/srv/irclogs.ubuntu.com/2014/07/01/#ubuntu-devel.txt

=== TheMuso` is now known as TheMuso
=== timrc is now known as timrc-afk
=== FourDollars_ is now known as FourDollars
Noskcajinfinity, Could you merge wxwidgets3.0 some time soon please? One of my merges is waiting for the new debian release03:48
=== saurik_ is now known as saurik
ikepanhc@pilot in04:54
=== udevbot_ changed the topic of #ubuntu-devel to: Archive: open | Devel of Ubuntu (not support or app devel) | build failures -> http://qa.ubuntuwire.com/ftbfs/ | #ubuntu for support and discussion of lucid -> trusty | #ubuntu-app-devel for app development on Ubuntu http://wiki.ubuntu.com/UbuntuDevelopment | See #ubuntu-bugs for http://bit.ly/lv8soi | Patch Pilots: ikepanhc
=== juliank_ is now known as juliank
pittiGood morning06:35
rsalvetidiwic: thanks for sending the patch, I sent a similar one last friday but it still needs to be approved it seems (got blocked by the ml)06:38
rsalvetiso thanks for taking care of that06:38
diwicrsalveti, aha, didn't know that06:38
rsalvetiit just accepts patches from subscribed members, was planning to send it again after subscribing with my canonical email, but you did that already06:39
rsalvetiso all good06:39
diwicrsalveti, I talked to tanuk upstream last Friday and he seemed to prefer changing PA to 200 rather than changing rtkit06:39
rsalvetiyeah, I saw that, that's why I decided to push that patch forward06:40
diwicok06:40
=== dbarth-afk is now known as dbarth
dholbachgood morning07:20
=== directhe` is now known as directhex
doko_tvoss, what's the status of the phone/4.9 transition?07:47
tvossdoko_, in silo, not done yet07:48
doko_tvoss, so no defaults change yet?07:48
tvossdoko_, nope, sorry for that. Getting the upstreams aligned is more difficult than anticipated07:48
sil2100dholbach: hi! A kind reminder to add me to the patch pilot calendar ;)08:14
dholbachsil2100, I can't believe I missed to do that08:15
dholbachsil2100, I'll take care of it once this call is finished :)08:15
sil2100Thanks ;)08:16
=== vrruiz_ is now known as rvr
=== lisca_ is now known as lisca
xnoxwgrant: thanks. didn't recognise corrupt WADL cache. for pkg_importer is that in just that users ~/.launchpadlib/api.launchpad.net/cache, and since it was etree parse error on incomplete tokens some wadl+xml files would be incomplete/invalid there.08:53
xnoxhm. shouldn't wadllib invalidate and redownload representations, if it fails to load them?!08:53
brendandmvo, cjwatson - you made it to the coverage dashboard :) http://162.213.34.64:8080/gaps/project/click/08:55
mvobrendand: yeah08:56
mvobrendand: I added the smoke test for building apps as part of the integration test too btw https://code.launchpad.net/~mvo/click/test-build-core-apps/+merge/22502208:57
cjwatsonbrendand: Cool, thanks.  Don't know if you noticed but the most recent entry there has coverage of the C code as well08:58
brendandcjwatson, i did, thanks!08:59
cjwatsonJust need to fix a few details per Francis08:59
brendandcjwatson, i think the last step before we go green is for you to have a look at the numbers and see if there's anywhere coverage can be improved08:59
brendandcjwatson, being sensible about it of course, so not just improving coverage for the sake of it09:00
cjwatsonYeah, I'm sure there are a few places09:00
ochosihi bregma, as one of the light-locker uploaders, would you mind to take a look at this (packaging) issue? https://bugs.launchpad.net/ubuntu/+source/light-locker/+bug/133606509:01
ubottuUbuntu bug 1336065 in light-locker (Ubuntu) "Does not fail gracefully when lighdm isn't running" [Undecided,New]09:01
wgrantxnox: yup, i just stopped, deleted, started, then requeued the failures09:02
mvobrendand: the big one that needed improvement was the chroot coverage but that improved quite a bit09:03
mvobrendand: but we will check for more :)09:03
dholbachsil2100, better? ;-)09:16
* sil2100 looks09:16
sil2100dholbach: better ;) Thanks!09:19
dholbach:-D09:19
