/srv/irclogs.ubuntu.com/2015/11/27/#ubuntu-ci-eng.txt

=== dobey_ is now known as dobey
robruoh god05:12
=== robru changed the topic of #ubuntu-ci-eng to: Train trouble? ping trainguards | CI problems? ping cihelp | Train: http://bit.ly/1hGZsfS | QA Signoffs: http://bit.ly/1qMAKYd | Known Issues: ci-train.ubuntu.com offline due to disk full, #webops has been contacted
=== robru changed the topic of #ubuntu-ci-eng to: Train trouble? ping trainguards | CI problems? ping cihelp | Train: http://bit.ly/1hGZsfS | QA Signoffs: http://bit.ly/1qMAKYd | Known Issues: -
=== robru changed the topic of #ubuntu-ci-eng to: Train trouble? ping trainguards | CI problems? ping cihelp | Train: http://bit.ly/1hGZsfS | QA Signoffs: http://bit.ly/1qMAKYd | Known Issues: train jenkins went offline, and when it came back *all* the job logs were lost.
Mirv:(06:58
robruyeah07:01
Mirvrobru: I've noticed that log disappearing before too on restarts, funny that the logs would be there but not shown07:03
robruMirv: yeah07:03
robruMirv: I just realized that all the diffs are inaccessible too, so I'm trying to get #is to tar them up for us so we can get to them07:05
robruMirv: so, WATCH_ONLY is gone, but if you're in a pinch and really need to regenerate diffs on a silo, you can add a phony source package, build that package in PACKAGES_TO_REBUILD, and when it runs the diff phase it diffs everything, then just drop the package from the silo config after that. but hopefully we'll get all the diffs and it won't be necessary07:07
robruMirv: or if the silo happens to have a manual source, just put that one in PACKAGES_TO_REBUILD and it'll do it07:07
Mirvrobru: oh right, thanks for the trick.07:10
robruMirv: I really don't want to bring back WATCH_ONLY but it seems like I'm going to need to provide some method of just regenerating diffs without rebuilding the packages.07:11
robruMirv: no luck from #is, apparently there's too many to feasibly tar them up, and it's difficult to grab just the latest one07:16
Mirvrobru: ok then07:29
robruMirv: actually he just made a tarball and it's only 340MBs...07:30
Mirv^ that's a first since March when the silo was started for vivid...07:49
Mirvmostly because no sense to work on the -gles packages before they are really needed07:49
Mirvbut surely there were some months of other build failure too, and then reappearing new build failures with new toolchains etc07:50
robruMirv: hmm? what's a first?08:02
Mirvrobru: first time silo 012 was fully and succesfully built08:07
Mirvall packages that are meant to land from there08:08
robruMirv: oooooh08:09
=== robru changed the topic of #ubuntu-ci-eng to: Train trouble? ping trainguards | CI problems? ping cihelp | Train: http://bit.ly/1hGZsfS | QA Signoffs: http://bit.ly/1qMAKYd | Known Issues: train jenkins lost all logs & artifacts; diffs are saved at https://chinstrap.canonical.com/~paulgear/jenkins-diffs.tar.gz
robruyowza08:13
robruMirv: so uh, you gonna land ticket 20 then? Would be great to get that one out the door eh? ;-)08:28
Mirvrobru: possibly next week :) it's starting to be as good as it's going to get at this point08:39
robruNice08:40
popeyMirv, https://requests.ci-train.ubuntu.com/#/ticket/710 is the clock update ticket for QA as requested yesterday.09:20
Mirvpopey: great!09:35
Mirvpopey: what about the dropping letters?09:35
popeynext on my list Mirv09:35
Mirvthank you09:36
diwichi, I click "Create new request" and get an error 401 - what team should I be a part of, and who can make me part of such a team?09:43
john-mcaleelysil2100, ^ ?09:55
brendanddiwic, it would be ci-train-users09:56
brendanddiwic, ci-train-ppa-service is the admin: https://launchpad.net/~ci-train-ppa-service/+members#active09:58
john-mcaleelyMirv, does that mean you can help add diwic?09:59
diwicbrendand, thanks, looking10:00
diwicall core devs are members of that team it seems like10:01
brendanddiwic, yeah should be lots of people who can do it10:02
brendanddiwic, unfortunately not me :(10:02
Mirvjohn-mcaleely: yes, diwic should just highlight "trainguards" so that we notice it more easily10:02
john-mcaleely:-)10:03
popeyMirv, dropping letters is done10:03
diwictrainguards: you are hereby highlighted10:05
Mirvdiwic: ok, you're now part of the ci-train-users team10:05
