2015/06/16/#ubuntu-ci-eng.txt

=== salem_ is now known as _salem
=== chihchun_afk is now known as chihchun
[06:41] <Mirv> al most the re
[07:31] <Mirv> sil2100: jibel: so great news. Riddell was ok in overriding the kde autopkgtests and I gave him a list of packages just before I went to sleep, but unfortunately both him and Scott were too busy last night. however, infinity heroically battled the Qt landing and also three KDE packages while still being able to leave most of the KF5 transition to KDE people's hands
[07:32] <Mirv> sil2100: jibel: so from our POV it's now done, we're still waiting for one KDE rebuild to transition but that's unrelated to us
[07:34] <sil2100> \o/
[07:34] <sil2100> Yaaay!
[07:37] <seb128> sil2100, http://people.canonical.com/~lzemczak/landing-team/ota/ota-4.changelog has "New nonohromatic" ... that's a typo for "mono" right?
[07:37] <sil2100> uuuh :)
[07:37] <sil2100> seb128: good catch!
[07:37] <seb128> :-)
[07:38] <seb128> sil2100, also the url in your email for the bugs doesn't work
[07:38] <seb128> https://launchpad.net/canonical-devices-system-image/+milestone/ww22-ota
[07:39] <seb128> https://launchpad.net/canonical-devices-system-image/+milestone/ww22-2015
[07:39] <seb128> works
[07:48] <seb128> sil2100, importing contact from sim was also added to that ota right? that might be worth listing in your changelog
[07:48] <seb128> sil2100, oh, there is still a typo in "monohromatic"
[07:48] <seb128> +c
[07:48] <seb128> :-)
[08:02]  * Mirv 's phone reported there's an update available!
[08:03] <seb128> Mirv, did you update yet?
[08:05] <Mirv> seb128: it's downloading, I'll update right after it's done
[08:05] <Mirv> ok, now it's installing/rebooting
[08:05] <seb128> Mirv, could be useful if some people who didn't install ...
[08:05] <seb128> k, not you then I guess
[08:05] <Mirv> how could one resist?
[08:05] <seb128> was going to say it could be useful to configure a reminder and see it if triggers after update
[08:05] <seb128> some users reported issues about that
[08:05] <Mirv> ah..
[09:01]  * Mirv 's first Go contribution is now merged :) https://code.launchpad.net/~phablet-team/ciborium/trunk
[09:03] <Mirv> bzoltan and others can ignore the above, in reality all trunks were updated already yesterday. I'm running watch only build job for 011 now to get the status back to correct.
[09:04] <Mirv> 041 will probably be correctly marked once it finishes its current build job.
=== vrruiz_ is now known as rvr
[09:23] <dbarth_> hey trainguards, i need a reconfig on silo 002 (line 13; yet anohter one)
[09:23] <Mirv> dbarth_: sure
[09:25] <sil2100> Damn...
[09:26] <sil2100> Having a cat is troublesome, we just had a 'crappy' incident
[09:26] <sil2100> Took us half an hour to clean the cat, another half an hour to clean the house
[09:26] <Mirv> :)
[09:27] <sil2100> We're all in scratches
[09:27] <sil2100> Grrr, still dealing with the aftermath...
[09:28] <Mirv> sil2100: the cat is not yet housebroken after 12 weeks teaching with the mother?
[09:28] <Mirv> I think our white cat knew how to do the business when he arrived
[09:28] <sil2100> Well, our knows as well, always uses the right place, but you know... he had some problems and it was liquidish
[09:28] <Mirv> anyway, a young cat will definitely consume some time, in both good and bad :)
[09:28] <davmor2> sil2100: just grab the cat by the scruff of the neck it goes comatose and you can clean it more easily
[09:29] <Mirv> sil2100: ah... right
[09:29] <sil2100> And then he was walking around with it and dirtying the whole house
[09:29] <Mirv> the correct neck holding indeed might help, although it might need a bit of training to do that correctly
[09:29] <sil2100> We wanted to clean him, but he's still afraid of water
[09:29] <Mirv> water + cats == extra fun
[09:29] <sil2100> And the long hair didn't help ;)
[09:29] <sil2100> Yeha
[09:29] <sil2100> *Yeah
[09:31] <davmor2> sil2100: https://www.youtube.com/watch?v=T9TmmF79Rw0
[09:31]  * ogra_ is afraid to click that after seeing the backlog
[09:32] <sil2100> hm, does that really work?
