/srv/irclogs.ubuntu.com/2014/01/13/#ubuntu-ci-eng.txt

vilasergiusens: (Too bad you're not there :-/) This should be fixed now (known issue, see .... damn it forgot to hit enter while chasing the url :-/ https://wiki.canonical.com/UbuntuEngineering/CI/IncidentLog/2014-01-11-otto-stop-no-name07:03
vilafginther: see above /me hides under a rock07:03
sil2100Morning!08:02
sil2100hm, brb, modem reboot08:02
sil2100Hi guys, do we know if the Terminal App keyboard-not-working bug is being targetted by anyone?08:38
=== jodh` is now known as jodh
sil2100didrocks: morning!09:02
didrockssil2100: good morning!09:02
sil2100didrocks: I just noticed that you're using a strange date for the landing team updates ;)!09:03
didrocksreally?09:03
sil2100didrocks: 09.11.14 and 10.11.14 ?09:03
didrocksah, yeah ;)09:03
didrocksthanks, will fix in the next one :)09:03
sil2100Looks like future to me! ;)09:03
didrocksheh, indeed ;)09:04
=== ev changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: ev | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: -
sil2100didrocks: I'll be right on09:31
didrockssil2100: waiting for you, tell us when you won your battle with hangouts ;)09:32
ogra_xnox, can you a) merge back your autopilot changes to trunk and b) please use a landing spreadsheet antry next time so stuff like http://people.canonical.com/~ogra/touch-image-stats/20140110.1.changes doesnt happen (and we dont have broken images over the weekend)09:44
ogra_(note the gigyntic amount of new packages)09:46
ogra_*gigantic09:46
didrocksogra_: are you going to revert the seed change so that we don't install python3-autopilot on the image?09:51
didrocks(I'm not sure why we have python-autopilot as well installed on the image, we shouldn't)09:52
sil2100didrocks: regarding that platform-api, qtubuntu and unity-mir - it indeed seems like one merge is still not in, so I guess the feature is not ready for release10:59
didrockssil2100: ok, just comment on the landing spreasheet please11:00
didrockson the PENDING as well, it was written ready when it wasn't11:00
ogra_didrocks, it was no seed change (sorry, had some network issues at home) but a dependency change in autopilot11:27
didrocksogra_: yeah, I saw that after the fact. We should look at why this metapackage is installed by default though11:27
ogra_oh. you mean we should drop autopilot-touch altogether ?11:28
=== MacSlow is now known as MacSlow|lunch
ogra_didrocks, http://paste.ubuntu.com/6744101/11:29
ogra_seems we carry it sinca august11:29
ogra_*since11:29
didrocksogra_: in your meeting, can you ask more info from rsalveti? Not sure I'm understanding why exactly11:30
didrocks(I guess we can still have an issue if utah doesn't install it automatically, but that's something we need to check)11:30
ogra_most likely because the package isnt installable even in rw mode i guess11:30
didrocksoh, the udev perm11:30
didrocksIIRC, we add a user11:30
ogra_(there are some that bail on the dpkg hardlinks)11:30
ogra_oh, right and the udev rule11:31
ogra_probably rather that than hardlinks11:31
didrocksok, we know why, still that one to figure out then :)11:31
ogra_will carry it to the meeting (but leave it seeded for now)11:31
didrockssounds legit, thanks ogra_!11:31
ogra_didrocks, do you want to revert xnox' dependency change (i think if you just build from trunk it will be reverted)11:33
didrocksogra_: no, let's wait for xnox to backport his change to trunk rather11:34
ogra_ok, so we keep all these python packages in the image11:34
didrocksogra_: I don't have opinions and we'll need to drop all this from the image anyway (all autopilot)11:34
didrocksyeah, that makes "this issue" even more visible for now :p11:34
ogra_oh. ok11:34
=== alan_g is now known as alan_g|afk
sergiusensdidrocks, ogra_ the reason it's preinstalled is to allow people to test click without going into rw12:16
