=== _salem is now known as salem_ === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun === salem_ is now known as _salem === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun [08:37] sil2100, hey, would you please ♻ https://requests.ci-train.ubuntu.com/#/ticket/1525 - it's a massacre there... [08:40] Saviq: on it [08:41] Saviq: ok, all recycled [08:43] sil2100, thanks, will keep an eye out on what happened there [09:22] sil2100: Any idea what landed in the latest images? All my adt-based tests are failing since it ends the user session as soon as adt returns. [09:23] (normally I use adt to set up the actual test, adt returns, then I measure the phone's behavior for a while) [09:23] ToyKeeper: hey! Recent images had repowerd from the important bits [09:23] Not sure if that could affect anything though [09:24] It's only the images from a few hours ago. Yesterday was fine. [09:24] http://people.canonical.com/~lzemczak/landing-team/ubuntu-touch/rc-proposed/bq-aquaris.en/krillin/383.commitlog [09:25] Looks like only repowerd could have such a big influence, nothing else was big enough [09:25] Hmm. [09:25] sil2100: Did that try to land once a few weeks ago, then get reverted? [09:26] Oh, wait, that was only krillin, and in image 371... [09:27] So this is probably brand new. [09:27] ToyKeeper: yeah, it was ready for landing for over a week but I didn't have time to check it [09:27] SO it landed yesterday only [09:28] sil2100: repowerd does look like a likely culprit though, since it's going to sleep just a few seconds after adt returns instead of going through the usual 60-second procedure. [09:34] jibel, sil2100, FYI vivid unity8 autopkgtest are dead atm due to https://bugs.launchpad.net/ubuntu/+source/repowerd/+bug/1602597 [09:34] Launchpad bug 1602597 in repowerd (Ubuntu) "package repowerd 2016.06+15.04.20160706.1-0ubuntu1 [origin: LP-PPA-ci-train-ppa-service-stable-phone-overlay] failed to install/upgrade: subprocess installed post-installation script returned error exit status 6" [Undecided,New] [09:35] Saviq: yeah, I noticed something similar for my qtmir landing, I wanted to take a look at that, glad to hear it's a known issue [09:36] hm, will check the details [09:36] it "works" on x and y because they're more relaxed about starting services in post-inst [09:43] sil2100, https://code.launchpad.net/~saviq/repowerd/+git/repowerd/+merge/299925 [09:45] sil2100, IMO we should fast-track this change as a packaging change, otherwise we'll be stuck in train due to britney failures everywhere [09:48] OMG that doesn't hepl [09:48] it's stuck starting [09:58] sil2100, bug #1602604 to [09:58] o [09:58] bug 1602604 in repowerd (Ubuntu) "package repowerd 2016.06+15.04.20160706.1-0ubuntu1 [modified: lib/systemd/system/repowerd.service] [origin: LP-PPA-ci-train-ppa-service-stable-phone-overlay] failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/1602604 [10:04] this is a MESS [10:05] we shouldn't have landed that without alf around ;P [10:07] I had no idea alf went for paternity leave ;p [10:09] sil2100, I'm worried we might need to revert === jamesh_ is now known as jamesh [10:27] I think the safest way would be indeed to revert the u-s-c change to get back to powerd === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun [12:19] Saviq: ok, so I would opt for a temporary revert now as well of the repowerd pieces [12:19] We can re-land them if needed [12:19] sil2100, I can see no other way without someone getting to know it [12:19] I just wonder, possibly I'll just do the soft revert on vivid [12:20] Since xenial and yakkety seem fine so far, and reverting it in yakkety would mean removing the package from the archives === chihchun is now known as chihchun_afk === _salem is now known as salem_ === chihchun_afk is now known as chihchun [15:43] sil2100, think we could