/srv/irclogs.ubuntu.com/2014/04/11/#ubuntu-devel.txt

xnoxsuperm1: i have a simple reproducer.... it's pointing at a gtk regression =)))01:48
hyperairRAOF: okay, i found it. turns out it's ccsm > opengl > enable fbc01:51
RAOFhyperair: You mean texture compression, or framebuffer object?01:52
hyperairRAOF: if that's enabled, compiz blacks out when trying to accomodate to a change in the display configuration01:52
hyperairRAOF: texture compression01:52
RAOFHuh, fun.01:52
hyperairmm01:53
hyperairdo you see the bug too?01:53
RAOFhyperair: Not exactly the same as you; I instead get seizure-provoking 60Hz flicker between a previous mostly-correct frame and black.01:55
hyperairugh01:55
=== wendar_ is now known as wendar
superm1xnox: well that's actually great to hear03:07
superm1can you share the simple reproducer?  whatever it is about what's happening in that page can probably be switched to something else to avoid hitting this bug in the short term03:08
superm1and long term get the GTK bug fixed03:08
xnoxsuperm1: yeah, i'm trying to work on it. I think if i tweak/fake environment enough it may just work. bug #130634103:09
ubottubug 1306341 in mythbuntu-common (Ubuntu) "ubiquity / mythbuntu assertion in gtkrecentmanager.c" [Undecided,New] https://launchpad.net/bugs/130634103:09
xnoxsuperm1: the problematic things appear to be the "filechooser" buttons in tab_remote.ui but i'm checking things further.03:10
superm1ah interesting03:10
xnoxsuperm1: looking at the glade file vs installer ui -> are the buttons to select configuration file actually used?03:11
superm1xnox: they're used in something else that uses the mythbuntu-common package03:12
xnoxsuperm1: oh, ok.03:12
superm1lemme double check if they can be ripped out though03:12
xnoxsuperm1: lemme check if that's the only issue.03:12
superm1i'd be surprised if that was the only issue; since taking the services page / plugin out of the picture fixed it, not the remote one03:13
xnoxsuperm1: probably the plugins were removed in the wrong order. So nuking myth-services would also nuke myth-remote.03:14
superm1ah03:15
xnoxsuperm1: since myth-remote is after myth-services.03:15
xnoxbut i'll check that as well.03:15
superm1the file chooser dialog was used for configuring a remote custom configuration; that whole thing is littered with bugs, i wouldn't mind just axing it if it fixes this03:16
superm1i'll take out the associated code and see03:17
xnoxsuperm1: i'm aiming for a one-line fix only =)03:17
Logan_xnox: to whom should I talk if I want to become part of ~udd?03:18
xnoxLogan_: anyone, but why would you want to? all you'll gain is commit access to lp:udd branch and that's it.03:19
xnoxLogan_: what are you after in terms of permissions/capabilities?03:19
Logan_requeue_package.py 😛03:19
Logan_oh god I really need to turn off this automatic emoji replacement03:19
xnoxLogan_: start here http://www.canonical.com/careers/all-vacancies ; once you get one of those, we could sort you out access =)03:20
Logan_if Canonical offered internships, I'd be all over that03:20
xnoxLogan_: it's running on a publically inaccessible machine, under an elevated celebrity account to push the branches to launchpad =03:21
xnox=/03:21
Logan_oh wonderful03:21
xnoxsuperm1: mid-air collision!04:03
xnoxsuperm1: http://bazaar.launchpad.net/~xnox/mythbuntu/fix-myth-passwords/revision/20104:04
xnoxsuperm1: but your way is fine as well. Please land that.04:04
xnoxsuperm1: the other bug, i'm fixing in ubiquity by doing this - http://paste.ubuntu.com/7233456/04:05
superm1xnox: hah wow talk about funny timing04:05
xnoxsuperm1: seconds difference upon push.04:05
superm1your way actually scales better for any changes in future04:05
superm1i've got a tear out of the filechooser stuff too that appears to resolve things as well04:06
=== FJKong_afk is now known as FJKong
xnoxsuperm1: don't.04:07
xnoxsuperm1: all works fine with http://paste.ubuntu.com/7233456/ applied to ubiquity04:07
