| === chihchun is now known as chihchun_afk | ||
| === chihchun_afk is now known as chihchun | ||
| === ljp is now known as lpotter | ||
| === jamesh_ is now known as jamesh | ||
| oSoMoN | good morning trainguards! silo 1 is good to publish, but I can’t mark it as such as there is no corresponding line in the spreadsheet | 06:43 |
|---|---|---|
| Mirv | oSoMoN: ok, thanks! | 06:48 |
| Mirv | oSoMoN: ok, core-dev ack needed. maybe ogra_ will be up soon to check the oxide dep bump + new qml dependency (in main) from https://ci-train.ubuntu.com/job/ubuntu-landing-001-2-publish/lastSuccessfulBuild/artifact/webbrowser-app_packaging_changes.diff and say "ack" :) | 06:53 |
| infinity | ... who keeps rebuilding nuntuim over and over again? | 07:39 |
| infinity | abeato: queuebot seems to want to blame you? | 07:40 |
| abeato | infinity, that's me | 07:40 |
| abeato | infinity, there is some bug that does not let me to build it for powerPC | 07:40 |
| abeato | compiler crashing | 07:40 |
| infinity | abeato: So, two things. 1) rebuilding over and over isn't going to fix a toolchain bug. 2) revving the version every time means you're building on 5 arches and wasting CPU cycles to test 1). | 07:40 |
| abeato | well, I actually found that some times the build result can vary | 07:41 |
| abeato | anyway, I opened a bug for gccgo | 07:41 |
| infinity | 7 times in a row might be a hint that they won't vary. :P | 07:41 |
| abeato | infinity, indeed :p | 07:42 |
| infinity | And either way, there are ways to retry just that one broken build, rather than rebuilding over and over with fresh source. | 07:42 |
| abeato | didn't think about the cycles though, sorry | 07:42 |
| abeato | infinity, ok, good to know | 07:42 |
| infinity | abeato: What's the bug number for your gccgo bug? | 07:45 |
| abeato | infinity, bug #1458761 | 07:46 |
| ubot5 | bug 1458761 in gccgo-5 (Ubuntu) "Cannot build nuntium on PowerPC" [Undecided,New] https://launchpad.net/bugs/1458761 | 07:46 |
| infinity | abeato: Ahh, wrong package. But I'll dupe it with LP: #1454183 | 07:46 |
| ubot5 | Launchpad bug 1454183 in docker.io (Ubuntu) "gccgo crash on powerpc" [Medium,Triaged] https://launchpad.net/bugs/1454183 | 07:46 |
| abeato | infinity, great, thanks... maybe the priority should be risen there | 07:47 |
| cjwatson | abeato: Yeah, in future ask the trainguards please if you don't have access to retry a single build in the PPA (which any member of ~ci-train-ppa-service can do). Some other architectures (notably arm64) are short of capacity and this sort of behaviour causes problems there. | 07:49 |
| abeato | cjwatson, thanks, good to know | 07:50 |
| cjwatson | sil2100: Is there some way we can improve the citrain UI to hint people away from doing this kind of thing? | 07:51 |
| cjwatson | like, "don't use this button to retry a single-arch failure" or some such | 07:51 |
| sil2100 | cjwatson: I have an almost finished branch that would allow people to retry single failed PPA builds | 07:52 |
| sil2100 | So this might help once people know about that | 07:52 |
| cjwatson | That sounds good! | 07:52 |
| cjwatson | I mean, if the whole thing weren't abusing the Launchpad permission model then it'd be right there in the LP UI, but still. | 07:53 |
| cjwatson | Hopefully with ephemeral PPAs we can give a grant to the ticket owner or whatever. | 07:53 |
| === mandel|eod is now known as mandel | ||
| sil2100 | hm, spreadsheet still broken, let me prepare a migration | 08:11 |
| sil2100 | ATTENTION! Spreadsheet being migrated, old spreadsheet is being disabled | 08:15 |
| sil2100 | jibel: will you be able to re-target the trello scripts to the new spreadsheet in a moment? | 08:19 |
| jibel | sil2100, sure | 08:20 |
| jibel | sil2100, ^ if it is the same that was here last week it can be published. ToyKeeper did the verification | 08:29 |
| jibel | sil2100, I flipped the switch the morning but the spreadsheet probably forgot it | 08:29 |
| jibel | this* | 08:29 |
| sil2100 | Yeah, it reverts since a day, I'll publish in a moment | 08:30 |
| oSoMoN | ogra_, hey, if you’re around I’d need a core-dev ack for https://ci-train.ubuntu.com/job/ubuntu-landing-001-2-publish/106/artifact/webbrowser-app_packaging_changes.diff , please | 08:31 |