Mirvdiwic: try again10:05
Mirvpopey: \o/10:05
robrudiwic: log out and log back in though10:05
Mirvdiwic: \o/10:05
diwicMirv, thanks, working now10:05
robrudiwic: https://wiki.ubuntu.com/citrain/LandingProcess documentation is here10:06
diwicrobru, thanks, I'll try to follow it10:07
robrudiwic: you're welcome!10:08
=== marcustomlinson is now known as marcustomlinson|
=== marcustomlinson| is now known as marcustomlinson
rvrjgdx: ping10:19
jgdxrvr, pong10:23
=== michi is now known as Guest1024
=== ahayzen_ is now known as ahayzen
=== _salem is now known as salem_
=== chihchun_afk is now known as chihchun
* sil2100 off to lunch12:19
=== ondra_ is now known as ondra
Elleosil2100 / robru: any idea what's causing the "not in PPA" errors for some (but not all) packages in silo 17?12:46
=== chihchun_afk is now known as chihchun
=== alan_g is now known as alan_g|lunch
=== joc_ is now known as joc|lunch
Saviqrvr, hey, just ACKed the two branches, silo 5 should be good for testing13:12
rvrSaviq: Ah, good13:13
Saviqthey were just missing top-ack after some tweaks13:13
rvrSaviq: I already installed it, and was checking every merge proposal in the (large) list13:14
Saviqrvr, yeah, we like our large silos ;)13:16
rvrSaviq: This one also lacks top approval https://code.launchpad.net/~aacid/unity8/clazy_run/+merge/27832613:18
Saviqrvr, oh sry13:18
jibelrvr, jgdx I'm failing silo 39. It doesn't fix the crash.13:19
rvrjibel: Ack13:20
jgdxhiggins, ack13:21
jgdxjibel, ack13:21
jgdxjibel, could you do apt-cache policy libsystemsettings1 ?13:21
jgdxjust to confirm13:22
jibeljgdx, you're right to not trust me ;) http://paste.ubuntu.com/13526210/13:22
=== ev_ is now known as ev
jgdxjibel, looks good though!13:25
jibelpopey, I approved the clock app13:28
popeyjibel, thanks :) I saw and uploaded it to the store.13:28
jibelpopey, great, thanks13:29
Mirvthanks p & j for clock!13:35
jibelxavigarcia, silo 6 looks good, approved13:40
jibelxavigarcia, I found one configuration that doesn't really make sense and needs design input: 'disable GPS' and select 'GPS only' in location settings.13:41
jibelxavigarcia, I'll file a bug it is an existing case.13:41
jgdxcihelp: I'm seeing "UbuntuClientIntegration: connection to Mir server failed. Check that a Mir server is13:42
jgdxrunning, and the correct socket is being used and is accessible. The shell may have13:42
jgdxrejected the incoming connection, so check its log file"13:42
jgdx(sorry) and wondering what to do13:42
jgdxcihelp: it's happening in a lot of system settings jenkins autopilot jobs13:43
jgdxphablet-test-run never (on 4th run now) reproduces the issue13:44
xavigarciajibel: cool, thanks!13:45
xavigarciajibel: yeah, I saw that as well13:45
jibelxavigarcia, bug 1520580 for info13:52
ubot5bug 1520580 in Canonical System Image "invalid configuration: 'GPS disabled' and location set to 'GPS only' " [Medium,New] https://launchpad.net/bugs/152058013:52
=== joc|lunch is now known as joc_
morphissil2100: what does happen if we have a vivid only and a dual landing MP in a silo?14:06
morphisdoes that work and land the vivid only manual upload to the overlay ppa only and the MP to both overlay and xenial?14:07
kenvandinejibel, rvr: the crash you guys found in silo 39, that isn't a regression14:13
kenvandinei just reproduced that with ubuntu-system-settings 0.3+16.04.20151109-0ubuntu114:13
kenvandinedifferent bug14:13
jibelkenvandine, it is the crash the silo is supposed to fix14:13
kenvandineno14:13
kenvandinei've never seen that crash14:13
jibelkenvandine, without the silo it's https://errors.ubuntu.com/oops/d3f94da4-9505-11e5-9c7f-fa163e525ba7 corresponding to bug 1519046 which is the bug attached to the MP14:19
ubot5bug 1517254 in ubuntu-system-settings (Ubuntu) "duplicate for #1519046 /usr/bin/system-settings:11:operator:call:call:QtPrivate::QFunctorSlotObject:call" [Critical,Triaged] https://launchpad.net/bugs/151725414:19
jibelkenvandine, and with the silo it crashes with the same test case (retracing failed unfortunately)14:19
jibelkenvandine, so that bug is not fixed14:20
jibelkenvandine, or bug 1517254 and 1519046 are not duplicates14:20
ubot5bug 1517254 in ubuntu-system-settings (Ubuntu) "/usr/bin/system-settings:11:operator:call:call:QtPrivate::QFunctorSlotObject:call" [Critical,Triaged] https://launchpad.net/bugs/151725414:20