superm1ok04:07
xnoxsuperm1: i'm landing that for ubiquity.04:08
xnoxsuperm1: if you are ok with myth-password change, then i can upload that as well.04:08
xnoxsuperm1: trumping your commit.04:08
superm1yeah go ahead and upload that04:08
superm1can you commit it to ~mythbuntu-dev bzr too?  I think all ~ubuntu-dev is able to04:08
Mirvif there's a core-dev awake, we'd need a packaging ack for messaging-app https://ci-train.ubuntu.com/job/landing-015-2-publish/10/artifact/packaging_changes_messaging-app_0.1+14.04.20140410.1-0ubuntu1.diff04:11
Mirv(tested + QA signed off)04:11
Mirvthere's not much to ack though04:12
xnoxsuperm1: both ubiquity & mythbuntu-live* packages uploaded. When they get accepted tomorrow, i'll ask for mythbuntu iso respin and i will test it out.04:27
superm1xnox: awesome, thanks so much04:27
xnoxsuperm1: no problem, the HOME fix should actually also resolve a few things across the board - e.g. pango, dconf etc complaining. none of them were fatal so far...04:28
xnoxsuperm1: so I'm happy about it. Still - why gtk decides to assert is beyond me.04:29
pittiGood morning05:15
pittibarry: :(05:15
pittiinfinity: mmmmm, "final release of Saucy Salamander in a week." :)05:37
pittiah, and the followup mail :)05:38
pittibarry: but thomi landed the AP branch to drop the deps05:41
* hyperair wonders where the zsh manpages are..05:44
StevenKhyperair: Replaced by info pages, as far as I can see.05:47
hyperairreally?05:47
hyperairi thought they used to be there together with the info pages05:48
hyperairin fact, zsh-common still has a dangling zsh5 manpage link05:48
hyperairsymlink*05:48
hyperairalso...05:49
hyperairhyperair@thinkpwn:~(1)% run-help autoload                                                                                                                                                                                     [ 1:49PM]05:49
hyperairautoload is a shell builtin05:49
hyperairmeh05:49
hyperairNo manual entry for zshbuiltins05:49
pittiinfinity: and the langpack fun begins :)05:54
hyperairxnox: ping. are the zsh manpages meant to be missing?05:57
StevenKhyperair: Are they in the source package?05:58
hyperairStevenK: dunno, i'll need to check.05:59
hyperairStevenK: but debian has its manpages intact at least05:59
hyperairStevenK: okay, for some reason we have a downstream ubuntu patch that disables building of docs06:01
hyperairmanpages included06:01
hyperairand a changelog entry from xnox adding the patch06:02
* hyperair has no idea why06:02
StevenKNo bug referenced or anything?06:02
hyperairnope06:03
hyperairno dep3 headers either06:03
hyperairjust.. patch.06:03
StevenKIt's in main, but I seriously doubt it's on the CD, so WTF06:03
* hyperair shrugs06:04
hyperairthat's why i pinged06:04
StevenKhyperair: It's a good plan, I want to know too.06:04
slangasekhyperair: the explanation in the changelog says:06:23
slangasek    - Keep using the upstream tarball that contains pre-generated docs as06:23
slangasek      yodl is required to build them but the MIR hasn't been approved.06:23
hyperairoh blargh06:23
slangasekso the docs are assumed to be present in the upstream tarball06:24
hyperairi see06:24
* hyperair didn't spot that entry06:24
pittiwgrant: I'm disabling the saucy langpack build cron job now; please feel free to do the same on LP06:40
=== liam_ is now known as Guest89383
=== zbenjamin_ is now known as zbenjamin
Mirvif http://bazaar.launchpad.net/~timo-jyrinki/ubuntu/trusty/lmms/1.0.0_packaging/revision/45 looks correct (chiefly silencing pkg-has-shlibs-control-file-but-no-actual-shared-libs), then https://code.launchpad.net/~timo-jyrinki/ubuntu/trusty/lmms/1.0.0_packaging/+merge/215131 would be ready07:28
hydeI just filed a bug report about patch (possibly duplicate, did not check....), but I wonder if someone here could shed some light on, how can command line argument handling of a program like patch get so broken, that it throws segfault?07:28