| sil2100 | Looks good to me, but yeah, it's a package in main so we need someone's thumbs up | 08:32 |
| sil2100 | jibel: new spreadsheet link: https://docs.google.com/a/canonical.com/spreadsheet/ccc?key=0Ahfru_NNQUKMdEZzQ3VHV0xHS3RKcERudVljcXQycWc#gid=0 | 08:35 |
| sil2100 | Or https://wiki.ubuntu.com/citrain | 08:35 |
| sil2100 | Dashboard retargetted, commitlog scripts retargetted (not working anyway) | 08:36 |
| sil2100 | Now only trello board scripts and queuebot scripts remain | 08:36 |
| sil2100 | For queuebot I need either stgraber or robru | 08:36 |
| jibel | sil2100, bot updated | 08:41 |
| robru | sil2100: nope only stgraber can update qbot (I'm but here either) | 08:41 |
| sil2100 | robru: ah ha! | 08:43 |
| sil2100 | robru: ok, thanks (even though you're not here) | 08:43 |
| sil2100 | jibel: thanks | 08:43 |
| jibel | pstolowski, Hey, is there a silo with a fix for bug 1457698? | 08:45 |
| ubot5 | bug 1457698 in Canonical System Image "Aggregator can't subsearch the same child twice to get different depts" [Critical,Confirmed] https://launchpad.net/bugs/1457698 | 08:45 |
| pstolowski | jibel, not yet, i've just requested it in line #66 of the sheet | 08:49 |
| sil2100 | Ah, my first official dual landing! | 08:55 |
| sil2100 | Let's give robru's code a spin :) | 08:55 |
| ogra_ | oSoMoN, lovely ! full ACK :) | 08:58 |
| sil2100 | ogra_: o/ | 08:58 |
| sil2100 | Ok, let me publish all those silos now since I'm done with migration | 08:58 |
| sil2100 | ogra_: sorry to bother, but could you take a look at one more quickie? https://ci-train.ubuntu.com/job/ubuntu-landing-019-2-publish/lastSuccessfulBuild/artifact/ubuntu-ui-toolkit_packaging_changes.diff | 09:01 |
| sil2100 | ogra_: looks good as well, it's Mirv's work so it has to ;) | 09:01 |
| ogra_ | sil2100, ACK | 09:02 |
| sil2100 | Thanks again :) | 09:02 |
| oSoMoN | ogra_, thanks :) | 09:03 |
| jibel | davmor2, I'm assigned silo 28 to you, you're the only one with a giffgaff SIM | 09:03 |
| Mirv | ;) | 09:06 |
| === chihchun is now known as chihchun_afk | ||
| === timp is now known as t1mp | ||
| === chunsang is now known as chunsang-away | ||
| === chihchun_afk is now known as chihchun | ||
| * Mirv away | 10:52 | |
| pstolowski | jibel, hey, the fix you asked about earlier is in silo 15 and tested by me | 10:57 |
| jibel | pstolowski, thanks. We'll do the verification asap. | 11:02 |
| * ogra_ wonders why his terminal app doesnt start on arale anymore after it was stated once | 11:07 | |
| ogra_ | *started once | 11:07 |
| ogra_ | aha | 11:08 |
| ogra_ | killing the hanging /usr/lib/arm-linux-gnueabihf/ubuntu-app-launch/cgroup-reap-all process makes it work again | 11:08 |
| ogra_ | tedg, ^^^^ | 11:08 |
| ogra_ | hmm ... and i can reproduce | 11:09 |
| sil2100 | uh | 11:11 |
| ogra_ | i'll not reboot til tedg shows up | 11:12 |
| ogra_ | quite some regressions in imae 15 vs 14 on arale, i wonder why | 11:13 |
| ogra_ | *image | 11:13 |
| ogra_ | (backlight not turning off when pressing power is another one) | 11:14 |
| ogra_ | and indeed image 15 had powerd changes ... http://people.canonical.com/~ogra/touch-image-stats/vivid/20150525.changes | 11:17 |
| === jamesh_ is now known as jamesh | ||
| jamesh | cihelp: ping. I made a request to the CI list asking for the jobs for the thumbnailer jobs to be reconfigured to point at wily last week. Has there been any progress on that? | 11:22 |
| * sil2100 needs to go for lunch | 11:48 | |
| === alan_g is now known as alan_g|lunch | ||
| ogra_ | sil2100, are you aware that the last three wily image builds failed ? | 12:09 |
| ogra_ | ... fix uploaded | 12:15 |
| === _salem is now known as salem_ | ||
| === chunsang-away is now known as chunsang | ||
| jgdx | hm | 13:03 |
| boiko | trainguards: I am trying to mark vivid silo 6 as tested but the spreadsheet is giving me server errors | 13:08 |
| ogra_ | are you using the right one ? | 13:09 |
| ogra_ | afaik they switched to a new one again | 13:09 |
| boiko | now that I looked at the title, seems no :) | 13:09 |
| boiko | ogra_: thanks | 13:09 |
| jibel | trainguards, can you publish silo 15? | 13:12 |
| josepht | jamesh: Yes, I'm in the process of getting the changes made. Sorry for failing to reply to your email. | 13:13 |
