[00:44] rsalveti: https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu-rtm/landing-002 packages still there, but dashboard no longer shows rtm ;-) === _salem is now known as salem_ === salem_ is now known as _salem [03:48] trainguards: hi. Could I get write access to the CI Train spreadsheet? My LP account is jamesh, and google account james.henstridge@canonical.com if that's needed. [03:49] jamesh: sure one sec [03:49] robru: thanks [03:50] jamesh: ok, you have write access on the spreadsheet and you're in the lp team necessary for running jenkins jobs. did you read the docs? any questions? [03:51] jamesh: spreadsheet moved recently, it's now https://docs.google.com/a/canonical.com/spreadsheet/ccc?key=0Ahfru_NNQUKMdEZzQ3VHV0xHS3RKcERudVljcXQycWc#gid=0 [03:51] robru: I'm reading through the new docs sil2100 posted on the wiki. If I have questions I'll ask them [03:52] jamesh: alright I'm winding down. Mirv should be around in an hour or so. [03:52] jamesh: goodnight! [05:08] mornings [05:14] jamesh: your first silo is 015! [05:33] Mirv: thanks [06:09] cihelp: is it possible that qmluitests job max time in unity8 has been decreased? [06:09] since the last jobs it's timing out [06:28] cihelp: or any idea of the [06:28] [0526/144740:FATAL:setuid_sandbox_client.cc(126)] Check failed: IsFileSystemAccessDenied(). [06:28] that we get on https://jenkins.qa.ubuntu.com/job/unity-phablet-qmluitests-wily/29/console ? [06:54] I got a weird build failure in silo 15: it passed everywhere for the vivid build, but on wily it failed to install dependencies on powerpc, choking on gdk-pixbuf. It's not clear what went wrong in the log: https://launchpadlibrarian.net/207580673/buildlog_ubuntu-wily-powerpc.thumbnailer_2.0%2B15.10.20150527-0ubuntu1_BUILDING.txt.gz [07:31] jamesh: there's most probably a real powerpc problem at the moment on wily [07:32] Mirv: okay [07:32] jamesh: I think you can ignore it for a moment and do your testing. [07:33] if we publish it for wily it'd get stuck on proposed, but then it'd get unstuck whenever the gdk/something problem on powerpc would be sorted out [07:33] Mirv: that's what I've started doing [07:34] Mirv: when there has been dependency issues in the past, I've usually seen LP fail the build before it got to that stage [07:34] and QA will take their time to verify the vivid part anyway [07:35] jamesh: it's probably not a missing dependency but a version mismatch, for example newer libgdk failed to build on powerpc and it's trying to install different -dev files (arch independent) and different library version === chihchun_afk is now known as chihchun [07:36] jamesh: I'm retrying the powerpc build right now, since it seems there was a gdk-pixbuf upload 2 hours ago that might be related https://launchpad.net/ubuntu/+source/gdk-pixbuf/2.31.4-2 [08:08] jamesh: it built fine now [08:08] sil2100, do we need to publish wily landings after marking as tested? or you trainguards do that for us? [08:08] thanks [08:09] abeato: we publish them [08:09] Mirv, ok, thx [08:09] abeato: we can be reminded, though, if we miss something :) [08:09] like.. 002 it seems :) [08:09] yep ;) [08:10] Mirv, anyway, not in a hurry, I just wanted to know [08:10] it's looking good [08:33] sil2100, arale #2 doesn't include the latest custom tarball, isn't it? [08:36] jibel: no... did it get signed off in the end? [08:36] Since tarball publishing is out of my control [08:36] cwayne (or someone else) needs to push some buttons which I don't know [08:37] sil2100, penk can do it? [08:39] jibel: not sure, is he the new custom tarball person? [08:39] In that case I suppose he should be able, but I don't see the tarball signed off yet [09:02] jibel: so, do you know if the tarball is good to go? Since Omer didn't seem to set any final status of testing [09:03] sil2100, I don't know he didn't leave any info :/ [09:06] hey there [09:06] sil2100, is there a flag in jenkins that can be used to ignore "missing changelog entry" erros? [09:14] oh, nice, the powerd fix made it in already [09:15] seb128: hm, yeah, let me just confirm how it was called [09:16] seb128: FORCE_REBUILD does everything right now [09:16] If you build with it, it will override any checks and let you do your thing [09:17] sil2100, thanks [09:17] In the past we had a separate flag for that, but it's all in one now [09:17] wfm [09:21] ogra_: yep [09:33] jibel: so I guess we'll need someone to re-validate the tarball... [09:51] sil2100, do you know why the custom tarball on the official server seems to be differenmt than on the last capomastro image ? i dont have the location at the top in the nearby scope since i flashed [09:51] (on capomastro the last few images had it) [09:52] ogra_: yeah, there's some confusion regarding the custom tarballs - we're waiting for Chris regarding that [09:52] The version numbers are confusing as well [09:52] yeah, i noticed that [09:54] sil2100, there seems to also be a bug in system-settings ... the custom version in the UI is completely different from system-image-cli -i [09:54] hm, now that's something new [09:55] i see 1430167525 in the UI and "version custom: 20150427-673-5-18" in s-i-c [11:12] o/ [11:13] Mirv: wait with publishing silo 15! [11:13] Mirv: it didn't