/srv/irclogs.ubuntu.com/2014/02/03/#ubuntu-ci-eng.txt

=== Ursinha-afk is now known as Ursinha
=== mandel` is now known as mandel
didrocksogra: coming?09:32
ogradidrocks, "you are not allowed to join this hangout"09:32
ograbeyond that ... yes, i would like to09:32
ogra:)09:32
didrocksogra: use your canonical account09:33
sil2100o_O09:33
didrockssee the email on various random ML09:33
didrocks:p09:33
ogradidrocks, thats what i'm trying, but both browsers seem to force me into my private one09:33
Mirvogra: enjoy the double accounts. logout from the private on in one browser.09:33
Mirvogra: I only use Chromium for this09:33
ograi even closed both completely ...09:34
Mirvogra: there was also some tip to use some url parameter somewhere, but I don't remember what was that..09:34
ogralet me try an explicit login/logout, probably that helps09:34
Mirvogra: so go to plus.google.com and logout from private one09:34
Mirvyes, that should help09:34
ogranope, no go09:35
Mirv:( that's how I've it working09:36
robruogra, if you see a 0 in the end of the url, change it to 1. or vice versa. that will make it choose the right account, but only if you're already logged into both09:36
ograrobru, oh, awesome tip, thanks, will try that the next tiome09:37
ograthat crap made me completely trash both browser sessions :(09:38
ogra(all tabs closed etc)09:38
Mirvoh, you got in09:38
ograChecking 'com.example.lp1245677_lp1245677_0.1.desktop availability ... [OK]09:42
ograChecking aa-exec-click ... [OK]09:42
ograChecking application upstart job ... !FAIL!09:42
ogradidrocks, ^^^^09:42
ograthats the error09:42
=== seb128_ is now known as seb128
=== davmor2_ is now known as davmor2
=== ubot5` is now known as ubot5
=== jasoncwarner__ is now known as jasoncwarner
=== vrruiz_ is now known as rvr
psivaamorning10:20
=== lan3y is now known as Laney
psivaadidrocks: morning. my irc was having issues, not sure if you saw my PM about any reruns. just let me know if you do.10:24
didrockspsivaa: just answered, sil2100 and robru are working on those, just keep the discussion going :)10:27
psivaadidrocks: sil2100: ack :)10:28
=== Ursinha-afk is now known as Ursinha
=== cjwatson_ is now known as cjwatson
=== Ursinha-afk is now known as Ursinha
=== Ursinha is now known as Guest40244
=== Ursinha_ is now known as Guest51202
=== Guest51202 is now known as Ursula
=== Ursula is now known as Ursinha
=== alan_g is now known as alan_g|lunch
=== fginther` is now known as fginther
=== 14WAB1TAR is now known as vila
elopio_cihelp, can you please check if I added all the required information for https://bugs.launchpad.net/ubuntu-ci-services-itself/+bug/1275012 ?14:23
ubot5Launchpad bug 1275012 in Ubuntu CI Services "Add a job to run all the image tests with qt5.2" [Undecided,New]14:23
dholbachhiya14:23
dholbachin the last comment of https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1237045 it says "Package build configuration needs to be changed to use trunk." - do you know who could do that?14:23
ubot5Launchpad bug 1237045 in Ubuntu UI Toolkit "Ubuntu UI Toolkit no longer builds on precise, quantal and raring" [Critical,Confirmed]14:23
cjohnstonelopio_: its a start.. when someone gets to it they will ask for any other needed info14:29
davmor2didrocks: are you looking to promote anything today, if so what image?14:32
dholbachsalut seb128 - comment ça va?14:32
seb128dholbach, salut, ça va ? et toi ?14:33
seb128ça va !14:33
dholbachseb128, in the last comment of https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1237045 it says "Package build configuration needs to be changed to use trunk." - do you know who could do that?14:33
ubot5Launchpad bug 1237045 in Ubuntu UI Toolkit "Ubuntu UI Toolkit no longer builds on precise, quantal and raring" [Critical,Confirmed]14:33
cjohnstondholbach: I'm chasing it from the CI side... I guess freenode is having issues this morning and not everyone on the team is able to get on14:34
dholbachcjohnston, thanks a lot!14:35
didrocksdavmor2: so, latest just built image is a candidate14:47
didrocksdavmor2: tests are running, but dogfooding would be good14:48
davmor2didrocks: 160 right?14:48
didrocksyep :)14:48
davmor2didrocks: no worries14:48
didrocksdavmor2: popey isn't around, so I guess you have double testing? :p14:48
davmor2didrocks: sadtrombone.com14:49
didrocksdavmor2: ahah14:51
kgunnsil2100: so what's the protocol? i see you want a new "landing line" do i need to copy paste entries down to the latest row in the spreadsheet?14:52
sil2100kgunn: the system doesn't allow re-using existing rows, so re-writing is the way I guess - you an copy paste but don't copy-paste the whole row ;)14:53
kgunnsil2100: ok, thanks...will do14:58
=== alan_g|lunch is now known as alan_g
sil2100kgunn: the slot for mir will soon be free \o/ But today we're blocked since we're waiting for the fixed-image test results sadly15:03
kgunnsil2100: ok, but we can at least build right ?...to know we're all ready to go when the pipe is unblocked15:04
sil2100kgunn: right, we could do that, but sadly... there's still platform-api in the silo (the fix for dbus-cpp) - it's ready to be released, but until it's not released we cannot assign a silo for mir since mir also requires platform-api to land (with the ABI bump)15:13
sil2100kgunn: while CITrain does not allow 2 silos using the same project ;(15:14