| jgdx | kenvandine, any idea why I can't rebuild silo 13? ( error msg at http://people.canonical.com/~platform/citrain_dashboard/#?q=ubuntu%2Flanding-013 ) | 13:26 |
| kenvandine | ugh | 13:27 |
| kenvandine | jgdx, 1) you shouldn't be working | 13:27 |
| kenvandine | 2) no freaking clue... | 13:27 |
| kenvandine | i'll look at it though | 13:27 |
| jgdx | kenvandine, I need to be more precise wrt email subjects/you need to read the email date :P | 13:28 |
| kenvandine | oh! | 13:28 |
| kenvandine | haha | 13:28 |
| kenvandine | yesterday was a holiday for me too :) | 13:28 |
| kenvandine | i just read your mail 2 minutes ago :) | 13:28 |
| kenvandine | jgdx, why are you trying to rebuild lxc-android-config? | 13:29 |
| kenvandine | i merged the changes that landed in it friday | 13:29 |
| jgdx | kenvandine, I didn't at first, but after I got that error message I did | 13:29 |
| jgdx | (… try to build lxc config pkg) | 13:29 |
| kenvandine | ok, you shouldn't need to rebuild that package | 13:30 |
| kenvandine | oh... i know why | 13:30 |
| kenvandine | you can only do watch only builds for that one | 13:30 |
| kenvandine | since i dput it | 13:30 |
| jgdx | kenvandine, okay, I have no idea what that means | 13:31 |
| kenvandine | haha | 13:31 |
| kenvandine | it's not built from a bzr branch | 13:31 |
| kenvandine | so i uploaded the package to the ppa manually and did a watch only build for it | 13:31 |
| kenvandine | so when you rebuild settings in that silo, be sure to always specify just that package name to rebuild | 13:32 |
| kenvandine | or it'll blow chunks :) | 13:32 |
| mandel | sil2100, one question, if I have a silo that will land stuff in trunk and the same stuff needs to be in vivid, a simple sync works, right? Or do I need to create mrs against a diff series | 13:41 |
| mandel | ?? | 13:41 |
| ogra_ | you should be able to use the new dual silos | 13:41 |
| ogra_ | (see mail) | 13:41 |
| mandel | ogra_, oh, I probably missed that email, on the the list? | 13:42 |
| ogra_ | yeah | 13:43 |
| mandel | ack | 13:43 |
| * mandel reads | 13:44 | |
| pstolowski | thostr_, ^ landing in progress it seems | 13:47 |
| Mirv | jibel: ok. | 13:47 |
| sil2100 | mandel: yeah, as ogra_ said ;) | 13:57 |
| sil2100 | mandel: robru finished the dual-landing feature | 13:58 |
| ogra_ | sil2100, new livecd-rootfs landed ... feel free to try a new wily image if you want one ... should build now | 13:59 |
| kgunn | sil2100: so, excuses shows nothing holding up mir....but it says still "stuck in proposed"....does someone need to do something extra ? | 14:07 |
| ogra_ | kgunn, new binary names ? (do youhave any packages that have the version in their package name ?) | 14:09 |
| ogra_ | (libs are typically a good candidate) | 14:09 |
| camako | sil2100, kgunn, Mir 0.13.1 was having problems landing on Fri due to UITK autopkg test failure which now landed (silo 1). Does it need to be prodded somehow? | 14:09 |
| kgunn | ogra_: yes, the one in silo 30 has a new versions | 14:10 |
| kgunn | in the name | 14:10 |
| ogra_ | kgunn, then you need to talk to an archive admin to approve it (in #ubuntu--remease) | 14:10 |
| ogra_ | *release | 14:11 |
| kgunn | :) | 14:11 |
| cjwatson | ogra_: New binary names affect entry into -proposed, not migration from -proposed to release. | 14:20 |
| ogra_ | oops, indeed | 14:20 |
| kgunn | ...and i got distracted anyway | 14:20 |
| kgunn | so what do i need to do now ? | 14:20 |
| cjwatson | http://people.canonical.com/~ubuntu-archive/proposed-migration/update_output.txt shows http://paste.ubuntu.com/11371646/ | 14:21 |
| cjwatson | So some of those packages are rendered uninstallable by the upgrade | 14:21 |
| cjwatson | You need to investigate that | 14:22 |
| cjwatson | kgunn: It basically looks like qtubuntu-android needs to be rebuilt against the new mir | 14:22 |
| cjwatson | kgunn: Er, qtubuntu-gles | 14:22 |
| cjwatson | kgunn: Also qtubuntu | 14:23 |
| cjwatson | The rest is fallout from that | 14:23 |
| kgunn | camako: ^ | 14:23 |
| kgunn | cjwatson: one question, we originally landed in vivid+overlay, does it not have the same test ? | 14:24 |
| cjwatson | it doesn't | 14:24 |
| kgunn | mmm | 14:24 |
| kgunn | ...once again, am scared by overlay setup | 14:24 |