pass QA sign-off and it's a dual-landing silo [11:14] I switched it to needing qa sign-off now [11:19] http://people.canonical.com/~jhm/barajas/ubuntu-rtm-14.09/device_vegetahd-testresuls-20150527-dd39a0f.ods [11:19] http://people.canonical.com/~jhm/barajas/ubuntu-rtm-14.09/device_vegetahd-20150527-dd39a0f.changes [11:19] http://people.canonical.com/~jhm/barajas/ubuntu-rtm-14.09/device_vegetahd-20150527-dd39a0f.tar.xz [11:19] sil2100, vegetahd tarball, pushed to ubuntu-touch/ubuntu-rtm/devel-proposed [11:20] john-mcaleely: excellent, thanks - do you need it copied to the rc channel already or only after you do some tests? [11:20] sil2100, only after joc has done some tests [11:20] I will let you know === tsdgeos_ is now known as tsdgeos === ondra_ is now known as ondra [11:49] sil2100, no sign of that having built. do you need to tickle something? === alan_g is now known as alan_g|lunch [12:00] Ok, ISP issues [12:01] sil2100, hi [12:01] john-mcaleely: hm, you mean on the ubuntu-touch/ubuntu-rtm/devel-proposed channel? A new image normally appears automatically when you push the tarball [12:01] john-mcaleely: it's not appearing? [12:01] john-mcaleely: are you sure the channel is properly configured? I can check in a moment [12:01] sil2100, yes, no sign yet of it appearing on that channel [12:01] which is odd [12:02] john-mcaleely: where does the tarball sit currently? I'll compare it with what's in the config [12:02] it's only ever done one build before, so I'm not sure of much [12:02] cwayne: hey! :) [12:02] cwayne: we'll need you to publish the tarball once we get QA confirmation (not yet) [12:02] sil2100, http://people.canonical.com/~jhm/barajas/ubuntu-rtm-14.09/device_vegetahd.build ? [12:02] and http://people.canonical.com/~jhm/barajas/ubuntu-rtm-14.09/device_vegetahd.tar.xz [12:02] hmmmmmm [12:03] sil2100, for arale? I got a +1 from om26er yeserday (but knew to hold off your a more official +1 from you :P) [12:03] Ok, hah, ubuntu-touch/ubuntu-rtm/devel-proposed is not configured anymore ;p [12:03] cwayne: aaaaaah! [12:03] sil2100, ogra_ the UI custom version string being different is not a new bug, let me find it [12:03] ha [12:03] Ok, so there were no issues? Since omer didn't switch the status on the spreadsheet [12:03] cwayne: then I suppose you're good to publish, right jibel ? [12:03] cwayne, that should really be fixed [12:04] john-mcaleely: let me check for another channel to use instead [12:04] was it ever configured? it always used to be ubuntu-touch/ubuntu-rtm/14.09-proposed :-) [12:04] (which seems to have gone) [12:04] seb128, any idea *why* we read it from /custom/build_id ? thats so inconsistent [12:04] john-mcaleely: ah! [12:04] sil2100, ogra_ https://bugs.launchpad.net/ubuntu/+source/ubuntu-system-settings/+bug/1393413 [12:04] Launchpad bug 1393413 in ubuntu-system-settings (Ubuntu) "Custom version string should come from channel info, not /custom/build_id" [Low,Confirmed] [12:04] ogra_, I agree, hence why I logged the bug :) [12:05] * ogra_ duplicates his bug then [12:05] john-mcaleely: so, we don't have that channel anymore, but we do still have the ubuntu-touch/ubuntu-rtm/14.09-factory-proposed channel - I'll just add vegetahd to it and it should appear there [12:05] We'll have to copy the last 14.09 rc image there though [12:06] john-mcaleely: can you wait ~15 minutes for me to do that? I'll just take a few bites of lunch and take care of it ASAP [12:06] sil2100, that sounds good [12:06] sil2100, factory-proposed will be very wrong. but for after your lunch [12:07] cwayne, thanks ! [12:07] sil2100, before publishing, I need to reflash and just double check stuff. if we're missing the 'where am i' that's a bug i'd rather not send out [12:08] it occurred to me that i may have not --wiped when i tested it quickly yesterday, redoing now to be more sure :) [12:09] sil2100, if omer +1'ed the image it's all good, once cwayne double checked stuff [12:15] ogra_, which channel had you flashed? I've got the 'where am i' bit [12:15] you scared me :) [12:15] ogra_, why do we have that file if it shouldn't be used? [12:16] seb128, custom-dconf-update upstart job uses it to determine if dconf should be updated (which is hwy its a timestamp) [12:16] sil2100, im good to push le button? [12:16] cwayne, why is the content wrong? [12:17] it's not wrong, it's just not the most informative way to version the custom bits [12:17] well, patches are welcome in any case [12:17] the one that system-image/channels.ini gets is easier to re-contruct if we ever need it to [12:17] unsure what to query/how to ask for it out of reading that file [12:18] do we have a dbus service providing that info? [12:18] how do you query the other bits (rootfs/device)? [12:18] we read /etc/media-info [12:19] jibel: anything agains publishing the custom tarball? [12:19] and fetch some android properties using property_get() [12:20] sil2100, ok to publish [12:22] john-mcaleely: so, I finished eating mostly, my proposition is to copy last 14.09 rc image to the ubuntu-touch/ubuntu-rtm/14.09-factory-proposed channel, then I'll enable the customs for vegetahd and fetch your newest device tarball for it [12:22] sil2100, we'll start the testing process and won't wait for silos 