sil2100kgunn: (as it might lead to conflicts)15:14
didrocksthis is to avoid people doing useless work15:14
didrocksas if you land A, you have basically to restart B15:14
sil2100Right'o15:14
seb128dholbach, sorry, I don't know, ask Mirv I guess15:14
didrocks(if A and B are using the same components)15:14
Mirvdholbach: check with bzoltan, he was AFAIK handling the precise earlier.15:16
Mirvit sounds like a one time newer build would do, based on satoris' comment15:17
Laneycihelp hi, can we have ubuntu-system-settings-autopilot run for touch smoke testing?15:19
kgunnsil2100: ok...but once ricmm & tvoss get their stuff landed (assuming that's today...) then we could build....or are there any other caveats ?15:20
cjohnstonLaney: I assume your asking for new testing?15:20
Laneycjohnston: like on this list http://ci.ubuntu.com/smokeng/trusty/touch/mako/159:20140203:20140115.1/6404/15:20
Laneywe want to add the system-settings tests there15:20
cjohnstonLaney: file a bug against ubuntu-ci-services-itself please15:21
Laneyah okay15:21
cjohnstonLaney: I assume things work and pass and such15:21
Laneywell, they work for us ...15:21
Laneybut I'd be more comfortable if someone else could do a manual run first in the same way that CI will15:21
cjohnstonfamous last words :-)15:22
dholbachcool15:22
Laneyhttps://bugs.launchpad.net/ubuntu-ci-services-itself/+bug/127578615:23
ubot5Launchpad bug 1275786 in Ubuntu CI Services "Please add ubuntu-system-settings-autopilot to touch smoke testing" [Undecided,New]15:23
kgunnsil2100: didrocks...yes of course guys...i'm just trying to figure out when i can get an eta on building...15:23
didrockskgunn: so the plan is:15:24
kgunnas the last build revealed an issue _only_ on calxeda15:24
didrocks- getting the test results from the current image15:24
didrocksto know if we can promote it or not15:24
didrocks(it's a revert)15:24
didrocksif green -> we can restart publishing things15:24
didrocksincluding this platform-api15:24
didrocks(probably tomorrow morning EU time, running the tests will still takes 5 hours)15:24
didrocksthen, we give you a slot ASAP15:25
kgunnwow15:25
davmor2didrocks, seb128: hmm I think I just rediscovered an old issue if you have your device on 3g and wifi is disabled if you check fo an update it does and it autodownloads it even though the setting say on on wifi15:26
didrockscyphermox_: mind looking? ^15:26
didrocksah sorry15:26
didrocksread to fast15:26
cyphermox_ack15:26
didrocksI think it's system-image15:26
didrocksbut system-image didn't change15:26
didrocksso either a signal in nm15:26
didrocksor in system-image15:26
seb128davmor2, seems like an issue for barry/system-image15:26
seb128we only use their dbus apis15:27
cyphermox_did system-image ever look  for the right signals?15:27
davmor2seb128: thanks15:27
davmor2cyphermox_: that's what I say I think it is an old issue that I just rediscovered15:27
davmor2cyphermox_: I'm too used to having my phone connected to my wifi15:28
didrocksok, so it should be the case for a long time already15:28
davmor2didrocks: indeed15:28
dholbachMirv, cool15:28
davmor2didrocks: this is definitely not a new issue I think I first saw it doing 3g testing on saucy15:29
sil2100NOT REGRESSION, *ignore*15:31
didrocksdavmor2: ok, keep us posted with any further infos15:32
davmor2didrocks: can you confirm something quickly please. Install a click app, once it is installed it seems to no longer show the open/uninstall option and instead leaps to the first app in the list (for me bbc sport)15:36
didrocks(checking on older and latest image)15:39
=== Ursinha_ is now known as Ursinha
didrocksdavmor2: ok… so…15:47
didrockswe looked at that with robru15:47
didrocksno issue on 15715:47
didrocksit's an issue on 16015:47
didrockshowever15:47
didrocks158 and 159 are empty15:47
didrocks160 only contains the revert of upstart-app-launch15:48
didrockswhich maybe, create that side-issue15:48
didrocksrobru is flashing 159 right now15:48
didrockslet's see15:48
davmor2didrocks: I'll leave it with you then let me know if you need a bug for it15:48
didrocksdavmor2: please create a bug anyway15:49
asacdidrocks: webbrowser green \o/ :) ... lets hope it continues that way15:52
didrocksasac: well, read the last 15 lines ^ :(15:53
asacdavmor2: in case you check 160, maybe see if this just went away15:54
asac(as didrocks said it might be side effect)15:54
didrocksasac: he's on 16015:54
asacdidrocks: sure, but at least we are closer to green again :)15:54
asacah15:54
davmor2asac: that is on 16015:54
asacdavmor2: regression from 159?15:54
didrocksasac: robru tested 15915:55
asacand?15:55
didrocksand same issue15:55
asacic15:55
asacdavmor2: is that something you do on every build?15:56
didrocksit's weird15:56
didrocks157 didn't get it15:56
asace.g. part of manual smoke tests?15:56
davmor2asac: it wasn't there on Wednesday which wa the last test I did15:56
davmor2asac: and yes15:56
ogra_heisenbugs all over the place15:57
didrocksso, let me sum up15:57
didrocks157 didn't get any issue15:57
didrocksbut it seems to flip back and force15:57
didrocksbetween the 1st and the click package just installed15:57
didrocksbut it doesn't end up that way15:58
didrocksthen, 158 has 0 change from 15715:58
didrocks159 has 0 change from 15915:58
didrocks158*15:58
didrocksand robru reproduced the issue on 159…15:58
didrocksand it works for me15:59
didrocksjust upgraded to 16015:59