ubot5bug 1517254 in ubuntu-system-settings (Ubuntu) "duplicate for #1519046 /usr/bin/system-settings:11:operator:call:call:QtPrivate::QFunctorSlotObject:call" [Critical,Triaged] https://launchpad.net/bugs/151725414:20
jibeland only the first one has been fixed14:20
psivaajgdx: Could you give me the link to the job?14:24
jgdxpsivaa, one example: http://s-jenkins.ubuntu-ci:8080/job/generic-deb-autopilot-vivid-touch/5409/14:29
=== tvoss is now known as tvoss|late_lunch
psivaajgdx: thanks14:29
jgdxpsivaa, filed here https://bugs.launchpad.net/canonical-devices-system-image/+bug/152060414:31
ubot5Ubuntu bug 1520604 in Canonical System Image "Autopilot tests are failing because Mir is refusing the connection" [Undecided,New]14:31
jgdxpsivaa, there's a ubuntu keyboard run in there as well. So it seems it's not system settings specific.14:31
psivaajgdx: did you say you weren't able to reproduce this locally?14:37
psivaaif that's the case, just wanted to confirm, on which device you're running locally14:37
jgdxpsivaa, I'm running mako locally and I can't reproduce. Elleo says he can repro sometimes during normal app launch.14:38
jgdxElleo, what device was that?14:38
psivaajgdx: in our case, we're  only running those tests on krillin14:39
jgdxpsivaa, I know. So this would be device specific, then?14:40
Elleojgdx: that was on a mako, I've only seen it a couple of times14:40
Elleojgdx: no where near as frequently as seems to be happening in the tests14:40
jgdxElleo, okay, and it seems random?14:41
Elleojgdx: yeah, and starting the app again after it happens results in a perfectly normal launch14:41
Elleojgdx: might be worth grepping your own app logs to see if its happened at all to you without you noticing14:41
psivaajgdx: more likely, i can not confirm though. It would be ideal if someone runs those tests locally on krillin. Since we're not doing anything special in jenkins than how you run locally, I dont think we, in CI can do much about this failure14:42
Elleojgdx: since unless you're looking at the logs at the time the only thing you notice is that the app didn't start when you click on it14:42
Elleojgdx, psivaa: I'll update my krillin and set the keyboard tests going on there and see if they hit it at all14:43
jgdxElleo, thanks!14:43
psivaaElleo: thanks14:43
jgdxpsivaa, maybe help debug though14:43
kenvandinejibel, i'm confused, rvr said when he followed the steps to test this bug it passed but it crashed when starting directly in updates with flight mode enabled, which is a different bug14:45
psivaajgdx: hmm, I dont think I can do much here though. If there is anything infrastructure related then I could help, but this is not related to any infra setup since there are other tests that run fine14:48
jibelkenvandine, the thing is if you load the update page with an update available then turn airplane mode on it crashes with the trace in 1519046, and same with or without the silo. So either it is a different test triggering the same crash or the crash is partially fixed or the 2 crashes are not duplicates.14:48
kenvandinejibel, i'm filing a new bug for that14:48
kenvandinejibel, oh?  rvr said it didn't crash14:48
jgdxpsivaa, okay14:49
jibelkenvandine, yeah I don't know why he didn't have a crash, but the crash is the same with or without the silo and whoopsie says it's the bug attached to the tiket14:49
kenvandinejibel, i can't make it crash when switching flight mode on while in the panel14:50
kenvandinei can without the silo of course :)14:50
jibelkenvandine, you need a U1 account registered and at least an update in the list14:50
kenvandinei have 3 updates14:50
kenvandinejibel, i did reproduce it with downgrading settings to what's in the overlay14:51
kenvandinebut it doesn't crash when i install the silo14:51
jibelkenvandine, hm, let me try again14:51
jibelkenvandine, on which device?14:51
kenvandinearale14:51
jibelkenvandine, I didn't try on arale with the silo but krillin, I tried without the silo on both though14:51
jibelI'm retrying on krillin then arale14:52
kenvandinejibel, i did the testing from the MR on krillin14:52
kenvandinebut silo testing on arale14:52
kenvandineit didn't crash on either for me14:53
kenvandinejibel, i filed bug 1520611 for the other crash14:53