ogra_yeah, i thought it was click related ... the commit message isnt so explicit12:16
didrockssergiusens: would be nice to find another way (or a simple switch to go into rw for them)12:16
Mirvev: qa-intel-4000 is offline12:17
didrockswe are installing dozen of package for testing in our product, doesn't make sense12:17
sergiusensdidrocks, if you go into read write and install dependencies you will never know if the system in ro mode would work though12:17
MirvI noticed it had had a kernel crash and a job was seemingly hanged, so I aborted it after which I got the report it's now offline12:17
didrockssergiusens: well, it's the same issues than for all the other tests TBH12:17
sergiusensdidrocks, yeah, I was hoping for QA to get us a path out of there12:18
didrockssergiusens: but I don't think it makes sense to ship autopilot in the image just for that12:18
didrocksyeah12:18
sergiusensit's not a good situation12:18
didrockssergiusens: adding a note for it, we need to fix that before 14.0412:18
sergiusensI asked for a self contained non deb autopilot once iirc12:19
didrocksthat can be a solution12:19
Mirvev: and it's up again, maybe an automagic recovery12:20
=== MacSlow|lunch is now known as MacSlow
psivaadidrocks: maguro tests with r125 just completed: http://ci.ubuntu.com/smokeng/trusty/touch/maguro/125:20140113:20140107.1/6035/12:44
davmor2psivaa: I've found an issue on the maguro post mir update, apps are hanging again so I'm not sure how much that is effecting results12:46
davmor2didrocks: ^12:46
=== alan_g|afk is now known as alan_g
Saviqev, hey, there's something weird in  http://naartjie/archive//head.unity8/ - there's a libunity-mir-dev 0.2+14.04.20140108.1bzr168pkg0trusty0-0ubuntu1, where lp:unity-mir only has 162 revs?12:47
davmor2didrocks: I'm going to be digging into it a bit more latter today if all goes well, but I discovered it testing on friday for balloons fix for calendar12:48
didrockspsivaa: thanks!12:51
didrocksdavmor2: can you give a little bit more details? And give them to kgunn?12:51
davmor2didrocks: didn't get much time to dig into it friday hence hoping to today once I have more info I will gladly pass it on,  I'm going to run a bunch of test from the coreapps ppa and see what fails where and why and then create a bug based on that info12:53
didrocksthanks12:53
didrocksdavmor2: this is from an older image, right? Since the Mir update? (so rev 119 or so)12:53
davmor2didrocks: yeah friday afternoon so 121 iirc12:54
didrocksdavmor2: Mir isn't new in 121, was it?12:57
sil2100didrocks: the evernote project is ok for preNEWing I guess, I added this MR to finish up some small things: https://code.launchpad.net/~sil2100/account-plugin-evernote/misc_and_bootstrap/+merge/20138412:57
sil2100(missing bootstrap in changelog)12:57
didrockssil2100: ok, reviewing at length at the same time12:58
=== alan_g is now known as alan_g|lunch
davmor2didrocks: no but it is when I hit the issue running tests, I think the new mir landed earlier in the week and we noticed issues on maguro from then12:58
davmor2didrocks: I can dig into when the new mir landed and see if I can still get that image and run against it, But I'm assuming I won't be able to13:00
didrocksdavmor2: it's the image just before IIRC, would be nice to ensure you are getting the issue on that one13:01
davmor2didrocks: no problems13:03
=== josepht changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: josepht | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: -
sil2100didrocks: strange, I can't globally approve it as well - I can only change the status to 'Merged', there's no 'Approved' o_O13:15
sil2100didrocks: which is strange, since I thought that when we're not added we cannot change the merge status in overall13:15
didrockssil2100: I guess you can mark it manually as merged as you are the branch owner13:16