didrocksbut I can see a ping pong between the 1st content15:59
didrocksand the newly installed one15:59
didrocks(I see the screen with the 1st content for half a second)15:59
=== xnox_ is now known as xnox
asachow does 155 look like?16:02
didrockswell, I didn't get it here16:02
didrocksI have a weird behavior16:03
didrocksbut not that one16:03
didrocksit just seems racy, but the question is, why robru gets it 100% of the time and I get the other behavior 100% of the time16:03
davmor2asac, didrocks: I wonder if it is just a race issue?16:03
didrocks(wonder if it has anything related to the network)16:03
asacdidrocks: what device is robru using?16:04
didrocksasac: same, mako16:04
asacmaybe check the emulator... if its racy that one will surely trigger the opposite of what folks want to see :)16:04
davmor2asac: and I'm looking at maguro and mako same issue on both16:05
asacdidrocks: maybe you have user data?16:05
asacthat changes runtime behaviour16:05
didrocksasac: no, fresh one, for both16:05
didrocksasac: well, give us a nice way to test on the emulator, not possible yet16:05
asacit is16:05
asacpeople showed it to me16:05
didrocksfor click installation?16:05
asacnot sure... havent tried that16:06
asacbut it surely was better to use than what i remember16:06
didrocksasac: can we get the QA team to help us?16:06
* didrocks feels really only to just fight on all those regressions16:06
ogra_bug 1274215 btw16:07
ubot5bug 1274215 in ubuntu-download-manager (Ubuntu) ""always download on WLAN" in the update manager does not work" [Undecided,New] https://launchpad.net/bugs/127421516:07
ogra_oops, wrong channel16:07
davmor2ogra_: thanks16:07
ogra_heh, pasted it in the right channel too :)16:08
didrocksok, they still get it, I still get the other behavior16:08
ogra_davmor2, i also filed bug 1273655 the same day16:09
ubot5bug 1273655 in system-image (Ubuntu) "system-image should propagate network errors to the UI " [Undecided,New] https://launchpad.net/bugs/127365516:09
asacdidrocks: davmor2 is from the qa team16:10
asac:)16:10
asacdidrocks: also you have mirv, sil, psivaa and plars16:10
asacand can surely ask ogra and popey also for help16:10
asacif you give them an image number to test16:10
* ogra_ only has a maguro for propsed testing atm16:10
asaci am sure you can convince 6 folks to test each one image number (after testing the current image to see)16:10
asacogra_: doesnt matter16:11
asacogra_: that thing has the problem as well16:11
asacogra_: its not proposed testing16:11
asacogra_: its bisecting and fixing16:11
asac(and yes, i know you still dont want to wipe your n4)16:11
didrocksasac: well, we have different results between people on the same image16:11
didrocksthis is THE issue16:11
ogra_asac, yeah, dogfooding ftw :)16:12
asacdidrocks: then get a second opinion16:12
asacdidrocks: often one of the tester doesnt get it right16:12
ogra_asac, auto-download never worked for me on any device though16:12
didrocksasac: we have 4 different people testing16:12
ogra_which is why i filed the two above bugs alst week after talking to mandel and barry16:12
ogra_*last16:12
didrocksso not sure what you are asking on the emulator16:12
asacdidrocks: first all have to see the same issue on the current build16:13
didrocksasac: yeah, and it's not the case, and that's what I'm trying to figure out16:13
asacdidrocks: everyuone who doesnt see it gets booted from the testers and gets a negative score for the next bonus payment :)16:13
asaclol16:13
didrocksasac: it's me who doesn't get it :p16:13
* mandel reads backlog16:13
didrocksI see the flipping though16:13
davmor2asac: and if I had 48 hour long days I might be able to fit it in, unfortunately between testing the new image on 2 devices, then I need to run the latest qt 5.2 branch and then I need to ....... :)16:13
didrocksit flips to the first item16:13
didrocksbut then, to the right one16:13
didrocksasac: yeah, I don't really "have" davmor2, psivaa, plars or popey to help where bad things happen :p16:14
ogra_davmor2, asac could change your contract to have 48h days :)16:14
didrocksyeah, let's asac fix this16:14
asacdavmor2: not sure about qt5.216:14
asactell your boss you cant do that16:14
asacand someone else should take that part16:14
ogra_you just need to find yourself how to squeeze them into a 7 day week then :)16:14
mandelogra_, I remember that bug, I'll ping barry to see what is going on, AFAIK in my integration tests I do send the signal16:14
asacdavmor2: is it close to be ready for production? otherwise, the sdk folks shoudl be able to test on their own?16:15
davmor2ogra_: living on Neptune should help16:15
ogra_mandel, note that i'mm missing the majority of my backlog (and the start of this conversation), didrocks or davmor2 can probably explain in more detail16:15
asacogra_: mandel: we know the app store is broken16:16
asacactually16:16
ogra_ah16:16
ogra_i thought it was about updates16:16
didrocksok, Laney got it as well16:16
asacok i bring in ralsina16:17
asacand his best guy16:17
asacthey are doing thsi click thing16:17
asacand if they see it they can probably help16:17
asaci feel there are multiple issues overlayed16:17
didrocksyeah16:17
mandelasac, oh.. ok, nevertheless if there is anything I can help with let me know, I'm always the first one to be blamed when things do not get downloaded and installed :)16:17
