/srv/irclogs.ubuntu.com/2015/06/30/#ubuntu-ci-eng.txt

fgintherbfiller, in case you didn't already see this: http://s-jenkins.ubuntu-ci:8080/view/click/job/gallery-app-click/263/artifact/out/com.ubuntu.gallery_2.9.1.1205_armhf.click01:00
fgintherbfiller, from http://s-jenkins.ubuntu-ci:8080/view/click/job/gallery-app-click/263/01:00
bfillerfginther: thanks02:52
=== chihchun_afk is now known as chihchun
=== chihchun is now known as chihchun_afk
=== chihchun_afk is now known as chihchun
bzoltan_robru:  do you know who should I ask if need help to make jenkins doing its job instead of blocking the integration?06:18
robrubzoltan_: which Jenkins? You need cihelp for anything other than ci-train.u.c06:19
robrubzoltan_: oh you mean the mp reviews? Yeah you need cihelp for that06:20
robrubzoltan_: i would recommend disabling that for train projects since it has no ability to handle deps across multiple MPs06:21
bzoltan_robru:  good point06:25
bzoltan_robru:  is there a separare channel for that or it should be a highlight here?06:37
robrubzoltan_: this is the spot, i don't know if any ci people are around just yet though. Maybe wait a few hours06:38
bzoltan_robru: OK, I will ping fginther or elopio or anybody who hears the cihelp call :)06:42
robrubzoltan_: Francis is us time, that might be a wait. Ev is uk, should be sooner06:43
thomirobru: ev is us too. psivaa is UK06:43
robruOoh right06:44
bzoltan_robru:  thomi: would be nice to have a GMT+N (N>=1) cihelper06:44
robruI forgot he moved to Houston06:44
robruAustin06:44
thomibzoltan_: psivaa06:44
robruThat place we just were ;-)06:44
ogra_hmpf07:46
ogra_http://i.imgur.com/B1bX5SO.png07:46
ogra_i assume this shouldnt actually be possible :P07:46
jibelI saw it once and some else reported it too (don't remember who though)07:48
psivaabzoltan_: hey08:06
psivaaWhich is the MP being blocked by jenkins?08:06
bzoltan_psivaa:  good morning, we have all kinds of problems with Jenkins... most of our MRs do not pass jenkins08:06
bzoltan_psivaa:  almost all of them https://code.launchpad.net/ubuntu-ui-toolkit/+activereviews08:07
psivaabzoltan_: morning, let me take a look08:07
bzoltan_psivaa:  this one I would like to land most https://code.launchpad.net/~ubuntu-sdk-team/ubuntu-ui-toolkit/qtmacros/+merge/25947708:07
* zbenjamin looks08:09
zbenjaminpsivaa: seems some tests (always different ones) can not get a Mir connection08:10
psivaabzoltan_: zbenjamin: yea as you said they are actual test failures, (albeit different ones) not related to the infrastructure08:12
psivaaso we'd not be able to do anything to fix them08:13
psivaabut if you want them the whole test to be disabled, we could do that08:13
zbenjaminpsivaa: i wonder how can not getting a mir connection be a test failure?08:13
zbenjaminthis one for example https://jenkins.qa.ubuntu.com/job/generic-deb-autopilot-runner-vivid-mako/2765/testReport/junit/ubuntuuitoolkit.tests.gallery.test_textinput/WriteAndClearTextInputTestCase/test_clear_textfield_must_remove_text_Nexus_4_standard_textfield_/08:14
zbenjaminpsivaa: especially this MR fails always with that error, but always in different tests08:15
zbenjaminpsivaa: or am i missing something?08:17
psivaazbenjamin:  if that (not getting mir connection) is causing an autopilot test to fail and the process depends on that test to pass for the MP to be merged.08:17
psivaazbenjamin: i dont know the depth of how this *test works and we in CI wont be able to debug that08:17
zbenjamin:(08:18
psivaazbenjamin: the process depended on each of the tests to pass so that has to be debugged by someone with in-depth knowledge of the tests08:18
psivaazbenjamin: sorry if I coming across straight, :), still haviing my first coffee08:19
psivaazbenjamin: do you by chance run the tests locally ?08:19
psivaaon a mako?08:19
zbenjaminbzoltan_: ^^ there is a script for that right?08:19
