/srv/irclogs.ubuntu.com/2014/01/24/#ubuntu-ci-eng.txt

pittigood morning07:43
pitticihelp: can I please have screen (or even better, byobu) on wazn?07:44
pitticihelp: unping, jibel installeld screen for me (thanks!)08:03
renatofginther, hi08:48
renatofginther, could you help me to understand what is happening on this build: https://jenkins.qa.ubuntu.com/job/address-book-service-trusty-amd64-ci/38/console08:48
renatoI do not have any idea what is causing the error,08:51
renatoonly if this is mixing different headers08:52
=== ev changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: ev | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: -
* cjwatson unsticks hud from trusty-proposed, again09:26
=== vrruiz_ is now known as rvr
didrockscjwatson: sorry, I didn't follow why it was blocked the first time (just mentionned it in the call), is there anything we should change?10:01
didrocksor is it again the tests being stuck while building? (not all processes killed)10:02
LaneyThe previous upload built packages for more architectures than the current one10:03
didrocksah ibgsettings-qt-dev10:05
didrocksl*10:05
didrocksand so Qt 5.210:05
didrocksit's all Mirv's fault! :)10:06
* didrocks chases dbus-cpp build-dep now10:07
didrockssil2100: I assumed that properties-cpp was Mir, wasn't it the case?10:08
didrockssil2100: that's why it can't see it's build-dep10:08
didrocksits*10:08
cjwatsondidrocks: it was only "again" because the revert brought the previous full set of architectures back after I removed them the first time10:10
cjwatsonI wasn't complaining, it was all reasonable, just mentioning10:10
didrockscjwatson: ah ok, was confused. Yeah, I noticed it this morning that it was blocked after the freeze. Thanks for looking at that! :)10:10
sil2100didrocks: aaah!10:12
sil2100didrocks: ok, makes sense - there's a MIR for that, but doko asked for it to be 'pulled in' by something so he can MIR approve it10:13
sil2100didrocks: https://bugs.launchpad.net/ubuntu/+source/process-cpp/+bug/127023410:13
ubot5Ubuntu bug 1270234 in properties-cpp (Ubuntu) "[MIR] process-cpp properties-cpp" [Undecided,Incomplete]10:13
didrockssil2100: so, I guess you can ping him :)10:13
sil2100didrocks: what does it mean 'needs a team subscriber'? ;p10:13
didrockssil2100: need a team to susbcribe to the bug reports10:15
sil2100didrocks: ah, in LP you mean?10:15
didrocksyeah, against the package10:15
sil2100didrocks: ok, this will sound really noobish, but how can I do that ;)?10:19
didrockssil2100: go to the package LP page: https://launchpad.net/ubuntu/+source/properties-cpp10:19
didrocksclick on subscribe to bug mail (in the upper right)10:20
sil2100Aaaaaaaaaa10:20
didrocksand you can subscribe bug mail recipient, (teams included)10:20
didrocksyou need to have someone part of that team subscribing of course10:20
sil2100So you can actually subscribe something else than just myself?10:20
sil2100Got it, thanks!10:20
Mirvdidrocks: Qt 5.2 should not yet be affecting anything unless the fixes made around the packages somehow break some archs. actually, after Qt 5.2 Qt should build on more archs because getting rid of V8 engine :)10:21
didrocksMirv: yeah, I was just kidding telling "we need it to be able to build on all archs" :p10:21
Mirvok :)10:22
cjwatsonbuild on more archs> that is indeed the point10:22
cjwatsonlooking forward to it for the arm64 and ppc64el stats :)10:22
tsdgeosguys, is there any way i can get shell access to the machines running https://jenkins.qa.ubuntu.com/job/autopilot-testrunner-otto-trusty/2262/? ?10:42
tsdgeosit's failing constantly in CI but i can't reproduce neither in my desktop nor in my phone10:42
tsdgeosev: ↑↑↑10:42
evtsdgeos: looking into your test failure now10:47
psivaadidrocks: so rerunning the maguro tests impacted by qmlscene crash does not help. the crash is reproducible and the tests are failing again.11:02
psivaatried twice.. running once more11:02
sil2100I jump out for some minutes to buy insurance for the trip11:03
didrockspsivaa: thanks for the head's up, seems a promotion blocker then11:04