| cjwatson | kgunn: in general, update_excuses is the first stage, and only considers each source package in isolation; the update_output stage considers interaction between packages | 14:24 |
| cjwatson | there probably ought to be two overlay PPAs, with proposed-migration set up to copy from one to the other | 14:24 |
| cjwatson | sil2100: ^- | 14:25 |
| kgunn | cjwatson: what would your recommendation be ?...i can add qtubuntu for this wily landing (which is a sync), and then just go rebuild qtubuntu in overlay ? | 14:25 |
| cjwatson | kgunn: I don't have a detailed recommendation, something like that sounds roughly reasonable ... | 14:26 |
| cjwatson | just doing your analysis :) | 14:26 |
| === chihchun is now known as chihchun_afk | ||
| === alan_g|lunch is now known as alan_g | ||
| thostr_ | ogra_: do you have more details about powerd issue | 14:41 |
| ogra_ | thostr_, not really, image 14 on aralle was fine, with 15 i see the backlight staying on for ~1min before it shuts off | 14:41 |
| ogra_ | -l | 14:41 |
| ogra_ | thostr_, seems to be reproducable by others ... but it is not easy to see in daylight ... | 14:42 |
| thostr_ | ogra_: when was image 14? | 14:42 |
| ogra_ | dunno, friday or so ... | 14:43 |
| ogra_ | image 15 had these changes: http://people.canonical.com/~ogra/touch-image-stats/vivid/20150525.changes | 14:43 |
| ogra_ | since powerd fiddles with the brightnness settings in that change i think it is a suspicious candidate | 14:43 |
| charles | ogra_, thanks | 14:44 |
| charles | ogra_, agreed, it seems the most likely | 14:44 |
| charles | ogra_, the actual diff looks innocent enough, but the other changes in image 15 look even less like candidates... | 14:45 |
| ogra_ | yeah | 14:45 |
| ogra_ | and the device or custom tarballs didnt change between arale 14 and 15 | 14:45 |
| thostr_ | seems there is never ever something like an innocent change.... | 14:46 |
| === chihchun_afk is now known as chihchun | ||
| charles | looks like the latest image on krillin is 14, let's see if that has the behavior... | 14:50 |
| jgdx | kenvandine, so libsystemsettings1 ubuntu-system-settings in PACKAGES… and what else? | 14:52 |
| kenvandine | no | 14:52 |
| kenvandine | just ubuntu-system-settings | 14:52 |
| kenvandine | it's the source package name | 14:52 |
| jgdx | okay | 14:52 |
| kenvandine | that's it | 14:52 |
| jgdx | thanks | 14:53 |
| kenvandine | do you need to rebuild again? | 14:53 |
| kenvandine | i fixed it earlier, just fyi | 14:53 |
| kenvandine | jgdx, to fix it i did a watch only rebuild, sorted it all out (for future reference) | 14:53 |
| jgdx | magic, got it | 14:54 |
| jgdx | kenvandine, yeah, found a bug with the latest hotspot thing using the serverpropsync code | 14:55 |
| jgdx | fixed and pushed | 14:55 |
| kenvandine | ok | 14:55 |
| sil2100_ | F**K | 15:12 |
| sil2100_ | Always when I need the internet most, I hate my ISP | 15:12 |
| sil2100_ | ogra_: could you take a look at https://ci-train.ubuntu.com/job/ubuntu-landing-028-2-publish/lastSuccessfulBuild/artifact/network-manager_packaging_changes.diff ? | 15:12 |
| === sil2100_ is now known as sil2100 | ||
| kgunn | sil2100: sorry about your isp, can you or a delegate hop a quick hangout with me and camako ? to sort this mir stuff out... | 15:22 |
| sil2100 | kgunn: sure, just need some time for my internet to stabilize | 15:22 |
| === tvoss is now known as tvoss|test | ||
| === tvoss|test is now known as tvoss | ||
| sil2100 | kgunn: in the meantime, let me take a look at the mir situation | 15:28 |
| kgunn | sil2100: do you see colin's posts around an hour ago ? | 15:30 |
| kgunn | regarding... | 15:31 |
| kgunn | http://people.canonical.com/~ubuntu-archive/proposed-migration/update_output.txt shows http://paste.ubuntu.com/11371646/ | 15:31 |
| kgunn | So some of those packages are rendered uninstallable by the upgrade | 15:31 |
| kgunn | You need to investigate that | 15:31 |
| sil2100 | Yeah, I already saw the migration output | 15:32 |
| sil2100 | Sadly we already discussed with slangasek the big issue of not having proposed-migration for the overlay PPA | 15:33 |
| sil2100 | We were thinking of running a separate britney instance just for that purpose, but it's actually a lot of stuff to setup | 15:33 |