27 and 5. We'll leave mms and location tests aside for the moment and run them once silos have landed. [12:22] john-mcaleely: does that sound ok? [12:23] jibel: ok [12:23] cwayne: publish teh tarball o/ [12:23] ZE BUTTON [12:23] sil2100, no [12:23] ogra_, cwayne, is that build-id issue a problem for anyone or just cosmetic? [12:23] 14.09-factory-proposed was for some wierdhotfix process that we don't want to touch [12:23] it sounds like 14.,09-proposed has been nuked [12:23] and 'replaced' by devel-proposed [12:24] seb128, i guess just cosmetic, as the other version string can be gotten from system-image-cli [12:24] and we need to heal that sil2100 ^ === _salem is now known as salem_ [12:24] Well, yeah, the ubuntu-touch/ubuntu-rtm/14.09-factory-proposed channel was and is for all hotfixing processes, we don't offer any rtm images now [12:26] sil2100, I don't see it that way at all. factory-proposed comes from some other place on people.c.c, and I don't want to touch that [12:26] if we've torn up RTM capability, that's a mistake [12:26] We still have that possibility, but we'd need to create a new channel for that [12:27] at some level, it's your server [12:27] but I think deleteing RTM 'normal channel' is a mistake [12:27] That's why I wanted to re-use an existing server [12:27] and should be fixed [12:27] Well, we simply don't build anything from RTM anymore, so the channels weren't needed [12:27] argh [12:27] hmmm [12:28] I *might* simply temporarily re-create the RTM channels [12:28] very argh [12:28] Let me check how things look now after slangasek's clean up [12:28] simply re-create rtm sounds good [12:29] (I'm going to give you an rtm krillin tarball shortly...) [12:29] tsdgeos: hi, I just got here, are you still having problems with your jobs timing out? if so I can have a look [12:30] sil2100, cosmetic but butt ugly :) [12:31] grrr, the problem is that ubuntu-touch/ubuntu-rtm/14.09-proposed is already a redirect to the rc channel [12:32] Which means we cannot re-use the existing ubuntu-rtm channels since all of them are now pointing on rc [12:32] something's a bit fucked up :-) [12:33] Let me check another thing [12:34] sil2100, well, you kind of have to keep the channels, or do you want to migrate users channel.ini on upgrade ? [12:34] john-mcaleely: not sure if it's fudged up, I know slangasek sent out multiple announcements and waited for comments before doing the switch [12:34] ogra_: I don't intend to remove anything [12:34] yep. I didn't see the bit 'deleting RTM capability'. I woudl have screamed :-) [12:34] sil2100, move, not remove :) [12:34] ogra_: all channels exist but are aliases/redirects [12:34] ah, k [12:35] Ursinha: we found out, was actually out code, thanks (well not our code, oxide, but ) [12:35] tsdgeos: okay, thanks :) [12:35] john-mcaleely: we still have RTM capability, slangasek in his proposition mentioned that the ubuntu-touch/ubuntu-rtm/14.09-factory-proposed channel can be used for fixes similar to this [12:36] wheee ! [12:36] my first proper OTA on arale [12:36] * ogra_ presses install and waits in awe [12:36] ok, well, if that's the plan. maybe I should let you tell me what you need. I'm scared though [12:37] john-mcaleely: so I guess the plan was, if we need to do a quick release for RTM when we're already switched to vivid on all other channels, we can use the -factory channel (I suppose) [12:38] john-mcaleely: ok :) I suppose what you need is the latest ubuntu-rtm rootfs and custom tarball (so the one from our rc channels) + the vegetahd device tarball you prepared, right? [12:38] Anything else? [12:38] hmm, nearby doesnt work at all anymore now [12:38] hang on (otp) [12:39] "please activate your locations services" ... [12:39] sil2100, that set sounds correct, yes [12:39] (funnily HERE finds my location just fine) [12:39] Now I need to figure out how to do this without blowing up the world [12:40] * sil2100 sighs [12:40] well, it is only wed. ... you have two more days to put out the fires ;) === alan_g|lunch is now known as alan_g [12:56] brrr [12:57] Almost ready [13:03] Ok, running importer manually to see what will happen === salem_ is now known as _salem [13:09] http://people.canonical.com/~jhm/barajas/ubuntu-rtm-14.09/device_krillin-20150527-dd39a0f.tar.xz [13:09] http://people.canonical.com/~jhm/barajas/ubuntu-rtm-14.09/device_krillin-20150527-dd39a0f.changes [13:09] http://people.canonical.com/~jhm/barajas/ubuntu-rtm-14.09/device_krillin-testresults-20150527-dd39a0f.ods [13:09] krillin peer of the vegetahd image above [13:09] sil2100, ^ not sure if you should just create an image, or if someone wants to +1 it first. I've emailed various folk as normal about that [13:11] john-mcaleely: I think for krillin we'll best need QA, but this won't block the vegeta one [13:11] sil2100, makes sense to me :-) [13:11] Still waiting for the importer to do its job ;) [13:11] next up, vivid-ppa [13:11] Thanks! [13:11] or whatever it's called these days :-) === _salem is now known as salem_ [13:20] john-mcaleely: how does this look? http://system-image.ubuntu.com/ubuntu-touch/ubuntu-rtm/14.09-factory-proposed/vegetahd/version-1.json [13:20] sil2100, looks good. will flash and let you know [13:21] john-mcaleely: the rootfs and custom tarballs are identical to what vegetahd has in the rc channel [13:21] yup, and device looks good [13:21] The device tarball should be from your location [13:21] let me make some factory bits [13:21] Ok, give me a sign if it's all cool after you flash :) [13:21] Sorry for all the confusion, the channel rearrangement made some bits a bit more wacky [13:22] But it's all for the future best === mzanetti is now known as mzanetti|otp [13:37] hey trainguards; is there a silo available for my request on line 70 ? thanks in advance [13:37] davidbarth: hey! Sorry, let me take a look, there's most probably a silo free for your needs [13:39] davidbarth: hm, just a thing worth noting - if you get a dual-landing silo, it will currently be blocked as we don't accept any new landings for vivid [13:39] ah [13:39] davidbarth: so you'd have to wait for QA sign-off till we have a release and open the gates [13:39] I know it's tempting to get a dual landing, but seeing how things are going, it'll take a while [13:40] yeah, understandably [13:40] So I would recommend landing for wily and then syncing it with the next landing and/or explicitly syncing [13:40] ok, let's go for wily then [13:40] right [13:44] o/ [13:45] davidbarth: hm, your branches are nor MPs ;) [13:46] Could you correct that? [13:49] Ok, I'll correct it for you [13:49] sil2100: ah sorry [13:51] jibel: the backlight fix from charles when in yesterday, right? [13:51] Yea [13:54] sil2100, I dont see any landing card for it [13:54] pmcgowan: landing card? You mean, on the trello? [13:54] yeah [13:55] Well, not sure about that [13:55] But it's landed as per info on the spreadsheet === mzanetti|otp is now known as mzanetti [13:57] sil2100, image looks good. We will spend around 2hr on QA [13:57] and then ask for a promotion to rc later today [13:57] Ok, give me a sign and I'll copy it to RC once it's done [13:57] (assuming that goes well!) [13:59] sil2100: 27 is probably good to reconfig for overlay and publish, as it was signed off by QA? [14:00] it fixes two High priority ota4 bugs [14:00] Oh, hm [14:01] sil2100: or are even QA sign-off:d High fixes banned now? [14:01] I think it's on the list of things to land [14:01] Let me double-check [14:01] is there another list of things to land than the milestone? :) [14:01] jibel: 27 is fine for this milestone still, right? [14:02] I'll reconfig it anyhow [14:02] sil2100, yes 27 on the list [14:02] Mirv: thanks :) [14:03] sil2100, yes it fixes bug 1361864 which was on the priority list [14:03] bug 1361864 in nuntium (Ubuntu) "add "Preferred" property to ConnectionContext interface" [High,In progress] https://launchpad.net/bugs/1361864 [14:05] jibel: is it known that there's an unrelated-to-those-bugs upstart change also in the silo? https://ci-train.ubuntu.com/job/ubuntu-landing-027-2-publish/lastSuccessfulBuild/artifact/nuntium_packaging_changes.diff <- abeato, mandel [14:07] so that's another ota4 bug #1456097 [14:07] bug 1456097 in nuntium (Ubuntu) "nuntium does not re-start" [Undecided,In progress] https://launchpad.net/bugs/1456097 [14:07] so seems good but checking since the description lists only those two other bugs [14:08] Sneaky silo [14:09] Mirv, yeah this is this MP https://code.launchpad.net/~alfonsosanchezbeato/nuntium/upstart-respawn/+merge/259383 and bug 1456097 [14:09] bug 1456097 in nuntium (Ubuntu) "nuntium does not re-start" [Undecided,In progress] https://launchpad.net/bugs/1456097 [14:09] jibel: if you're happy the silo includes also that fix, I'm also happy [14:09] davmor2 probably wasn't aware of that bug fix though when testing [14:10] Mirv, the MP was mentioned in the landing request [14:10] yes, just not in the description [14:13] Mirv: it restarted as I had to test that it came up I did 20 restarts for the bug on it starting before ofono, only issues I hit were sending mms yesterday which turn out to be a server issue and today receiving mms [14:13] which is a phone issue and exists without the silo [14:17] davmor2: ok, so it sounds it's good to go then? [14:21] pmcgowan: for Qt 5.4 new features, you could add this recent addition: Two new QML modules on images: Declarative State Machine, QML Models [14:23] sil2100: hi - yes, that's exactly right, if an out-of-band update needs to be done to the factory image, we can inject it just as you've done here [14:23] bzoltan: ^ or you if you have edit right to that doc that lists them [14:28] Mirv: so I guess 27 is fine [14:28] Right? [14:28] slangasek: thanks for confirming [14:28] :) [14:33] Ok, not waiting for double-confirmation, publishing 27 [14:43] sil2100: looks so [14:43] sil2100: it was about quad-confirming it at that point already :) [14:44] ;) [15:02] uh, brb [15:06] sil2100, can I get a silo for line 71 please? [15:06] On it [15:06] jhodapp: just remember that the dual-landing will wait before publishing, since vivid landing gates are closed for now [15:07] sil2100, yes I know, this is a critical fix anyway so someone may override it [15:08] sil2100, that is on crit list if we can land it [15:08] jibel, ^^ [15:08] jhodapp: hmm, no merges [15:08] jhodapp: column F has only some text in it [15:08] sil2100, oops sorry, wrong paste buffer in that line :) [15:08] ;) [15:08] there we go [15:08] sorry [15:10] Should be assigned [15:10] sil2100, thanks [15:10] which silo #? [15:11] there we go :) [15:11] 20 ;) [15:11] bfiller, a fix for the uncaught exceptions making unity8 crash is building in silo 20 right now [15:13] jhodapp: ok thanks [15:18] seb128: do you usually publish your silos yourself? Since I want to avoid a double-publish ;) [15:18] sil2100, yes, just did for settings [15:18] Ok, great, thanks! [15:18] sil2100, thanks for watching ;-) [15:47] \o/ [15:47] Publishing! [15:48] hm, I lost connection to the canonical IRC and canonistack [15:48] Anyone else having issues like this? [15:48] Can't access the train too [15:48] ogra_, jibel: ^ ? [15:49] fine here (canonical IRC) [15:49] Ok, then it's on my side... eh [15:49] vpn dropped ? [15:50] No VPN, seems like my ISP has route problems to canonical servers (WTH?!), not sure, I'll use my proxy to publish silo 29 [15:52] Yeah, traceroute says packages get lost somewhere on the 5th hop [15:53] Great, not only to the Canonical servers, also to the US - so my main proxy is useless [15:53] I'll try through my other PL server, but this is absurd [15:53] sil2100: I can publish 29 since I glanced here.. [15:54] Mirv: please :) Thank yoU! [15:54] oh, needs reconfig to overlay too [15:54] Oh, ok, it's back now [15:56] sil2100: I'm trying to unravel the mir-0.13.1 landing. AIUI we have mir, qtmir[-gles] and usc already in wily-proposed and we need a corresponding update to qtubuntu[-gles]. Do I need to set up a silo with no-change branches for the existing ones? Or just for qtubuntu? Or something else? [15:56] alan_g: a silo for just qtubuntu[-gles] is enough [15:56] alan_g: all the others are in -proposed already, so they're just waiting for the fixed qtubuntu [15:58] sil2100: thanks. I'll do that then === chihchun is now known as chihchun_afk [16:06] sil2100: the watch job doesn't really seem to finish (and I restarted it already), so feel free to continue staring at it.. but 029 reconfigured to vivid-overlay already [16:07] so it'd be ready to publish after something like https://ci-train.ubuntu.com/job/ubuntu-landing-029-1-build/50/console would finish.. [16:09] Mirv: looking [16:09] Wait, sometimes it was taking a while [16:09] Not sure why, but sometimes it was hanging like that [16:11] Mirv: btw. you subscribed to the lt mailing-list, right? [16:14] Mirv: I just setup my scripts on a safe remote server now and, besides bugs being closed, changes e-mails are sent to this list too === chihchun_afk is now known as chihchun [16:21] Mirv: what's up with silo 29? QA just marked it tested but looks like build failed? [16:21] sil2100: yes, usually the wait was 5min, I waited over 10. [16:22] sil2100: yes I subscribed [16:22] The build failed [16:22] bfiller: it just needs targetting overlay, no changes. [16:22] now it claims it's trying to search for wrong version.. [16:22] Mirv: is that normal that it failed? [16:22] ;/ [16:23] Mirv: why was rebuild necessary? [16:23] bfiller: not a rebuild, but watch_only build which is needed after reconfiguring from vivid main -> vivid overlay [16:23] Mirv: ah, got it [16:23] salem_: ^^^ [16:24] bfiller, ah ok, thanks [16:24] sil2100: nopes, it's trying to search for wrong version. robru has fixed some/many of the cases, but I try to now specify oen package only. now it's again waiting. [16:25] fginther, ping [16:29] Mirv: thanks for taking care of this :) I rarely re-configure vivid->overlay silos so I didn't see this case yet [16:36] sil2100: ^ well we might need robru to tell us how to make the train find the correct version from the PPA... [16:37] without a real rebuild [16:40] sil2100, +1 on vegeta image promotion to rc [16:40] uh [16:40] robru: ping [16:40] john-mcaleely: ok, on it in a minute! [16:40] thanks sil2100 === chihchun is now known as chihchun_afk [16:42] IOError: [Errno 2] No such file or directory: '/srv/cdimage.ubuntu.com/scratch/ubuntu-touch/wily/daily-preinstalled/live/boot-armhf+grouper.img' [16:42] rsalveti, sigh ... [16:42] moar changes deep in cdimage :/ [16:42] hm, iirc it was kind of self contained in one place [16:43] but yeah, this is painful =\ [16:43] snappy FTW :) [16:47] rsalveti, yeah, was only one line to drop ... [16:47] (in config.py) [16:47] john-mcaleely: copy in progress [16:47] trainguards can you reconfigure silo 2 ? [16:47] sil2100, ack [16:58] sil2100, can you reconfigure silo 20 for me please? [16:58] One moment everyone, OTP [16:59] k [17:00] sil2100: hey sorry, wha'ts up? [17:01] robru, sil2100 is OTP, can you quickly reconfigure silo 20 for me please? [17:01] jhodapp: sure [17:02] Mirv: 21.1 is indeed not in the silo. still digging on that one [17:03] jhodapp: is the only change that it's wily->dual? or did you change other stuff too? I'll have to fuss with this a bit. === alan_g is now known as alan_g|EOD [17:06] jhodapp: ok it's dual now [17:06] robru, I added another MR [17:06] robru: hey! Could you look at silo 29? [17:06] robru, awesome thanks [17:06] sil2100: yeah hang on [17:06] robru: we reconfigured it for overlay and the train doesnt' see the packages [17:12] john-mcaleely: image 2 in ubuntu-touch/rc/bq-aquaris.en for vegetahd should be your thing [17:12] sil2100, looks