ubot5bug 1520611 in ubuntu-system-settings (Ubuntu) "[system-updates] Starting in updates panel with flight mode enabled crashes" [Undecided,New] https://launchpad.net/bugs/152061114:53
jibelkenvandine, my case is slightly different, you start the update panel then enable flight mode.14:54
kenvandineyeah14:54
jibelI don't know if it makes a difference14:54
kenvandinewhich is what i was fixing14:54
kenvandinewhile it's doing a check, enable flight mode14:55
kenvandinethat was triggering the crash in 151904614:55
kenvandinejibel, what are you doing?14:55
jibelkenvandine, what I said, with the silo installed, open settings, and enable flight mode, it still triggers this bug14:56
kenvandinein which page?14:57
jibelkenvandine, in the updates page14:57
kenvandinejibel, and you said the stacktrace matches?14:58
* kenvandine doesn't think that could be possible14:58
kenvandinethat code path wouldn't be hit after the check finishes14:58
jibelkenvandine, no errors.u.c says it's the same14:58
kenvandinewith the silo installed?14:59
kenvandinegot a link to the errors page for the crash with the silo?14:59
jibelyes. I'm redoing it to double check14:59
kenvandinejibel, if that's the same crash, that's not what i was looking at while working on this bug :)15:01
kenvandinei was only working on the case where it went into flight mode(or lost network) which the check was being done15:01
jibelkenvandine, I confirmed that with the silo installed, open system-settings/updates, wait until the 3 updates are displayed, then enable flight mode, it crashes15:01
kenvandinei was getting that same crash when it was actively checking for the updates15:02
jibelconfirm*15:02
kenvandineand this branch fixes that15:02
kenvandineso i need another fix for your case15:02
jibelkenvandine, https://errors.ubuntu.com/oops/d23833f6-9517-11e5-b5b7-fa163e22e46715:04
kenvandinejibel, thx15:05
=== balloons is now known as Guest87033
=== Guest87033 is now known as balloons_
Elleojgdx: here's the crash from unity8 running autopilot: http://mikeasoft.com/~mike/_usr_bin_unity8.32011.crash15:14
jgdxElleo, asum15:14
kenvandinejibel, ok, i pushed a fix that i think should fix the case you hit, i need a build to confirm it15:15
jibelkenvandine, I was hoping it would take you more time and finish the week with empty queues of silos for QA ;)15:18
kenvandinehaha :)15:18
kenvandinelets hope you still do... just with this silo landed :)15:18
kenvandinejibel, i'm confident this will fix it :-D15:19
bzoltan_ping cihelp15:53
t1mpcihelp: hello, I got failures in my MR that I think are CI related https://code.launchpad.net/~tpeeters/ubuntu-ui-toolkit/10-subheader/+merge/27871315:53
bzoltan_we have seen the same failure with  more and more MRs15:53
bzoltan_https://jenkins.qa.ubuntu.com/job/ubuntu-sdk-team-ubuntu-ui-toolkit-staging-vivid-i386-autolanding/675/consoleFull15:53
t1mpIt includes errors like this: /tmp/buildd/ubuntu-ui-toolkit-1.3.1688+15.10.20151018.1bzr1733pkg0vivid1268/tests/checkresults.sh: line 53: bc: command not found15:54
psivaabzoltan_: let me take a look15:55
bzoltan_psivaa: thank you15:55
psivaabzoltan_: t1mp: that ^ error 'bc: command not found' appears to be a red herring16:10
psivaawhat's bothering me more is the output after 'Fatal server error:'16:10
psivaaEE) Cannot establish any listening sockets - Make sure an X server isn't already running(EE)16:10
psivaa_XSERVTransSocketUNIXCreateListener: ...SocketCreateListener() failed16:10
psivaa_XSERVTransMakeAllCOTSServerListeners: server already running16:10
t1mppsivaa: what can cause that? The MRs are for changes in UI toolkit, no changes directly related to X16:16
t1mpthe tests pass on armhf16:16
t1mpalso they pass on amd64, at least for my MR16:17
=== elopio_ is now known as elopio
kenvandinejibel, i've confirmed my fix fixes the crash in both cases16:25
psivaat1mp: i can't think of the cause, yet. this is only happening in i386 builders. digging a little more to see if there is any history with it16:31
oSoMoNdavmor2, how is the validation of silo 13 going ?16:35
davmor2oSoMoN: nearly done16:36
t1mppsivaa: okay, thanks16:36
* t1mp gotta go now16:36
davmor2oSoMoN: just finished infact with the download of the next track for tubular bells :)  all present and correct qa granting16:44