cjwatsonrbasak: Would you mind merging ming?  It's one of the last three automake1.10 users09:20
cjwatsonrbasak: (Now the last one left.)09:47
dokomvo, python-apt ping09:56
jibelmvo, just a confirmation, for bug 1311396 it's a translation fix and there is no change for this in u-r-u or update-manager, correct?10:07
ubottubug 1311396 in update-manager (Ubuntu Trusty) "broken translations results in traceback in new release notification" [Undecided,Confirmed] https://launchpad.net/bugs/131139610:07
dokotvoss, is duflu on irc?10:12
tvossdoko, in #ubuntu-mir10:12
tvossdoko, but he might well be offline10:12
apwis compiz crashing on login on utopic a known issue10:13
dokotvoss, ok, added my comment to the merge proposal (I think this is bikeshedding now)10:14
tvossdoko, thanks10:14
apwjodh, hey ... upstart as a user session, where are log job logs for those jobs ?10:19
jodhapw: $XDG_CACHE_HOME/upstart/*.log (documented in init(5)).10:19
jodhapw: or ~/.cache/upstart/ if var not set.10:20
jibelmvo, bdmurray langpacks for ja, id, eo and ug must be refreshed and SRUed in Precise in order to verify 1311396.10:21
apwjodh, great thats them, trying to work out why unit7 is just dieing in a heap on me10:21
seb128Noskcaj, dholbach: hey, your libgtop update has a soname change without binary renaming, making rdepends not start, including unity :/10:32
seb128shrug10:33
apwok ... yeah my desktop seems to be blank, because unity7 no longer starts, becase bamf-daemon doesn't start, because it depends on libgtop2-7, and10:33
apwthe latest migration replaces that with .1010:34
seb128Noskcaj, dholbach_: is one of you around to sort that out? we need to get that resolved before more people upgrade and get a non working desktop, just wondering who should look at fixing it10:35
alexbligh1I am trying to debug why debootstrap is doing something slightly different making a trusty image to a precise image. The symptom I'm seeing is /dev/net/tun does not exist after boot, yet it is mentioned in /var/log/udev, and udevadm trigger --action=add does not actually create it. Any ideas how /dev/net/tun is meant to be mknod'd and how to debug?10:36
Wiziledoi need to know as a ubuntu user why nzoom.com tv on demand doesnt work10:36
seb128k10:37
darkxstWiziledo, try #ubuntu10:37
Wiziledothey won't give me voice because its not important as they put it10:37
seb128I'm going to revert libgtop, fixing it the proper way would mean blocking the fix until all the rdepends are ported to 1010:37
seb128Laney, ^ does that make sense to you?10:37
darkxstWiziledo, this channel is for development of ubuntu, not user support10:38
Wiziledook10:38
Wiziledosorry :)10:38
Wiziledodoes mark shuttleworth come here?10:38
Laneyseb128: that seems fair10:38
seb128k, doing that10:39
seb128Laney, thanks10:39
xnoxWiziledo: he holds qa sessions often during ubuntu online summits and user days, i don't think there is one comming up shortly, as we just finished online summit.10:39
Wiziledosweet, i look up to that guy10:41
Wiziledoi get what he is doing10:41
seb128Noskcaj, dholbach_: I uploaded a revert10:44
* Laney adds a hint to make it skip testing and get faster to release10:44
WiziledoQuestion, is ubuntu literally made in this channel?10:45
cjwatsonMost of the main developers are here, yes10:46
apwseb128, and confirmed just reverting that library restored my desktop10:46
seb128k10:47
seb128https://launchpad.net/ubuntu/+source/libgtop2/2.30.0.is.2.28.5-0ubuntu1 is building10:47
apwseb128, will test it once it has built10:47
seb128thanks10:47
mvojibel: hm, is that somethat that pitti or dpm can help with? the langpack refresh for precsie for bug 1311396 i mean10:48
ubottubug 1311396 in update-manager (Ubuntu Trusty) "broken translations results in traceback in new release notification" [Undecided,Confirmed] https://launchpad.net/bugs/131139610:48
WiziledoAre all developers paid?10:48
cjwatsonNo10:48
milissahttp://adf.ly/pyduc10:49