[09:32] <sil2100> I mean, is that universal?
[09:32] <nik90> lol that trick in the video is awesome!
[09:33] <nik90> ogra_: you gotta take a look ;)
[09:33] <sil2100> Anyway, need to put some things into the washing machine as everything was in... you know
[09:33] <sil2100> ogra_: yeah, it's nothing disgusting
[09:33] <ogra_> ouch !
[09:35] <davmor2> ogra_: doesn't hurt it replicates the mother picking the cat up, the cat goes comatose so the mothers teeth don't tear the skin
[09:35]  * ogra_ guesses this works with something less strong too, that clip is quite strong
[09:35] <ogra_> davmor2, sure, the method is great ... i just doubt the tool is
[09:35] <davmor2> ogra_: you can just use your hand
[09:36] <ogra_> thatrs what i do
[09:36] <ogra_> but thats a nice way to make it permanent ...
[09:36] <ogra_> just not with such a strong clamp
[09:37] <ogra_> (thats surely way above the jaw pressure a cat-mom would produce)
[09:38] <davmor2> ogra_: probably not don't forget mom is picking up with teeth that is a really small pressure point
[09:39] <davmor2> sil2100: and yes it works :)
[09:40] <cjwatson> I'm not sure it lets my 3yo get away with picking up kittens by their heads though, much as he would like to
[09:41] <cjwatson> Our kittens have retreated to my study, so I have a pile of mini-cats <-- that way
[09:43] <ogra_> davmor2, yeah ... well, i would still use my hand
[09:43] <ogra_> especially with a kitten like sil2100's
[09:43] <ogra_> my female cat was diabetic for a while, i had to give her three shots a day ... that kind of teaches you how to do it right over time :)
[09:43] <ogra_> (and how to not give yourself an insulin shot while holding her at the same place you have to place the needle too :) )
[09:44] <sil2100> hah
[09:45] <Mirv> dbarth_: so removing ubuntu-system-settings-online-accounts from the silo is correct?
[09:45] <sil2100> Anyway, I hope this was just a one-time problem, for the last few days everything was fine
[09:45] <Mirv> dbarth_: , oh 002, not 001..
[09:45] <dbarth_> Mirv: yes
[09:45] <davmor2> ogra_: oh indeed I was just showing the way to sil2100 so he knew how to control his cat when it comes to dealing with water in future and leave him with less scratches
[09:45] <Mirv> dbarth_: line 13 is 001, line 24 is 002
[09:46] <dbarth_> Mirv: 002 is hopefully good once its rebuilt; it's 001 i needed a reconfig on
[09:46] <Mirv> dbarth_: right, you just said "i need a reconfig on silo 002 (line 13; yet anohter one)" so I was confused :) 001 is now reconfigured and needs a proper build to get the signon built
[09:46] <dbarth_> Mirv: but i'm trying to see if what's in 010 can't also be regrouped
[09:47] <dbarth_> since this one never got to the qa plate
[09:47] <ogra_> davmor2, yeah
=== mvo__ is now known as mvo
[10:31] <popey> sil2100: Mirv http://imgur.com/gallery/G3U4rsV
[10:31] <sil2100> hah!
[10:32] <Mirv> that might work...
[10:38] <sil2100> Ok, back to work
[10:39] <sil2100> Great way to waste a lot of work hours ;/
[10:39] <sil2100> jibel: do you think we could also promote the ubuntu community images?
[11:03] <jibel> sil2100, no one tried it, we can probably promote it after we verified that it boots
[11:07] <sil2100> Actually, hm, it might be a bit troublesome in overall
[11:08] <sil2100> We don't really even have ubuntu images for OTA-4
=== chihchun is now known as chihchun_afk
[11:11] <sil2100> eh
[11:29] <oSoMoN> trainguards: can I haz a silo for line 54, please?