Mirvbut I hesitate with the shlibs, whether it should be there at all or if there should be some manual shlibs file07:28
LocutusOfBorg1https://bugs.launchpad.net/ubuntu/+source/network-manager/+bug/125402807:41
ubottuLaunchpad bug 1254028 in network-manager (Debian) "[precise] network manager set incorrect /64 prefix from dhcpv6 client" [Unknown,Confirmed]07:41
LocutusOfBorg1hi, anybody wants to sponsor this?07:41
Mirvmvo: would you have time to give DDTP a health check? ie. syncs from Debian (looks like last in Nov/Dec) + uploads to archive/mirrors (not sure how to check if done recently).08:54
Mirvor if there are other people who know the DDTP magic08:55
cariboupitti: can I come and bug you about apport for a few minutes ?09:13
=== vrruiz_ is now known as rvr
pitticaribou: just ask :)09:16
cariboupitti: PM09:17
xnoxhyperair: bug #124210809:42
ubottubug 1242108 in zsh (Ubuntu) "all zsh manpages are missing" [Medium,Triaged] https://launchpad.net/bugs/124210809:42
hyperairxnox: ah, thanks.09:42
hyperairoh crap, my launchpad.net session was expired09:42
* hyperair tries to remember password09:43
infinitypitti: The one to -announce was slightly more right, the -release readers get a treat.10:07
=== MacSlow is now known as MacSlow|lunch
Chipacawho should I pester about https://bugs.launchpad.net/ubuntu/+source/unity/+bug/1304265 ?11:28
ubottuLaunchpad bug 1304265 in unity (Ubuntu) "Unity bugs out when changing screen size" [Undecided,New]11:28
seb128Chipaca, try asking bregma about it11:29
Chipacaseb128: ta, will do11:29
seb128yw11:29
Chipacabut first i think i'm going to grab me some lunch11:29
Laney#ubuntu-unity more generally11:29
Chipacaokie doke11:29
ChipacaLaney: good point :)11:30
seb128the channel is mostly ovetaken by unity8 discussion though, so unity desktop questions might not get a reply, check with bregma if that happens, he's leading the team working on the desktop issues11:30
dokopitti, uploaded python3.4. fixed all the new testsuite errors, caused by moving around binary test cases which don't appear in the diff. there is one still failing, which I documented and disabled for now so that the autopkg tests pass11:54
=== FJKong is now known as FJKong_afk
=== _salem is now known as salem_
=== tkamppeter_ is now known as tkamppeter
=== dpm is now known as dpm-afk
infinitymvo_: Since when does apt allow # as a comment char in configs?13:52
infinitymvo_: It certainly didn't used to (which makes the aptdaemon change odd)13:52
cjwatson  * [ABI break] support '#' in apt.conf and /etc/apt/preferences13:53
cjwatson    (closes: #189866)13:53
cjwatsonapt 0.7.2213:53
infinitycjwatson: Oh.  So maybe what I'm thinking of is that aptdaemon's tests failed when we tried to use #, not that apt didn't allow it. :)13:54
infinity(I remember we added a config file with # and switched to // because of a failure somewhere)13:54
infinityProbably the kernel autoremove stuff.13:55
apacheloggerany has a clue why kde-runtime is blocked from proposed migration?13:58
apacheloggerexcuses says it's waiting for autopkgtest for libreoffice 1:4.2.3~rc3-0ubuntu2: RUNNING13:58
apacheloggeralthough that seems to have failed on a tooling exception13:58
mvo_infinity: indeed, the kernelautoremove stuff14:00
infinitymvo_: Right.  We switched to // thinking # wasn't supported, but that was probably because of the bug in aptdaemon. ;)14:01
mvo_infinity: yeah, that sounds likely :)14:02
mvo_infinity: anyway, thats fixed now and the other failures should be fixed too, one real bug in the py3 switching that probably prevernts dpkg from installing single debs, otherwise mostly missing dependencies etc14:02
ScottKpitti: Would you please have a look at the latest libreoffice test on i386.  Something seems to have gone horribly wrong in Jenkins.14:10
DevilsOwnHello14:11