didrockspsivaa: can you try to revert the content of the image one by one?11:04
didrocks(and then, the previous image)11:04
didrockspsivaa: weird that's it's "reliable" :/11:04
sil2100;/11:06
psivaadidrocks: ack, will do that.11:07
didrockspsivaa: keep us posted! thanks! ;)11:08
didrocksMirv: tell us when you got more infos/debugs from ricmm11:09
Mirvsure11:10
tsdgeosev: is there no change i get access to the machine instead of you looking at it? It'd be probably faster if i do the research than if you do (no offense) since i know more about unity8 and it's stuff11:28
tsdgeoss/change/chance11:28
renatohey guys, could you help me with this build: https://jenkins.qa.ubuntu.com/job/address-book-service-trusty-amd64-ci/38/console11:37
renatothe error is very strange, looks like jenkins is trying to use a old header .h with a new .cpp file11:37
renatolooking at the build log looks like jenkins does some file copies and updates, what I guess is that the header file has moved in the new version but jenkis did not notice that and now I have 2 headers11:40
renato'11:44
evtsdgeos: no, I really do not want to start handing out access to these machines. I'd much rather we solve the underlying problem of there being inconsistencies between what you're seeing on your local machine and what we're seeing in the CI infrastructure.11:44
renatorenato11:44
renatoalo123ola11:44
tsdgeosev: ok11:44
renatosudo reboot11:44
evIt should be dead-simple to reproduce the same environment.11:44
tsdgeosrenato: that looks bad :D11:44
renatoalo123ola11:44
tsdgeosev: yes, should is always the magic keyword :-)11:45
ev:)11:45
renatotsdgeos, hi11:46
tsdgeosrenato: i think we saw your password or something11:47
renato:D11:47
renatoyeah, my machine freezes :D, I changed that already11:48
tsdgeosoki11:50
=== MacSlow is now known as MacSlow|lunch
didrockspsivaa: any luck yet?12:14
psivaadidrocks: not yet. i remember we had qmlscene crash with 143 as well so going thorough http://people.canonical.com/~ogra/touch-image-stats/20140122.1.changes as well12:15
didrockspsivaa: yeah, but it' wasn't 100% of the time like that one12:16
didrocksdavmor2: hey, meanwhile, can you dogfood latest of latest? :)12:16
psivaadidrocks: yea, just to make sure if any combination of this list in 20140122.1 and those in 20140123.1 is causing it12:17
didrocksok ;)12:18
davmor2didrocks: sure12:20
didrocksthanks!12:20
davmor2seb128: on the bluetooth setting page if you turn blutooth off should you see the spinner for searching still?12:47
seb128davmor2, I guess not12:47
davmor2seb128: I'm assuming it isn't searching and it's just the spinner spinning for no reason just looks odd12:48
seb128davmor2, seems like  a bug, please open one ;-)12:49
davmor2seb128: certainly I'll do it at the end of the testing :)12:49
didrockspsivaa: seems this time was better on the dashboard?12:51
psivaadidrocks: the failures have gone down but the crash still occurred12:52
didrockspsivaa: hum, interesting…12:52
psivaadidrocks: yea, quite confusing as to what's causing this crash..12:52
tsdgeosev: so did you have any luck finding out what is wrong?12:53
didrockspsivaa: maybe, we just got lucky yesterday?12:53
didrockspsivaa: try upgrading everything again, and run, BUT with fingers crossed? :p12:53
evtsdgeos: no, it's taking forever and a day to bootstrap my environment12:53
didrockspsivaa: to see if it's really 100% or if crossing fingers influence ;)12:53
evI'm going to have to hand this off to the next vanguard as I'm coming up on the end of my shift12:53
psivaadidrocks: :), quite possibly12:54
evtracking it here https://app.asana.com/0/8736198969650/978255354031112:54
tsdgeosoki12:54
tsdgeosev: how do i login there, with my canonical sso?12:56
evoops, need to add you as a follower on the task12:56
evbut yes, canonical sso12:56
evshould work now12:58
tsdgeostx12:59
=== MacSlow|lunch is now known as MacSlow
davmor2seb128: feel free to unsub yourself but you seem the obvious target for system settings :) https://bugs.launchpad.net/ubuntu/+source/ubuntu-system-settings/+bug/127231713:01