didrockseven if you don't have the rights to approve it13:16
sil2100didrocks: ah, makes sense13:18
=== alan_g|lunch is now known as alan_g
sil2100didrocks: could you approve the invite of ubuntu-unity to be part of reminders-app-dev ? ;)13:40
* sil2100 poked Mirv to add us there13:40
didrockssil2100: done13:41
sil2100didrocks: thanks! Approved the merge now13:41
didrocksgreat ;)13:42
timpI got a unity8 crash here. Any ideas why? https://jenkins.qa.ubuntu.com/job/generic-mediumtests-runner-mako/4625/13:50
timpgreyback: ^?13:51
greybacktimp: think mir failed on startup, I see "what():  error during hwc set()" which causes crash13:53
timphmm - I see something similar in another MR https://code.launchpad.net/~nicolas-doffay/ubuntu-ui-toolkit/fix-1240019/+merge/19931214:08
asacdidrocks: so are the failures on unity8/notes/rssreader flaki or hard regressions?14:25
asachttp://ci.ubuntu.com/smokeng/trusty/touch/mako/125:20140113:20140107.1/6032/14:25
didrocksasac: notes/rss -> flaky, still the same14:25
didrockssee my previous emails14:25
didrocksand unity8, regression due to pygi14:25
didrockslook at #ubuntu-touch and #ubuntu-unity14:25
asacdidrocks: ok we stop the line for that, right?14:25
didrocksyep14:25
asacok good.14:25
didrocksupstream issues, not sure why pitti doesn't revert the faulty upstream release14:26
didrocksat least, there is a workaround for it and unity8 is implementing it14:26
asacdidrocks: workaround or fix?14:26
didrocksworkaround14:27
asacdidrocks: have you explained pitti that we now do a workaround?14:27
didrockssee bug #126857814:27
ubot5bug 1268578 in pygobject (Ubuntu) "Notification callback causes exception in gi" [High,Triaged] https://launchpad.net/bugs/126857814:27
asachave you asked him if we couldnt back this out rather than doing a workaround?14:27
didrockshe suggested the workaround14:27
=== vrruiz_ is now known as rvr
evSaviq: did you get a response to your query about naartjie? Sorry, I was at lunch when you asked.15:29
=== vrruiz_ is now known as rvr
=== alan_g is now known as alan_g|tea
rsalvetididrocks: when are we integrating another mir revision?15:50
* rsalveti wants the nexus 7 fix15:50
didrocksrsalveti: we didn't get any sign since Thursday, so when they are ready. Note that davmor2 is investigating a Mir-related potential regression, so we need more infos on that first15:51
didrocksdavmor2: did you open a bug about it btw? will  be easier to track15:51
rsalvetididrocks: right, need to ping someone from the mir team then15:51
rsalvetididrocks: who usually request a mir landing, kgun?15:52
davmor2didrocks: I was looking into it more first but being as it is on maguro will you care any more ? :)15:52
=== alan_g|tea is now known as alan_g
didrocksrsalveti: indeed, kgunn15:52
rsalvetidavmor2: no, we want a fix still15:52
rsalvetiwe'll drop it soon, but we want to have it supported until we officially drop it :-)15:52
rsalvetiand have one last stable image supported by UE15:53
rsalvetibefore we move to community15:53
davmor2rsalveti: no worries then I'm assuming it is a memory issue which hangs the app on start up but hadn't had chance to dig any yet15:53
rsalvetialright15:53
davmor2so a mix of memory and gfx possibly15:54
davmor2rsalveti: when running some tests for balloons I basically got this http://ubuntuone.com/2ixQVLJb711JUr5m7OLXpB15:55
rsalvetidavmor2: hm, weird15:55
rsalvetidavmor2: anything useful in syslog/logcat?15:56
davmor2rsalveti: not that I could see but after I'm going to do a fresh install so the logs are mint and try running the test again so if there are any issues they stand out more15:57
davmor2but again it was late friday for me so didn't get to dig too much, I'm just clearing down the last of the apps for today and then dealing with this for the rest of the day15:58
rsalvetidavmor2: right, let me know if you're able to reproduce it again16:03