[11:29] <Mirv> oSoMoN: you can haz
[11:32] <oSoMoN> Mirv, thanks!
[11:53] <sil2100> Interesting that system-image made a delta for such a big diff
[11:53] <sil2100> Not sure if that's super safe, but people seem to be having working phones
=== MacSlow is now known as MacSlow|lunch
[12:01] <sil2100> jibel: btw. do we have a test-case in the OTA tests that would check if old alarms work after upgrade?
[12:01] <sil2100> If not, could we have something like that added?
[12:05] <ogra_> sil2100, people have working phones, the s-i screwup will only take effect with the upgrade to OTA5 indeed
[12:05] <seb128> sil2100, hey, did you read my other comments earlier?
=== alan_g is now known as alan_g|lunch
[12:06] <sil2100> seb128: ah, sorry, because of those problems with my cat I completely missed my backlog ;/ Looking now!
[12:07] <seb128> sil2100, no worry, it was around 9:48 european time
[12:07] <sil2100> seb128: fixing and adding! Thanks!
[12:07] <seb128> sil2100, yw! thanks for updating those ;-)
[12:07] <sil2100> Yeah, I browsed through the image and changes and tried getting as much good info as possible, since the milestone bugs weren't really helpful ;) As those were from the vivid baseline too, so many of them weren't even visible in stable
[12:08] <sil2100> But I missed a lot of changes probably
[12:09] <sil2100> seb128: next OTA we'll also have a better place for proper release notes
[12:09] <seb128> great
=== mzanetti is now known as mzanetti|run
[12:09] <sil2100> I already had that in plans but there's so much other stuff to do! :)
[12:09] <rvr> sil2100: I don't remember any test that checks specifically an alarm.
[12:09] <rvr> Let me see
[12:10] <rvr> sil2100: Is there an issue with OTA4?
[12:10] <sil2100> rvr: since people report that old alarms are not working when people upgrade to OTA-4, new ones work fine
[12:10] <sil2100> I saw 2 people mentioning that at least
[12:11] <rvr> Weird
[12:12] <jibel> sil2100, we don't have a test for the alarm
[12:12] <jibel> after OTA
[12:12] <rvr> https://docs.google.com/spreadsheets/d/1j91dnYhnffw_6BQCebNncXl4lcS6HMo-qVCdn_YdGmY/edit#gid=1054971369
[12:12] <rvr> Missing test for alarms
[12:13] <rvr> jibel: I'm adding it
[12:13] <rvr> Oh, I can't, not edit permissions
[12:13] <jibel> rvr, can you give me edit rights?
[12:13] <jibel> davmor2, ^
[12:24] <sil2100> Thanks guys
=== mzanetti|run is now known as mzanetti
[12:44] <davmor2> jibel: that should give you edit
=== _salem is now known as salem_
=== alan_g|lunch is now known as alan_g
=== MacSlow|lunch is now known as MacSlow
[13:08] <davmor2> jibel, rvr: both of you should be able to edit now
[13:18] <sil2100> barry: ping!
[13:18] <sil2100> barry: hey!
[13:18] <barry> sil2100: hiya!
[13:18] <sil2100> barry: we seem to have a slight s-i problem in OTA-4 ;)
[13:18] <barry> wonderful
[13:19] <barry> sil2100: what's up?
[13:19] <sil2100> Remember 2.5.1 you published for 14.09 and the 3.0 you published for wily? :)
[13:19] <sil2100> None of them has been published to the vivid-overlay ;p
=== pete-woods1 is now known as pete-woods
[13:20] <barry> sil2100: huh?  2.5.1 has been out for ages
[13:20] <sil2100> barry: yes, but not in vivid or the overlay
[13:20] <sil2100> It was only in ubuntu-rtm/14.09
[13:20] <sil2100> No one forwarded it to vivid or the overlay, it's still on 2.5.0
[13:21] <sil2100> So all users of the OTA-4 now have a reverted version of s-i, and in 2.5.1 I saw some phased-upgrade fixes
[13:21] <barry> sil2100: well, that's no good.  2.5.1 at least should have been in vivid.  i guess that got missed somewhere in the crush of channels
[13:22] <barry> sil2100: yes, that's the difference in 2.5.1.  it has better phased update algorithm
[13:22] <sil2100> Yeah... would you be fine with simply getting 2.5.1 released to vivid-overlay then?