DevilsOwnI wanted to ask something related to the ubuntu-server boot process14:12
DevilsOwnanybody here?14:12
dobeyDevilsOwn: #ubuntu-server is probably a better place14:17
dobeyDevilsOwn: and just ask, don't ask to ask14:17
DevilsOwnsure!14:20
bdmurraymvo_: Do you know what would call "pkexec /usr/lib/update-notifier/package-system-locked"?14:44
seb128bdmurray, http://ubuntu-codesearch.surgut.co.uk/search?q=package-system-locked suggests nothing14:46
bdmurrayseb128: hunh, that's a neat site14:47
seb128bdmurray, indeed ;-)14:47
bdmurrayI'd received and authentication dialog with the message that search returned.14:48
mvo_bdmurray: ubuntu-system-service might be14:53
dokomvo_, suggestion here from PyCon ... lp: #1306682  can we still do this for trusty?15:26
ubottuLaunchpad bug 1306682 in command-not-found (Ubuntu) "recommend to use python3 when python is not found" [Undecided,New] https://launchpad.net/bugs/130668215:26
jrwreni don't think python would ever not be found. ubuntu doesn't run without python, does it?15:27
xnoxjrwren: both python and python3 have been removed from minimal ubuntu core installations a few cycles back.15:29
jrwrenxnox: good to know. thanks.15:30
xnoxdoko: suggestion here from Greenwich, implement "from __future__ import unicode_str" =)15:30
dokoynox: that's backward =)15:32
jrwrenhahaha, on 12.04 if you try to remove python-minimal, apt sasy "You are about to do something potentially harmful."15:32
xnoxdoko:i think we do have overrides in command-not-found to do such a trick.15:33
xnoxjrwren: yes, it used to be marked essential, it isn't anymore. I think we did that in saucy or raring.15:33
jrwrenxnox: understood. I didn't mean to suggest you were wrong. 12.04 is definitely more than a few cycles back.15:34
jrwrenI just thought it was funny.15:34
jrwrenI've not seen that messge before.15:34
rbasakI don't follow how this would work. "python"..."apt-get install python3!"...ok...<does it>..."python"..."apt-get install python3!"...?15:34
superm1xnox: all the pieces look to have landed in the archive, can you respin the ISO now?15:34
rbasakhttp://www.python.org/dev/peps/pep-0394/ is pretty clear that python must mean Python 2.15:34
rbasakWell, "should"15:35
rbasakdoko: ^^?15:35
xnoxrbasak: it would work like this "did you mean python3 ? for python2 install package python"15:35
xnoxrbasak: we will not make python symlink to -> python315:35
rbasakxnox: ah. That makes sense :)15:36
dokorbasak, we only make aunannounced version updates with ruby =)15:36
rbasakdoko: yeah I didn't think that you were suggesting changing the symlink. I just wondered how the message would help without causing an infinite loop. xnox's explanation works for me :)15:37
dokorbasak, yep, just wanting to point people to the new version15:39
jhenkehi, somebody here know why the daily installer from today installed libc* packages in version 2.19-0ubuntu5 while the archive just has 2.19-0ubuntu4?15:40
infinityjhenke: Look again?15:40
infinity libc6 | 2.19-0ubuntu5      | trusty           | amd64, arm64, armhf, i386, powerpc, ppc64el15:40
infinityjhenke: If "the archive" only has ubuntu4, you're using an out-of-date mirror.15:41
jhenkeinfinity I had it query several time already15:41
jhenkenever had that before, was by default on mirror for germany15:42
jhenkemaybe somebody wants to check that?15:42
infinityjhenke: Looks like their upstream (ftp.uni-kl.de) hasn't updated for ~8h15:44
infinityjhenke: Probably just an temporary problem, but switching from de.archive.ubuntu.com to gb.archive.ubuntu.com for now would fix you up.15:45
jhenkewell yes I am not to the main mirror for the time being15:45
jhenkejust want to point out, so somebody might be able to tell the mirrors admin to check the machine15:46
mvo_doko: let me check15:46
infinityjhenke: Given the heartbleed madness this week, people taking services offline for a few hours here and there isn't something worth bugging them about, to be fair. :P15:46
