[00:06] there is no libqt5opengl5-gles-dev in the armhf repo [00:06] so yeah, it's only x86 stuff [00:06] is there any way to simulate pkgstriptranslations (as it acts on the buildds) in pbuilder? [00:07] sergio-br222: what RAOF was saying is that libqt5opengl5-dev on armhf *is GLES only*. [00:07] sergio-br222: and that the packaging layout is just really confusing for $reasons [00:07] heh [00:07] OK, now I'm understanding [00:07] $reasons, heh [00:08] RAOF: might be something to fix :) [00:08] RAOF: since the reasons seem fairly shallow === anthonyf is now known as Guest70961 [03:55] The awesome software is published, You can use the trial version of Sticky Notes. [03:55] http://korsoftware.com === Guest97037 is now known as Tm_T === Guest50701 is now known as Tm_Tr === Guest47071 is now known as pitti [04:53] Good morning === pitti is now known as Guest48990 === Guest48990 is now known as pitti_ [06:23] What's the best way to get a patch into the fbi package? It's just a one liner so that it actually picks up the correct jpeg lib and doesn't just abort on run === geser_ is now known as geser === ttx` is now known as ttx [06:46] good morning === sidi is now known as Guest51001 === mthaddon` is now known as mthaddon [07:56] morning === jamespag` is now known as jamespage === davmor2_ is now known as davmor2 === sidi is now known as Guest47548 [08:43] Laney, I see you're piloting sometime today. Please may I kindly request you take a look at the package updates: [08:43] https://bugs.launchpad.net/ubuntu/+source/ubuntu-mate-artwork/+bug/1466521 [08:43] Ubuntu bug 1466521 in ubuntu-mate-artwork (Ubuntu) "ubuntu-mate-artwork 0.4.9 new nelease [debdiff attached]" [Undecided,New] [08:43] https://bugs.launchpad.net/ubuntu/+source/ubuntu-mate-settings/+bug/1466785 [08:43] Ubuntu bug 1466785 in ubuntu-mate-settings (Ubuntu) "ubuntu-mate-settings 0.4.6 release [debdiff attached]" [Undecided,New] [08:48] infinity: good morning, may I ping you to approve this x11-utils SRU that's in the precise queue? https://bugs.launchpad.net/ubuntu/+source/x11-utils/+bug/1463663, https://launchpad.net/ubuntu/precise/+queue?queue_state=1, https://wiki.ubuntu.com/StableReleaseUpdates#Publishing [08:49] Ubuntu bug 1463663 in x11-utils (Ubuntu) "[SRU] Enable setting property of type UTF8_STRING" [Medium,In progress] === alkisg is now known as work_alkisg [09:13] pitti_: libkdegames seems stuck on a test server error, can you retry it? https://jenkins.qa.ubuntu.com/job/wily-adt-libkdegames/lastBuild/ARCH=amd64,label=adt/console [09:14] t/win 23 [09:18] Riddell, hey, did you notice bug #1461987? [09:18] bug 1461987 in kpackage (Ubuntu) "package kpackagetool5 5.9.0-0ubuntu1 failed to install/upgrade: trying to overwrite '/usr/share/kservicetypes5/kpackage-packagestructure.desktop', which is also in package libkf5package5:i386 5.9.0-0ubuntu1" [Medium,Confirmed] https://launchpad.net/bugs/1461987 [09:18] seb128: hmm I did not, thanks [09:18] Riddell, yw! === yecril71pl1 is now known as yecril71pl === rbasak_ is now known as rbasak === gammax90 is now known as gammax === anthonyf is now known as Guest35024 === MacSlow is now known as MacSlow|lunch === marcusto_ is now known as marcustomlinson [11:30] LocutusOfBorg1: debfx: have you seen any problems with cmake 3.2 and CMAKE_INSTALL_LIBDIR? qtmir has started FTBFSing because it doesn't point to the multiarch libdir any more. [12:03] Laney: I don't know how it worked before but -DCMAKE_INSTALL_PREFIX=$(TMP1_DIR)/usr/ is just wrong === pitti_ is now known as pitti === pitti is now known as Guest84030 === Guest84030 is now known as pitti_ === sidi is now known as Guest53135 [12:14] Riddell: I doubt that this helps -- the -proposed version has failed consistently 8 times in a row, it's a real regression (in code or test) [12:15] pitti_: for which one? [12:15] Riddell: libkdegames [12:15] ah libkf5kdegames6 [12:17] mm I think that's just not been changed since the kdelibs4 version so of course it's wrong [12:17] debfx: this seems weird === owlman_ is now known as owlman === sidi_ is now known as Guest55217 [12:29] Laney: cmake uses multiarch paths only when the prefix is /usr, except when you patch it too much. [12:30] so I'd suggest to a) fix the cmake invocation in qtmir [12:30] b) fix the cross compile patch to not alter behavior in the default case [12:30] c) install MultiArchCross.cmake to the correct dir === coreycb` is now known as coreycb === MacSlow|lunch is now known as MacSlow [13:03] good morning! [13:04] hey cyphermox, how are you? [13:04] good, you? [13:04] did you have good holidays? [13:04] I'm good thanks ;-) [13:04] yep :) [13:17] Previously, I was able to upload to a release pocket in my ppa other than the one specified in the changelog via "incoming = ~r0lf/ppa/ubuntu/trusty" in dput.cf [13:17] Is that no longer supported? It was very nifty. [13:22] Laibsch: Should still work, but