davmor2rsalveti: should be able too what logs are useful to you and I'll grab the lot,  the test is against calendar so I'm assuming ~/.cache/upstart/calendar.log and /var/log/syslog anything else?16:04
rsalvetidavmor2: /system/bin/logcat as well (run as root)16:06
davmor2rsalveti: will do16:06
didrockssergiusens: hey, can you ensure latest notes-app will be soon in the archive (deb side is fine)16:44
sergiusensdidrocks, sure, let me do that now16:44
didrockssergiusens: also, can you fix usensord to be priority optional?16:46
didrockssergiusens: I wrote that on the landing spreadsheet for robru, but I guess he missed it…16:47
sergiusensdidrocks, yup16:47
=== gatox is now known as gatox_lunch
didrocksthanks ;)16:47
sil2100Meeting time!17:00
didrocksrobru: ogra_: coming?17:01
ogra_nope :P17:01
davmor2rsalveti: hmmm this is interesting on a fresh install the tests are running fine that failed on friday, so I'm wondering if it was the fact that I had accounts setup apps installed etc that was breaking stuff so I'll add some data to the phone and try it again then look at what changed between the images17:06
kenvandinesil2100, don't rerun the webcred stack until this is merged https://code.launchpad.net/~ken-vandine/libaccounts-glib/no_libtool_bin/+merge/20143917:09
sil2100kenvandine: sure :)17:09
kenvandinethe ppa build will block forever without it17:10
sil2100plars: ^ regarding the jenkins unblocking for webcreds17:13
kenvandinesil2100, merged17:13
sil2100plars: ^ nevermind!17:13
sil2100kenvandine: thanks for the info!17:13
kenvandine:)17:13
kenvandinenp17:13
plarssil2100: I was just talking to fginther about it, not sure why that file is missing, but it is indeed missing. Any chance it was just missing because of the other breakage?17:14
sil2100plars: yes, I think the breakage might have happened here: http://q-jenkins.ubuntu-ci:8080/view/cu2d/view/Head/view/WebCreds/job/cu2d-webcred-head-1.1prepare-uoa-integration-tests/364/console17:15
sil2100plars: seems to be some random failure?17:15
sil2100plars: it was the previous run of webcreds, automatic build I guess17:15
kenvandineoh, i killed a webcred build over the weekend, it was never going to finish because of libaccounts-glib dep wait17:16
kenvandinemaybe that caused it?17:16
kenvandinethe job had been running for 2 days, waiting for webcred build job to finish, which was in dep wait in the ppa17:18
plarssil2100, kenvandine, fginther: I can restart http://q-jenkins.ubuntu-ci:8080/job/cu2d-webcred-head/ now and we'll see if it's better then?17:18
kenvandineworth a shot i guess17:19
sil2100plars: sure17:19
plarssil2100: I've not restarted one of these, which packages or other parameters should I feed it? or just leave all of them blank?17:21
sil2100plars: you can leave all blank, then it will do a standard run with rebuilding what needs to be rebuilded17:21
plarsack17:21
plarssil2100, kenvandine: looks better so far :)17:23
kenvandinegood17:23
plarshttp://q-jenkins.ubuntu-ci:8080/job/cu2d-webcred-head-1.0prepare/17:23
kenvandineplars, stx17:23
popeysil2100: / didrocks sorry I missed the meeting, I have a personal appt at this time.17:23
popeydidrocks: was there an update to the online accounts plugin ask?17:23
didrockspopey: no worry, as there is no promotion under way ;)17:24
popeyok17:24
didrockspopey: the evernote one, but not anything else17:24
popeydidrocks: any idea when we might see it landing?17:25
popeyor at least reviewed ☻17:25
didrockspopey: evernote? soon soon, when sil2100 will succeed in running prepared (see the failure discussion above ^)17:26
popeyah17:26
didrocksif everything goes smootly before sil2100's EOD I guess ;)17:26
* popey scrolls17:26
sil2100popey: yes, we wait for the webcred stack to finish17:26
popeyk17:26
sil2100Once it finishes, I'll just press the send button and violla17:26
didrocks"voilà" ;)17:27