like it \o/ [17:12] robru, did you abort silo 20? [17:13] jhodapp: yes, you clicked build too soon and it stomped over my reconfigure. so your extra mp wasn't going to be built and it wasn't going to be dual either. [17:13] thank you sil2100 [17:13] robru, ah bummer, is it ready now? [17:14] jhodapp: no [17:14] jhodapp: ok now [17:14] awesome thanks [17:15] jhodapp: yw! [17:16] cihelp, can I get http://91.189.93.70:8080/job/ubuntu-clock-app-ci updated to use vivid as the testrunner, and not utopic please? [17:17] sil2100, has QA started in on testing the latest image for the next OTA? [17:17] jhodapp: yes [17:17] jhodapp: but they skip testing MMS and location for now until all fixes for that land [17:17] For MMS we still need silo 29 [17:17] sil2100, ok great, and my silo [17:17] 20 [17:18] balloons: is that urgent? I can have a look in a short while, if that's okay [17:18] One more? [17:19] jhodapp: well, QA is not aware of that from what I know, probably best if you poke jibel or davmor2 about it [17:19] jhodapp: but we're REALLY REALLY low on time [17:19] sil2100, hmm, pmcgowan did this morning [17:19] not sure if they actually connected or not though [17:19] Like, seriously, I'm worried we won't make it right now [17:20] And new fixes keep coming and coming [17:20] sil2100, it's up to pmcgowan, he wanted to try and get this into the next OTA [17:20] which one? [17:20] the SMS fix for media-hub [17:20] pmcgowan: silo 20 [17:20] and causing Unity8 to crash [17:21] sil2100: Mirv: guys. [17:21] sil2100: Mirv: guys. srsly. [17:21] *SMS notification fix that is [17:21] robru: wazzup? [17:21] sil2100, yes thats a very nice fix if we can have it, but I don't see a silo 20 [17:21] sil2100: Mirv: https://ci-train.ubuntu.com/job/ubuntu-landing-029-1-build/44/consoleFull clearly shows gsettings-ubuntu-touch-schemas being deleted and then failing to build. v21.1 was simply never built. just build it. [17:21] pmcgowan, it's building right now, try a refresh [17:22] wily or vivid [17:22] both [17:22] robru: wait, but QA signed-off this silo, it was built correctly and then just reconfigured [17:22] robru: so did QA test a not-fully-built silo? [17:22] boiko: ping [17:22] sil2100: yes apparently. v21.1 is simply not in the PPA. [17:22] jhodapp, it just shows under wily apparently,a nd there is no trello card [17:22] robru: maybe it's fallback from the spreadsheet issues [17:23] sil2100: just read through the logs. v21 was uploaded in a successful build, then they tried to build v21.1 but it had a merge conflict and so failed to be uploaded. [17:23] sil2100: 27 and 29 are both done [17:23] sil2100: nope, no spreadsheet issues. build job failed due to merge conflict and then nobody bothered to fix it, and then everybody's wondering why the failed build isn't in the ppa. [17:23] robru: but if the build failed, the silo wouldn't get marked as ready for testing [17:23] pmcgowan, weird, robru had to do something to it so it'd actually be targeted to both...maybe it's not quite there yet? [17:24] sil2100: although it looks like 29 failed [17:24] sil2100: oh, well maybe that's a spreadsheet issue I guess. but that doesn't change the fact that v21.1 rightfully doesn't exist due to merge conflict. [17:24] boiko: we need your councelling, are you around? [17:24] pmcgowan, it's definitely building for both [17:24] davmor2: what do you mean it failed? Failed QA sign-off? [17:25] pmcgowan, just verified [17:25] jhodapp: pmcgowan: yeah I've just discovered that "dual" silos can't be reconfigured at all, I had to poke it by hand, will fix that shortly. [17:25] robru, ah nice, a good test case :) [17:26] sil2100: o/ [17:26] sil2100: sorry, I was on the phone [17:26] boiko: hey! So, we see some strangeness in silo 29 [17:26] jhodapp: https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-020 looks good anyway with wily and vivid versions [17:26] yeah exactly [17:26] sil2100: what's going on there? [17:27] boiko: is the gsettings-ubuntu-touch-schemas that's in PPA 29 the right one? [17:27] sil2100: https://code.launchpad.net/~tiagosh/gsettings-ubuntu-touch-schemas/enable-group-chat/+merge/259682 [17:27] sil2100: yep, it just changes the default value [17:29] sil2100: ok, v21 in the silo contains that MP [17:29] robru: btw. where do you see a merge conflict in gsettings-ubuntu-touch-schemas? I don't see it in the logs [17:29] boiko: what happened is that bfiller tried to rebuild the silo, but it failed with a merge conflict, and then he never fixed it, so the silo is in an inconsistent state. [17:30] robru: the merge conflict was in messaging-app [17:30] robru: so gsettings-ubuntu-touch-schemas was built correctly, and messaging-app was rebuilt later on [17:30] So not sure what left the silo in an inconsistent state [17:30] sil2100: yeah the merge conflict was in messaging-app but it still failed the build job, and then nobody ever fixed it, so the silo is in an inconsistent state [17:30] sil2100: robru: oh, I see, let me ask salem_ to merge trunk again on all messaging-app MRs [17:31] sil2100: what happened is that build with the merge conflict was going to rebuild the whole silo, so a new version number was generated for gsettings, but then gsettings was never actually