/srv/irclogs.ubuntu.com/2013/10/17/#ubuntu-quality.txt

phillwI'd be much more use checking the desktop installer, as that's where we know there were issues... but, get the alternates out of the way 1st as they are our "hey, it works" series :P00:00
phillwAhh, Jeff Wayne; War of the Worlds... Calming on the ears :D00:27
phillwthanks que-bot... Let me just grab the only person I know to get it tested :P00:30
=== charles_ is now known as charles
=== _salem is now known as salem_
=== salem_ is now known as _salem
phillwhi, anyone on ubiquity partitioning crashing bug watch?03:57
phillwbug 124079404:12
ubot5bug 1240794 in ubiquity (Ubuntu) "In partioner 'select something else' deleteing and adding a partition causes crash" [Undecided,New] https://launchpad.net/bugs/124079404:12
sonuhi, little help with info: what is "Hardware profile? URL to the hardware profile." while reporting test result?05:23
Noskcajsonu, It helps with diagnosing hardware specific issues. If possible, have one05:56
sonuNoskcaj, agree. But how to create one?06:14
NoskcajI think phill made a guide, one sec06:14
Noskcajsonu, https://wiki.ubuntu.com/QATeam/Hardware should work06:16
NoskcajIf you use a VM, just add a comment saying something like "Tested on Vbox 4.2"06:16
sonuNoskcaj: wonderful! thanks!06:17
Noskcajno problem, it's great to get more testers06:17
DanChapmanmorning all06:49
elfymorning DanChapman06:49
DanChapmanmorning elfy, how are you?06:50
elfyok thanks - be glad when it's Friday and Smaug is released and I can forget about it for a day06:52
DanChapman:-)06:53
pittieveryone feeling saucy today? :-)06:59
elfydefinitely smaug here :p07:04
* DanChapman is feeling extra saucy! :-)07:04
elfyand I think we should keep that to another channel - smartboyhw or nosckaj might be about :P07:05
DanChapmanhehe ;-)07:05
om26erubuntu-qa Oy! please can anyone confirm bug 1240841 ?08:02
ubot5bug 1240841 in mir (Ubuntu) "[Mir]In-App scrolling is lagging much" [High,New] https://launchpad.net/bugs/124084108:02
slickymastermorning all08:45
smartboyhwGuys, we have the Xubuntu i386 desktop and Lubuntu amd64+mac and ppc images needing testing:)09:05
=== vrruiz_ is now known as rvr
davmor2Morning all Happy release day10:01
jibelGood morning davmor2 !10:02
pittihm, adb doesn't want to come up for me today (still worked yesterday, and tried rebooting)11:06
pittican I use phablet-flash from fastboot?11:06
jibelpitti, you can boot to recovery and use phablet-flash for there if you specific -d <device name>11:11
pittijibel: thanks, finally got recovery to start; trying now11:20
pittijibel: hm, still nothing in "adb devices" for recovery; I'm beginning to think it's not my phone but my workstation..11:25
pittiit's in lsusb, but adb doesn't see it11:26
jibelpitti, did you try to disconnect/reconnect the usb cable? last thing is to restart the desktop11:26
jibels/desktop/workstation11:27
pittijibel: yes, I rebooted phone, dis/reconnected cable11:27
pittiand killed/restarted adb11:27
pitti(daemon on desktop)11:27
pittiso I guess I'll need to reboot once I'm done with my VM11:27
jibelpitti, as user or root?11:27
pittijibel: I never started it as user, has worked fine so far11:27
pittijibel: indeed, works as root; wth?11:28
jibelpitti, I've a machine that I broke but didn't find how, and I can only start adb as root to make it work11:28
pittijibel: argh, c'était moi, sorry11:29
pittiI did "sudo stop udev" earlier for testing my bug fix..11:29
pitti*brown paperbag*11:30
sonuquit12:23
pittifginther: just logged into the otto test box; last saucy version is from Oct 1, isn't that a tad old for testing?12:41
pittifginther: anyway, I'm afraid I can't debug the (supposed) at_console failure remotely, I'd actually need to be "in" the running session12:43
sergiusenspitti, perhaps you can polish the udev rule there for adb12:43
pittisergiusens: what? where?12:43
pittisergiusens: this is about the saucy/desktop autopilot tests, not the ones on the phone12:43
sergiusenspitti, would be great i you could take a look; dpkg -L android-tools-adb12:43
sergiusenspitti, I was referring to what you and jibel were talking about above :-)12:44
pittisergiusens: oh; yeah, I had udev stopped still12:44
sergiusenspitti, oh sorry, didn't read all of it :-)12:44
pittisergiusens: looks quite alright to me12:44