asacand depending on whether you are in france and have a bad ping or are in the UK and have a decent pipe16:17
asacyou see different things16:17
ralsina_asac: I'm here, alecu will be here shortly16:17
ogra_do we have logs uploaded somewhere ?16:17
asacmandel: you can also help16:17
asacralsina_: so the click store in latest image behaves badly16:18
asacralsina_: we tried to bisect when it started, but founid that it behaved super weird the days before as well..16:18
asacand everyone sees slightly weird - but all bogus - behaviour16:18
davmor2ogra_: I haven't16:18
ralsina_asac: strange, we have not landed anything there in a couple of weeks16:18
asacdidrocks: can you describe to ralsina_ what we see?16:18
davmor2I can write a bug up though and upload some16:18
didrockssure, so:16:18
didrockslatest image (we're still bisecting)16:18
asacralsina_: right. i think something is bogus somewhere, so you might help us figuring what changed to cause this racy thing etc.16:18
didrocks1. install a click app16:19
didrocks(not the 1st item)16:19
alecuhi16:19
didrocks2. once the click app is installed, the UI flip back to the 1st element16:19
asacalecu: hi.. .read the 20 lines above or so :) ... didrocks is summarizing16:19
didrocks(bbc one app)16:19
didrockseveryone reproduces it16:19
didrocksbut me16:19
didrockswhere, on the same hw, I have a different behavior16:19
didrocks(which seems related)16:20
didrocksI get:16:20
didrocks1. install a click app (not the 1st item)16:20
didrocks2. once the click app installed, I see the UI flipping to the 1st element (bbc one app)16:20
didrocksbut then, I see the UI flipping back to the just installed app16:20
didrocksok, robru can't reproduce it in last promoted image16:21
* ogra_ cant either on mako 15316:23
ralsina_grrr I am device-less today. That does sounds like we are getting signals from download-manager out of order or something just as weird16:23
=== balloons is now known as Guest54791
didrocksso, robru is installating 154 (+ libunity-mir1 to have the UI starting)16:23
didrocksand sil2100 is installing 15516:23
didrocksSaviq: you probably want to be in, unity8 changed ^16:23
didrocks(in 154)16:23
mandelralsina_, well, udm has not been updated in the image for a while...16:24
didrockscan be as well unity-scopes, but I doubt it16:24
davmor2didrocks, asac: I couldn't on 151 (after figuring out the image he tested on Wednesday) and since then I was on holiday so haven't need to install any apps :)16:24
mandelralsina_, didrocks I'd be very surprised it is udm..16:24
didrocksmandel: I don't think it is, we just need all available help in debugging16:24
ogra_mandel, must be udm ... because it is easy to blame :P16:24
mandelogra_, hahahaha16:24
mandeldidrocks, ok, can you send me a video, I don't have a device.. or can it be reproduced in the emulator???16:25
didrocksmandel: can you try the emulator, I couldn't yet test it16:25
mandeldidrocks, sure thing.. lets hope is not too slow :)16:25
didrocksmandel: yeah, at least, it will help to understand if it's CPU related or anything…16:26
* ogra_ tests on a nexus7 android 4.4 cdimage install 16:27
ogra_(since thats what i have around atm)16:27
alecudidrocks, asac: has anything changed in the "network status" detection code?16:28
alecuthat's what the click scope uses to decide when to refresh16:29
didrockscyphermox_: maybe you can sync up with alecu ^16:29
ogra_hmm, intresting ... on the above (unsupported) setup it failed completely, just dropping me back to the dash16:29
ogra_after downloading16:29
didrocksso, 155, same issue16:30
didrockslet's wait for 15416:30
ogra_aha, this time it worked and dropped me to "MyMaps" after neing done16:30
ogra_*being16:30
ogra_(which is the first app in the store for me)16:30
asacsorry guys, i have to go on another fire :/16:31
asacyou guys will sort it ... everyone is now here16:31
didrocksasac: not the unity8 team, which I think will be needed16:31
asacralsina_: doesnt this work on the emulator?16:31
didrockskgunn: can we get someone from unity8 around?16:31
asackgunn: ^^ can you bring someone in?16:31
ralsina_asac: AFAIK it does work on the emulator, I only tried on my tablet before because it's faster16:32
mandeldidrocks, the emulator is not working for me at the moment.. boots to a terminal and shows nothing in the ui, I'm guessing it is the issue with the dropped lib (cannot record the name but was related to mir)16:34
didrocksso 154 is bad16:34
didrocksmandel: try to install in it libunity-mir116:34
didrocksto check if it's that one16:34
asacmandel: you need latest build afaik16:35
asacnot something old16:35
* didrocks tries to revert unity8 with robru16:35
mandeldidrocks, aha, that is probably the issue16:35
mandelasac, ack16:35
didrocksok16:40
didrocksunity816:40
didrocksmandel: ^16:40
didrocksthat's the cause of the issue16:40
didrocksthe "flush trunk"16:40
didrockssad we don't have from unity8 around, let's revert it then16:40
didrocksdavmor2: did you open a bug?16:40
mandel:-/16:41
davmor2didrocks: not yet I asked if someone wanted me too,  any idea what to file against?  I'm assuming a scope/unity816:42
didrocksdavmor2: so, yeah, it's unity816:42
didrocksasac: FYI ^16:42
davmor2didrocks: I'll file it now then16:42
* didrocks prepares the second revert from the day16:42
ralsina_didrocks: so, we don't know exactly what change breaks it, it's just something in a large revision of unity8?16:44