| sil2100 | kgunn: anyway, I suppose the only way to unblock is indeed get a silo assigned and rebuild qtubuntu/qtubuntu-gles against the new mir | 15:38 |
| sil2100 | kgunn: you can try getting a dual landing silo for this I suppose | 15:38 |
| sil2100 | It would build binaries for both vivid and wily, since both need fixing | 15:38 |
| cjwatson | sil2100: Really? I did most of the prep work to make that kind of thing pretty straightforward before I left Foundations. | 15:39 |
| kgunn | sil2100: actually....would we need to rebuild everything ? mir+ all it's rdeps | 15:39 |
| cjwatson | kgunn: Why would you need to rebuild everything? | 15:39 |
| kgunn | e.g. wouldn't we need to back out that 13.0 from proposed since it's broken? | 15:39 |
| cjwatson | Er, no | 15:39 |
| cjwatson | You're fixing it | 15:39 |
| cjwatson | Backing it out and putting it back in again is a waste of everyone's time | 15:39 |
| kgunn | ok, my confusion, i thot possibly there might be an issue with the silo | 15:40 |
| cjwatson | It's already out of the silo | 15:40 |
| kgunn | happy to just add qtubuntu | 15:40 |
| sil2100 | kgunn: could you prepare a new landing request that only has qtubuntu and qtubuntu-gles? I'll make sure the PPA will build against -proposed | 15:40 |
| kgunn | cjwatson: ok, so it just sits in proposed and waits until you fix it? | 15:40 |
| kgunn | camako: ^ | 15:41 |
| cjwatson | kgunn: Well, not me, but yes | 15:41 |
| kgunn | cjwatson: sure, you=me | 15:41 |
| sil2100 | Yeah | 15:41 |
| sil2100 | That's why we run most tests on -proposed, as this is the place where things can get a bit broken and wait for repair | 15:41 |
| camako | kgunn, I think alan_g is doing the mir 0.13.0 landing | 15:42 |
| alan_g | camako: I thought it was done? | 15:43 |
| cjwatson | sil2100: It would require tweaking archive-reports to mirror the PPA(s), the britney1 wrapper scripts to assemble Packages files from the appropriate mirror, either coming up with some way to run autopkgtests/boottests or disabling them in the interim, and teaching promote-to-release to copy between PPAs | 15:43 |
| cjwatson | I think that's about it | 15:43 |
| camako | alan_g, me too | 15:43 |
| cjwatson | sil2100: autopkgtests/boottests are potentially complicated, indeed, but bearing in mind that they aren't being run for the vivid overlay right now, it would still be an improvement to run the rest of proposed-migration with those disabled | 15:43 |
| cjwatson | Should be under a day's work all told if you omit the test runner bits | 15:44 |
| sil2100 | hmmm | 15:45 |
| alan_g | camako: so what do you think I'm doing? | 15:45 |
| sil2100 | cjwatson: that's good news then, since from what I talked with slangasek it seemed a bit troublesome to setup, requiring some input from CI - but maybe that was because of the boottests | 15:46 |
| kgunn | alan_g: long story short, there's some package interaction | 15:46 |
| kgunn | testing that's only done in the official archive | 15:46 |
| kgunn | not currently in vivid+ovly | 15:46 |
| cjwatson | sil2100: the CI input would be for autopkgtests/boottests | 15:46 |
| camako | alan_g, mir 0.13.0 failed to land on wily is the bottom line | 15:46 |
| cjwatson | sil2100: I agree that starts to get into things that are less than trivial, but as I say, it's surely better to have proposed-migration with no test runners than no p-m at all | 15:47 |
| kgunn | alan_g: so it seems that it caught the need for us to rebuild qtubuntu/qtubuntu-gles | 15:47 |
| sil2100 | alan_g: it looks like qtubuntu/qtubuntu-gles will have to be rebuilt with the latest mir | 15:47 |
| sil2100 | alan_g: I would propose preparing a new silo using the dual-landing functionality | 15:47 |
| sil2100 | This would also unblock mir wily migration | 15:47 |
| kgunn | right ^ | 15:47 |
| kgunn | broken in both | 15:47 |
| alan_g | cjwatson: I'm playing catchup here. What problem are you having with qtubuntu? | 15:57 |
| sil2100 | alan_g: it seems to be causing trouble causing uninstallable packages with the new mir in -proposed | 15:57 |
| alan_g | Do we know why it is "uninstallable" or do I need to reproduce that? | 15:59 |