ogra_bzoltan_, oh, just FYI, sil2100 manages the metapackages nowadays (i saw the discussion in the other channel)08:20
bzoltan_ogra_:  Good to know. Thank you.08:21
bzoltan_zbenjamin:  the test plan script can run UITK AP tests of course08:21
zbenjaminbzoltan_: the script also might need to adapt to the new build structure?08:22
bzoltan_zbenjamin: what new build structre? Why?08:23
bzoltan_psivaa:  sorry I missed to understand the Mir display issue... how is that not an infrastructure problem?08:23
bzoltan_psivaa:  zbenjamin: I am running dozens of times the UITK AP tests locally on krillin and never seen that failure.08:24
zbenjaminbzoltan_: because the plugins are now compiled into a folder named "qml" and not "modules"08:24
zbenjaminbzoltan_: thats a result from using the qt macros08:24
bzoltan_zbenjamin: so? how does that effect the test plan?08:25
zbenjaminbzoltan_: well i assumed its using the push_to_phone script. But maybe not :D08:26
bzoltan_zbenjamin:  absolutely not08:26
psivaabzoltan_: sorry, what should we in the infrastructure, do to fix 'Mir display issue'08:26
zbenjaminbzoltan_: ok then just forget what i said :D08:26
psivaai'm putting this in quote is because that's all what i understand :)08:26
bzoltan_psivaa:  no idea ... I do not see it locally08:27
psivaabzoltan_: also do you run this test on Mako?08:27
bzoltan_psivaa:  no08:27
bzoltan_psivaa: we work on OTA updates what target bq devices08:27
psivaabzoltan_: ok, so running these tests in mako do not make much sense08:28
psivaa?08:28
bzoltan_psivaa: sense it might make ... but it would not be enough to pass the QA validation08:28
psivaabzoltan_: if that's the case, we'd need to get an agreement if08:29
psivaa1. we should disable the tests on Mako08:29
psivaa2. Fix the tests for mako even though the target is BQ08:29
psivaa3. Change the devices to krillin08:30
bzoltan_psivaa: 3.08:30
bzoltan_psivaa:  testing changes on mako has value ... but blocking landings on krillin because of mako does not sound good to me08:30
psivaaI'd need to discuss 3 with the team, but especially whilst we're in the process of moving away from a single entity running all the product/ project tests towards each team, i'm not sure if that will be done in time08:31
bzoltan_psivaa:  do you think the same tests would pass on krillin?08:31
psivaabzoltan_: I thought that's what you said above :)08:31
psivaa' I am running dozens of times the UITK AP tests locally on krillin and never seen that failure.'08:31
bzoltan_psivaa:  I strongly doubt that the only difference between the Jenkins testing is the device type :)08:32
psivaabzoltan_: but that's all we in CI have control over :)08:32
bzoltan_psivaa:  but I find it very very strange that the CI infra is not testing on the primary target device08:32
bzoltan_psivaa:  what would it take to replace those makos with krillins?08:33
psivaabzoltan_: (whilst running the risk of discussing something which I do not think will happen soon, due to change of focus as i said above)08:34
psivaawe'd need more krillins08:34
bzoltan_psivaa:  anyhow... right now running AP tests on UITK MRs does not make much sense as I am doing super massive 4 time full AP tests before landings.08:34
bzoltan_psivaa:  The UITK test plan covers more AP test than any QA/CI process ...08:35
psivaabzoltan_: ok, shall we disable that then?08:35
bzoltan_psivaa:  we run _all_ available AP tests ...08:35
bzoltan_psivaa: temporally I would do that08:36
bzoltan_zbenjamin: ^ are you OK with that?08:36
zbenjaminbzoltan_: psivaa: +108:37
psivaabzoltan_: zbenjamin: ok, i'll work on it then08:37
bzoltan_psivaa:  Thank you a bunch!08:37
zbenjamin\o/08:37
psivaazbenjamin: bzoltan_: thanks for the info08:37
bzoltan_psivaa: at the same time I will enforce the QA process of the UITK project. It means we will do more tests on our staging branch.08:39