Wiziledoanyway I'm from NZ I'm a big fan of ubuntu just want to get our national TV website tv on demand to work lol10:50
Wiziledoyoutube works, just not national tv on demand10:51
WiziledoAlso the desktop is dying10:53
darkxstWiziledo, many of the devs, like me are volunteers10:55
darkxstWiziledo, and as I said, this is not the channel for user support10:55
Wiziledooh thats awesome10:59
xnoxWiziledo: there is a NZ ubuntu LOCO group http://loco.ubuntu.com/teams/ubuntu-nz/ there is an IRC channel #ubuntu-nz maybe you can find out more about the national TV website working on ubuntu there?11:05
pittimvo, jibel: yes, I can help with the SRU, if the translations fixed on LP for precise/saucy11:22
=== MacSlow is now known as MacSlow|lunch
bluesabreHello sponsors, is anybody available to upload these packages to trusty-proposed to begin SRU verification?11:46
bluesabrehttps://bugs.launchpad.net/ubuntu/+source/lightdm-gtk-greeter/+bug/133187111:47
ubottuUbuntu bug 1331871 in lightdm-gtk-greeter (Ubuntu) "[SRU] Please backport lightdm-gtk-greeter 1.8.5 to trusty" [Undecided,New]11:47
bluesabrehttps://bugs.launchpad.net/ubuntu/trusty/+source/menulibre/+bug/132340511:47
ubottuUbuntu bug 1323405 in menulibre (Ubuntu Trusty) "[SRU] Please backport menulibre-2.0.4 to trusty" [Undecided,New]11:47
jibelpitti, I checked indonesian and the fix is not in LP, so this SRU is invalid.12:00
=== dholbach_ is now known as dholbach
dholbachseb128, sorry - I was out for lunch12:06
dholbachseb128, thanks a lot for your work on this - I'm sorry I didn't notice this when sponsoring :(12:15
seb128dholbach, hey, no worry !12:17
=== psivaa_ is now known as psivaa-lunch
dholbachseb128, IIRC that's not the first time for them, right?12:19
seb128dholbach, you mean not the first time Noskcaj has updates with issues?12:21
dholbachno, libgtop not bumping the soname12:21
seb128oh12:21
seb128dholbach, they did bump the soname, Noskcaj just didn't rename the binary in the packaging to match the version12:22
dholbachohhhhh ok, then indeed I should have noticed this :-/12:22
seb128like they did .7 -> .10 and the lib is still named lib7 and ships 1012:22
mlankhorstdpkg -L libosmesa6 ;-) (on utopic, for historic reasons)12:24
=== MacSlow|lunch is now known as MacSlow
bluesabrehey dholbach, thanks for the endorsement :)12:25
dholbachbluesabre, anytime12:26
bluesabreif you get a free moment, do you think you can upload the lightdm-gtk-greeter and menulibre packages I have in the sponsor queue?12:26
bluesabre(trusty-proposed_12:26
dholbachbluesabre, I'm afraid I'm quite busy right now12:28
dholbach^ can anyone else help out a bit with sponsoring?12:28
bluesabreok, np. gotta run to work now, be back tonight12:30
=== _salem is now known as salem_
shadeslayermvo: ping, I was wondering what the status of appstream is in ubuntu12:57
shadeslayerand whether USC and Synaptic are going to move to it12:57
=== psivaa-lunch is now known as psivaa
mpt“Du har ikke lov til at redigere denne side.”14:02
mptI thought MoinMoin had stopped giving me error messages in random languages in 200914:03
juliankmpt: It just wanted to have some fun :)14:05
=== ogra_` is now known as ogra_
ev_dobey: have you come across any good solutions for doing checkout rather than export in Tarmac's verify_command plugin, short of writing a new plugin? I'm just surprised no one has run into this as an issue before. Surely we can't be the only people populating version information from bzr?14:54
=== dpm__ is now known as dpm
dobeyev_: what issue?14:58
ev_dobey: that tarmac does bzr export for verify_command, rather than bzr branch or bzr co. Our setup requires the branch information to be present, since we send that as part of a deployment of this code. Our tests then validate that.14:59
ev_I could just bzr init as a hack around it, I suppose14:59
dobeyev_: tarmac used to just run the tests in the checked out branch, but it was changed to doing an export to a temp dir for security, so that rogue branches can't commit themselves and such14:59