ubot5Ubuntu bug 1272317 in ubuntu-system-settings (Ubuntu) "Bluetooth spinner spins when bluetooth is disabled" [Low,New]13:01
seb128davmor2, I'm subscribed to the package so I get the new bugs anyway, not need to subscribe me, but thanks13:02
davmor2seb128: assign no subscribe sorry wrong term13:03
seb128davmor2, oh ok, it's one for charles (he wrote the bluetooth panel), I'm got to reassign, no worry13:04
davmor2seb128: thanks13:05
=== ev changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cihelp | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: -
davmor2didrocks: 146 is looking pretty stable here looks like popey is seeing the same13:10
popey+113:10
davmor2popey: have a look at you national rail app and see if you can set a date for travel :)13:12
popeyk13:13
popeyworks for me13:15
davmor2popey: last night I wanted to check times and prices for the trip down I had a floating list view of numbers13:16
popeyscreenshot?13:16
davmor2popey: I'll grab one in a second13:16
=== alan_g is now known as alan_g|lunch
didrockspopey: davmor2: great! thanks guys :)13:18
didrockspsivaa: it's on you now, to know if we can promote (like the crash magically disappeared and we were just unlucky) or really reliable13:19
davmor2popey: oh so now the date works but doesn't change unless you change another bit13:19
psivaadidrocks: ack, give me about 30-45 mins..?13:19
didrockspsivaa: perfect :)13:20
didrocksricmm: hey, did you get any luck looking at psivaa's stacktrace on qtubuntu?13:37
davmor2popey: slowly the script gets more refined, now errors out nicely when there is no device attached :)13:37
didrocks(qmlscene crash)13:37
popeydavmor2: make it work with more than one device attached ☻13:37
popeye.g. two phones or one phone and the emulator13:37
davmor2popey: it's probably possible but I doubt my head can handle it :)13:38
popeydavmor2: ping me the link to the code I'll look13:38
davmor2popey: https://code.launchpad.net/~davmor2/+junk/is_there_an_update  nice and simple13:39
popeykk13:40
Mirvdidrocks: status update with bug #1271879 is that we replied to ricmm about that yes it's multiple apps and on maguro, no further updates at the moment13:43
ubot5bug 1271879 in qtubuntu (Ubuntu) "qmlscene crashed with SIGABRT in qt_message_fatal()" [Undecided,Confirmed] https://launchpad.net/bugs/127187913:43
didrocksMirv: ok, thanks, let's see once ricmm is freed13:43
psivaadidrocks: qmlscene the crash keeps occurring, no luck with reverting packages. Since the crash has been occurring even before 144, i dont see why we should not promote 14614:04
didrockspsivaa: well, it wasn't happening everytime in previous images14:05
didrocksnow, it does14:05
didrocksricmm_: this is blocking promotion FYI14:05
fginthermorning14:16
sil2100didrocks: a batch of packaging ACK's required o/14:18
=== alan_g|lunch is now known as alan_g
sil2100didrocks: http://paste.ubuntu.com/6808577/ :)14:22
sil2100ogra_: maybe you have a momento for some packaging ACKs? These all seem to be related to the introduction of unity-control-center14:25
sil2100ogra_: I prepared for non-VPN access14:25
sil2100ogra_:14:25
sil2100ogra_: http://paste.ubuntu.com/6808584/14:25
didrockssil2100: +1 for all14:27
ricmm_psivaa: what apps does it happen with14:36
ricmm_all? or just that one14:36
psivaaricmm_: calendar and clock apps easily reproducible on maguro14:36
ricmm_can you reproduce on mako?14:36
davmor2psivaa: balloons was looking at calendar yesterday with some testing help from me I think he got the issue mostly resolved then tried to tweak it and broke it again :(14:37
davmor2ricmm_: ^14:38
psivaadavmor2: but it does not explain the crash during clock app tests14:39
psivaaricmm_: no, the crash has not occurred on mako14:39
davmor2psivaa: ah sorry are you talking crashes not tests fails sorry14:40
psivaadavmor2: np :)14:40
psivaaricmm_: the qmlscene crash that we see now has never occurred on mako iirc14:41
ricmm_but it doesnt happen every single time, right?14:41
balloonspsivaa, clock is crashing on maguro?14:42