I can't see any evidence of you having tried that recently in the logs ... [13:23] Laibsch: Though the preferred form now is ~r0lf/ubuntu/ppa/trusty; the earlier form is deprecated [13:25] cjwatson: it was rejected by dput, I guess. It could be that this is because it's debian's dput in this case === dpm_ is now known as dpm [13:26] Laibsch: I don't see why dput would care. Do you have a transcript? [13:26] cjwatson: http://paste.debian.net/252618/ [13:27] Laibsch: That has nothing to do with the incoming directory. See line 28. [13:27] dsc-upload-$target is a script I wrote for uploading to ppa: http://paste.debian.net/252619/ [13:28] "Error: uploading files for distribution UNRELEASED to trusty not allowed." [13:28] $ dpkg -L dput | xargs grep -s 'not allowed' [13:28] /usr/bin/dput: raise dputhelper.DputUploadFatalException("Error: uploading files for distribution %s to %s not allowed."%(distribution, host)) [13:29] well, but I am uploading to my PPA. At least that is what i am trying to do and what worked until a few days ago. [13:29] well, but you need to read the error message [13:29] it's not relevant where you're uploading to [13:30] except in that the host definition in dput's config has an allowed_distributions regex which is checked [13:30] but UNRELEASED normally simply means what is says; I'd recommend running dch -r first [13:30] *what it says [13:30] releasing things (to PPA or otherwise) which are marked UNRELEASED is terribly confusing and should be avoided [13:31] yes, I now checked /etc/dput.cf and this indeed seems to be Debian-specific [13:32] checking for UNRELEASED [13:32] it's not Debian-specific [13:32] well, I am working on releasing the package officially and just making some testing packages available now [13:32] the default allowed_distributions in Ubuntu's dput.cf for all hosts is (?!UNRELEASED) [13:32] how about HALFRELEASED ;-) [13:32] I care not what you write there, but dput's config does [13:33] apparently, but IIRC this used to work when I did this kind of stuff in now-EOL lucid [13:33] anyway, the root of the problem was found. thanks! [13:33] that was rather a long time ago. anyway dput's config is sensible and if I were you I'd improve your practices to be less confusing :) [13:34] this dput change appears to predate lucid, see dput 0.9.2.30's changelog [13:34] but hard to debug something that long ago [13:36] jamespage: hi. why did you upload intermediate broken versions of unittest2/testtools to wily, rather than the latest versions? [13:36] debfx: looks like (b) is enough ;) [13:36] I saw " * Fix install path of MultiArchCross.cmake to be cmake-3.0." but didn't twig that this was for 3.0 and not 3.2 [13:36] dobey, that was an interim step towards newer versions, without the requirement for two new bits of packaging right now [13:36] so discounted it as a possible breakage source [13:39] jamespage: well it seems that it breaks @testtools.skip decorator, causing previously working tests to fail in autopkgtests; and it seems wily has python 3.5, so the latest versions could probably work without the other new packages in wily, by depending on 3.5? [13:39] dobey, urgh - sorry - obviously un-intended [13:42] jamespage: any idea when this can be fixed? https://jenkins.qa.ubuntu.com/job/wily-adt-unity-scope-click/lastBuild/ARCH=amd64,label=adt/console for a reference to see the skip failure [13:43] dobey, looking now [13:51] mvo, online? [13:53] doko: sprinting, sort of online === smoser` is now known as smoser === sidi is now known as Guest78284 === sidi_ is now known as Guest57893 === sidi is now known as Guest22933 === mfisch` is now known as mfisch === mfisch is now known as Guest82290 [14:39] @pilot in === udevbot changed the topic of #ubuntu-devel to: Archive: wily open | Devel of Ubuntu (not support or app devel) | build failures -> http://qa.ubuntuwire.com/ftbfs/ | #ubuntu for support and discussion of lucid -> utopic | #ubuntu-app-devel for app development on Ubuntu http://wiki.ubuntu.com/UbuntuDevelopment | See #ubuntu-bugs for http://bit.ly/lv8soi | Patch Pilots: Laney [14:40] dobey, looks like some sore of change in behaviour - even with latest I still see the same issue with the scopes test [14:42] dobey, still looking [14:43] * dholbach hugs Laney [14:43] jamespage: hmm, i am seeing the tests passing with what's in ppa:dobey/testtols (installed the trusty versions on top of wily) [14:43] dobey, ok - I have a fix - but its in the scope-harness package [14:44] dobey, ScopeHarnessTestCase needs to inherit from testtools.TestCase - currently it goes off unittest.TestCase [14:44] this may be a bug in testtools - but that at-least gets you past this problem [14:45] hmm, ok [14:45] dobey, I'll raise a bug and see what upstream think [14:46] jamespage: i guess we can land that change. i don't have a problem with it [14:47] it seems like an unnecessary dep