| cjwatson | alan_g: *I'm* not having any problem personally; I'm helping kgunn out with analysis. qtubuntu has not been rebuilt against the new libmircommon ABI and needs to be. | 15:59 |
| cjwatson | alan_g: You don't need to reproduce anything, it's trivially uninstallable per its dependencies. | 15:59 |
| cjwatson | alan_g: Or rather would be once the old libmircommon is removed. Detecting this kind of problem is one of the core functions of the proposed-migration system. | 16:00 |
| alan_g | cjwatson: that's odd, libmircommon ought to be ABI compatible. | 16:00 |
| alan_g | Sorry, thinko | 16:00 |
| cjwatson | libmircommon3 | 0.12.1+15.04.20150324-0ubuntu1 | wily | amd64, arm64, armhf, i386 | 16:00 |
| cjwatson | libmircommon4 | 0.13.1+15.10.20150520-0ubuntu1 | wily-proposed | amd64, arm64, armhf, i386 | 16:00 |
| cjwatson | alan_g: ^- | 16:00 |
| cjwatson | qtubuntu-desktop etc. still depend on libmircommon3 | 16:00 |
| cjwatson | Therefore they need to be rebuilt | 16:01 |
| alan_g | Ok. The real problem is that qtubuntu binds to libmircommon at all | 16:01 |
| cjwatson | How you fix it is up to you | 16:01 |
| cjwatson | But mir ain't getting through -proposed until it doesn't increase the uninstallable count :-) | 16:01 |
| alan_g | Sure. Just trying to understand the difficulty | 16:01 |
| alan_g | kgunn: did you say we have a silo? | 16:03 |
| kgunn | alan_g: no, we don't have a silo atm | 16:06 |
| kgunn | to correct this | 16:06 |
| kgunn | alan_g: and...looks like what i was thinking, we have a dep we shouldn't | 16:06 |
| sil2100 | I can give you guys a silo if anything, we have plenty | 16:09 |
| alan_g | kgunn: it's a guess, but I know libmirclient (prior to 13) had a public transitive dependency on libmircommon and (I think) some entry points were in the wrong library at some point prior to 012. I suspect that the dependency comes from that. | 16:09 |
| alan_g | sil2100: I'm in my last hour of the day, but if you won't run out... | 16:11 |
| === chihchun is now known as chihchun_afk | ||
| pmcgowan | jibel, fyi the last tarball had the wifi wakeup workaround from alex | 16:13 |
| === alan_g is now known as alan_g|EOD | ||
| robru | sil2100: hey how'd it go today? | 17:10 |
| robru | been a long time since I did a cu2d rollout ;-) | 17:12 |
| cwayne | fginther, ping | 17:13 |
| cwayne | cihelp pingerino | 17:15 |
| sil2100 | cwayne: give me a ping-back if you have everything setup for the arale customs :) | 17:17 |
| sil2100 | robru: nothing exploded so far! | 17:17 |
| robru | sweet | 17:17 |
| sil2100 | robru: let me check the status of one of the landings, one moment | 17:17 |
| sil2100 | robru: ah, btw. we migrated the spreadsheet, so I changed the dashboard locally on platform but didn't commit to your git repo :) | 17:18 |
| plars | cwayne: did you need something? | 17:18 |
| cwayne | sil2100, will do, having a quick hiccup with jenkins not wanting to build customs though.. | 17:18 |
| cwayne | plars, yeah, trying to figure out what's up with this: http://s-jenkins.ubuntu-ci:8080/view/cambridge/job/Arale%20Custom%20Tarball/21/console | 17:18 |
| cwayne | i.e., jenkins not being able to branch stuff from hanloon (that worked yesterday) | 17:19 |
| sil2100 | robru: so, the silo looks goodish, although it seems something funny happened to the wily upload | 17:19 |
| sil2100 | robru: not sure if that's caused by anything CI Train doing | 17:19 |
| plars | cwayne: I've been getting that on some branches here too, I think something is wrong with launchpad right now :( | 17:19 |
| robru | sil2100: what looks goofy? | 17:20 |
| robru | sil2100: oh did you get queuebot updated with stgraber? | 17:23 |
| sil2100 | robru: so it seems it published it to -proposed correctlish but the train can't track it | 17:23 |
| sil2100 | robru: not yet, will poke him soon | 17:23 |
| sil2100 | robru: anyway, take a look at silo 15 | 17:24 |
| sil2100 | It's a dual-landing one | 17:24 |
| sil2100 | The package is in vivid-overlay and wily, but it cannot find it anywhere it seems | 17:24 |
| robru | sil2100: is that the only dual silo that's been published? or have others been successful? | 17:25 |
| sil2100 | I don't remember if we had any others, since basically dual-landings will be low on usage currently due to the vivid landing gates closed - so we only accept selected fixes | 17:25 |