sergiusensgreat12:45
fgintherpitti, having an older snapshot helps to expose dependency issues (although it's not a perfect solution)12:54
pittifginther: oh, you mean versioned dependencies12:54
pittifginther: I see12:54
pittifginther: I answered to your otto email12:54
pittifginther: so if I can abuse the CI system to do a few debugging steps in my proposed branch, see what the otto output is, and revert, that's ok for me (but it'd waste the runs on mako/maguro)12:55
fgintherpitti, that should be ok. there isn't much going on right now (in fact it's very quiet on jenkins)12:57
fgintherpitti, if you wanted to just isolate your experiments to otto I can help with that12:58
fgintherpitti, you would just need to push to a branch without a merge proposal and trigger the generic-mediumtests-saucy job manually12:59
pittifginther: mako/maguro fail as well, but for different reasons (mostly maliit and mir crashes), so starting debugging just for otto would be good12:59
pittifginther: oh, cool; how would I do that?12:59
fgintherpitti, ok, first create an account on http://10.97.0.26:8080/ and I'll enable job building access13:00
pittifginther: done, "pitti"13:01
fgintherpitti, when you go to http://10.97.0.26:8080/job/generic-mediumtests-saucy/, you should now see a build link13:03
fgintherpitti, landing_candidate is your lp branch, target_branch is the branch to merge into (i.e. lp:dialer-app) and test_suite is the name of the autopilot suite, "dialer_app" in this case.13:04
fgintherpitti, every other parameter can be left blank for dialer-app13:05
pittifginther: build link -> the "Build now" on the left top menu?13:05
fgintherpitti, yes13:05
pittifginther: ah, splendid13:05
pittifginther: thanks!13:05
pittifginther: how do you specify a branch? lp:~... or http://bazaar... or https://code... etc?13:29
smartboyhwGood job everybody for testing 13.10!13:30
fgintherpitti, lp:~pitti... works13:31
pittifginther: do I need to specify anything in addition after all? http://10.97.0.26:8080/job/autopilot-testrunner-otto-saucy/1254/13:46
pitti/var/local/autopilot/autopilot.log: autopilot run: error: too few arguments13:46
pittifginther: like "messaging_app" (test name)13:46
fgintherpitti, yes, you do need the test_suite == messaging_app13:47
pittifginther: ah sorry; you already said so, missed that13:47
fgintheryou can always look at the jenkins parameters used by an automated build13:47
pittifginther: http://10.97.0.26:8080/job/autopilot-testrunner-otto-saucy/1255/console14:07
pittifginther: ok, what I suspected: there is no logind session, i. e. none of the "local foreground" functionality by d-bus or policykit will work14:07
pittifginther: curious that this didn't already come up, as these are rather common (device access, suspend, bluetooth, whatnot)14:08
pittifginther: is that session spawned by lightdm autologin? or something manual?14:08
fgintherpitti, the tests are launched via an upstart job14:09
melodiehi14:10
pittifginther: ah, that would do it14:10
pittifginther: I mean, explain why they aren't attached to a console14:10
melodieI would like to have a tip/information about discussions in the blueprint sections14:10
melodiecan someone help me?14:10
melodiemy question is : is it ok to post long texts in that section?14:10
fgintherpitti, it's an xdg autostart job, I would have that that would have access to most of the user session14:12
pittimelodie: if you have more than two or three paragraphs, consider writing a wiki page and linking it in the blueprint, as that's much easier for sectioning, formatting, and reading14:12
fgintherwould have *thought*14:12
pittifginther: oh; those would, but I mean how is that session itself started? apparently not through lightdm?14:12
pitti(or any dm really)14:13
* fginther grammer on IRC gets badder over time14:13
pitti*chuckle*14:13
pittifginther: I thought you meant "the whole unity session is started from a (system) upstart job"14:14
fgintherpitti, I thought unity was started via auto-login, but I'll admit my ignorance here. This is all based on jibel's work14:15
pittifginther: so this is by and large using otto as it is?14:15
fgintherpitti, yes14:15
pittifginther: I guess I should try checking what breaks otto on current saucy then (at least on my workstation), and see there14:16