infinityjhenke: (Heck, my local government has shut down half their public-facing services for a good chunk of the week)15:47
jhenkehmm, if you say so15:48
bdmurraydobey: I feel like the SRU for bug 1306225 could be more informative, like have a hyperlink to the announcement.15:53
ubottubug 1306225 in Ubuntu One Client "The user should be informed about the u1 file service shutdown" [Critical,In progress] https://launchpad.net/bugs/130622515:53
xnoxsuperm1: i don't have powers to respin, can you click the respin buttons on http://iso.qa.ubuntu.com/ ?16:09
xnoxsuperm1: or i'll poke ~cdimage people.16:10
xnoxsuperm1: i did upgrade in the livecd and tested and that worked fine.16:10
LaneyI don't think mythbuntu has that set up16:11
Laneysuperm1: xnox: you want one?16:11
xnoxLaney: yeah i want mythbuntu i386 and amd64 respin.16:11
Laneyok16:11
Laneythere16:12
xnoxLaney: ta!16:12
dobeybdmurray: a link where? you're welcome to discuss it with beuno and others. i didn't make the patch, i'm just doing the packaging for it16:14
beunodobey, to the blog post?16:15
beunothat seems to be what bdmurray is implying, which sounds sane16:15
dobeyyes, i was asking where in the UI16:16
dobeyubuntuone-client has no UI, and links in notifications is not a nice thing16:16
bdmurraybeuno: right, there is no easy way for people to find out why or get more information.16:18
xnoxdobey: beuno: i presume the control panel shows "error syncing" maybe that could show a better message after 1st of june e.g.  "these services are no longer offered"16:20
xnoxditto indicator-sync and/or cmd line utilities / logs that generated for syncing.16:20
dobeyxnox: i'd prefer to just not do anything16:21
dobeybdmurray: well we've already e-mailed every user we have16:21
xnoxdobey: fair enough =) if people ignore planet, blogs, news sites, emails they recieved and refunds..... it's really their own fault by 1st of june ;-)16:22
dobeyxnox: well then they can figure it out and go to the web site and download a zip of all their data or something16:22
bdmurraydobey: I think emails are easily missed, I don't recall receiving one and its not much work to point to the blog announcement is it?16:25
beunobdmurray, emails are still going out16:25
dobeybdmurray: it depends16:25
dobeybdmurray: and we couldn't do it in ubuntuone-client really, we'd have to put it somewhere there is actual UI16:26
=== bfiller is now known as bfiller_afk
=== happyaro1 is now known as happyaron
=== dpm-afk is now known as dpm
rohanhi: can someone familiar with xorg tell me if this bug needs any more info?18:07
rohanhttps://bugs.launchpad.net/ubuntu/+source/mesa/+bug/128286718:07
ubottuLaunchpad bug 1282867 in linux (Ubuntu) "Many bugs in rendering with lockups, likely caused by SNA" [High,Confirmed]18:07
rohanit is a pretty serious bug affecting intel GPUs on trusty18:07
dobeyyeah, don't file one bug that has a title of "many bugs" and don't say "likely caused by"18:07
dobeyassumptions aren't helpful18:08
rohandobey: and that was helpful how?18:11
rohandid you even bother looking at the bugs, and the logs provided?18:12
rohanand "many bugs" refer to graphical glitches, not launchpad bugs.18:12
=== roadmr is now known as roadmr_afk
dobeyno, because i'm not an intel driver developer, and the summary is not helpful to me even if i was18:13
dobeyyou asked for help, i gave you some helpful advice; if you don't like it, don't complain about it18:13
rohansorry, but how is the summary not helpful?18:13
dobeyyou're free to ignore if you wish18:13
rohanno, you crticised without giving any constructive advice.18:13
dobeyas i said, it says "many bugs" and it is making an assumption about the cause18:14
rohanthe summary has the dmesg log outputs, and the corresponding driver and package versions18:14
dobeywhatever18:14
rohanif i can do something to improve it, let me know, otherwise you are the one making assumptions :)18:14