[13:22] <sil2100> And then SRUed to vivid
[13:22] <sil2100> Well, actually, not sure we can SRU it to vivid, as wily only has 3.0
[13:22] <barry> sil2100: that's probably the safest route.  at some point we want people to get on 3.0, but probably not right now under pressure
[13:23] <sil2100> barry: ok, if you're fine with that I'll re-version s-i 2.5.1 for the overlay and prepare a silo for it
[13:23] <barry> sil2100: that should be fine.  we should indeed sru it to vivid
[13:23] <sil2100> Ok :)
[13:23] <barry> sil2100: +1  i can work on an sru for vivid
[13:25] <sil2100> We might think about getting it released as a OTA-4.1 ;)
[13:25] <barry> sil2100: "it" == 2.5.1?  what's the best way to do that?
[13:26] <ogra_> push it to the overlay ... thats the fastest way to get it in
[13:26] <sil2100> Will release it to overlay first, then copy to the snapshot PPA, build an image an maybe try releasing as an update - we could do that without phasing then, since users with OTA-4 could have bad phasing experience right now
[13:26] <sil2100> But we'll discuss that with the product team today
[13:27] <sil2100> barry: do you know if there were any other critical bugs in 2.5 which got fixed by 2.5.1, or only a better phasing algorithm?
[13:27] <barry> sil2100: cool.  q: do you think i need to go through the train to sru into vivid, or just do the "normal" sru process?
[13:27] <barry> sil2100: no, just LP: #1383539 and a couple of related improvements to the cli
[13:28] <ubot5> Launchpad bug 1383539 in Ubuntu system image "[TOPBLOCKER] phased update support does not give idempotent answer for each (machine,update)" [Critical,Fix released] https://launchpad.net/bugs/1383539
[13:28] <sil2100> barry: need to think about the SRU, since if I prepare a quick overlay landing for it, QA will be signing it off there anyway, so the SRU part could go through the normal SRU process
[13:29] <ogra_> why do you want an SRU at all ?
[13:29] <ogra_> the overlay is enough to get it to the phone image ... and nobody else uses vivid system-image stuff
[13:30] <barry> ogra_: that's fine with me too
[13:30] <sil2100> ogra_: right, but archive admins prefer SRUing everything anyway
[13:30] <ogra_> SRU is just extra paperwork for no benefit
[13:30] <sil2100> The idea of the overlay was to use it as a 'quick-landing', but in the end by principle stuff should be SRUed ;)
[13:30] <ogra_> at least in this case
[13:30] <sil2100> Well, true...
[13:30] <sil2100> Anyway, for now overlay
[13:30] <barry> sil2100: only if there's a reason to have the newer version in vivid.  nothing on desktop or server uses it, so it's only for touch and snappy anyway
[13:31] <ogra_> and snappy doesnt use vivid anymore
[13:31] <barry> ogra_: right, and we *still* need to unfork snappy's si
[13:31] <ogra_> oh, wait., we do
[13:31] <ogra_> but we also use system-image-snappy anyway iirc
[13:32] <barry> ogra_: i was waiting to hear from mvo_ about si 3.1 and plans to unfork it there, but haven't heard anything.  i could do a quick 3.1 release to wily if that helps
[13:32] <ogra_> https://launchpad.net/ubuntu/+source/system-image-snappy/
[13:32] <barry> ogra_: yes i know.  that's what we need to unfork
[13:32] <ogra_> it is at 2.5.1-0ubuntu2
[13:32] <sil2100> The thing is...
