, as a result of that MIR bug being processed. And third, packages get *dropped from main* if they're not being used [17:55] anpok: please revert this packaging change [18:01] this will take about an hour [18:01] slangasek: shall I ping you again as soon the landing ppa is updated again? [18:14] anpok: yes, that's fine [18:22] fyi, not sure what is going on with the spreadsheet for telepathy-mission-control-5. the package is in stable-phone-overlay but the spreadsheet's Status cell is empty and the ppa is empty [18:40] robru: ^^ spreadsheet explosion? [18:41] jdstrand: what row? [18:41] 59 [18:42] jdstrand: are you surprised that the PPA is empty? [18:42] jdstrand: sounds like you published it, and it published fine, but the spreadsheet just displays wrong/no status [18:43] robru: no, it was successfully copied over [18:43] robru: that is what happened. I just didn't do a merge/clean [18:43] so was mildly surprised it was empty [18:43] jdstrand: yeah merge/clean happens automatically after a successful publish [18:43] so wanted to make sure it did the right thing [18:43] ah, ok [18:43] cool [18:43] jdstrand: yeah the spreadsheet is a steaming pile, generally jenkins is working really well but the spreadsheet has trouble updating itself properly. [18:56] robru, how do I tell citrain command I want vivid version of a package [18:56] or dont I [18:56] pmcgowan: point it at a silo containing vivid packages? [18:57] pmcgowan: what's the scenario? are you trying to downgrade after having installed a silo? [18:58] robru, I want silo 3 [18:58] which is dual landing [18:58] do I just use ubuntu for the distro? [18:59] pmcgowan: yes if your device is vivid you'll get the vivid packages, if your device is wily you'll get the wily packages [19:00] robru, very good [19:00] pmcgowan: thanks [19:03] robru, is it supposed to get updates to every package that landed? [19:04] pmcgowan: not sure what you mean. the tool will install every package in the silo onto the phone. [19:04] pmcgowan: did you pin the silo first? [19:04] pmcgowan: https://wiki.ubuntu.com/QATeam/TestingProcessesUbuntuPersonal [19:05] robru, it updated everything in the overlay [19:05] pmcgowan: oh yeah the version of the tool in vivid has a bug, if you branch the trunk version from lp:phablet-tools it should work [19:05] pmcgowan: I believe that's expected, yes. [19:05] pmcgowan: yes expected, but you probably didn't get the silo stuff unless you pinned it if you are running vivid on your desktop [19:05] robru, I have the phablet-tools ppa enabled but dont see an update [19:06] davmor2, did not know about pinning, but seems I got the package [19:06] davmor2: the trunk version does the pinning. [19:06] robru: oh nice [19:06] robru, can you publish a citrain package in the tools ppa? [19:07] pmcgowan: I'll have a look. never done that before [19:09] sergiusens: who maintains phablet-team/tools ppa? [19:27] robru: you mentioned it last, so you do. (I don't think anyone does) [19:27] popey: cool, I'll just build some packages quick and cram 'em in there then [19:29] anpok: lol, you missed: https://ci-train.ubuntu.com/job/ubuntu-landing-004-1-build/lastSuccessfulBuild/artifact/mir_packaging_changes.diff/*view*/ [19:29] slangasek: new packaging diff w/abi-compliance-checker restored: https://ci-train.ubuntu.com/job/ubuntu-landing-004-1-build/lastSuccessfulBuild/artifact/mir_packaging_changes.diff/*view*/ wanna ack? [19:34] mandel: https://ci-train.ubuntu.com/job/ubuntu-landing-009-2-publish/101/console need these merges top-approved [19:39] davmor2: pmcgowan: ok I have an updated phablet-tools-citrain package uploaded to phablet-team/tools PPA, can you guys update to that and check if the silo pinning works correctly in vivid? [19:49] mandel, i did find a problem with silo 9... your unity-scope-click branch used in the silo has been superseded, and the new branch has conflcts [19:50] mandel, i think we can just change the status from superceded though [19:50] but we should have something from the right team set it to approved [19:51] mandel, ok, i changed it to needs review... which is weird because it still says it's superseded :) [19:51] but i think that's ok, just need