pittifginther: thanks, doing that then14:16
melodiepitti is there a preferred wiki page?14:16
pittimelodie: if you write one, you should use https://wiki.ubuntu.com/SpecTemplate14:16
fgintherpitti, let me know if you find anything, I'd like to move the jenkins otto machines to saucy eventually14:17
melodiepitti I can also put it on a forum I run or on a website I run, if needed. Here is the text, can you have a look and tell me what would be best? http://pastebin.fr/2915814:17
pittifginther: they currently run on raring?14:17
fgintherpitti, raring is what's installed on the bare metal, the overlay FS is a saucy ISO14:18
pittimelodie: well, anythign which is comprehensible, explains what you want to do, and is actionable; so whether it's a link to a wiki or a forum doesn't matter that much, but folks generally prefer wiki.u.c. as then other people can make updates etc.14:19
pittifginther: *nod*14:19
pittibbl, otto testing14:20
melodiepitti I am not used to the Ubuntu ways when it comes to contributions, so I would like you to tell me if the text I wrote would be ok as is in the wiki side you just pointed to for me?14:23
melodiepitti I give up the wiki, it has disconnected me: Erreur OpenID : Nonce already used or out of range.14:25
melodielast time I tried to edit a page in the wiki I was never able to do so14:26
pittimelodie: hm, no idea about the OpenID thingy; you might have some old cookies around in your browser or something?14:35
pittimelodie: it's okay for a first start; proper drafting happens after UDS discussions anyway14:36
pittijibel (FYI fginther): I found out what's wrong with otto here: http://paste.ubuntu.com/6251313/14:36
pittijibel: so the first failure is due to that kernel option not being enabled by default, so I just commented out the limit stuff14:37
pittijibel: the second is a bug in otto; $RELEASE is indeed not defined anywhere14:37
pittijibel: curiously, even after destroying the container, adding that hack (to the checkout) and re-creating the container the phablet-prepare.sh is still the original version (without the export RELEASE)14:37
pittijibel: so eventually I added the export RELEASE to /var/lib/lxc/.../phablet-prepare.sh14:38
* pitti files a bug, to track that14:38
pittijibel: oh of course, I added it to phablet-prepare, silly me14:39
jibelpitti, this is defined in a configuration file, but if the configuration file is not found then it's definitely not defined, thanks for finding this14:40
pittijibel: filed bug 1241021 with details14:43
ubot5bug 1241021 in Otto "premount script fails with $RELEASE not defined" [Undecided,New] https://launchpad.net/bugs/124102114:43
pittifginther: but the main point of the exercise, that gives me a normal session, with a logind session and all bells and whistles14:44
pittifginther: so I guess I now need to locally reproduce how you start a test, i. e. what you throw into the otto config14:44
jibelpitti, hm, http://bazaar.launchpad.net/~jibel/otto/phablet/view/head:/lxc.defaults/scripts/phablet-prepare.sh#L2714:47
jibelpitti, I don't know where the phablet-prepare.sh script comes from14:47
pittijibel: ah, different branch14:48
pittijibel: lp:otto is http://bazaar.launchpad.net/~otto-dev/otto/trunk/files14:48
pittijibel: you have a ~jibel branch14:48
pittijibel: http://bazaar.launchpad.net/~otto-dev/otto/trunk/files/head:/lxc.defaults/scripts/ has phablet-prepare14:49
jibelpitti, yes the phablet stuff has never been merged in trunk14:49
jibelwell partially merged14:50
pittijibel: err, it *is* in trunk!?14:50
pittijibel: at least lp:otto points to the ~otto-dev one, not the ~jibel one14:50
pittifginther: ^ which one do you use in CI?14:50
jibelpitti, that's something I never had time to merge completely because I've been swamped into phone testing14:51
pittijibel: ah, ok; so I guess in the DC we don't run lp:otto, but your branch?14:52
jibelpitti, yes, and last time I touched it was in August14:53
jibelpitti, in summary otto for desktop -> everything is in trunk, for phone -> I must merge my work14:54
elopiogood morning.14:56
balloonsWelcome to Ubuntu Quality | http://wiki.ubuntu.com/QATeam | Celebrating Saucy's Release!14:58
=== balloons changed the topic of #ubuntu-quality to: Welcome to Ubuntu Quality | http://wiki.ubuntu.com/QATeam | Celebrating Saucy's Release!