sil2100didrocks: just in case - is the whitelist updated? ;)17:27
sil2100didrocks: ;p17:27
didrockssil2100: doing for safety ;)17:27
didrocks(done)17:28
=== gatox_lunch is now known as gatox
sergiusensdidrocks, notes app should be in any minute; tested on mako and generic (emu)17:52
sil2100didrocks, kenvandine, popey: evernote published, it's in the queue now18:06
didrockstranks guys :)18:06
sil2100See you tomorrow :) !18:07
popey\o/18:11
kenvandinefginther, your upstart-app-launch rebuild failed, same failure18:29
dobeyrobru: hi18:29
robrudobey, hello18:29
dobeyrobru: i see your name in the PoCs column for unity-scope-click/ubuntu-purchase-service landing. just wondering what sort of time frame we should expect to see it done, as we have "frozen" trunk right now to avoid piling more stuff on, and will need to un-freeze it when the current stuff is landed18:31
robrudobey, i was just going to start on that right now18:32
dobeyrobru: great. i'm just curious about timeframe, because i have no idea how long these landing requests typically take. i guess this one will take a bit longer than normal as it's quite a bit of fixes, and a new package18:32
robrudobey, yeah, could be a couple of hours. first i have to kick builds, then then I have to run a bunch of manual tests, then I have to poke at it. probably 4 hours tops, depending on how long the builds & the tests take to run.18:33
dobeyrobru: ah ok. fwiw, the latest versions seem to be built in the daily-build ppa already, from saturday18:34
robrudobey, oh, no new commits since saturday? so that speeds things up then ;-)18:34
dobeyrobru: nope, last commit was on friday18:35
robrudobey, confirmed. ok, i just have to run the tests then. shouldn't be more than an hour or two18:36
dobeyrobru: great! please ping me when it's done. and thanks. :)18:36
robrudobey, ok, no worries18:37
=== fginther changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: fginther | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: -
fgintherkenvandine, looking18:40
kenvandinefginther, thx18:40
fgintherkenvandine, argh! there were two dh packages that machine was missing, I only installed one of them.18:42
kenvandine:)18:42
fgintherkenvandine, I'll restart it in a moment18:43
sergiusenskenvandine or robru care to check https://code.launchpad.net/~sergiusens/usensord/package_prio/+merge/201492 ?18:59
robrusergiusens, approved19:00
sergiusensrobru, ty19:03
kenvandinefginther, now the i386 build failed :)19:05
robrukenvandine, cyphermox: does anybody know what the 4 regressions for unity8 are? I'm currently trying to test purchase-service and seeing http://paste.ubuntu.com/6746370/ but not sure if those are the known regressions or if purchase-service broke those19:12
fgintherkenvandine, same problem, different host. sorry19:12
fgintherI assumed that host was already updated19:12
cyphermoxrobru: http://ci.ubuntu.com/smokeng/trusty/touch/mako/125:20140113:20140107.1/6032/unity8-autopilot/ ?19:13
cyphermoxbut that would only be two of them I guess19:13
cyphermoxrobru: check the reports in general19:13
robrucyphermox, hm, yeah, that's two of them...19:14
davmor2rsalveti: https://bugs.launchpad.net/unity8/+bug/1268693 that's the info gathered so far19:15
ubot5Ubuntu bug 1268693 in Unity 8 "Possible bug in mir/memory on maguro" [Undecided,New]19:16
davmor2rsalveti: don't know if there is anything of any use19:16
robrudobey, ok, published unity-scope-click and ubuntu-purchase-service. feel free to land merges in trunk now.19:55
dobeyrobru: great. thanks!19:56
robrudobey, y19:56
robrudobey, you're welcome19:56
rsalvetidavmor2: cool, let me check19:59
rsalvetidavmor2: just updated the bug, do you have the unity8 log as well?20:37
rsalvetionly interesting line is coming from the app:  QUbuntu: Could not create application instance20:37
