[00:13] rsalveti: you have to use the private link from update_excuses to retry autopkgtests [00:13] I've poked a build now [00:16] cjwatson: ah, thanks [00:32] hmm, same error, let's see [00:40] seems to come down to http://paste.ubuntu.com/8102268/ [00:45] (poking) [00:58] robru,rsalveti: screw it, I know roughly what needs to be fixed here, but it will take a certain amount of wallclock time, it's 2am here, and you need to be unblocked. I've overridden that test failure for the next proposed-migration run [00:58] cjwatson: thanks a ton! and you got it in time for the image build too, so we might have a sane image in a few hours! [00:59] well, it still has to go through p-m and a publisher run [00:59] but yeah, it probably ought to make it [00:59] cjwatson: yeah but there's like 2 hours before the image build [00:59] one and a bit [00:59] hm [01:00] it should be OK if everything isn't terrible for some reason [01:00] oh right, it's 6. looked at the clock and saw "5:59" and rounded that to "5" ;-) === salem_ is now known as _salem [01:24] wahoooooooo! [01:25] that was quick, good [01:25] dobey: bzoltan1 bfiller veebers: you all have outstanding requests for silos. there are now 2 silos available. a) who of you is actually around to test a silo if I give you one, and b) which of your landings represents the highest priority fixes? [01:25] (though will still be waiting for publishing) [01:25] cjwatson: thanks again, you really saved the day [01:25] robru: I would say the autopilot-legacy request is low priority [01:26] bfiller: yours looks important but are you around to test now? [01:26] robru: making progress on the root cause, though probably won't finish until tomorrow [01:27] cjwatson: well don't rush on my account. ;-) [01:27] plenty of other accounts to care about this on :) [01:33] robru: hi :) [01:33] robru: does it matter if there's noone from QA around to do the qa approval? [01:43] dobey: have you asked ToyKeeper ? [01:44] popey: well, apparently robru is no longer around to give me the silo anyway :) [01:44] Got a nice blocker fix/workaround approved, hoping to get it into the next image and get it promoted. Hopefully no more traincon in a few hours. [01:46] ToyKeeper: which one's that? [01:46] ToyKeeper: any chance you could sign off on landing-018 (row 33 in the spreadsheet)? it's already built and tested, and getting that through to landing should at least free up another silo [01:46] Silo 005 has workarounds specifically to get an image promoted. [01:46] Though I don't think there's time to publish something now and get it published in utopic before the automatic image build === _salem is now known as salem_ [01:47] cjwatson: well it already says "landed" so i presume it's already in utopic [01:47] It might not be a good idea to try to add anything else to the next build, since the next one is likely good to go for promotion now. [01:47] Oh, right, 5 was published already [01:48] i doubt unity-scope-click would get published before the next build at this point, but eh. if we can have normalcy restored soon, then nevermind :) [01:48] and that pulseaudio has definitely landed too [01:49] Oh, pulse landed? Should I be worried? [02:06] dobey: sorry, trying to juggle cooking dinner and working [02:09] dobey: ok you got silo 2 [02:09] veebers: hey, your request on line 22 conflicts with silo 1, any chance those can be merged? if not, best to wait until 1 is released before assigning the other one [02:10] robru: I'm just looking now [02:11] robru: in what way does it conflict? It should be attempting to merge into lp:autopilot/legacy [02:12] veebers: they're both merges against the autopilot source package in utopic [02:12] veebers: robru: they shouldn't be both against autopilot. [02:12] autopilot-legacy has it's own source package [02:13] robru: line 22 should be against autopilot-legacy [02:13] oh, what thomi said too :-) [02:13] huh, citrain thinks they're the same [02:15] ugh, citrain doesn't like it when the lp project name doesn't match the source package name [02:16] veebers: ok, I put it in silo 5, please build [02:16] robru: sweet, thanks [02:17] veebers: you're welcome === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha [03:02] ToyKeeper: ^ not sure if you wanna do that one or just wait for