ricmm_tedg: did we (you/gerry) ever resolve that timing issue with upstart signals and the shell's authroization or processes?14:42
ricmm_tedg: sounds like it could be it ^14:42
psivaaricmm_: with 146 almost every calendar and clock AP tests14:42
psivaaballoons: it's not clock but qmlscene crashes during clock AP tests14:42
psivaaand calendar14:42
tedgricmm_, Yeah, we created two events.  We block on Unity replying before starting the app.14:43
tedgIt does have a timeout of 2 seconds.14:43
tedgPerhaps that timeout is getting hit on the emulator.14:43
ricmm_its maguro thats crashing14:43
tedg(seems like a long time though)14:43
tedgHmm, probably not timing out.14:44
tedgIs there anything in the application log?14:44
davmor2psivaa: I think this might be the issue I saw. Due to the low memory on the maguro if you close an app and the memory level is not fully released and you reopen the app it locks up and your screen goes grey but the app never fully opens.14:44
ricmm_why are we testing maguro14:45
ricmm_if asac deemed it EOL'd14:45
ricmm_side question :)14:45
=== ricmm_ is now known as ricmm
asacricmm: because we didnt move over yet14:46
asacuntil its dead its alive :)14:46
psivaadavmor2: probably. any sign of crash files in there when that occurs?14:46
ricmmasac: understood14:46
davmor2psivaa: only really in the apps log no really crash report.  I've not had it recently though.14:47
asacricmm: once the emulator is ready we do the 4.4. migrate14:47
asaci hope noone thinks we go earlier to 4.414:47
asacat leaast not without discussing14:47
* asac hopes for next week14:48
davmor2psivaa: if you have a look at the log for calendar it tells you if it failed to open the app let me dig out the bug it had all the logs I could get in place14:48
davmor2psivaa: https://bugs.launchpad.net/unity8/+bug/126869314:49
ubot5Ubuntu bug 1268693 in Unity 8 "Possible bug in mir/memory on maguro" [Undecided,New]14:49
didrocksdavmor2: here the qmlscene is crashing on every try, so it's not random anymore14:51
ricmmim charging a maguro14:52
didrocksthanks ricmm14:52
davmor2didrocks: was less frequent when I reported it though :)  Let me grab htop and run the calendar tests and see what happens14:53
didrocksdavmor2: great!14:53
balloonsdavmor2, I guess I'll try my heand again at calendar before you EOD14:57
balloonsI'll let you know14:57
psivaadidrocks: i've rerun those tests after reverting all the packages that changed between 144 and 146 and still it's the crashes occur with the same freq btw. so we should have been lucky with 144 i guess14:58
davmor2balloons: cool I'm here till 20:00 UTC14:58
didrockspsivaa: yeah, really weird, can you try to reinstall directly 144 and rerun?14:58
psivaadidrocks: ack, will try. (dint think of that :))14:59
didrockspsivaa: let's see, as rw + revert != flashed image14:59
didrocksthanks psivaa, keep going :)14:59
ricmmdidrocks: alright so which app is easiest to crash15:07
ricmmclock? or calendar15:07
ricmmpsivaa: davmor215:15
davmor2ricmm: I had issues running tests on calendar15:16
psivaaohh ricmm any of the two would do to repro the crash15:16
ricmmand you were able to reproducing by running qmlscene manually from terminal?15:17
psivaaricmm: no, i've been running the AP tests15:18
=== alan_g is now known as alan_g|tea
=== gatox is now known as gatox_lunch
davmor2ricmm, psivaa, didrocks: I have the grey screen, I have the _usr_lib_arm-linux-gnueabihf_qt5_bin_qmlscene.32011.crash and the memory look like it was climbing whoever the shutdown and startup of the app was making unity8 (I'm assuming mir) leap to the top of the htop and was maxing out the cpu cores. So still looks like a mir/gfx memory/memory/cpu issue which is possibly why we don't see it on mako15:49
davmor2it took 3 runs for me to grey the screen and get the crash15:49
ricmmI dont know what you mean by grey scren15:50
ricmmscreen15:50
davmor2ricmm: https://bugs.launchpad.net/unity8/+bug/1268693 like in the image in here15:50
ubot5Ubuntu bug 1268693 in Unity 8 "Possible bug in mir/memory on maguro" [Undecided,New]15:50