davmor2sil2100: did the bot die?16:51
sil2100davmor2: which one?16:51
davmor2sil2100: https://requests.ci-train.ubuntu.com/#/ticket/709 passed about 8 minutes ago hasn't been flagged yet here16:52
sil2100You mean, queuebot? There was a netsplit16:52
sil2100Queuebot did not return yet16:52
davmor2sil2100: ah that would explain it then thanks :)16:53
=== alan_g|lunch is now known as alan_g
popeysil2100, for OTA8.5 - how does it work - just like OTA(N) you pluck the latest rc-proposed and promote after QA?17:05
popeysil2100, or is there some special process which takes OTA8 and adds a few cherry picked items?17:05
sil2100popey: hey! It's a bit different17:06
sil2100popey: we prepare an OTA-8 snapshot in a PPA and cherry pick packages that we need to have in the 8.5 release17:06
sil2100So more like your second idea17:06
popeyok, so it isn't the things that are in rc-proposed now plus fixes, but ota-8 plus fixes?17:06
popeyok, good17:06
rvrSaviq: Approving silo 517:07
popeyCan we assume silo 9 will _not_ be part of that? (media-hub recent landing) ?17:07
sil2100Yes, since we want to make sure that there's only a controlled number of fixes/changes in the image we want to release17:07
sil2100I think it wasn't part of the OTA-8.5 plan17:07
popeyok17:08
jhodappit wasn't indeed17:08
jhodappOTA917:08
sil2100So the music-app changes still need to wait a bit ;)17:08
ahayzen:-)17:09
jibelpopey, 8.5 are all the fixes targeted to OTA9 and tagged 'hotfix'17:25
popeyooh, ta17:25
jibeland fixed of course :)17:26
bzoltan_psivaa:  could you figure out anything about that odd failure?17:45
psivaabzoltan_:  looks like one of the nodes is misbehaving. still figuring out how. i have offlined that node so the runs are passing17:47
psivaahttps://jenkins.qa.ubuntu.com/job/ubuntu-sdk-team-ubuntu-ui-toolkit-staging-vivid-i386-autolanding/17:47
bzoltan_psivaa:  nice, thank you17:50
psivaabzoltan_: i have kicked off a couple of failed ones, it would help if you're able to rebuild any failed ones that you were waiting on.17:51
bzoltan_psivaa:  it is this one - https://code.launchpad.net/~zsombi/ubuntu-ui-toolkit/resolve_qpointer_data/+merge/278766 what I need the most17:51
psivaaright, that one is still in progress after my rebuild. thanks17:52
bzoltan_psivaa:  other ... i see this package is in the proposed pocket https://requests.ci-train.ubuntu.com/#/ticket/711 Is anything holding it or should I expect to land at some point without anything?17:55
psivaabzoltan_: since the particular node is offlined, it should not impact any more tests17:56
bzoltan_psivaa:  cool, thanks. what about that qtcreator-plugin-ubuntu landing?17:57
psivaabzoltan_: sorry, i dont see any autolanding job for qtcreator-plugin-ubuntu?17:58
psivaabzoltan_:  i only see qtcreator-plugin-ubuntu-ci17:58
bzoltan_psivaa:  in this silo https://requests.ci-train.ubuntu.com/#/ticket/71117:59
psivaabzoltan_: we dont deal with ci-train stuff, should be for trainguards i suppose17:59
psivaaapologies for my confusing answer before18:00
bzoltan_psivaa:  ohh, sorry :)18:00
bzoltan_psivaa:  my bad, thank you for your help with the other issue. That was a clear blocker.18:00
psivaabzoltan_: yw :)18:01
=== joc_ is now known as joc|away
=== alan_g is now known as alan_g|EOW
=== psivaa is now known as psivaa-afk
davmor2popey: ^19:04
davmor2ahayzen: ^19:04
ahayzen\o/19:04
popey\o/19:33
popeythanks davmor2 ahayzen19:33
popey\o/ store has updated weather19:36
=== salem_ is now known as _salem
dobeyis fginther around today?21:26
dobeytrainguards: can i get someone to manually ack some autopkgtest failing that has never worked?22:17
robrudobey: what, in proposed? that's a job for #ubuntu-release22:17
dobeyyes22:18
robrudobey: also proposed migration is not supposed to block on something that always failed.22:18
dobeyrobru: i think maybe something is wrong with how that is determined then? this fails at installing the built binaries on some archs, which i guess we didn't have autopkgtests running for until recently22:19
dobeywell, one of the built binaries22:20
robrudobey: dunno then. not my area, sorry22:20
dobeyok22:20
dobeyprobably not the best time to be pinging people either, late on a friday, and with most US people on vacation today22:21

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