traincon to be over ;-) [03:09] robru: Assuming an image builds soon (shouldn't it be in progress now?), I'd like to avoid landing more silos in it because that increases the risk of not promoting the image. [03:13] robru: I could use a silo and release it in 2 hours. [03:15] robru: if you need a silo re-use the autopilot-legacy one as it's going to take an hour or so to test it, I can do that tomorrow if needed [03:27] Sorry guys just having dinner [03:28] ToyKeeper: yeah it won't make this image no matter what [03:33] ... and I think something in my notebook shorted out. Tonight might be fun. [03:36] * rsalveti back [03:37] cjwatson: cool, nice to know (about the links to retry the autopkg tests) [03:38] cjwatson: robru: thanks for unblocking pulse [03:39] hm, no bot message announcing that the build started [03:39] finished already, 44 mins ago (livebuild at least) [04:05] rsalveti: seems imgbot isn't even here, blame ogra_ ;-) [04:05] OK, eod! Goodnight everybody! === salem_ is now known as _salem [04:31] Okay... is the bot broken or is there actually just no new build? [04:32] Duh, scrollback. [04:40] :) [04:41] === trainguard: IMAGE 202 DONE (finished: 20140821 XX:XX) === [04:41] === changelog: http://people.canonical.com/~ogra/touch-image-stats/202.changes === [04:41] * Mirv the bot [06:10] Mirv: any chance to get a Silo for line 35? The MR is tested already and it is a desktop only small change [06:41] bzoltan: not right at the moment, 0 silos, not even a spare one [06:42] so let's just have it queued and land it later today when at some point there's a silo free [06:47] Mirv: This might help. I've just set testing to done for silo 1, also silo 005 is super critical priority so can be re-used and I'll pick it up tomorrow if needed [06:50] Hmm, no one running the train right now? [06:50] davmor2: I'm not sure about image 202. [06:51] davmor2: On the one hand, the current blockers seem to be fixed or at least worked around now. [06:51] davmor2: On the other hand, other things weren't working quite right... [06:52] One device had its UI lock for about 30 seconds in the middle of a call (~30s after receiving the call, while I was playing with call setting buttons). [06:52] The other device crashed right after a call ended (received the call, caller hung up). [06:52] ToyKeeper, were you playing with the proximity sensors at the same time? [06:52] (unity restarted) [06:52] jibel: I'm not sure. [06:53] ToyKeeper, unity8 crashed at the end of the call is known and not new [06:53] ToyKeeper: I am [06:53] The welcome wizard crashed on my mako the first two times I tried to select 'swipe' as a security method. [06:53] (couldn't repro later while attempting again) [06:53] ToyKeeper, it's bug 1359258, Saviq said there is another report already [06:53] bug 1359258 in unity8 (Ubuntu) "/usr/bin/unity8:6:__gnu_cxx::__verbose_terminate_handler:__cxxabiv1::__terminate:std::terminate:__cxxabiv1::__cxa_throw:core::dbus::Bus::remove_match" [Critical,New] https://launchpad.net/bugs/1359258 [06:54] My other device couldn't show any local music in the music scope, even though it was there and the music app could play it. [06:54] (even after a couple reboots) [06:55] veebers: is there a test report on that autopilot landing? it won't of course break anything directly on the device, but the worry always is if it'd get us worse dashboard results. [06:55] OTOH, in-song seeking works again, and I saw no issues with thumbnailing videos, "accounts -> back" works now, and the volume controls aren't jittery on the rtm device. [06:55] Mirv: ah right, I was just finishing off the wee text report that I put together, will update [06:56] I guess we'll need davmor2's opinion also on #202 then, to try to confirm whether there's a real regression or something that has already been there. [06:56] veebers: thanks [06:56] davmor2, sil2100: So, I'd like more opinions on whether 202 can be promoted. [06:57] Normally I'd just say no, but since we need to unblock everyone I'd call it a maybe. [06:57] ToyKeeper: we'll probably come up with a plan to do more testing in 1.5 hours in the meeting. the calls definitely will need more testing from your experiences. [06:59] As far as I can tell, all the current blockers are resolved though... [06:59] It's other stuff which