davmor2ricmm: the app holder opens but the app itself doesn't15:51
ricmmoh that15:51
ricmmyea thats a shell component, nothing solid or to do with mir15:51
ricmmjust the app when it first fails to start, that stays up there15:51
ricmmbecause unity/upstart get out of sync on whats currently running15:51
=== alan_g|tea is now known as alan_g
psivaadavmor2: that sounds like the crash that we see15:54
bfillerfginther: seems like some CI problems, can you check: https://code.launchpad.net/~bfiller/address-book-app/disable-predictive-text/+merge/20277616:16
psivaadidrocks: davmor2: ricmm: so i have run clock app test on a maguro with 144 thrice and was able to reproduce the crash each time16:16
fgintherbfiller, looking16:17
fgintherbfiller, some automatic translations were added 2 days ago and it looks like the packaging wasn't updated to match16:18
bfillerfginther: oh really, shoot16:19
fgintherbfiller, any idea how that is  supposed to work. Appears to be a good way to break the build16:19
bfillerfginther: no, dpm any idea what is needed for address book packaging to pick up the translations? enabling them seemed to break the build ^^^16:20
dpmbfiller, otp, will have a look as soon as I'm done. Where's the build log?16:21
bfillerdpm: thanks, https://jenkins.qa.ubuntu.com/job/address-book-app-trusty-armhf-ci/69/console16:22
bfillerfginther: sorry to bug you on this one, I should have looked at the log more closely.16:23
fgintherbfiller, no worries. It's always good to keep an eye on these things for the next time it comes up16:25
bfillerdpm: think I got it, just missing this in address-book-app.install16:48
bfillerusr/share/locale/*/LC_MESSAGES/address-book-app.mo16:48
=== gatox_lunch is now known as gatox
dpmbfiller, ok, I'm off the phone. Did that solve your issue, or do you still want me to look at it?16:58
bfillerdpm: solved16:58
dpmok, cool16:58
davmor2popey, didrocks, cyphermox: just confirming I will be there Tuesday \o/ so see you all then :)17:00
popeyyay17:00
didrocksogra_: plars: coming?17:02
ogra_yep17:02
didrocksdavmor2: looking forward to see you!17:02
cyphermoxdavmor2: cool17:02
davmor2cyphermox: anything other than the bluetooth headset you want me to bring down?17:02
davmor2I'll put it on charge Monday :)17:02
plarsdidrocks: yes, brt17:03
cyphermoxdavmor2: tbh, any bluetooth thingy is cool17:10
cyphermoxI'm bringing two headsets that can do a2dp and hsp, one LE SDK device, and one BT keyboard17:11
davmor2cyphermox: bt headset/dongle/laptop/ps3 controller it is then17:11
cyphermoxmmm17:11
davmor2haha17:11
cyphermoxperhaps I could bring a wiimote then17:11
cyphermoxnot that those things are very much useful to test17:12
cyphermoxbut at least they prove input support works17:12
=== retoaded changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: retoaded | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: -
didrocksricmm: if you get any progress, please check to release that with cyphermox/robru17:12
cyphermoxthe worst I guess is going to make sure we have batteries17:13
=== Ursinha is now known as Ursinha-afk
robrukenvandine, cyphermox: anybody have time for a trivial review? https://code.launchpad.net/~robru/cupstream2distro-config/friends-packages/+merge/20314318:16
kenvandinerobru, done18:18
robrukenvandine, thanks18:18
=== retoaded changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cihelp | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: -
=== retoaded changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cihelp | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: -
=== Ursinha-afk is now known as Ursinha
mterryI have a unity8 autopilot test that is failing in jenkins but not on my device.  Is there a special setup to get a more jenkins-like environment?20:52
mterryfginther, ^ ?21:01
fginthermterry, is this a touch device?21:01
mterryfginther, mako21:01
fginthermterry, I'll send you something, give me a few minutes please21:04
mterry fginther, sure thanks!21:04
robruricmm, any updates on qtubuntu? ready for testing?21:15

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