[13:32] <sil2100> We don't want to create a boundary between 'desktop' and 'touch'
[13:33] <ogra_> why would we
[13:33] <sil2100> That's not the convergence way of things
[13:33] <barry> ogra_: 3.x has all the stuff mvo_ requested, so really that's the way to go
[13:33] <ogra_> barry, sure, i only look at vivid though
[13:33] <barry> ogra_: yeah
[13:33] <ogra_> and in vivid sanppy is fine as is ... for snappy an SRU would be no benefit
[13:33] <ogra_> and touch will get it from the overlay
[13:34] <barry> ogra_: okay, then we don't need to sru 2.5.1 into vivid
[13:34] <ogra_> right
[13:34] <ogra_> just do the overlay dance
[13:34] <mvo_> barry: nice, I was waiting for a release with that, but if its close I can use 3.0 that already has everything except for the error reporting so if 3.1 has that I'm good
[13:34] <sil2100> Just make sure infinity doesn't hear this ;)
[13:34] <sil2100> Or slangasek
[13:35] <barry> mvo_: yep.  i'll do a release of 3.1 to wily today.  any help you need for the unfork, just holler!
[13:35] <sil2100> They were anyway a bit mad that none of the overlay stuff got SRUed at all
[13:35] <barry> sil2100: tbh, it's all a big ball of huh? to me ;)
[13:39] <sil2100> Well, I like the idea, the archive should be self sufficient and we should aim to have both touch and desktop working from one single place - the overlay is a convinence device to work-around the long paperwork, but it's only a step in the long way
[13:40] <sil2100> At least that's what I understood from infinity's and slangasek's mails/conversations
[13:44] <sil2100> I *might* have misunderstood though ;)
=== chihchun_afk is now known as chihchun
[14:36] <boiko> trainguards: I have re-targetted silos 43 to 46 from wily to dual, could you please re-configure them?
[14:42] <Mirv> boiko: it can be done, but the silos need to be emptied, ie packages removed, is that ok? and 43 + 46 only, no 45?
[14:43] <boiko> Mirv: 43, 44, 45 and 46 :)
[14:43] <jhodapp> sil2100 or Mirv, can you please reconfigure silo 19?
[14:43] <boiko> Mirv: emptying the silos is fine
[14:43] <Mirv> boiko: ok, handling all of them
[14:44] <boiko> Mirv: thanks!
[14:50] <jhodapp> Mirv, around?
[14:51] <Mirv> jhodapp: I should say no ;)
[14:51] <jhodapp> hehe
[14:51] <jhodapp> just need a quick reconfigure on silo 19
[14:52] <Mirv> jhodapp: ok
[14:52] <jhodapp> thanks
[14:53] <Mirv> jhodapp: done
[14:53] <jhodapp> Mirv, thanks much
[14:53] <Mirv> you're welcome much
[15:12] <Mirv> rsalveti: note that you have some gst-plugins-bad in https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-014/+packages without a line in CI Train sheet, can the silo be freed?
[15:13] <rsalveti> it had a line eventually
[15:14] <rsalveti> but will check in a few
[15:30] <bzoltan> Mirv:  is there an chance to push the qtc-p-u to w so it lands on trunk and I can setart to bckport and test on LTS/Vivid?
[15:40] <robru> Mirv: you don't need to empty the silos to reconfigure series anymore, that went away when dual silos was implemented.
[15:44] <Mirv> bzoltan: sure
[15:46] <bzoltan> Mirv: thank you
[15:47] <Mirv> robru: oh cool, then I can finally reconfig my Qt 5.5 silo from vivid to wily too! I started in early April and haven't dared to reconfig since I've my test packages there (and they're nowadays even for wily)
[15:47] <robru> Mirv: heh, yeah
[15:48] <rvr> sil2100: I haven't got the update on my personal krillin yet, should I?
[16:43] <davmor2> rvr: if you type in system-image-cli -i what channel does it say you on out of interest?