doesn't look so great. [07:00] In any case, it's bed time and I'm off tomorrow. If we're still in traincon 0, brendand agreed to cover for me... but I'm sure he'd rather not have to. :) [07:01] let's hope for the best [07:02] Mirv: I've added the pastebin link for the testing results in the spreadsheet (this one: http://pastebin.ubuntu.com/8104052/), [07:04] Mirv: It's a bit of a mouthful but for each gatekeeper run I compare the failing tests against those shown in the smokeng dash to see if they are either 'known' i.e. show up in the dash too, 'new' i.e. apparent from the changes being tested or 'transient' i.e. the test is known to fail sometimes [07:05] veebers: Ooooooooooh you're here. Confirm with Mirv whether you can finish your legacy silo before your eod please. [07:06] veebers: yes, I've seen your reports in the past, the format is good. all in all it seems to prove there are no regressions. [07:06] robru, Mirv: oh I mean "please do trash the ap legacy silo" it's fine, I've already EOD :-) [07:07] Mirv: sweet. This release had many gatekeeper runs as there were so many test issues it was hard to sort the wheat from the chaff :-) [07:07] veebers: as usual, I hit the same using qt gatekeeper recently (I'm battling with you on who gets to use mako-07...) [07:08] but eventually it was possible to run everything needed [07:08] Mirv: ah hah :-) Well I won't be using it for the next day or so at least, so have at it [07:08] nice [07:08] veebers: ah, the MP is not approved [07:08] :-\ [07:08] * veebers fixes [07:09] Mirv: rats, sorry about that. It has been approved now [07:10] Mirv, ToyKeeper is right, on 202, the music scope doesn't show any local media [07:10] veebers: thanks, published, good night! [07:11] jibel: ouch :( there was a music-app update in it, we need popey to analyze it [07:11] Mirv, and 2 unity-scope updates [07:12] right, those too [07:20] Mirv: awesome thanks. Now, tea time o/ [07:26] bzoltan: I assigned you a silo and also kicked a build https://ci-train.ubuntu.com/job/ubuntu-landing-005-1-build/21/console [07:29] Mirv, sil2100 popey bug 1359582 [07:29] bug 1359582 in unity-scope-mediascanner (Ubuntu) "Music scope doesn't show any local music" [Undecided,New] https://launchpad.net/bugs/1359582 [07:35] :| [07:35] jibel, ToyKeeper: thanks [07:37] jibel: hmm, I copied some files over and they are seen and play fine for me [07:38] immediately after copying [07:39] jibel: ..but not anymore after reboot! [07:40] Mirv, right, sometimes the scope doesn't even show online music [07:41] Mirv, it's similar to bug 1281706 [07:41] bug 1281706 in unity-scope-mediascanner (Ubuntu) "Music scope intermittently displays locally stored music" [Critical,Fix released] https://launchpad.net/bugs/1281706 [07:41] jibel: yeah, after 2nd reboot visible again [07:42] so we need to at least make sure it's not worse than before [07:42] Mirv, I haven't seen that on 200/201, did you? [07:43] Mirv, https://launchpadlibrarian.net/182844560/empty_music_scope.png [07:44] * jibel flashes 201 [07:46] jibel: I didn't notice, but I don't use the music scope normally either [07:48] * sil2100 checks his 201 [07:51] After first reboot it's still there [07:52] Mirv, jibel: just so you know, I'm testing on krillin right now [07:54] did you put the music on after going to 202 though? [07:54] Mirv, jibel: on second reboot I still have local music viwible [07:54] popey: I'm on 201 right now [07:54] sil2100, problem is the same on krillin for me on 202. Newly uploaded music is not visible in the scope [07:55] but what was there before 202 is visible [07:55] * sil2100 checks the commitlog [07:55] http://people.canonical.com/~lzemczak/landing-team/202.commitlog [07:59] i saw music on #201, upgraded to #202 and see none, not local or online. online video also missing, local video shows [08:01] Damn, I can't find anything that could obviously break it [08:02] sil2100, there are 2 unity-scopes updates on 202 [08:03] jibel: I'm downgrading those now [08:03] left a comment [08:03] I mean, upgrading [08:03] search for music [08:03] something you know you have... for me, searching triggers it to appear [08:04] morning guys at the office now train wifi died just catching up on stuff [08:04] popey: where did you leave the comment? [08:05] in the bug [08:05] bug 1359582 [08:05] bug 1359582 in unity-scope-mediascanner (Ubuntu) "Music scope doesn't show any local music" [Undecided,New] https://launchpad.net/bugs/1359582 [08:06] sil2100: flashing devices now [08:07] davmor2: thanks! [08:08] sil2100, clearly a regression in 202, I'll upgrade my device again to 202 and downgrade unity-scopes-* [08:08] davmor2: hi [08:08] hah [08:08] davmor2: do you know if omer had issues yesterday with silo 15? [08:08] jibel: I think I just confirmed that the scopes landing breaks it [08:09] sil2100, great, so I don't have to do it :) [08:09] davmor2: ie, i see it is still blocked waiting for qa signoff [08:09] jibel, popey, Mirv: on my #201 I installed the 2 scope-packages and now my music scope doesn't find my tracks [08:09] Let me try uploading something new [08:09] * sil2100 prepares for an aggressive revert [08:09] good man! [08:10] thats the spirit! [08:10] Just uploaded a new track and nothing [08:10] popey: ;p [08:11] search for it? [08:11] which packages did you upgrade? [08:11] ah the two listed in #202, got it [08:12] popey: yeah, searching unblocks it basically [08:12] But that won't save it from reverting! [08:12] hah [08:12] You're going dooown scopes [08:13] dbarth: he didn't pass anything on I'm still waiting for my email to catch up so I can check that. [08:13] ok, let me know; apparently there was a rebuild, but i didn't trigger this; so the status is unclear [08:14] hmm, i wish i knew why curl and libu1db1 were dropped [08:16] ogra_, u1db was dropped because of unity-scopes-shell (0.5.4+14.10.20140820-0ubuntu1) [08:16] Switch to QSettings, removing U1DB dependencies Support new location [08:16] setting. [08:17] hmm [08:17] i thought QSettings would make use of U1 [08:17] as a backend ... [08:19] ogra_, and curl probably unity-scopes-api (0.6.3+14.10.20140820-0ubuntu1) Switch from u1db to QSettings and drop u1db and related dependencies. [08:19] ah, ok [08:19] well, seems that didnt work so well :P [08:21] popey: do you want to double-confirm that this fixes it before I upload a revert? [08:21] It's ready and waiting for one button press [08:22] sil2100: eh? I dont understand [08:24] Saviq, since apparently you need to fix/rebuilding thing, can you sneak https://code.launchpad.net/~mterry/unity8/dismiss-old-pam-prompts/+merge/231363 in your unity8 silo? ;-) [08:24] sil2100: lets discuss it in teh call in 5 mins [08:24] popey: do you want to double confirm that reverting the scopes packages fixes the bug ? [08:25] hmm [08:25] seb128, http://people.canonical.com/~platform/citrain_dashboard/#?distro=ubuntu&q=landing-016 [08:25] Before I do the actual revert in the archive [08:25] I'd rather not make my phone rw [08:25] it's my main device [08:25] music app doesnt have any music for me either [08:25] ACK [08:25] Saviq, thanks [08:25] pete-woods: see above we're hitting problems with the scopes and considering revert [08:25] ogra_: yeah, I have a revert for that [08:25] ogra_: try reverting the scopes landing and it should be fixed [08:25] sil2100, ah, ok so its a different issue [08:26] or just search for music in the scope [08:26] sil2100, hmm, did you try just installing one of the two dropped packages instead ? [08:26] which will trigger a refresh, then music app will see it [08:26] i assume reverting breings back the U1DB [08:26] i wouldnt see why else music app would be affected by scope changes [08:27] sil2100, I'm doing a revert to confirm [08:27] Mirv: I see local media in the local music scope [08:27] mediascanner populates the music database [08:27] and in the video scope [08:27] ogra_: no, but scopes was enough to cause the issue [08:27] pete-woods: add more music [08:27] popey: okay [08:27] ogra_: since what I did is, while running vanilla #201 I installed scopes and it was broken [08:27] pete-woods: https://launchpad.net/bugs/1359582 is the bug nr for it [08:27] Ubuntu bug 1359582 in unity-scope-mediascanner (Ubuntu) "Music scope doesn't show any local music" [Critical,Confirmed] [08:27] sil2100, scopes pulls these packages back in [08:28] ogra_: while on #201 it was broken [08:28] ogra_: well, anyway, if the scopes pull in these packages then a revert will make it no longer pull them in [08:28] So it's a win [08:29] jibel: thanks [08:29] sil2100, other way round [08:29] the new scopes dropped two packages [08:29] Mirv: hmm. given there were no changes to the media scope. I'd say that the revert is reasonable :( [08:30] pete-woods: no worries, it's just a revert in the archive, you'll have trunk with latest changes that you can look into and try fixing [08:30] sil2100, installing libcurl3-nss and libu1db1 on 2092 makes the music app work [08:30] *on 202 [08:30] pete-woods: so, once you think it's fixed, you can just prepare a landing with that fix alone and get everything back to how it was [08:31] sil2100, and the scopes too [08:31] sil2100: oh, wait. it's not just the missing dep then? [08:31] Might be! [08:31] that would seem strange that it works intermittently? [08:31] sil2100, i dont think the revert is right ... just seed these two packages [08:32] ogra_: I would say seeding is bad, we need the scopes to dep on those then [08:33] So maybe a landing that explicitly adds those dependencies in? [08:33] ogra_: the meeting btw! [08:33] sil2100: it works at times, doesn't really sound like missing deps [08:33] ogra_, I confirm that libunity-scopes3 depends on libcurl3-nss and that's why curl was dropped from 202 [08:34] ogra_, Mirv, jibel: I would recommend a revert anyway then, as this would get things up to shape [08:34] And then we can get things right later [08:35] hmm, I thought we only dropped u1db as a dependency. I guess it has transitive dependencies [08:35] sil2100: yes sure revert, that's what I mean better to revert than just adding some packages back [08:35] ogra_, jibel, pete-woods: I actually see that the latest landing from unity-scopes-api actually drops those 2 dependencies [08:35] So they had to be dropped for a reason I guess? [08:36] ogra_: are you sure it's reliable after adding the packages back? [08:36] that's what I'd like to know for sure [08:37] well, I'll try too [08:37] pete-woods, libunity-scopes3 0.6.1+14.10.20140809-0ubuntu1 directly depends on libcurl3-nss [08:38] sil2100, I downgraded libunity-scopes3 and unity-plugin-scopes on #202 and the music scope behaves as expected. [08:41] indeed, installing those extra deps does nothing to make the music scope more reliable. I don't know how I didn't see this during testing. but it's happening somewhat regularly now :( [08:41] jibel: its only the local-music that isn't shown? I can see the local videos correctly in the video scope. [08:42] it seems to vary [08:42] it's aggregated queries that seem to fail [08:42] i.e. scopes that contain other scopes [08:42] ah ok [08:44] is there a way to find out which packages were not explicitly depending on libcurl, so we can get them fixed? [08:44] as there seems to be two issues here [08:44] davmor2, ha this is the silo i tested :) [08:44] nik90, I only saw this problem for the music scope, but it can be local or online music that isn't shown [08:44] davmor2, for shame [08:44] pete-woods: I think the key is that click packages don't have deps, they just rely on the platform having been defined well enough (via ubuntu-touch meta package) [08:45] davmor2, was it found by ToyKeeper in dogfooding? [08:45] pete-woods: for example music-app is a .click only, not in archives [08:45] ahh [08:45] wonder why it uses libcurl. not exactly the most friendly library for QML [08:45] also, I thought that net-cpp used libcurl [08:46] apparently not [08:46] brendand, ToyKeeper mentioned it briefly in the channel before heading to bed, so I gave it a try. [08:46] at least not the nss one [08:46] brendand: yes apparently [08:47] pete-woods: you wonder why which uses curl? ms? [08:47] the music app [08:47] i dont believe music app uses curl [08:47] thumbnailer does, doesn't it? to get album art from 7digital? [08:48] it uses libsoup, which may possibly in-turn use libcurl [08:49] no, apparently not [08:50] pete-woods, is this the test plan that was run on the silo? https://wiki.ubuntu.com/Process/Merges/TestPlan/scopes [08:51] brendand: yes. I seriously tried everything, though. including the things that fail. I even ran this plan more than once. I think we need to get some more rigorous integration testing around scope aggregation to prevent this happening again. as I already had to delay our release because of another issue in this area [08:52] pete-woods, that test plan doesn't say much [08:52] pete-woods, if it is the one you ran it probably needs expanding [08:52] it doesn't. but we really do test all the various scope functionality [08:52] sure [08:52] iI agree it could do with saying what we *actually* do [08:53] pete-woods, okay but it's better if it's written down [08:53] * pete-woods will add this to the list [08:53] pete-woods, we're not trying to apportion blame. i'm sure it did work when you tested it but we need to try and find out why it broke in the image [08:53] yeah [08:53] I think we have a plain ol' bug in unity-scopes-api [08:54] pete-woods, there are a variety of possibilities and you not testing it is only one of them - and probably the least likely [08:56] I also agree that we should probably revert this release. given that the scopes go reliable again with a downgrade [08:57] pete-woods, do you know which image you tested on? do you think you can install that, then upgrade just the broken packages and see if you can reproduce it. i could also that if you can tell me which image you tested on [08:57] 201 [08:58] cjwatson: we would very much like https://launchpad.net/ubuntu/utopic/+source/unity-scopes-api/0.6.3+14.10.20140820.is.0.6.1+14.10.20140809-0ubuntu1 to be bumped in build importance (KDE 4.14 just got in before it) [08:58] sil2100: ^ [09:01] sil2100: so are we sure unity-scopes-shell does not need reverting? [09:01] just wondering [09:01] brendand: well for what it's worth, I'm flashing 201 now, then trying the upgrade of the suspect packages [09:01] it was part of the same landing [09:01] Mirv: it has to be reverted at the same time, as it has a dep on that version of libunityscopes [09:02] sil2100: ^ [09:02] Mirv: done [09:03] thanks Colin! [09:06] cjwatson: thank you! [09:06] ogra_, Mirv - davmor2 tells me it wasn't happening on each boot? [09:07] Mirv: let me check [09:07] brendand: no, I had one boot where music app worked fine [09:07] brendand: sorry, I mean music scope [09:07] brendand, i'm currently rebooting a few times in a row here to see [09:07] Mirv, i'm also pretty sure that when i tried it it worked [09:08] Mirv: ah, didn't upload that one yet, right! [09:08] (though i have re-installed the missing libs) [09:09] Mirv, my music was on the sd card though [09:09] brendand, if you upload new music to the device it is not shown in the scope. It might after a reboot or not. But never immediately after upload [09:09] ok, first reboot i got no music and no online videos ... searching gets me local music shown ... [09:09] same thing for online video ... searching fixes it [09:10] * ogra_ reboots again [09:11] same thing [09:11] looks pretty consistent to me [09:12] jibel: it did show up immediately for me... [09:13] Mirv: unity-scopes-shell uploaded [09:13] ok, in one of five reboots online video behaved different and showed up without searching ... the rest of the time it was consistently broken [09:13] sil2100: ping colin about that too once it shows in LP [09:13] ACK ;) [09:16] I can confirm that flashing r201, then upgrading to the new scopes packages causes the reliability on some reboots (about 1 in 3 for me anecdotally) [09:16] so revert was definitely the right thing [09:17] cjwatson: so, once those two packages build and migrate we'll kick a new image and have it dogfooded as a promotion candidate - if it passes (which is highly probable) we should have a new promoted image around afternoon [09:17] cjwatson: would that be fine? [09:17] sil2100: well, I don't have a better alternative :) [09:18] so it'll have to be [09:19] I've scored up unity-scopes-shell [09:20] cjwatson: thank you! [09:21] beats trying to figure out why my wife's laptop isn't IPv6-routable, argh [09:47] sil2100, hey, got a minute? [09:55] Saviq: what's up? [09:57] sil2100, I've been thinking... how about some hook mechanism in train [09:57] sil2100, immediate use case is .pot file generation [09:57] sil2100, which everyone and their mother forgets daily [09:58] sil2100, it doesn't make sense for per-MP generation 'cause it's just cruft there [09:58] hmm [09:58] sil2100, so ideally it would happen just after train merges all MPs [09:58] But where would you think those hooks should be defined? [09:58] sil2100, project source tree I'd say [09:59] sil2100, debian/ maybe even [09:59] sil2100, caveat