didrockscihelp: anything happened to q-jenkins?16:44
didrocksthe tests are not running anymore16:44
didrockson latest image16:44
plarsdidrocks: yes16:45
didrocksralsina_: yeah, we have ~20 mp bundled16:45
ralsina_ouch16:45
didrocksplars: can you continue the runs?16:45
plarsdidrocks: there was a problem with the storage, it will need to be restarted16:45
didrocksplars: can we continue them, to not rerun the first ones?16:45
didrocksplars: I would like to have as much info as possible while we continue16:45
plarsdidrocks: I'm not sure it's worth it, doesn't look like we made it that far into 160 when problems started16:47
plarsdidrocks: it would be much easier, in this case, to restart16:47
didrocksplars: well… we won't get any useful results then before the next image is published16:48
didrocksplars: so, better to juts get as much as infos16:48
didrocksotherwise, we will have nothing and if yet-another-regression happens…16:49
plarsdidrocks: I can start running things one at a time where it left off - this could leave things in a weird place when we get a new image though and require some manual actions16:49
davmor2didrocks: https://bugs.launchpad.net/unity8/+bug/127583216:50
ubot5Launchpad bug 1275832 in Unity 8 "REGRESSION: Installing apps open the first click app once installed rather than their own app" [Undecided,New]16:50
didrocksplars: hum, why, the whole system won't restart everything?16:50
plarsdidrocks: or I can restart and we'll get a full run - if a new image comes in the next 3-4 hours, it'll be delayed starting until we either end, or cancel the job on 16016:50
didrocksplars: ah, it will be delayed…16:50
didrocksplars: so yeah, please rerun everything16:50
plarsdidrocks: how long do you expect until 161?16:50
plarsdidrocks: ack16:50
didrocksplars: time to publish unity816:51
didrocksthanks davmor216:51
didrockssecond revert of the day proceeding16:51
didrocksthostr_1: FYI, so slots are delayed ^16:52
didrocks* revert uploaded *16:57
=== balloons is now known as Guest51231
davmor2didrocks: does that make continuing to test r160 pointless?17:00
didrocksplars: coming?17:01
didrocksdavmor2: if you can get any new fixes17:01
didrockssorry17:01
didrocksregression detected17:01
plarsdidrocks: I'm there17:01
davmor2didrocks: I'll carry on then :)17:01
plarsdidrocks: I don't see you though17:01
davmor2didrocks: hmmm the main apps, dialer/messaging/settings etc seem to swap places from the top to the bottom of the screen I think this might be related to if the network is detected before or after the device is unlocked17:07
=== Ursinha is now known as Guest4888
=== Guest4888 is now known as Ursula
davmor2didrocks: hmm it looks like it reverts if you open one of them17:09
didrocksdavmor2: interesting…17:19
asacdidrocks: ok, came off that fire/prep call ... you have all the folks helping you need?17:21
asacwill be on next call in 10 :)17:21
didrocksasac: well, we did found the regression ourself, it's unity817:22
didrocksfound the exact version17:22
asacok so answer is yes17:22
didrocksnobody was around, so reverted17:22
asac:P17:22
asacsure17:22
asacdidrocks: whole unity8 landing or what got backed out?17:22
asace.g. was that the trunk flush?17:22
didrocksasac: it was the trunk flush, maybe more than 30mps17:23
didrocksso whole unity817:23
ogra_two revisions actually17:23
didrocksogra_: you mean, 2 versions17:23
ogra_well, from an archive POV two package revision (yes, or versions)17:24
sil2100ogra_: you're inserting confusion!17:24
=== Ursula is now known as Ursinha
ogra_i know, i know ... my brain is still way to much into debs :P17:24
ogra_sil2100, yeah. i'm good at that :)17:25
ogra_(sorry)17:25
sil2100;p17:25
davmor2sil2100: turns out it's not just google plus with the links issue so I'm going to assume it is the browser :(17:44
ogra_links issue ?17:45
* sil2100 sighs17:47
davmor2ogra_: https://bugs.launchpad.net/webbrowser-app/+bug/127585517:48
ubot5Launchpad bug 1275855 in webbrowser-app "REGRESSION: non direct links nolonger open" [Undecided,New]17:48
davmor2didrocks: https://bugs.launchpad.net/webbrowser-app/+bug/127585517:49
ogra_davmor2, G+ doesnt use the browser properly17:49
ogra_it is heavily broken17:49
ogra_dont use it for testing such stuff17:49
didrocksis there anything new?17:49
didrocksrelated to webbrowser-app17:49
davmor2ogra_: no if you open google plus from the browser it does it as does facebook and twitter17:49
ogra_(G+ uses the web view directly from QML, that cant open external links at all)17:50
ogra_oh17:50
ogra_yeah, if you dont use the app thats indeed different17:50
davmor2ogra_: I noticed it at the weekend on the app so tried the browser direct same thing17:51
ogra_dont test with the G+ app though17:51
didrocksdavmor2: can you try on latest promoted image?17:51
* ogra_ tries a webapp that he knows works for external links17:51
ogra_davmor2, yeah, doesnt work here either ... it highlights the link when tapping but doesnt open anything17:53
didrocksogra_: latest promoted image?17:53
ogra_didrocks, no, my mako is dead atm ... need to charge first17:54
ogra_didrocks, fridays cdimage install on the flo (new N7)17:54
didrocks ogra_: which version? 155?17:55
ogra_didrocks, fridays cdimage zip17:56
ogra_one sec, let me find out what system-image that corelates to17:56
didrocksthanks ;)17:56
ogra_15617:57