| robru | sil2100: that's really weird, the primary mp-based build I didn't change any codepaths. I just added some hooks to also do vivid stuff on top of all the existing wily code that was already going | 17:25 |
| sil2100 | Not sure if Mirv had published anything besides this one | 17:26 |
| cwayne | sil2100, so jenkins+lp is being a bit finnicky, but the jobs themselves are created | 17:26 |
| cwayne | sil2100, http://s-jenkins.ubuntu-ci:8080/view/cambridge/job/Arale%20Custom%20Tarball/lastSuccessfulBuild/artifact/out/bq-international/build/custom.tar.xz would be the meizu.en-proposed | 17:27 |
| cwayne | and http://s-jenkins.ubuntu-ci:8080/view/cambridge/job/Arale%20Tarball%20Vivid%20-%20Gated/lastSuccessfulBuild/artifact/custom.tar.xz would be meizu.en | 17:27 |
| sil2100 | ACK | 17:27 |
| sil2100 | Thanks :) | 17:27 |
| robru | sil2100: uh I can't branch any lp branches it seems. great | 17:27 |
| robru | cjwatson: hey, uh.... http://paste.ubuntu.com/11374362/ how would I troubleshoot this? | 17:30 |
| robru | cjwatson: happening on public branches and 'bzr whoami' and 'bzr lp-login' have sensible values... | 17:31 |
| boiko | trainguards: can I have a silo assigned again for row 24? | 17:40 |
| robru | boiko: sure one sec | 17:41 |
| robru | boiko: are you wanting to land in JUST wily or do you want wily + vivid in one go? | 17:41 |
| === robru changed the topic of #ubuntu-ci-eng to: Need a silo or CI Train support? ping trainguards | Need help with something else? ping cihelp | Train Dashboard: http://bit.ly/1mDv1FS | QA Signoffs: http://bit.ly/1qMAKYd | Known Issues: otto is broken for a while now and in process of being removed; Landing gates closed for vivid-overlay | ||
| boiko | robru: that's a good question actually, bfiller ^ | 17:42 |
| bfiller | boiko: which silo? | 17:43 |
| boiko | bfiller: address-book splitting | 17:43 |
| bfiller | boiko: vivid overlay is frozen now for ota-4, so if we want to land that it will have to be wily only for now. | 17:44 |
| boiko | robru: so wily only :) | 17:44 |
| boiko | bfiller: thanks | 17:44 |
| robru | bfiller: sil2100: i wonder though will it be ready to land before vivid unfreezes? | 17:44 |
| bfiller | boiko: don't know if I'd bother retargetting until we at least get it working first? | 17:44 |
| boiko | bfiller: well, for two of the problems QA found I already fixed, there is only one or two left | 17:45 |
| boiko | bfiller: but yeah, maybe we can start testing it using vivid | 17:45 |
| bfiller | boiko: I would | 17:45 |
| boiko | robru: so, let me change it to vivid only | 17:45 |
| bfiller | boiko: after fully working and ready to land lets see where to land it | 17:45 |
| boiko | bfiller: yep, makes sense | 17:46 |
| robru | bfiller: well the issue is that you can't just go back and change it; the packages would need to be rebuilt if you were going to change from vivid to wily or wily to vivid. | 17:46 |
| bfiller | robru: understand | 17:46 |
| robru | bfiller: we could set it for both and then you have the option to delete the vivid packages later if you want to land in wily without them | 17:46 |
| bfiller | robru: ok | 17:48 |
| robru | bfiller: ok so I'll assign for both then? | 17:48 |
| bfiller | robru: yes please | 17:48 |
| robru | bfiller: oh, lp is broken | 17:49 |
| robru | bah | 17:49 |
| robru | bfiller: ok, got you silo 16. | 17:52 |
| robru | boiko: bfiller: so trigger the build normally in 16 and you'll see both vivid and wily packages in the silo. when you're ready to publish you can choose whether you want to publish just one or both (will need a trainguard to delete the unwanted packages and reconfigure silo) | 17:56 |
| bfiller | robru: ok thanks, boiko ^^^ | 17:56 |
| boiko | robru: nice! thanks! | 17:57 |
| robru | you're welcome! | 17:57 |
| sil2100 | robru: anyway, besides the migration check, everything works like a charm it seems | 17:58 |
| robru | sil2100: well, with the migration check not working, auto-merge-clean is also broken ;-) | 17:59 |
| sil2100 | robru: hah, right, although everyone had their share of manual merge-and-cleaning in the past already, so we can live with it until it's not fixed ;) | 18:00 |
| sil2100 | s/not fixed/fixed | 18:00 |