backport bug #1585517 to xenial overlay? or would you rather say this would be a SRU? [15:43] bug 1585517 in python-setuptools (Ubuntu Xenial) "Can't install for cross-building" [Undecided,New] https://launchpad.net/bugs/1585517 [15:44] slangasek, hey, you mentioned ↑ was due to a "lost" patch to apt? do we know how it got lost yet? [15:46] Saviq: it was lost by the apt folks deciding to get Ubuntu in sync with Debian [15:47] slangasek, ack, so the right solution these days is to actually fix M-A: fields, understood [15:47] Saviq: it's the more robust solution anyway, since then it works for both apt-get build-dep and things like mk-build-deps; it just requires more work to annotate individual packages [15:48] slangasek, ack, thanks! [16:16] Saviq: Silo 32 approved [16:17] rvr, awesomes, mterry ↑ [16:17] rvr, thanks! [16:17] Saviq: I didn't test on the desktop, though [16:17] turbo and frieza are ok [16:17] rvr, oh ok, we've done quite some testing there [16:17] Ack [16:18] as this was the biggest impact actually === nuclearbob is now known as nuclearbob_afk === chihchun is now known as chihchun_afk === nuclearbob_afk is now known as nuclearbob [18:27] trainguards I summon you! [18:28] Kaleo: what ppa you want copied from and to? [18:28] I'd love for the packages in https://launchpad.net/~fboucault/+archive/ubuntu/dh-exec-backport/+packages to go in https://requests.ci-train.ubuntu.com/#/silo/075 [18:29] robru, unfortunately the build in the origin ppa are not complete quite yet [18:29] Kaleo: no worries, I'm just copying sources, not binaries [18:29] robru, cool [18:29] robru, thanks [18:29] robru, and one depend on the other FYI [18:29] Kaleo: what's bats? lol [18:29] a bash unit testing tool [18:29] apparently [18:29] .. [18:30] Kaleo: ok it's copied, at some soonish point please run the build job again as it will generate diffs. [18:31] robru, thanks [18:31] Kaleo: wait why do we need a bash unit testing tool in vivid overlay? [18:31] robru, how soon? [18:31] robru, because it's a build dep of dh-exec [18:31] robru, of a newer version of dh-exec to be exact [18:31] Kaleo: probably now-ish is fine. it just diffs sources so you don't need to wait for the builds to finish [18:31] ok [18:32] robru, and why we need a newer version dh-exec you might ask :)à [18:32] robru, because I'd like to get fancy with some debian packages (and ultimately make the toolkit crossbuildable) [18:33] Kaleo: that's cool. you'll need core dev ack for the packaging changes so maybe seek those sooner rather than later. [18:33] robru, oh really [18:33] robru, who would that be? [18:34] Kaleo: anybody in https://launchpad.net/~ubuntu-core-dev/+members#active but my go-to peeps are kenvandine or mterry [18:34] Kaleo, yo! [18:34] ;-) [18:35] Kaleo, anything for you dude! [18:35] :D [18:35] Kaleo: watch out kenvandine is after something he's being nice [18:35] kenvandine, I am attempting to land a backport of dh-exec in vivid overlay [18:36] davmor2, he is just a friendly guy :) [18:36] kenvandine, and I changed one line in the debian/control [18:36] :) [18:36] cool [18:36] silo? [18:36] kenvandine, https://requests.ci-train.ubuntu.com/#/silo/075 [18:37] Kaleo, the packaging diff is more than a line... i guess that's the backport [18:37] what is different than yakkety? [18:38] kenvandine, I took the source from xenial which I believe is the same as in yakety [18:38] kenvandine, which is the one line [18:38] ah [18:38] kenvandine, I debumped the dependency version on debhelper [18:38] Kaleo: nah dude the diff is huge: https://objectstorage.prodstack4-5.canonical.com/v1/AUTH_39a8dbb93caf4ec889f8a1b7f69885db/bileto-1671/2016-07-13_18:32:23/vivid_dh-exec_content.diff [18:38] that's the diff against vivid though [18:39] robru, yeah [18:40] - ${perl:Depends}, perl (>= 5.14.2~), debhelper (>= 8.9.13~) [18:40] + ${perl:Depends}, perl (>= 5.14.2~), debhelper (>= 9.20151004~), [18:40] kenvandine, yeah mostly :) [18:40] oh I see bats is being ripped out, so it's not a "new" thing, just newly packaged separately. [18:40] Kaleo, you said you debumped the debhelper dep? [18:44] debhelper | 9.20150101ubuntu1 | vivid | source, all [18:45] Kaleo, that's no going to work [18:45] s/no/not [18:53] kenvandine, it's 9.0 [18:54] kenvandine, I made it go from 9.20151004~ to 9~ [18:55] kenvandine, http://pastebin.ubuntu.com/19297295/ [18:55] Kaleo, not in the diiff i'm lookingg at [18:56] no, that's build dep [18:56] i'm talking about the depends [18:56] - ${perl:Depends}, perl (>= 5.14.2~), debhelper (>= 8.9.13~) [18:56] + ${perl:Depends}, perl (>= 5.14.2~), debhelper (>= 9.20151004~), [18:56] kenvandine, ooops [18:57] are you sure dropping that to 9~ is safe? [18:57] kenvandine, pretty sure from looking at the git log [18:57] it'll afect all vivid packages built in the overlay [18:57] kenvandine, yes, all those using dh-exec which I think/hope is 0 [18:58] kenvandine, and only at rebuild time [18:58] yeah, just worried we'll suffer from new build prooblems :) [18:58] damn... this keyboard is pissing me off [18:58] laggy and keeps repeating keys [18:59] * kenvandine wonders wtf is going on with thhat [18:59] grr [18:59] kenvandine, das keyboard is the answer [18:59] kenvandine, yeah I am too [18:59] kenvandine, but mostly not [19:00] kenvandine, and xenial is "around the corner" [19:00] * Add support for DH_CONFIG_ACT_ON_PACKAGES, see #698054 for details. [19:00] As a consequence, depend and build-depend on debhelper (>= 9.20151004~). [19:05] Kaleo, ^^ [19:07] kenvandine, yeah that's right [19:07] kenvandine, would need to remove that feature heh [19:07] feck [19:07] :/ [19:07] not that anyone will ever care about it [19:07] by the time people who know about dh-exec will start using it in the vivid overlay, we will have switched to xenial [19:08] (using that feature I men) [19:08] mean* [19:08] I'm tired [19:08] gett some rest! [19:08] f'iingg keyboaard! [19:10] kenvandine, hang on I think it's optional [19:10] kenvandine, ah yes it is [19:10] kenvandine, that's why I thought it was fine [19:10] cool [19:10] kenvandine, originally it only recommended debhelper (>= 9.20151004~) [19:11] kenvandine, then the author decided it would be nice for people to have the feature automatically [19:11] kenvandine, and made it an actual dep [19:11] After some consideration, depend and build-depend on the newer [19:11] debhelper, so that packagers that want to use the [19:11] DH_CONFIG_ACT_ON_PACKAGES feature will only have to build-depend on [19:11] dh-exec (>= 0.18~), instead of both a versioned dh-exec and a debhelper. [19:13] robru, I uploaded a fix to dh-exec in https://launchpad.net/~fboucault/+archive/ubuntu/dh-exec-backport/+packages [19:13] robru, can you copy dh-exec - 0.24ubuntu2 to landing 75 please? https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-075/+packages [19:14] kenvandine, http://pastebin.ubuntu.com/19298864/ [19:14] kenvandine, should be better [19:16] Kaleo: sure. Ken can also do it since he's core dev 😎 [19:17] Kaleo: ok copied. please re-run build job to update diff [19:19] robru, cheers [19:24] kenvandine, dependency waits will automatically be re-built? [19:29] Kaleo, nope [19:29] kenvandine, so that would be you? :) [19:29] i can kick retry :) [19:29] or me clicking build mayube? [19:30] nah, i'll do it [19:31] kenvandine, thanks [20:58] Kaleo: you clicking build won't do anything to the packages already in the ppa (build would upload new packages from MPs but you don't have MPs). kenvandine: depwaits are retried automatically but it takes 1.5-2 hours so manual retrying is probably justified. [20:58] robru, ok thanks [20:59] Kaleo: you're welcome === salem_ is now known as _salem