ev_yeah, I can understand that motivation14:59
ev_yeah, this bzr init should work15:01
ev_sorry for disrupting you :)15:01
dobeyev_: i think if the build system/tests requires .bzr to exist with valid data, then the build system/tests are broken. i'd have to look at what you're doing exactly to give better feedback/recommendation for how to do it better though.15:01
ev_well, it's driving our integration tests15:02
ev_which validate the deployment, which puts copies of the code under directories with the revno in the name15:02
ev_so that we can switch between a few on production15:03
ev_right now I have it running both the deployment and the tests under tarmac15:03
ev_deployment to a temporary location, obviously :)15:03
dobeyi totally understand why you'd want to do that in an actual deployment15:03
dobeyhmm15:03
ev_I don't think IS would like us doing CD with Tarmac15:04
ev_sorry, continuous deployment. Hate acronyms.15:04
ev_yeah, I'll hack around it for now15:04
dobeyi guessed that :)15:04
ev_we don't have any tests that validate the bzr revno matches the other side15:04
ev_so I can just fix this once we do :)15:04
ev_though I guess we could just validate that as 015:05
dobeyev_: if you want to file a bug against the project this is for, and sub me to the bug, i can try to look at it sometime when i have some free time (which i have extremely little of lately)15:06
ev_sure, will do15:07
* dobey wonders who to ping about debugging upstart jobs15:07
dobeyhmm, or maybe i can just stop the job and run it manually under gdb and it'll just do what i want to do15:08
alexbligh1I'm having problems with systemd-udevd on 14.04 that didn't occur with udevd on 12.04. This is a debootstrap'd image. It appears not to be creating device nodes (specifically /dev/net/tun). As far as I can tell udev is matching the appropriate rule, but the node is not being created. udevadm test produces what I think is the expected result, but udevadm trigger does nothing. Details here: http://pastebin.com/4U15:13
alexbligh1CGmXWx Any ideas?15:13
alexbligh1Argh, URL is http://pastebin.com/4UCGmXWx15:13
Unit193ari-tczew appears to have nuked the entire Ubuntu history in a changelog, and improperly re-enabled a feature by missing a build-dep: https://launchpadlibrarian.net/177154250/lightdm-gtk-greeter_1.8.4-0ubuntu1_1.8.5-1ubuntu1.diff.gz15:21
elopioHello!15:33
elopioI need a core-dev to review a packaging change, here: https://code.launchpad.net/~elopio/address-book-app/qmltest1/+merge/22126315:33
=== timrc-afk is now known as timrc
xnoxelopio: looks good, can i use qmltestrunner against system installed address book app?15:45
xnoxelopio: are qmltests executed in jenkins with results publised on ci.ubuntu.com? if yes, then where?15:46
xnoxI wonder how to make that qmltests to be executed as dep8 test as well.15:46
elopioxnox: you can't run against the installed because the qml import paths I'm using are like:15:47
elopio../../src/imports/ContactEdit15:47
=== tkamppeter_ is now known as tkamppeter
elopioxnox: and the tests are executed by jenkins on MPs, but not during the smoke tests, so not on ci.ubuntu.com15:48
elopioxnox: I first tried making it an autopkgtest, but pitti recommended to it on the dh_auto_test15:49
elopioxnox: on dep8, wouldn't it be better to run the autopilot tests?15:49
xnoxelopio: they way you did, it must be in dh_auto_test.15:53
xnoxelopio: dep8 would be in-addition, running against system app to satisfy the removal of the autopkgtests which are currently published on ci.ubuntu.com, until qmltests are also published there.15:53
elopioxnox: I added an autopilot test that will be on ci.ubuntu.com and test the creation of a new contact.15:54
elopioso I'm just removing the tests that are just duplicating some things from that more general test15:55
elopioif on qml tests we check all the combinations that will enable the save button, on autopilot we can just assume that the save button will be enabled.15:56