* pitti ^5s balloons14:59
smartboyhwballoons, let's celebrate?14:59
smartboyhw!14:59
balloons^^ never give that balloons OP powers your channel :-)14:59
smartboyhwballoons, I suppose we now can really discuss about classroom sessions:P14:59
* slickymaster throws same champagne all around 14:59
balloonsI'm struggling to put together long sentences, but yes, totally14:59
smartboyhwballoons, how do you want the classroom sessions to be done?15:00
balloonssmartboyhw, well I don't know15:04
smartboyhwballoons, I'm thinking something simliar to Ubuntu Open Week / Dev week, but I wonder if that is too ambitious15:05
balloonsI'm open to ideas. I'd like to think about trying something new, but if we don't, that's ok too15:05
balloonsI'd like to seperate the content out of the sessions.. So the content can be updated and the session can be more hands on15:06
fgintherpitti, using lp:~otto-dev/otto/trunk15:06
balloonsdoing hands on things is interesting on the internet of course15:06
balloonsvideo, text, both?15:06
smartboyhwballoons, dunno.15:07
pittifginther: ok, thanks; so I now need to know how the tests are auto-started through otto15:07
melodie<pitti> melodie: hm, no idea about the OpenID thingy; you might have some old cookies around in your browser or something? // I don't think so. I have met with this before and it happens only in the wiki ubuntu, with my Firefox browser (didn't try other browsers yet)15:09
jibelpitti, it's an autologin with an autostart15:09
pittijibel: hm, then I have no idea why these don't have a logind session :/15:10
jibelpitti, http://bazaar.launchpad.net/~otto-dev/otto/testsuite_autopilot-unity/view/head:/target-override/etc/xdg/autostart/autopilot.desktop15:11
pittijibel: thanks15:12
jibelwhich executes http://bazaar.launchpad.net/~otto-dev/otto/testsuite_autopilot-unity/view/head:/target-override/usr/local/bin/run-autopilot.sh15:12
pittijibel: thanks; that all looks fine15:13
balloonssmartboyhw, we can lay out the schedule. There are some other changes I'd like to broach that might make it more interesting. We have the problem of giving training before it can be used by folks.. IE, testing is done largely at the end of the cycle15:15
melodiepitti thank you, I have been able to finish writing it here: https://blueprints.launchpad.net/lubuntu-brainstorming/+spec/zram-config15:22
jibelpitti, where is the branch you want to test?15:30
pittijibel: lp:~pitti/messaging-app/phonesim-tests-debugging15:30
pittijibel: I pushed two extra commits there to debug why there is no logind session; running another CI test now15:30
* pitti kicks off http://10.97.0.26:8080/job/generic-mediumtests-builder-saucy-amd64/966/15:31
pittijibel: originally, it's https://code.launchpad.net/~pitti/dialer-app/enable-phonesim-tests/+merge/19124815:31
pittijibel: I pushed a separate branch for the debugging stuff to kick off CI builds on otto manually, to avoid wasted runs on phones15:32
smartboyhwballoons, agreed15:37
pittineed to leave for today, will continue with this tomorrow (although I don't have a great idea now); current build shoudl show me syslog, auth.log and so on15:41
=== jfunk is now known as jfunk-afk
=== _salem is now known as salem_
melodiebbl19:03
thomimorning19:08
cgoldbergmorning thomi19:10
=== salem_ is now known as _salem
elopiorobotfuel: I don't really understand bug #123193919:48
ubot5bug 1231939 in Ubuntu UI Toolkit "selecting an element in itemSelector or OptionSelector rapidly shows disabled text for selected element on touch devices." [Undecided,Confirmed] https://launchpad.net/bugs/123193919:48
elopiomaybe I'm not doing it fast enough. But I don't understand where and why the text should be disabled.19:48
robotfuelelopio: the text is disabled a disabled grey instead of black19:48
robotfuelelopio: the text should not be disabled19:49
elopiorobotfuel: can you make a screenshot? I seee the selected text always black.19:49
robotfuelelopio: maybe it was fixed with #1231936? I'll reproduce now19:50
elopiorobotfuel: I'm using image 100.19:51
robotfuelelopio: I just tried it on my desktop and it's there in the examples on the option selector19:51
elopiorobotfuel: I might not be doing it right, but on the gallery I also see the text black.19:53
robotfuelelopio: http://imgur.com/a/Du1ON19:58