psivaabzoltan_: that sounds great08:40
popeysil2100: on todays image on krillin and arale, neither of them will stop buzzing after a notification...08:51
popeyhaving to restart both phones to shut them up08:51
sil2100popey: oh, hm, let me check the commitlog on what landed08:52
popeybrb08:52
sil2100popey: yeah, looks like we had unity-notifications landing08:53
sil2100"Make unity8 notifications server more robust against misbehaving clients"08:53
* sil2100 waits for pete-woods08:54
popeyok08:55
popeywant a bug filed?08:55
sil2100popey: yeah, please, might use that if I want to revert it08:57
sil2100Thanks!08:57
popeyok will do08:57
jibelpopey, after a calendar notification or any notification?09:05
oSoMoNtrainguards: hey, I need to reconfigure silo 16 to target both wily and vivid overlay, can someone do that for me please?09:08
=== vrruiz_ is now known as rvr
sil2100oSoMoN: you mean, you want it to be a dual-landing silo, right?09:17
oSoMoNsil2100, yes09:17
psivaabzoltan_: just one more question, do you only want the AP tests to be disabled in mako or even in amd64 as well?09:17
sil2100oSoMoN: on it09:17
oSoMoNthakns09:18
oSoMoNthanks09:18
sil2100oSoMoN: done!09:18
oSoMoNsil2100, thanks09:19
popeyjibel: calender here09:19
popeysil2100: jibel bug 147003109:25
ubot5bug 1470031 in unity-notifications (Ubuntu) "Phone won't stop vibrating on calendar notification" [Undecided,New] https://launchpad.net/bugs/147003109:25
mzanettisil2100, hey, did you have a chance to look into the wily builds?09:26
mzanettistill failing it seems09:26
jibelpopey, did you file a bug? I saw that last week but don't remember if it was with or without the silo09:27
popeysee above :)09:27
popeyI am not on any silos, unmodified largely.09:27
jibelpopey, heh sorry :)09:28
bzoltan_psivaa:  on mako as start ... if I start to see green Jenkins jobs then the amd64 can stay09:34
sil2100mzanetti: I think for this one we'll need some LP/builder specialists09:34
psivaabzoltan_: ack, that's what i've done, MP waiting for approval. thanks09:34
sil2100cjwatson, wgrant: could you help us with something? We have https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-017 that seems to fail to build basically all its packages stating that dependencies couldn't be met09:35
sil2100While the new packages don't even change their dependencies09:35
psivaabzoltan_: zbenjamin: https://code.launchpad.net/~ubuntu-sdk-team/ubuntu-ui-toolkit/qtmacros/+merge/259477 is now green ( from a manual run skipping mako tests)09:36
jibelsil2100, I confirm popey's bug, very likey silo 18 that landed yesterday09:37
bzoltan_psivaa: \o/09:37
sil2100jibel: could you try reverting qtdeclarative5-unity-notifications-plugin to 0.1.2+15.04.20141104-0ubuntu1 ?09:37
jibelsil2100, on it09:37
cjwatsonsil2100: Looking09:38
cjwatsonYikes09:38
cjwatsonsbuild: warning: can't parse dependency g++-4.9:native09:38
cjwatsonSo there are two possible fixes for this09:39
cjwatsonOne is a thing I'm working on today anyway, to have sbuild use apt-get build-dep - it's non-trivial because it requires rearranging how we detect dep-waits09:39
cjwatsonThe other is to have sbuild strip the qualifier there, which I kind of thought it already did09:40
cjwatsonLet me have a look at why the latter isn't working09:40
sil2100cjwatson: thanks! hah, actually with the new output I somehow missed this particular error09:40
sil2100The output is too clean for me and it confuses me09:40
cjwatsonI wonder if precise's libdpkg-perl is too old for this09:42
cjwatsonwgrant: Right, so precise's dpkg is lacking https://anonscm.debian.org/cgit/dpkg/dpkg.git/commit/scripts/Dpkg/Deps.pm?id=40d51dc36b2375679a00ed9ff64ee017dc78b0a9, possibly other related things.  I propose we hack the precise sbuild backport to strip off architecture qualifiers before deps_parse sees them.  Any objections?09:53