built. [17:31] boiko: no messaging-app was built afterwards [17:31] Ah, uh, hm, that's bad [17:31] That's not something a normal user would expect [17:32] robru: sil2100: yeah, my bad, I fixed the merge conflict but got distracted with other tasks and forgot to rebuild the silo [17:32] Since they saw an issue in messaging-app and then rebuilt messaging-app, they wouldn't expect having to rebuild anything else too [17:32] Ok, so we need to get it rebuilt then [17:32] sil2100: well I don't know any way around this. a new build for gsettings was started, and then it was halted due to the job failing, and then nobody finished the build, so gsettings in the silo is in a half-built state [17:33] sil2100: all I can say is, if you don't want to rebuild the whole silo, don't rebuild the whole silo. [17:33] robru: is it safe to do an empty Build, or should I specify gsettings? [17:33] sil2100: I would specify gsettings [17:33] I mean, now [17:33] Ursinha, yes. no crazy urgent, but today :-) [17:33] sil2100: robru: and messaging-app? [17:33] robru: no, messaging-app has been rebuilt correctly [17:33] boiko: no messaging-app was already rebuilt successfully [17:33] I mean, boiko [17:33] ;) [17:33] ah ok [17:34] robru: sil2100: sorry for the mess :/ [17:34] Ok, this shouldn't require a re-test, as there was no change in the branch itself [17:35] boiko: it's ok [17:36] sil2100: Mirv: on an unrelated note, did either of you reconfigure any silos today? particularly people trying to change from wily to dual or the other way around? [17:37] robru: hm, no, not from dual landings, maybe Mirv reconfigured though? [17:37] sil2100: I discovered a bug this morning that dual landings just cannot be reconfigured. because a "dual" series is really "wily", but when you reconfigure there's a check that says "hey dual isn't wily, you can't change the series!" and aborts. [17:38] uh [17:38] hah ;) [17:38] sil2100: I've got a branch prepped, will test in staging [17:38] sil2100: this is a result of the prepare job being untestable ;-) [17:39] Aww, well, yeah, these parts would be hard to test ;) At least my sync-parts were easily rip-able so I was able to write some tests for it [17:39] When making changes [17:39] But well, reconfigures... this would require the rewrite, otherwise too much wasted work [17:47] sil2100: you all sorted now? [17:54] boiko: ok, ACKing your packages but please make sure you and your team writes all dependency changes in changelog entries please! [17:55] bfiller, pmcgowan sil0 20 is ready to try out for bug 1457129 [17:55] bug 1457129 in media-hub (Ubuntu) "All sound stops playing after receiving SMS messages" [Critical,In progress] https://launchpad.net/bugs/1457129 [17:55] sil2100: yep, ok, actually it is the first time someone asks me to do so... good to know :) [17:59] jhodapp, did you figure out a good way to reproduce it [18:00] pmcgowan, no, just keep trying [18:01] jhodapp, you mean keep sending texts until it happens [18:01] pmcgowan, it's a very rare event, but based on the backtraces I was able to get, my improvements should prevent the situation [18:01] davmor2: I should be good [18:01] pmcgowan, right...I need some good testing from several people to make sure this is fixed [18:01] davmor2, jibel, ToyKeeper: you want me to kick a new image with the 2 MMS silos in? [18:02] sil2100: yeap [18:10] sil2100, is it expected that there is no 'delta' image for: [18:10] http://system-image.ubuntu.com/ubuntu-touch/rc/bq-aquaris.en/vegetahd/index.json [18:10] makes for a *huge* OTA at the moment [18:14] Interesting [18:15] john-mcaleely: not sure what went wrong, copy-image should generate the delta from previous images in the channel automatically [18:16] sil2100, https://bugs.launchpad.net/ubuntu-system-image/+bug/1459371 [18:16] Launchpad bug 1459371 in Ubuntu system image "The channel ubuntu-touch/rc/bq-aquaris.en for device vegetahd appears to have no delta images" [Undecided,New] [18:16] john-mcaleely: let me look into that tomorrow, today it's already a bit late :) [18:16] sil2100, yeah, no-one will actually ota between those as a customer anyway [18:19] john-mcaleely: but good that you found that, we need to make sure that when we copy it to stable it will normally create the delta [18:20] sil2100, well, stable has 0 images for this device :-) [18:20] still something is wrong though [18:20] john-mcaleely: hmmm, touche' [18:20] ;) [18:20] anyway. enough for today :-) [18:23] jibel, davmor2: kicking the new rootfs in the meantime === oSoMoN_ is now known as oSoMoN [18:53] trainguards, could someone please reconfigure line 16, i.e., silo 5? [18:55] robru: ^ could you take care of it? [18:55] * sil2100 is half-EOD already [19:02] robru, this is kinda urgent, and I'm well past my eod [19:03] tvoss: sorry, missed the first ping somehow [19:04] robru, sure, no worries :) [19:05] tvoss: you've got a branch in there, need MP urls: https://ci-train.ubuntu.com/job/prepare-silo/5026/console [19:05] robru, sorry, fixed :) [19:07] tvoss: ok good to go [19:52] brb [20:03] rsalveti, ^ since you're the only one online, is silo 5 ready for QA? [20:11] jibel: yup [20:13] rsalveti, can you mark it so, and someone will proceed with the verification. [20:13] jibel: sure [20:13] thanks [20:14] ToyKeeper, ^ can you take silo 5, Omer is on 20. [20:15] Sure, I take it that's unblocked now? [20:16] ToyKeeper, it is unblocked apparently. It is the same than yesterday with an additional 4 lines patch to fix location accuracy [20:16] yeah, I'm changing the state in a minute [20:24] ToyKeeper: it's unblocked as in packages are available, since the developers who created the mrs are all gone now [20:24] we're trying to test in parallel [20:25] but not getting good results here [20:28] So far I haven't gotten it to find me at all. [20:31] $ sudo apt-get install ubuntu-location-provider-here=0.1+15.04.20141110-0ubuntu1 [20:31] try downgrading provider-here [20:31] it seems to be busted [20:31] get only the location-service binaries [20:35] The only package changed was ubuntu-location-provider-here, pulled from the silo. [20:35] Granted, I can't always trust 'citrain device-upgrade' to do the right thing. It's too automagic for its own good. [20:37] I'll upgrade it manually and see what happens... and maybe submit a patch to the citrain tool to handle apt sources differently. [20:38] ToyKeeper: right, that's because of the pinning [20:38] so if we have a package with the same name in the overlay ppa, it will always use that instead [20:38] Yeah, I find it usually works better to temporarily disable all other apt sources so only the PPA can be installed. [20:38] even if your new ppa got a more recent version [20:41] rsalveti, I dont think that fix did anything [20:44] ToyKeeper, even if you disable all apt sources you have to pin the silo with a higher score than the overlay otherwise packages already on the image from this origin won't be upgraded [20:47] ToyKeeper, jibel silo 5 seems to not fix anything [20:47] I would not bother with it [20:47] pmcgowan, ack. [20:58] yeah, hard to say if it changed anything [21:29] om26er: what's the deal with silo 20? is that needed in vivid? [21:29] jhodapp, ^ ? [21:30] robru, I think yes, vivid [21:30] om26er, yes [21:30] oops robru [21:31] jhodapp: om26er: just that vivid is "closed" so I'm not sure if this has approval to get in. I don't see it listed on the bug [21:31] robru, it does: it's on pmcgowan's list [21:31] jhodapp: ok [21:32] jhodapp: https://code.launchpad.net/~phablet-team/media-hub/fix-1457129/+merge/260343 please approve the merge ;-) [21:32] robru, I shouldn't top approve my own, rsalveti care to top approve for me? [21:33] jhodapp: well it doesn't matter to *me* who top approves it... [21:33] alright, done :) [21:33] om26er, thanks for testing that so quickly [21:34] jhodapp, code looked "safe" to me, plus it worked fine. [21:35] om26er, awesome...I also hammered it not only with SMS, but telegram as well just to make sure [21:40] hey [21:41] jhodapp: I approved it to avoid possible issue later on regarding self approvals [21:41] rsalveti, thanks man! [21:42] getting threads and locks right is hard [21:42] rsalveti, indeed, that's why I try and avoid using them at all costs :) [21:42] yeah [22:22] trainguards: does anyone know where I can find debug debs for the vivid overlay ppa? [22:23] boiko: as far as I know those don't exist. IIRC there was some talk of enabling it but I'm not sure if that actually happened or not [22:23] boiko: cjwatson or wgrant would know more about that I think ^^ [22:24] robru: ah ok, I needed them to retrace a crash file to debug :/ === salem_ is now known as _salem [22:25] boiko: like I said I'm not sure. double check with those guys I mentioned before giving up. [22:25] robru: ok, thanks [22:25] boiko: np [22:37] robru: Oh, is that PPA not dbgsym-enabled, I wonder? [22:37] cjwatson: dunno, can you check? [22:37] hmm which one? [22:38] In [3]: ppa [22:38] Out[3]: [22:38] slangasek: boiko needs ddebs from overlay ppa... [22:38] In [4]: ppa.build_debug_symbols [22:38] Out[4]: True [22:38] they should be available through LP [22:38] looks enabled to me [22:38] publish_debug_symbols is on too, so they're even on ppa.launchpad.net [22:39] boiko: what package do you need it for? [22:39] they're only available for packages built after *handwave some date a few weeks ago* [22:39] robru: history-service at first, but if you or cjwatson can point me to the URL to the debug syms it should be enough [22:39] but e.g. http://ppa.launchpad.net/ci-train-ppa-service/stable-phone-overlay/ubuntu/pool/main/a/address-book-app/ [22:39] boiko: that was built a little too early I'm afraid [22:39] ok, so it wasn't specifically a setup problem with the new silos? [22:40] you'll get .ddebs in http://ppa.launchpad.net/ci-train-ppa-service/stable-phone-overlay/ubuntu/pool/main/h/history-service/ for your next build that winds up in that PPA [22:40] no [22:40] cjwatson: ok, thanks [22:40] or you can get them from https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/stable-phone-overlay/+packages for the relevant package, which is probably marginally better because https [22:41] boiko: sounds like you need a rebuild. want a silo? [22:41] though "deb http://ppa.launchpad.net/ci-train-ppa-service/stable-phone-overlay/ubuntu vivid main/debug" would work too, and be secure provided that you have that PPA's key trusted by apt [22:42] cjwatson: nice! thanks a lot!