is obviously dependencies of such a hook [09:59] Saviq: we *could* add a new custom tag to the control file that would say 'use hooks' or even 'use this hook' or something [09:59] sil2100, and potential for leaving them around [10:00] sil2100, yeah, but every project needs to be in control of those, they might have different requirements [10:01] Saviq: right, I would think of a flag such as 'if this flag is set, use the hook that's provided in this directory' or 'if this flag is set, use the hook from the path that we gave here' or something [10:02] Yeah, that's a bit more complex anyway [10:02] Saviq: could you fill in a bug for that anyway? It's a good proposition I would say [10:02] sil2100, will do [10:03] sil2100, did it get reverted already? [10:03] brendand: it's still in proposed === psivaa changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | CI Train support: trainguards | Vanguard (general help): psivaa | CI Train Status: #179 promoted | CI Train Dashboard: http://bit.ly/1mDv1FS | Known issues: we're in TRAINCON-0! Also: citrain struggles with source packages, don't WATCH_ONLY until *after* the source is built in PPA. http://youtu.be/-Rnw0D2AdYU. === vrruiz_ is now known as rvr [10:33] sil2100: looks like we're waiting for autopkgtest executors again [10:33] jibel: is there any way to beef up our autopkgtest executor farm? we've run into load issues there a few times recently [10:34] cjwatson: yeah, I saw that it was staying on autopkgtests for a while [10:34] sil2100: it's numbers 3-6 in the relevant queue, so hopefully soon [10:34] But at least scopes-shell migrated [10:35] Yep [10:38] cjwatson, no, it's working at full capacity. yesterday evening it was excessively slow because tests were running on disk instead of memory, but it should be faster today. I'll do another check to make sure it's not that again. [10:39] jibel: Right, I mean extending its capacity [10:40] It's usually fine but then sometimes KDE lands or something [10:47] cjwatson, the trend is more to "cloudify" the farm rather than adding more physical hardware. I'll see with the CI team if they can allocate more machines to autopkgtest. With the increasing number of packages I agree it's becoming a bottleneck. [10:47] cjwatson: do I see it right or did the autopkgtests for unity-scopes-api fail? [10:47] :| [10:48] sil2100, it never passed [10:48] Ah, ok, so it will be skipped with Always failed ? [10:48] it fails with: cp: cannot stat ‘scopetest.cpp’: No such file or directory [10:48] pete-woods, ^ [10:49] jibel: thanks [10:54] sil2100: yes. they have always failed. I have nagged our tech lead to get them fixed (and indeed show him how to even know they are failing, as it's not obvious). we have merge in that should fix it [10:55] fortunately the same check they perform is at least carried out as part of our unit tests [10:55] I think a lesson to take away from all of this is to have more people on our team exposed to the "debiany stuff" besides muggins [10:55] (i.e. me) [10:58] is it possible to get our jenkins jobs to tun our autopkgtests after build? [10:59] *to run [11:39] sil2100: it's migrated, publishing nowish [11:46] sil2100: that's published. want to kick an image build? [11:46] om26er: ping? [11:47] om26er: did you run into issues with silo 15 yesterday? [11:47] sil2100: actually I'm going to assume you're at lunch or similar and kick it myself [11:47] hey, silo 19 would probably need manual inspection as there are debian/control changes [11:48] please see the commit message in: https://code.launchpad.net/~unity-api-team/connectivity-api/connectivity-service-source-package-shuffle/+merge/231431 [11:51] psivaa: any takers? :) --^ [11:52] Wellark: for this you'd need trainguards [11:54] === trainguard: IMAGE 203 building (started: 20140821 11:55) === [11:55] whee [11:58] dbarth, no, it was fine for me [11:59] dbarth, was there a regression ? === psivaa changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | CI Train support: trainguards | Vanguard (general help): cihelp | CI Train Status: #179 promoted | CI Train Dashboard: http://bit.ly/1mDv1FS | Known issues: we're in TRAINCON-0! Also: citrain struggles with source packages, don't WATCH_ONLY until *after* the source is built in PPA. http://youtu.be/-Rnw0D2AdYU. [12:03]