cjwatsonWe wouldn't have seen this in PPA testing, of course.09:53
cjwatson(because trusty)09:53
wgrantcjwatson: Argh.09:55
wgrantBut sounds like the best solution.09:55
jibelsil2100, I confirm that it's a regression introduced by silo 1809:55
sil2100jibel: ok, best way is to soft-revert and poke pete for a real fix09:56
sil2100Let me prepare everything, would need to revert it for both archives09:56
* rvr tested silo 18, not happy :-/09:57
rvrpopey: Good catch09:59
popeynp09:59
jibelsil2100, can you rebuild an image once it is reverted?10:01
sil2100Sure10:01
jibelthanks10:01
cjwatsonmzanetti: I think I have a patch, just setting up a suitable test environment10:10
sil2100cjwatson: thanks \o/10:10
mzanetticjwatson, awesome! thanks a lot!10:12
sil2100jibel: the revert is building in the overlay and archive, will be tracking once it's done and kick an image10:23
sil2100jibel: the image is building11:10
=== MacSlow is now known as MacSlow|lunch
cjwatsonwgrant: Could you review http://paste.ubuntu.com/11798585/, please?11:36
cjwatsonSeems to work here, tested with unity-api from that silo.11:36
cjwatsonThe regex is basically the start of the one from Dpkg::Deps.11:39
wgrantcjwatson: Looks sane to me, as long as it also works.11:42
cjwatsonThanks.11:43
popeysil2100: do we not have an image bot anymore?12:05
sil2100popey: it's not working right now12:06
ogra_well, it could tell you about old images ...12:06
ogra_but thats about it ... the re-swetting of the image versions caused quite some havoc i havent found time to fix12:06
cjwatsonmzanetti,sil2100: https://rt.admin.canonical.com/Ticket/Display.html?id=82496, shouldn't be long now12:10
sil2100cjwatson: thanks for that!12:11
mzanettigreat :)12:12
cjwatsonmzanetti: Um, I didn't say you could rebuild yet :)12:17
mzanetticjwatson, sorry, I triggered the rebuild before your last messages12:17
cjwatsonmzanetti: Also, don't rebuild it like that.12:18
cjwatsonmzanetti: Unless you were actually making source changes.12:18
mzanettididn't know I get those nice status updates so I thought I just retry every couple of hours12:18
mzanetticjwatson, how should I build then?12:18
mzanettior rather: what exactly is the difference?12:18
cjwatsonmzanetti: When it's upgraded, I'll retry it for you in the PPA, without needing a new source upload (which is what the "Build" button in the train amounts to)12:19
mzanettiohh. that makes sense12:19
mzanettiack12:19
=== chihchun is now known as chihchun_afk
=== MacSlow|lunch is now known as MacSlow
=== _salem is now known as salem_
brendandsil2100, when was the last image built?13:08
brendandsil2100, for vivid13:08
brendandactually let me check the image server13:08
brendandthis morning13:09
jibelbrendand, there was one this morning and another is currently buildiong13:10
brendandjibel, how do you find if one is in progress? my connection is not reliable so i often miss the pings here13:10
jibelbrendand, because sil2100 said so 2 hours ago13:11
jibelbrendand, otherwise https://launchpad.net/~ubuntu-cdimage/+livefs/ubuntu/vivid/ubuntu-touch/13:11
jibelyou can trust this more than what I say :)13:12
sil2100I was building an image in the morning, it should be in the channel13:30
psivaabzoltan_: the change to disable the touch tests on UITK has landed in the infra. if you re-run any failed MP, they should pass13:43
cjwatsonmzanetti,sil2100: That's all fixed for you now, PPA builds retried, and ci-train is doing a watch-only build13:48
sil2100\o/13:51
mzanetticjwatson, thank!13:53
mzanetti+s13:53
cjwatsonmzanetti: also, it may be relevant to your interests that :native now works in recipes, as of a day or two ago13:54
mzanetticjwatson, that is interesting indeed. thanks for that too :)13:54
cjwatsonno more pbuilder in recipe builds!  phew13:54