didrocksok, so, we need to test on 15317:57
sil2100davmor2: what app can I use to test this?17:57
* ogra_ cant really downgrade flo ... 17:57
sil2100davmor2: one that doesn't require some specific account17:57
ogra_sil2100, the BBC apps from popey should work17:58
ogra_then click something that links externally in an article17:58
* ogra_ just tesed with his heise.de app, but thats all german ... you dont want that :)17:58
ogra_davmor2, try to "tail -f" the log of the app you are running to see if it prints something17:59
ogra_(while tapping the link)17:59
didrocksworked for sil2100 in 155 though17:59
ogra_hmm, works here now too ...18:00
ogra_it just opened zdnet in the browser from a link18:00
didrocksurgh18:00
ogra_but it doesnt seem to work all the time for all links18:00
ogra_aha18:00
sil2100davmor2: ^18:00
ogra_Invalid message received: {"event":"newtab","url":"http://www.zdnet.com/more-on-microsofts-sku-morphic-windows-vision-7000024092/"}18:00
didrocksyeah… "fun"18:00
ogra_thats what i see in the log18:00
ogra_for the *working* link18:01
didrocksbut it work to create a new tab for others, right?18:01
ogra_the non working ones dont print anything at all18:01
didrockshum18:01
ogra_it fired up the browser with a new tab showing the zdnet article18:01
ogra_so ... something thats more reproducable for non germans:18:03
ogra_BBC news app18:03
ogra_clicking the artictle abour woody allen18:03
ogra_there is a link to an open letter18:03
ogra_tapping it opens the browser with the new york times open lette for me18:03
=== alan_g is now known as alan_g|EOD
ogra_no issues in the log when it opens the browser18:05
didrocksogra_: opened letter?18:06
didrocksopen*18:06
* didrocks doesn't find18:06
ogra_didrocks, scroll down ...18:06
ogra_there is a woody allen article with a pic of him18:06
didrocksah ok18:06
didrocksfirst paragraph18:06
didrocksworks here18:06
ogra_here too18:06
didrocksso, it's "random"18:07
didrocksI'm on r16018:07
didrocksogra_: mind updating the bug report?18:07
ogra_didrocks, added18:10
didrocksthanks18:11
ogra_hmm, in the nelson mandela article from the top there is a link to his "40 page will" ... that opens only a blank page18:12
ogra_ah, that prints alsa errors in the log ... ignore it ... its specific to the flo installl18:13
ogra_(seems it wants to play sound, thats not supported on that HW yet)18:14
=== Guest51231 is now known as balloons
davmor2didrocks, ogra_, sil2100: so is this just how the link is coded then I wonder?18:20
ogra_probably18:20
ogra_i wonder if the links you click are opened by a javascript function or so18:20
davmor2ogra_: well they used to work, they certainly did on tuesday on the way down in the train18:22
* ogra_ blames british trains then :P18:23
ogra_webbrowser-app (0.23+14.04.20140122-0ubuntu1) trusty; urgency=low18:25
ogra_  [ Ted Gould ]18:25
ogra_  * Add URL Dispatcher configuration file.18:25
ogra_hmm18:25
asacdavmor2: how did you find https://bugs.launchpad.net/webbrowser-app/+bug/1275855 ?18:25
ubot5Launchpad bug 1275855 in webbrowser-app "REGRESSION: non direct links nolonger open" [Undecided,New]18:25
ogra_thats in the last webbrowser-app18:25
asacdavmor2: did you find that as part of our smoke tests that we have agreed to be run on every build? or as part of exploratory testing?18:25
ogra_i wonder if url-dispatcher gets in your way18:25
asaceverything tedg uploads is causing issues by definition18:26
asacback it out18:26
asacsee if it fixes it18:26
davmor2asac: mix of dogfooding the device daily and noticing it stopped working and then digging into it18:26
didrocksasac: who is going to back it out? I have nobody left around me18:27
asacdavmor2: its an operational nightmare to treat things that you dont run daily as blocking our promotion ... unless its a clear CRITICAL issue18:27
asacdavmor2: so would you have hit this problem every day?18:27
asacdavmor2: do we know it wasnt in 158?18:27
asacfor instance?18:27
asac:)18:27
asacdidrocks: i am not saying we should back itout18:27
ogra_asac, well, we know it worked once18:27
ogra_so it is a regression18:27
asacogra_: doesnt help our daily effort18:27
asacogra_: sure, but its not critical if its not discovered early18:28
asacwe cant hold the line for stuff that maybe happened 2-3-4 weeks ago18:28
ogra_i would call it a "normal" severity regression bug18:28
asacright18:28
asacogra_: was it in our last promoted image?18:28
davmor2asac: so it's something I use everyday not in the confides of testing, but being on holiday has slowed my usage down on thursday and friday18:28
ogra_it needs to be noted down and upstream should try to solve it asap18:29
ogra_but we shouldnt block18:29
asacright18:29
asacthats what i was trying to say18:29
asacits a bad upstreawm bug18:29
didrockswell anyway, an image is kicked18:29
asacbut nothing we as the CI team should throw our scarce resources in bisecting in18:29
didrockswe'll get results in 8 hours now18:29
davmor2asac: I leave the blocking to didrocks I just let him know what I hit18:29
asacright18:30
asacdavmor2: just saying we need to categorize tests somewhat18:30
didrocksdavmor2: but we always need to know if it was in latest promoted image first18:30
ogra_asac, well, it is QA team work ... and davmor2 is QA18:30
didrocks(which I asked I guess)18:30
ogra_the CI team might just be the wrong address for the bisecting18:30