elopiobut it will fail anyway if for some weird reason, this works while building the package but not while running the autopilot tests.15:56
xnoxelopio: i see.15:59
=== jsalisbury_ is now known as jsalisbury
=== alexisb is now known as alexisb_afk
argesslangasek: hey... trying to figure out bug 1274444. Do you know if there is a simple way to ensure that kernel messages with <N> n>=12 are logged with rsyslog/whatever? $ConsoleLogLevel 14 doesn't seem to do anything17:57
ubottubug 1274444 in linux (Ubuntu) "echo string to /dev/kmsg fails to appear on /var/log/syslog" [Medium,In progress] https://launchpad.net/bugs/127444417:57
=== timrc is now known as timrc-afk
=== dpm is now known as dpm-afk
slangasekarges: I really don't know, sorry18:08
slangasekarges: maybe pitti knows the interfaces here?18:08
argesslangasek: figured id ask before digging even more. i've been playing with the settings but not being very successful18:09
smoserinfinity, around ?18:49
smosersome folks are interested in getting a d-i build with linux-keystone for trusty18:50
smoser https://launchpad.net/ubuntu/+source/linux-keystone18:50
smoserexpecting that to land at18:50
smoser https://code.launchpad.net/~smoser/maas/maas-ephemerals-v218:50
smoserer..18:50
smoser http://ports.ubuntu.com/ubuntu-ports/dists/trusty-proposed/main/installer-armhf/current/images/18:50
smoseri'm not familiar with how that request is made or how that is done.18:51
smoseroh. and its canada day, and you're not here. rightfully so.18:51
smoserslangasek, ^ ?18:51
stgrabersmoser: do you have a d-i branch adding support for that?18:53
smoser"i'm not familiar with how that request is made or how that is done."18:53
smoser:)18:53
smoserso, no.18:54
smoserdannf, maybe has done that in the past?18:54
stgrabersmoser: adding a new ARM board/platform to d-i isn't extremely difficult (we reworked some of that last cycle to be simpler) but you typically want to send a merge proposal for d-i adding the required dependencies and code to generate a bootable image for your target and you likely also need a matching change to flash-kernel and the libdebian-installer18:54
stgraberI'm not familiar with this specific target, so could be that some of those bits have been done but not others18:55
smoserstgraber, thanks.18:56
stgraberall of that is obviously better done by someone who has access to the actual hardware as all of those bits require some knowledge about how the bootloader is setup, what kind of storage it uses and its devicetree dtb file (and whether it needs to be appended to the kernel or resides in flash)18:57
=== dpm is now known as dpm-afk
dannfsmoser: manjo has MPs for that iirc - at least for the components, if not for d-i itself18:58
dannfstgraber: ^18:58
smoserdannf, thanks18:58
smosermanjo, ^18:58
smoseri just hit 'send' on the mail that started me bothering people. i'm going to be gone soon for a few days.18:58
dannfwell, i don't see a d-i MP - but i know he's produced one for testing18:59
manjosmoser, I have a d-i building in a PPA for keystone ... I can submit to you ... install images are in http://ppa.launchpad.net/marcola-team/ppa/ubuntu/dists/trusty/main/installer-armhf/current/images/keystone/netboot/19:00
dannfmanjo: can you propose an MP for d-i? infinity said he'd take a look at MPs by next week, i owe him a list19:01
manjodannf, ack19:01
=== alexisb_afk is now known as alexisb
=== dpm-afk is now known as dpm
robert_ancellkentb, do you know the correct URL for efibootmgr?20:07
robert_ancellIs it https://github.com/vathpela/efibootmgr?20:07
kentbrobert_ancell, yep. that's  it.20:08
robert_ancellkentb, ta20:08
kentbnp20:08
=== salem_ is now known as _salem
directhexinfinity, any idea on ppc64el mono status? upstream indicate they haven't seen anything relating to it20:27
directhex(which could be an error)20:27
=== tgm4883__ is now known as tgm4883
infinitydirecthex: I'll have to talk to the IBMers about where their upstream submissions have (or haven't) gone.21:12