sergiusensfginther, thomi if I want something in trusty's autopilot; what branch should I create the MR against?21:34
fgintherkenvandine, finally! http://s-jenkins.ubuntu-ci:8080/job/upstart-app-launch-trusty-i386-ci/64/21:34
fginthersergiusens, lp:autopilot unless thomi disagrees with me21:36
sergiusensack21:36
sergiusensfginther, hmmm, lp:autopilot looks nothing like what is release :-/21:42
fginthersergiusens, I think it's been a while since lp:autopilot has made it into distro21:44
sergiusensfginther, I sort of need something in sooner than later so I'll do a debian patch if it comes to it21:45
fgintherbut I don't know if that explains what you are seeing21:45
sergiusensfginther, meh, not even :-)21:45
sergiusensfginther, yeah, it's been a while, so debian patch it is ;-)21:46
thomisergiusens: lp:autopilot, and we'll stabilise it into the release series for the next + 1 release21:46
sergiusensthomi, what is the plan for stuff we need now?21:46
sergiusensas in distro21:47
thomisergiusens: OTP, one second21:47
kenvandinefginther, yay... thanks!21:57
thomisergiusens: hey, I'm back22:36
thomisergiusens: what's the patch?22:36
sergiusensthomi, http://paste.ubuntu.com/6747393/22:39
sergiusensthomi, or just the patch http://paste.ubuntu.com/6747399/22:40
thomisergiusens: is this really urgent enough to justify shipping a distro patch?22:41
sergiusensthomi, I can do that or disable all the gallery and media player tests22:41
thomisergiusens: OK, so if you ship it as a distro patch, please also make a MP against lp:autopilot with tests to cover the new functionality22:42
thomithat way we can roll it into a regular release22:42
thomias it is, it looks like this code is untested22:42
sergiusensthomi, hmm, but the code has changed so much22:42
sergiusensthomi, I'll log a bug perhaps22:42
thomisergiusens: what I'm saying is: please dont' ship a distro patch without at least making an attempt to upstream the change22:43
thomithe new code is actually easier to modify, not harder22:43
thomiplus, it has 100% unit test coverage22:43
thomiso your change should be easy to add :)22:43
sergiusensthomi, please release it into the distro then22:44
thomisergiusens: believe me, we're trying22:45
thomisergiusens: we're in the middle of the release process right now. If you have concerns, you need to speak to jfunk about it :)22:45
sergiusensthomi, you should release more often; this just complicates things; I'll talk to jfunk22:48
thomisergiusens: dude, we're trying... really. If it were up to me I'd release once a week at least22:49
sergiusensthomi, so what is blocking you now?23:02
thomisergiusens: you really need to talk to jfunk about that - he's managing the release process. AIUI, we're waiting on some response from the release team, but I may be wrong about that23:08
=== fginther changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cihelp | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: -
cjwatsonthomi: not seeing anything relevant either on the release list or in our subscribed bug list23:47
thomithere's a release list? anyway, as I said earlier, jfunk said he was sending the email - maybe he doesnt know about the list either?23:48
thomior maybe he just hasn't sent it yet? I'm guessing now23:48
cjwatsonhttps://lists.ubuntu.com/mailman/listinfo/ubuntu-release23:48
thomicjwatson: that's not a replacement for the spreadsheet though, I take it?23:49
cjwatsonthis is a six-year-old list, it can't possibly be a replacement for something less than a year old23:49
cjwatsonit's not touch-specific23:49
cjwatsonjust as the release team (without qualifiers) is not touch-specific23:49
cjwatsonif you mean the touch landing team then that's something different ...23:50
cjwatson(or whatever the latest name for it is, but at any rate we should take care to avoid name collisions with something that already exists)23:50

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