[16:43] <rvr> davmor2: Let me check
[16:44] <rvr> channel: ubuntu-touch/stable/bq-aquaris.es
[16:45] <davmor2> rvr: thanks
[16:57] <dbarth_> trainguards o/ you can remove silo 010; i have moved its content to other silos to streamline testing and landings
[16:57] <dbarth_> ie, don't merge / clean, just delete the silo please
[16:57] <dbarth_> sorry for the churn
=== alan_g is now known as alan_g|EOD
[17:13] <rvr> seb128: Approved silo 6
=== salem_ is now known as _salem
[18:00] <robru> kenvandine: content-hub was already released to wily. You can't just trunk release to vivid now as that would be going backwards; either branch for vivid or do a dual landing.
[18:01] <kenvandine> robru, huh?
[18:01] <kenvandine> robru, oh... mandel's silo?
[18:01] <robru> kenvandine: your build in silo 9 failed because you're a bad person and you should feel bad.
[18:01] <slangasek> who would know the reason for this build failure, related to libmirclient-dev and libubuntu-application-api-headers in wily? https://launchpad.net/ubuntu/+source/gst-plugins-bad1.0/1.4.5-1ubuntu4/+build/7539540
[18:01] <kenvandine> robru, haha
[18:01] <slangasek> /usr/include/ubuntu/application/instance.h:28:44: fatal error: mir_toolkit/mir_client_library.h: No such file or directory
[18:02] <slangasek> looks like a mir regression, the mir_client_library.h header still exists but is apparently not on the include path
[18:02] <kenvandine> robru, looks like he never reconfigured it for the 15.04 branch
[18:02] <robru> kenvandine: ah ok
[18:02] <kenvandine> but i'd prefer a dual landing... just not the only thing in the silo
[18:02] <kenvandine> i know settings isn't ready for a dual landing
[18:02] <robru> kenvandine: so the spreadsheet has the right MPs but the silo doesn't?
[18:02] <robru> slangasek: no idea
[18:03] <kenvandine> robru, not sure, but the content-hub branch is superseded
[18:04] <robru> kenvandine: ah, no spreadsheet has the same MP. so yeah, you'll need to replace that MP with one not targetting trunk
[18:04]  * kenvandine wonders if they are all wrong
[18:05] <kenvandine> yup
[18:05] <robru> kenvandine: it's possible. You can check by examining debian/changelog file in the respective MP targets, if the first line contains 'wily' then it's all bogus for a vivid landing.
[18:05] <kenvandine> mandel, has udm branched for 15.04?
[18:06] <kenvandine> i know settings has, so we need a new MP for system-settings
=== _salem is now known as salem_
=== salem_ is now known as _salem
[19:05] <balloons> fginther, did you try / have any luck with using old versions of llvm / kernel on vivid.. re: core apps?
=== _salem is now known as salem_
[19:09] <fginther> balloons, no, I had not finished that experiment yet, let me try again
=== salem_ is now known as _salem
=== _salem is now known as salem_
[19:43] <jhodapp> robru, can you reconfigure silo 19 for me please?
[19:44] <robru> jhodapp: done
[19:44] <jhodapp> thanks
[19:46] <robru> you're welcome
[20:11] <fginther> balloons, any idea what might be happening here? http://91.189.93.70:8080/job/generic-mediumtests-vivid-test/25/console
[20:29]  * balloons looks
[20:30] <balloons> fginther, sorry looking
[20:44] <balloons> fginther, my guess was the app just crashed on launch. Looking at the videos confirms this.
[20:44] <balloons> no app ever appears
[20:46] <fginther> balloons, hmmm, qmlscene is running, could it be not rendering to the wm?
[20:48] <balloons> fginther, did you try running clock?
[20:48] <fginther> balloons, will try that next
[20:48] <balloons> fginther, yea, if qmlscene isn't running than that certainly shows the app didn't start
[20:49] <balloons> should make sure /usr/share/music-app/app/music-app.qml exists, and you can run it with /usr/lib/x86_64-linux-gnu/qt5/bin/qmlscene
=== chihchun is now known as chihchun_afk
=== tvoss is now known as tvoss|test
=== tvoss|test is now known as tvoss