someone to approve it [20:10] kenvandine, ok, that is weird.. [20:11] kenvandine, sorry, I was having dinner.. [20:11] kenvandine, is 22 here, I'm going to rest for the rest of the day, will do as much as possible in the morning to fix that [20:23] robru: yes, ack on the mir packaging now [20:26] robru: trying to publish fails because trust-store was already published to the archive, is that the bit that sil2100 was saying needs forced? [20:27] slangasek: let me check [20:28] slangasek: yeah seems so. I should really fix that check so that it doesn't block on packages dangling in -proposed [20:28] robru: er, that doesn't sound right [20:29] slangasek: hm? [20:29] the purpose of this check is to avoid accidentally clobbering changes from another silo that have already been published [20:29] skipping the warning for packages which are in -proposed just means you're letting things be clobbered [20:29] slangasek: don't we already have a silo dirty check for that? the purpose of this check is to not clobber manual archive uploads done by non-train-users [20:30] slangasek: anyway I'll force publish for now [20:30] robru: the version numbers shown there are not manual uploads [20:31] and being in -proposed vs. wily again does nothing to invalidate this check [20:31] slangasek: no, but that's because the check just says "this version is different than it used to be!" it doesn't inspect what the version number is [20:31] "ERROR Version of trust-store at dest changed from 1.1.0+15.04.20150213-0ubuntu1 to 1.1.0+15.10.20150629-0ubuntu1 since packages built". [20:31] mandel, no problem [20:32] 1.1.0+15.10.20150629-0ubuntu1 wasn't a manual upload, that came from another silo landing [20:32] mandel, i think we just need someone to give the current branch an ack and set it to approved [20:32] slangasek: or an earlier publish of this same silo [20:32] even though it's superseded, this is the branch we tested and it should be fine to land [20:32] ah [20:32] slangasek: point is, that check was written in order to protect manual uploads by other people. my argument is that because this isn't a manual upload, it's a false positive, and should not be blocking publishing [20:33] robru: ok, but that has nothing to do with "packages dangling in -proposed" which was what you first said [20:34] steve pls [20:34] slangasek: "packages dangling in -proposed" means "we're fixing a previous publish that failed and got stuck in -proposed" [20:36] slangasek: I feel like this is a false positive on that test. sometimes silos get published and then fail in -proposed and need to be rebuilt & republished. but *every time* that happens, that check explodes because it finds it's own version in -proposed and says "hey, that's not the right version!" and prevents you from publishing [20:36] right [20:36] I agree this is a false positive [20:37] slangasek: so if that test was changed to ignore versions in -proposed, it wouldn't false positive like this ;-) [20:43] robru: that's still the wrong fix. [20:43] just because the package is in -proposed at the time you click 'publish' does not mean it's ignorable [20:44] (or stuck, or from this silo...) [20:45] slangasek: right === salem_ is now known as _salem [23:07] anpok, robru, FYI unity-system-compositor from http://people.canonical.com/~platform/citrain_dashboard/#?q=ubuntu%2Flanding-004 is having trouble in boottest [23:08] it appears that adb fails to come up after the reboot. Testing locally to confirm [23:10] * fginther goes offline for dinner [23:21] fginther: I'm seeing a mix there. it used to be they all had boottest regressions now it seems only some do. Did you retry all of them? [23:24] robru, the *gles packages failed due to a known bug so they can be ignored. I've only retried unity-system-compositor locally so far [23:24] robru, anpok, and can confirm that unity-system-compositor boottest fails locally. After rebooting, the device is stuck on the Google screen [23:24] fginther: hm, that is worrying. I guess we'll have to leave for anpok to investigate, it's not my area [23:25] robru, right, just trying to pass on the failure info since boottest has been less then reliable [23:25] * fginther leaves again