asaci am just thinking out loud18:30
asacogra_: right. unless its something that slipped through CI, its a normal bug. reporter can help if he narrows down finding18:31
asacetc.18:31
asacbyu bisecting on his own18:31
asacotherwise its just one thing that will sit and wait for someone to prioritize :)18:31
asacand regression tests should be better prioritized than wishlists etc.18:31
davmor2asac, didrocks, ogra_:  so I can tell you it worked up till wednesday last week which I think was 151 and wasn't last night. When it broke exactly between those I have no idea. I can dig into it but not today and I'm at a funeral tomorrow18:33
asacdavmor2: we should work on ensuring that your work doesnt get dropped if you are off18:33
asaci think that means we need someoen else also doing what you do every day18:33
asacas calibrating for short term leaves just doesnt work18:33
asacdavmor2: but dont worry... think its a jfunk thing :)... you might want to raise it18:34
davmor2asac: I think the biggest thing is the fact that I use the devices as daily drivers not just for testing.18:34
ogra_the last change in webbrowser-app was on image 14418:35
asacdavmor2: daily drivers? you mean dogfooding?18:35
davmor2asac: I have no other phone, I only have a xoom for tablet. I use my n4 now was maguro as my daily phone and I use the n7 downstairs and the xoom upstairs xoom is running android18:37
davmor2asac: it basically means that I'm using devices for 4-6 hours each day instead of just testing them for an hour18:38
=== fginther changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: fginther | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: -
didrocksok, so if we don't know and nobody tries on the latest promoted image, let's call it a non issue18:42
ChrisTownsendfginther: Hi, I have some questions about the Jenkins AP machines.  Are you the right person to ask for this?18:50
fgintherChrisTownsend, go ahead18:50
ChrisTownsendfginther: Well, it may be easier for me to send an email because there are a few.  Then if there are questions, then maybe we can chat about it.18:51
ChrisTownsendfginther: But I wanted to make sure you could be a point person for it.18:51
fgintherChrisTownsend, that works.18:51
ChrisTownsendfginther: Ok, thanks.  I'll send one in a bit.18:51
* didrocks really doesn't understand why he has to do QA work18:53
davmor2didrocks: how about we call it a regression and not a blocker :)18:53
didrocksdavmor2: well, being able to know if it was introduced since last promoted image will help knowing that18:54
didrocksok, can reproduce on latest image18:54
didrocksnow, reverting to latest promoted one18:54
davmor2didrocks: yeap I was about to look at that18:54
didrocksit's only the 13th time of the day I'm flashing it to do QA and testing…18:54
didrocks(bisecting included)18:54
davmor2didrocks: :(18:58
ogra_didrocks, your laszt image build failed19:03
didrocks"nice"19:03
didrocksogra_: you get traces?19:03
ogra_i got a failure mail19:03
ogra_didrocks, http://people.canonical.com/~ubuntu-archive/livefs-build-logs/trusty/ubuntu-touch/20140203.2/livecd-armhf.out19:04
ogra_The following packages have unmet dependencies:19:04
ogra_ ubuntu-touch : Depends: ubuntu-sdk-libs but it is not going to be installed19:04
ogra_E: Unable to correct problems, you have held broken packages.19:04
ogra_WOW !19:04
ogra_how did that go past britney ?19:04
didrockswhat was uploaded?19:05
didrocksI don't find anything related to that19:05
didrocksdavmor2: ok, so the bug was already in the latest promoted image19:06
didrocksnow the image build failure19:06
didrocksogra_: any idea?19:06
ogra_s/pritney/proposed-migration/19:06
davmor2didrocks: mine is still flashing fscking maguro so slow at updating19:06
ogra_didrocks, gimme a sec (sorry, had to care for the cats)19:07
ogra_didrocks, that smells like a seed issue ... i.e. someone changing seeds without uploading meta or some suchg19:08
* didrocks bzr update19:08
didrocksogra_: sdk-libs-dev installs cmake19:09
ogra_didrocks, yes, but it does so since xnox made the change on the 22nd19:09
didrocksyeah19:09
ogra_cant be it19:09
didrocksso no19:09
ogra_lets ask in -release19:10
davmor2ogra_: britney was too busy singing19:11
veebersdoanac, fginther: ping19:54
fgintherveebers, howdy19:54
veebersfginther: Hi! How are you ?19:54
doanacveebers: hey19:55
veebersdoanac: hey o/19:55
veebersdoanac, fginther: query: what would be required to take this job I pilfered off doanac (http://q-jenkins:8080/job/autopilot-release-gatekeeper/) and add the steps that are here: http://pastebin.ubuntu.com/6826187/ (lines 9-27)19:55
doanacveebers: so you want to inject that pastebin logic somewhere into the process?19:58
fgintherveebers, looks like you're wanting an interface to remove a set of packages?19:58
fgintherveebers, is this specifically for qt 5.2 testing?19:59
fgintherveebers, oh, I see there is a bit more to it then removing packages, there is that "rm"20:00
doanacfginther: i think we scared him away :)20:02
fgintherdoanac, it's the whole time travel thing, we have to wait a day for him to see our questions :-)20:03
fgintherdoanac, FTR, I've been asked to do special setups for testing of a planned QT bump, so if this request is specifically for QT5.2, it'll probably come up again.20:09
veebersdoanac, fginther sorry was distracted. I would like to add the logic of lines 9-27 to the job (probably a cloned job) as I suspect the ppa and package install stuff is already handled20:09