elopiorobotfuel: got it!19:59
robotfuelmaybe it's the animation hasn't finished and the level remains the same19:59
elopioI was using the installed package, not from trunk.19:59
elopiomust be a recent regression.19:59
robotfuelelopio: the level of grey/black is depending on how quickly I click the item in the combobox/list20:00
robotfuelI've updated the bug20:03
elopiorobotfuel: thanks. I think I'm done with your bugs. Let me know if I missed something.20:03
elopiowe should be meeting weekly with the SDK team to have a voice on the priority of these bugs.20:03
* elopio walks with the dog.20:04
robotfuelelopio: it looks like you got them all.20:05
robotfuelelopio: thanks!20:05
Letozaf_balloons, hi20:18
balloonsLetozaf_, hello20:19
Letozaf_balloons, I re-flashed my device as I has too many weird things happening20:19
balloonsLetozaf_, ahh..20:20
Letozaf_balloons, I wanted to ask you what do I need on it after re-flashing to make the click packes stuffu work20:20
balloonsI've got build 100 running now20:20
Letozaf_balloons, so have I, finally the first official one :P20:20
balloonsI reflashed at one point to transition out of the rw mode with all the edits I had20:20
Letozaf_balloons, what do I have to install on it to make the click packages install work ?20:21
balloonshttp://pastebin.com/rDhysJ2J20:22
balloonsLetozaf_, nothing... you should install nothing on it20:22
balloonsrun that script and it should work20:22
Letozaf_balloons, ok I will try it now, thanks20:22
* elopio <- lunch20:25
Letozaf_balloons, :(  http://paste.ubuntu.com/6253131/20:33
Letozaf_balloons, maybe I'm missing something20:33
balloonsrssreader_app20:38
Letozaf_balloons, do I have to install ubuntu-rssreader-app-autopilot on the device20:39
balloonshmm20:39
balloonsno20:39
balloonsthe module name is ubuntu_rssreader_app20:39
balloonsI blame the script author20:39
balloons<--20:39
Letozaf_balloons, :D20:39
balloonsI don't use it20:39
balloonsahh I see one issue20:40
Letozaf_balloons, what ?20:41
balloonsthe script sucks20:47
balloonsregardless the point is if the click package installed it would find the tests20:47
Letozaf_balloons, yes I was looking under /home/phablet/autopilot ...20:48
Letozaf_balloons, well if you got time, otherwise I will be back tomorrow evening or when you can20:49
balloonsi'm playing with it but the script has nothing to do with your issues20:52
balloonsis there a click package in /tmp?20:52
balloonsyou can see it installed, so it should run20:52
balloonssimple as that20:52
Letozaf_balloons, I'v got com.ubuntu.shorts_0.2.120_all.click in /tmp20:53
Letozaf_balloons, it's a file20:53
balloonsyep20:53
balloonsand your log shows it installed20:53
Letozaf_balloons, but it does not run :(20:53
Letozaf_balloons, it's owner and group are root while all the other files in /tmp hav phablet as owner and rott20:54
Letozaf_balloons, sorry root20:54
Letozaf_balloons, could that be an issue ?20:54
Letozaf_no owner and group20:55
Letozaf_balloons, I will be back tomorrow, going to bed now, it's getting late for me :D21:00
balloonsLetozaf_, best to ask sergio I think21:00
balloonsyou installed the click package it should work21:00
balloonsend of story21:00
Letozaf_balloons, ok maybe tomorrow I will ask him21:00
balloonsyea, sorry I've really no idea why it doesn't21:01
balloonsbut I think I made a better script anyway..21:01
Letozaf_balloons, doesn't matter, I was curious to see it run on my device that-s all21:01
Letozaf_balloons, thanks anyway :D21:01
Letozaf_balloons, going to bed now, good night!21:02
balloonsLetozaf_, night :)21:02
balloonshere's the fixed version: http://pastebin.com/nFcF8RpR21:02
Letozaf_balloons, ok I will try this tomorrow, thanks21:02
phillwballoons: (and any one else) "so is there a do-releases-upgrade log somewhere? the error "an unresolvable problem occurred while calulating the upgrage." is pretty much as useful as just failing without any error." A Question that has just been asked.21:14
balloons/var/log/dist-upgrade21:16
balloonsphillw, ^^21:16
phillwballoons: thansk!21:16
jfunkping veebers - can you meet now?21:21
veebersjfunk: sure can21:22
jfunkcool, see you there21:22
slickymastergood night all23:23

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