directhexinfinity, thanks21:13
infinitydannf: If you get me d-i MPs (libdi and d-i itself), I can blind review them.  I've done more than enough new platforms in d-i by now to know if it'll work without even trying (that might be a bit of a sad statement...)21:14
dannfmanjo: ^21:15
dannfthe libd-i one is there- https://code.launchpad.net/~manjo/ubuntu/utopic/libdebian-installer/HP-m800 - https://code.launchpad.net/~manjo/ubuntu/trusty/libdebian-installer/HP-m80021:16
manjoinfinity, ack21:16
manjoinfinity, there is an MP For libdi already21:16
manjoinfinity, I will MP di soon21:16
infinitydannf: Great.  Add it to that email full o' MPs that I asked for, so I don't lose it in holiday backscroll. ;)21:17
dannfinfinity: will do21:17
infinity<321:17
LeonBoI've debbuild the libav package to add some extra encoders21:32
LeonBoThe original version is 6:9.13-0ubuntu0.14.04.121:32
LeonBoMy version is 6:9.13-0ubuntu0.14.04.1~ppa121:32
LeonBoWhen installing the ~ppa1 version I get the warning: 'dpkg: warning: downgrading libavcodec54:amd64'21:33
infinityLeonBo: That makes your version lower than the archive version.21:33
LeonBoYeah I thought so21:33
infinityLeonBo: "~" means "just before".  As in, -1~foo is "just before -1"21:33
LeonBoOK21:34
infinityLeonBo: You might want something like -0ubuntu0.14.04.1+ppa121:34
LeonBoShould I just use +ppa1 then?21:34
LeonBoinfinity: you were quicker21:34
LeonBo:)21:34
infinityLeonBo: Or -0ubuntu0.14.04.1+leon1 so it's obvious to people where it came from.21:34
LeonBoinfinity: awesome, I'll use that then21:34
LeonBoAnd is there a way to keep my version always higher?21:35
LeonBoSo If 9.14 of the package is released, my version isn't automatically removed21:35
infinityLeonBo: Yeah, you could bump the epoch, but you may well not want that.21:36
infinityLeonBo: Other packages that have versioned deps will break a bit in that case, and I wouldn't recommend it.21:37
infinityLeonBo: Since you're building a package that *is* 6:9.13, if something in the archive depends on >> 6:9.14, you don't want you package to satisfy the dep by accident, you want to update your package.21:37
LeonBoYeah, you're right21:38
LeonBoSo release my + version21:38
LeonBoAnd then pin it?21:38
infinityLeonBo: That's the saner way to go, yeah.21:38
infinityLeonBo: Then when apt starts whining about "Not upgrading 4 packages", you know you need to update your PPA. ;)21:39
LeonBoExactly21:39
LeonBoThanks a lot!21:39
ari-tczewcjwatson: I guess there's existing problem again with M-o-M. It hasn't been updated since a few hours. "Generated at 2014-07-01 15:43:30 UTC."22:26
Unit193ari-tczew: You seem to have nuked the Ubuntu changelog of lightdm-gtk-greeter (Used to be an Ubuntu package, later Debian picked it up.)  As well as missing a build-dep on ido.22:39
slangasekarges: bug #1274444> what do you mean, "pending merge"?22:46
ubottubug 1274444 in rsyslog (Ubuntu) "echo string to /dev/kmsg fails to appear on /var/log/syslog" [Medium,In progress] https://launchpad.net/bugs/127444422:46
cjwatsonari-tczew: thanks, hopefully fixed22:47
ari-tczewUnit193: I've dropped the whole changelog before initial release of Debian because I think it's not needed so much keeping a lot of changes in d/changelog as there are explanation of delta in newer releases.23:02
ari-tczewUnit193: B-D on libido3-0.1-dev is dropped because the reason of that addition wasn't mentioned in d/changelog and package builds fine without this one.23:03
Unit193It'll build fine without libindicator-dev too. :P23:04
ari-tczewUnit193: can I ask, is there any special reason to that investigation?23:04
Unit193ari-tczew: Normally some indicators need that, icon related.23:11
Unit193By default it's enabled and checks for it.23:11
ari-tczewUnit193: if this B-D is really needed, please fill a bug, attach a patch and subscribe to ~ubuntu-sponsors.23:14
* Unit193 sighs.23:17

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