=== pat_ is now known as Guest96981
kenvandine2015-06-30 14:20:34,372 ERROR Uncaught exception: ServerNotFoundError: Unable to find the server at api.launchpad.net14:21
kenvandinetrainguards: ^^ tried to rebuild a silo14:21
kenvandinecitrain isn't happy14:22
sil2100hmmmm14:34
sil2100kenvandine: could you re-try?14:34
kenvandinesure, i'll re-try again14:35
kenvandinesil2100, boom14:35
kenvandinesocket.gaierror: [Errno -2] Name or service not known14:35
sil2100uuuh14:40
sil2100Some network issues on the jenkins server it seems14:40
mzanetti /subscribe to when this is resolved ^ :)14:49
cjwatsonsil2100: being chased in #is-outage internal14:50
cjwatsongeneral DNS problem in PS4.514:50
sil2100cjwatson: hah, just asked on #webops, thanks ;)14:50
=== sil2100 changed the topic of #ubuntu-ci-eng to: Need a silo or CI Train support? ping trainguards | Need help with something else? ping cihelp | Train Dashboard: http://bit.ly/1mDv1FS | QA Signoffs: http://bit.ly/1qMAKYd | Known Issues: general DNS problem in PS4.5 causing CI Train to be out-of-order
=== cjwatson changed the topic of #ubuntu-ci-eng to: Need a silo or CI Train support? ping trainguards | Need help with something else? ping cihelp | Train Dashboard: http://bit.ly/1mDv1FS | QA Signoffs: http://bit.ly/1qMAKYd | Known Issues:
cjwatsonsil2100: should be fixed now, you can retry stuff14:56
sil2100Yay14:57
brendandogra_, sil2100 - it seems that -updates and -security are in the sources.list for these vivid images, but not used during the build?15:04
sil2100They should15:04
brendandsil2100, that's what pitti tells me15:05
sil2100Interesting, the build config doesn't have anything specific that would say it's like that15:05
ogra_brendand, they definitely shoudl eb used ...15:06
ogra_unless someone explicitly hacked them out in livecd-rootfs ... which i dont think happened15:06
brendandogra_, ok i'm trying to figure out why i have oxide-qt-codecs-extra 1.7.8 installed, but 1.7.9 is in -security15:09
brendandi mean -updates15:09
brendandogra_, i just moments ago did the image update15:10
ogra_had it migrated already when the image was built ?15:10
ogra_s/was built/started building/15:10
brendandogra_, that's one possibility15:15
brendandogra_, pitti pointed me to - https://launchpad.net/ubuntu/+source/oxide-qt/1.7.9-0ubuntu0.15.04.115:16
brendandogra_, that was last week15:16
* sil2100 goes lay down for a while15:30
boikotrainguards: can I get the telepathy-qt5 package removed from silo 39? it has already landed in wily15:39
bfillertrainguards: silo 6 can be freed and I need a silo for line 62 please15:56
bzoltan_psivaa: thank you16:02
brendandogra_, sil2100 - would it be okay to ask for another vivid image build? this oxide landing is giving us bother16:14
ogra_brendand, well, someone should check the log and see if -security is actually enabled during builds ...16:14
brendandogra_, who would be able to do that?16:15
ogra_i guess someone who has time :P16:15
ogra_(dunno if we have anyone who is responsible for phone image builds already, i can perhaps look later but i dont plan to do a long day today)16:16
cjwatsonmzanetti: ^- that looks like an actual failure this time16:23
mzanettilooking16:23
cjwatsonmzanetti: if you think it's transient we can retry just that one build16:23
mzanetticjwatson, indeed16:24
mzanettino, it's a test failure16:24
brendandogra_, i'd do it myself but i don't even know where the logs are16:35
ogra_https://launchpad.net/~ubuntu-cdimage/+livefs/ubuntu/vivid/ubuntu-touch perhaps16:36
brendandogra_, ok then it does appear to be using -updates and -security16:37
ogra_Get:44 http://ports.ubuntu.com/ubuntu-ports/ vivid-security/main liboxideqtcore0 armhf 1.7.8-0ubuntu0.15.04.1 [21.3 MB]16:37
brendandgo build go!16:37
brendandogra_, that's looking like it was a timing issue. i think it migrated only 3 hours ago16:38