| robru | sil2100: just looking over the code, I'm not sure why it wouldn't work... this is a bit curious. | 18:00 |
| robru | sil2100: nm, I figured it out. .project file gets clobbered with vivid version number, so migration phase is expecting wily to have vivid version | 18:02 |
| sil2100 | Ah, right, double build and the vivid one is second | 18:03 |
| robru | sil2100: ok should have a fix soon, just testing. | 18:09 |
| cjwatson | robru: Sorted by a rollback, I believe, but I'm still trying to figure out how this makes any sense | 18:25 |
| robru | cjwatson: yes it seems to be working now, thanks for that. | 18:26 |
| cjwatson | robru: Well, I was at dinner, sysadmin handled the rollback :) | 18:27 |
| robru | cjwatson: heh. do you need a guinea pig to test on? | 18:28 |
| cjwatson | no | 18:30 |
| charles | trainguards, can I get a silo for line 67 for https://bugs.launchpad.net/powerd/+bug/1458936 | 18:38 |
| ubot5 | Launchpad bug 1458936 in powerd "[arale] screen backlight remains on " [Critical,In progress] | 18:38 |
| robru | charles: ah, a dual landing, excellent choice ;-) | 18:39 |
| sil2100 | ;) | 18:39 |
| sil2100 | robru: \o/ | 18:39 |
| robru | charles: silo 19 | 18:39 |
| * charles hides | 18:39 | |
| charles | robru, thanks | 18:39 |
| robru | sil2100: ok I have a fix for migration, with tests. will go live and hope for the best. | 18:39 |
| robru | charles: you're welcome | 18:40 |
| sil2100 | Right, not much possibility of testing it in staging | 18:40 |
| robru | sil2100: yeah, but the test coverage is *really* *good* ;-) | 18:40 |
| robru | sil2100: like 110% coverage. | 18:40 |
| charles | only 110%? | 18:42 |
| robru | charles: I know I can only do so much. | 18:42 |
| sil2100 | Wow | 18:42 |
| robru | sil2100: ok, migration phase looking good | 19:17 |
| robru | cihelp more boottest hiccups? can somebody retry? http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html#unity-scopes-api | 19:28 |
| * thomi looks | 19:28 | |
| robru | thomi: thanks | 19:29 |
| thomi | robru: both retried and running now | 19:30 |
| robru | thomi: thanks! | 19:31 |
| === salem_ is now known as _salem | ||
| === _salem is now known as salem_ | ||
| robru | sil2100: good god are you still around? ^^ silo 15 migrated & merged successfully thanks to my train fix & thomi's help with boottest. | 20:38 |
| sil2100 | \o/ | 20:39 |
| sil2100 | That's excellent news, then it means we're ready for post-OTA | 20:39 |
| sil2100 | No more worries about syncing things wily<->vivid! | 20:40 |
| sil2100 | robru: thanks! | 20:40 |
| jibel | sil2100, I verified silo 19 on arale (powerd fix) and charles did it on krillin, I think you can publish | 20:41 |
| charles | \o/ | 20:41 |
| sil2100 | Another dual landing landing | 20:41 |
| sil2100 | ^ ignore that | 20:42 |
| charles | heh | 20:42 |
| sil2100 | queuebot got switched to the new spreadsheet | 20:42 |
| jibel | lol, busy night for QA :) | 20:42 |
| sil2100 | charles, jibel: published, waiting for packages to fully settle and kicking a new rootfs | 20:42 |
| sil2100 | For vivid | 20:42 |
| awe | sil2100, hey... just noticed that there's a note about reviewing packaging changes in the NM silo ( 28 )? | 20:43 |
| charles | sil2100, jibel, om26er, thanks for the expedited service | 20:43 |
| sil2100 | awe: argh! Yeah, poked ogra_ about those and didn't get an answer, need that published too | 20:44 |
| sil2100 | awe: on it | 20:44 |
| awe | k | 20:44 |
| awe | let me know if you need anything from me | 20:44 |
| sil2100 | No worries | 20:45 |
| sil2100 | Anyway, waiting for both silos and *then* kicking the new image | 20:45 |
| ToyKeeper | alecu: FWIW, I see anywhere from 1s to 24s of backlight on image 15. | 20:47 |
| ToyKeeper | ... wrong channel, d'oh. :) | 20:47 |
| sil2100 | New image build running | 20:52 |
| sil2100 | Going now o/ | 20:52 |
| robru | but but... did he wait for it to show up in rmadison? that seems too fast... | 20:56 |
| === salem_ is now known as _salem | ||
| robru | oh, vivid would be copied instantly, nm ;-) | 21:07 |
| robru | rsalveti: http://people.canonical.com/~platform/citrain_dashboard/#?q=14.09 what's going on here? can I free this? we're sunsetting rtm support in the train | 23:56 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!