though, since it doesn't seem python3-scope-harness depends on testtools for anything else [14:50] dobey, https://bugs.launchpad.net/testtools/+bug/1467558 [14:50] Ubuntu bug 1467558 in testtools "testtools.skip decorator not functional with unittest.TestCase as base class" [Undecided,New] [14:51] oh [14:52] jamespage: i guess the issue is because the decorator is on setUp there? [14:52] quite possibly [14:52] I think its the mixing of unittest/unittest2 which is causing the problem - this may have worked by accident before [14:53] jamespage: i'm re-running the jenkins test builds on https://code.launchpad.net/~dobey/unity-scope-click/more-harness-tests/+merge/255211 now. i'm at least not seeing the problem there, with the testtools from my ppa; we'll see what happens with proposed [14:54] ok [14:54] dobey, apologies for the break [14:55] jamespage: well, if it works ok with this branch, i'll fast track it through landing to get things back on course [14:57] oh, that jenkins doesn't build with proposed it seems === wendar_ is now known as wendar === pitti_ is now known as pitti === pitti is now known as Guest85442 === Guest85442 is now known as pitti_ === sidi_ is now known as Guest73040 [15:24] jamespage: looks like the issue is gone with that branch and the testtools in proposed, so I'll try to get it pushed through asap. [15:25] jamespage: so i guess the issue is that it doesn't like the decorator on setUp() in newer versions for some reason [15:32] dobey, looks that way [15:32] does anyone know how often the component mismatches report is mean't to refresh? [15:32] http://people.canonical.com/~ubuntu-archive/ [15:32] last refresh at 0850 ish today - I was hoping that my junit4 upload this morning might have made that a little shorter [15:42] Laney, sorry I was afk, seems you already have the solution :) [15:43] LocutusOfBorg1: ya, merge mistake [15:44] I moved the offending file up a bit in .install and added a comment but I worry this won't be enough at the next merge [15:47] Laney, I guess it will be enough if I prepare the next merge by myself (at least I hope to) [15:47] :) === Guest82290 is now known as mfisch === jdstrand_ is now known as jdstrand [16:35] infinity: any chance of SRU review for docker.io please? [16:36] Maaaaaybe. [16:54] infinity: thanks a lot :) [17:03] @pilot out === udevbot changed the topic of #ubuntu-devel to: Archive: wily open | Devel of Ubuntu (not support or app devel) | build failures -> http://qa.ubuntuwire.com/ftbfs/ | #ubuntu for support and discussion of lucid -> utopic | #ubuntu-app-devel for app development on Ubuntu http://wiki.ubuntu.com/UbuntuDevelopment | See #ubuntu-bugs for http://bit.ly/lv8soi | Patch Pilots: [17:03] TBC === _salem` is now known as _salem === highvolt1ge is now known as highvoltage [18:00] wgrant, cjwatson: how "scalable" is stalingstack right now? i. e. how many instances could/should I be using in parallel for autopkgtesting? or should I ask IS that and not you? === anthonyf is now known as Guest16072 === _salem is now known as salem_ === Elbrus_try_again is now known as Elbrus [18:48] who is the right person to talk with for more privileges on errors.ubuntu.com? cboltz is very helpful to the apparmor project, he's trying to track down something that's been reported in errors.ubuntu.com, and it'd be nice if he had privileges to view all the *apparmor* reports, at a minimum.. [19:01] anyone know of something that does what 'annotate-output' does but does so without running 'date' for every line of input ? [19:07] smoser: is it the calls to date you don't like or the timestamps themselves? maybe pass a format like +"" ? [19:08] the fact that it calls date. ie, forks for every line of input. [19:08] ouch [19:08] strftime is right there... [19:08] in bash ? [19:08] this thing is written in _bash_?? [19:08] yeah. its really simple. and works really well. [19:08] but no 'date' builtin in bash means fork. [19:09] it's shorter and cleaner than I expected :) impressive === anthonyf is now known as Guest26733 === salem_ is now known as _salem === _salem is now known as salem_ === elmo_ is now known as elmo === lionel_ is now known as lionel === salem_ is now known as _salem [21:46] arges: bug 1425619 - i think i'm about to become very cross [21:46] bug 1425619 in ubuntu-cloud-archive "Migration fails between QEMU 1.5 and QEMU 2.0" [Undecided,New] https://launchpad.net/bugs/1425619 [21:47] hallyn_: what happened now [21:52] hallyn_: so the missing bit was allow_incoming_qemukvm [21:52] pitti_: You should ask IS that; we don't have direct visibility. [21:56] arges: yeah. so the guy who was marking it incomplete wasn't fully up on the issue. now i guess the test case wasn't fully documented ... [21:56] grrrr! [21:57] hallyn_: well we can get this into the next update [22:00] had you deleted the qemu pkg or can that still be used alongside a new libvirt pkg?