doanacfginther: k. i'm not sure about how to handle it. i think there are probably 2 or 3 ways:20:09
veebersYeah, it's specifically for qt5.220:09
doanac1) don't call run-smoke directly in the job. instead call the things it does, provision.sh and run-autopilot-test.sh20:10
doanacwith this custom stuff somewhere in the middle of them20:10
doanac2) fork off run-smoke for this and basically do what i just described in 120:10
doanac3) add some sort of "hooks" logic to run-smoke to handle something like this20:11
fgintherdoanac, I was thinking along those lines as well, particularly 3...20:11
fgintherhmm20:11
doanacveebers: so i'd probably vote 1 or 3.20:11
veebershmm, 1 seems quicker :-) but 3 seems more complete, especially if that functionality will be needed for other things20:12
doanaci don't know how often such specific things might come up. but doing 3 might not be too bad. it would be in python-code (run-smoke). so adding args with custom actions might be fairly easy20:15
doanaci could see it being easier than 1, once you get your hands dirty on the problem20:15
fgintherdoanac, would you use a bzr branch as the source for the hooks file (which then becomes a jenkins job parameter)?20:15
doanacfginther: i'd probably make the parameter to the script just be a path (relative or absolute). then the jenkins-job itself can deal with how it wants to get the hook20:17
fgintherthen you just need to define the hook points (A* run before provision, B* run after provision, etc)20:17
doanacyeah. something like --post-provision-hook=./foo.sh20:17
fgintherdoanac, right, makes more sense20:17
fgintherthe dir thing20:18
doanacit could be an "append" arg in argpase so you could append multiple hooks20:18
fgintherdoanac, there might be an easier way to do that...20:18
fgintherpbuilder runs hooks based on name only, you don't need to explicitly pass any on the CLI.  For example, during the build phase, it will run an B* hooks it finds in the hook directory (in sorted order)20:20
fgintherso you just need to pass in the hook directory and the hooks names will take care of when they are executed20:20
=== josepht__ is now known as josepht
mterryfginther, heyo!  If/When you have time, I'm available to talk otto shop20:24
doanacfginther: that would work also. i'm fine with that appoach.20:25
doanacveebers: is this something you are okay patching. or were  you hoping we'd add that function for you?20:26
doanacveebers: either answer is "safe" :)20:26
fginthermterry, I'll be ready in a few minutes. Can you look at http://s-jenkins.ubuntu-ci:8080/job/autopilot-testrunner-otto-trusty-debug-fjg/16/ and confirm it is the same issue?20:27
veebersdoanac: I was hoping that you brilliant examples of men would do that work for me :-)20:27
doanacveebers: i've got some big MPs going in this week near this code, so its probably not a bad idea for me to give it a stab20:28
doanacveebers: we are targetting friday for an update to the daily-image testing scripts. is that too far out for you?20:29
mterryfginther, looks like it20:29
veebersdoanac: awesome :_) Um, sounds great, I haven't been given a deadline, I'll let those involved know that's the timing we haev20:31
doanacveebers: ack.20:32
doanacfginther: i'll circle back with you on the best way for me to do this hooks logic. probably tomorrow20:32
fgintherdoanac, works for me20:33
=== kgunn is now known as Guest438
fgintherkenvandine, cyphermox_, is daily release still in manual mode?20:56
slangasekhmm, how do I map 's-jenkins.ubuntu-ci' to an IP?21:38
josephtslangasek: if you're trying to connect to it, it's best to get your VPN/DNS configured21:40
slangasekjosepht: yes, so which DNS server for which VPN do I need to configure?21:40
josephtslangasek: https://wiki.canonical.com/UbuntuEngineering/CI/VPN step 4 should get you going21:41
=== ev_ is now known as ev
=== jasoncwarner__ is now known as jasoncwarner
slangasekjosepht: ta21:42
slangasekah well, it seems I don't actually have perms for the action the email suggested anyway. :) http://s-jenkins.ubuntu-ci:8080/job/autopilot-1.4-ci/11/rebuild21:43
balloonsping fginther, can you ensure this merge builds; it's an old request :-) https://code.launchpad.net/~nskaggs/ubuntu-ui-toolkit/emulator-docs-try3/+merge/19935322:21
balloonsI updated it, just want to make sure it gets picked up. ty22:21
=== fginther changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cihelp | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: -
fgintherballoons, I'm not a member of the sdk team, if it fails, there isn't much I can do22:29
fginthershould probably correct that situation22:29
Saviqfginther, hey, "All nodes of label 'trusty&&amd64' are offline", known?23:27
Saviqhttp://s-jenkins.ubuntu-ci:8080/job/unity-phablet-qmluitests-trusty/23:27
fgintherSaviq, looking23:27
fgintherSaviq, the connection to the VM service is screwed up, let me see if I can fix it without restarting jenkins23:34
Saviqfginther, thanks23:37
fgintherSaviq, thanks for bringing this to my attention. Looks like I'll have to restart jenkins, but I'll need to wait just a little bit for the phablet flash jobs to go through. It should be working again a little later23:39
Saviqfginther, that's fine, thanks23:40

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