brendandogra_, perhaps during that build16:38
ogra_well, at least it is definitely pulling the packages from -security16:38
brendandogra_, yep so we can do a build16:39
popeysil2100: I'd like to request a debian package be added to the phone, how do I do that these days?16:40
ogra_popey, file an MP against the seed and point sil2100 to it16:40
ogra_oh, wiat16:40
ogra_no seed :P16:40
popey:)16:40
popeyglad I asked16:40
ogra_i guess then by simply asking sil2100 :)16:41
ogra_(or via a bug ... to prevent typos of the package name )16:41
popeyhaha, that's the voice of experience.16:42
popeyfiled against what?16:42
ogra_ubuntu-touch-meta16:44
popeykk16:44
* popey tickles sil2100 with https://bugs.launchpad.net/ubuntu/+source/ubuntu-touch-meta/+bug/147016616:47
ubot5Launchpad bug 1470166 in ubuntu-touch-meta (Ubuntu) "Please add libsmbclient to seed" [Undecided,New]16:47
popeyshould keep studio_ happy :)16:47
ogra_oh man ... how many deps does that pull in ?16:47
ogra_did anyone check ?16:47
* popey will add some detail to the bug :)16:51
pmcgowanuh oh16:51
* ogra_ has no intention to make studio_ happy ... where would be the fun in that :P16:51
popeyspoilsport16:51
popeyAfter this operation, 31.0 MB of additional disk space will be used.16:52
popeytiny :)16:52
ogra_rrright ... tiny ...16:53
ogra_i guess thats a bottle of mccallan 18 for pmcgowan to get that in16:53
pmcgowanooo16:53
popeypython 2.7 :S16:54
ogra_we need a "bribes" field in LP bugs ;)16:54
popeyhah16:54
ogra_awww16:54
davmor2popey: no no, not landing no python 2.7 on phone16:55
ogra_python 2.7 is rather a no ...16:55
popeyright, so how do we do this then16:55
popey(I prefer solutions then flat out 'no')16:55
popeyand no, "snappy fixes this" is not a solution I accept :)16:55
ogra_obey !16:56
ogra_:)16:56
popey:)16:56
davmor2popey: but but but but "Snappy fixes everything"!!!!!!!  ogra_ said so16:56
popeyfact is someone using a converged device is quite likely to want this16:56
ogra_well, i guess you ask slangasek if foundations could work on porting libsmbclient to python 316:56
popeybe good to test this _before_ then16:56
cjwatsonpopey: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=78191316:57
ubot5Debian bug 781913 in src:samba "please build bindings for Python3 and let samba-common-bin use them" [Normal,Open]16:57
popeythanks cjwatson16:57
ogra_ha !16:57
popey"yay"16:57
* popey goes outside to eat. will come back when the flames die down.16:58
davmor2popey: man you I love the fact that you know you are going to combust in the heat of outside but you are still doing it ;)17:00
charlestrainguards, ^ yes, please17:07
robrucharles: silo 4117:08
charlesrobru, thanks17:08
robrucharles: you're welcome!17:08
charleson it17:25
bfillerrobru: I don't need silo 6 anymore so it can be freed17:57
charles...2.0.0+15.10.20150619.1-0ubuntu1 is the current version number for 15.04...?17:59
charlesah17:59
robrucharles: hm?18:00
charlesrobru, fixable on my end, I think it's an MR error -- this should be going into the 15.04 branch but was MRed for trunk18:01
robrucharles: ah ok18:01
charlesrobru, well actually, since we're still testing it, maybe the shorter path would be to switch line 64 from vivid to wily...18:04
robrucharles: can do18:05
charleslooks like that's doable via Landing Tools18:05
robrucharles: yep, I made some changes recently so that you're able to take care of that yourself. but I just did it already so don't worry18:07
charlesrobru, :-)18:07
karniping tranguards - I'd like to have Telegram re-QA'ed, we've updated translations (which has been the only comment from last QA pass, as the scope was missing some strings)23:32
karni(FTR I'm off on Wednesday)23:33
robrukarni: you'll need to ping qa about that, ask jibel.23:49

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!