rohanalso, my assumption was not stupid: disabling SNA actually fixed the problem, so it *is* likely caused by SNA18:15
dobeysigh18:15
dobeyi didn't say it was stupid18:15
=== bfiller_afk is now known as bfiller
dobeyi said it was an assumption18:15
dobeyand i don't have time to have a stupid argument about that with you18:15
rohandobey: well then, you needn't have bothered looking at it and giving unconstructive suggestions, right?18:19
rohandobey: in the spirit that you're a developer and probably know more than me, i've updated the title of the bug.18:25
pittidoko: yay, thanks18:32
pittiScottK: seems it's back to normal, jibel kicked it18:32
pittiScottK: and yes, jenkins is acting up a lot recently, mostly due to the unreliable ppc64el boxes; I'm working on a replacement system to avoid jenkins entirely :)18:33
ScottKpitti: Thanks.  I overrode it so we weren't blocked, but it's good to know it's working again.18:34
pittiScottK: yep, https://jenkins.qa.ubuntu.com/job/trusty-adt-libreoffice/333/ is the latest18:35
pitti334 is currently running18:35
pittiinfinity, smoser: can you please kick wolfe-05? ssh takes ages, and sudo reboot is just stuck; I guess the zero-page memory corruption bug wreaked enough havoc18:42
pittiI'd like to reboot, dist-upgrade, refresh the container, and put it back on jenkins duty18:42
infinitypitti: Lemme have a look.18:55
infinitypitti: 5, you say?18:56
infinitypitti: Should be back.18:58
=== saurik_ is now known as saurik
smoserthanks infinity .19:41
smoserpitti, did you want an entirely clean system ?19:41
=== roadmr_afk is now known as roadmr
=== beisner is now known as beisner-food
=== beisner-food is now known as beisner
jtaylorcjwatson: broken grub-probe with lvm snapshots should probably go into the release notes for 14.0420:23
jtaylordisabling the os-prober is probably a reasonable workaround for isntallation/upgrade20:24
jtaylorI had a short look at why it breaks, but this issue is over my head :/20:24
cjwatsonjtaylor: I uploaded the fix today ...20:25
jtayloroh20:25
cjwatsonanalysis in https://bugs.launchpad.net/ubuntu/+source/grub2/+bug/128743620:25
jtaylorI just tried it, though maybe I did apt-get upgrade to early20:25
ubottuLaunchpad bug 1287436 in grub2 (Ubuntu) "grub-probe fails in the presence of snapshot logical volumes" [Undecided,Fix released]20:25
jtaylorah my mirror hasn't got it yet20:26
jtaylorthanks20:26
jtaylorI'll test it when its available20:26
cjwatsonThanks.  I tested in a Xen guest with unstable20:26
jtaylorits pretty easy to reproduce if you are using lvm, just create a snapshot and do update-grub220:27
cjwatsonIt wasn't obvious - first two or three guesses as to the cause turned out to be entirely wrong when I actually compared against 2.00's behaviour in detail20:27
cjwatsonRight, that's what I did20:27
cjwatson2.00's grub-probe actually didn't understand snapshots properly either, it's just that grub-mkconfig didn't fail in that situation20:28
cjwatsonI was initially confused because it worked with upstream 2.00, but that's because an entirely different bug in grub-mount (patched in my packaging) masked it20:29
bdrungxnox: can you update app-install-data-ubuntu again for picking up the fixed sonic-visualiser (bug #1161283)?20:32
ubottubug 1161283 in sonic-visualiser (Ubuntu) "Problems reported with .desktop files for sonic-visualiser" [Medium,Fix released] https://launchpad.net/bugs/116128320:32
bulletxtthanks for updating cups 1.7.2 in ubuntu 14.0420:38
=== Ursinha is now known as Ursinha-afk
=== Ursinha-afk is now known as Ursinha
=== salem_ is now known as _salem
darkxstcan somebody upload bug 1300464 for me? it was ok'ed by infinity and cjwatson in -release22:50
ubottubug 1300464 in ubiquity-slideshow-ubuntu (Ubuntu) "Ubuntu GNOME Trusty Slides Update" [Undecided,Confirmed] https://launchpad.net/bugs/130046422:50

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