[02:54] <shadeslayer> yofel: do you have any experience making apt-cacher-ng cache custom repos? like a ppa?
[03:26] <shadeslayer> yofel: nvm, figured it out
[04:59] <Mirv> sitter: re; qtscript the packaging branch is still at 5.3.2 and requires no changes besides version bumps. if it worked before, maybe you didn't bump the build deps before or something? (s/5.3.2/5.4.0/ in debian/control would lead to 5.4.0+dfsg qtbase dependencies)
[06:59] <bukai> Good Morning!
[07:15] <soee> good morning
[08:07] <sitter> Mirv: with the not updated branches I do absolutely nothing as jenkins enforces order ^^
[08:07] <sitter> most peculiar
[08:10] <sitter> aha!
[08:10] <sitter> http://anonscm.debian.org/cgit/pkg-kde/qt/qtscript.git/commit/?h=experimental&id=97fb9702ddc5e9cd1a122d8793de3e0f0307a2ef
[08:10] <sitter> all mitya57's fault xD
[08:24] <sitter> ah yeah, I used the git version because we have a conflicting patch in launchpad
[08:24] <sitter> Mirv: do you by any chance already have something to push for qtscript?
[08:32] <Mirv> sitter: well I have a local diff I can push. I'm not going to need it myself now that I can sync qtscript 5.4.0+dfsg-1 from debian experimental instead.
[08:32] <sitter> Mirv: nevermind then
[08:33] <sitter> I just made a branch on debian git. good enough for now
[08:33] <Mirv> ok, sounds good
[09:02] <lordievader> Good morning.
[09:28] <bukai> Riddell: ping
[09:29] <Riddell> hi bukai 
[09:35] <bukai>  Hi, I am very confused now. I asked ovidiu-florin  for setting up wordpress he asked me to install it on some other private server what do I do?
[09:35] <bukai> Riddell: ^^
[09:40] <Riddell> set it up on that test2 domain I made
[09:41] <sitter> and of course plasma5 isn't starting with qt5.4 anymore
[09:43] <Riddell> uh oh
[09:43] <Riddell> does other stuff?
[09:43] <Riddell> most of the plasma devs use qt5.4
[09:54] <sitter>             if (d.n == 0) {
[09:54] <sitter> why use names when you can use letters
[10:10] <sitter> bad feeling I have
[10:17] <sitter> http://paste.ubuntu.com/9549049/
[10:19] <Riddell> dlopen, I've no idea how to debug that
[10:19] <Riddell> except recompile everything and hope it helps
[10:19] <Riddell> but then why would it if it's all binary compatible
[10:19] <yofel> plugin ABI mismatch?
[10:22] <sitter> that would be very scary
[10:22] <sitter> I only switched out qt5.4 core pieces that were required to make the packaging ABI dependencies resolved
[10:24] <sitter> so either there is a problem with the ABI tracking magic or 5.4 has a BIC or a framework leaks private qt symbols through its own API
[10:24] <sitter> none of them explain why it would crash on kStoreUmask though
[10:24] <sitter> and yet rebuilding kdelibs4support fixes it
[10:27] <sitter> Q_CONSTRUCTOR_FUNCTION
[10:28] <sitter> could be that this changed 
[10:28] <sitter> Q_CONSTRUCTOR_FUNCTION(kStoreUmask)
[10:28] <sitter> other than that kstoreUmask is a posix function operating on pod statics
[10:31] <sitter> hasn't changed in 2 years -.-
[10:42] <sitter> Mirv: please add kdelibs4support to the nochange rebuild upload list
[10:47] <sitter> plasma-nm also has flaky symbols
[10:47] <sitter> -.-
[10:48] <Mirv> sitter: hmm, why it'd need a rebuild? it does not seem to depend on the private headers and does not depend on qtbase-abi-5-3-2 or qtdeclarative-abi-5-3-2 -> it should just work? or was it that you have some different behavior anyhow #ifdef:d in there?
[10:48] <Mirv> sitter: bug #1395651 :)
[10:48] <ubottu> bug 1395651 in plasma-framework (Ubuntu) "plasma-framework needs symbols updates for Qt 5.4.0" [Undecided,New] https://launchpad.net/bugs/1395651
[10:49] <Mirv> sitter: I'm reading the backlog..
[10:49] <Mirv> that's quite weird + unfortunate behavior..
[10:50] <sitter> yeah, not sure how to debug it either
[10:50] <Mirv> I can add a new section "rebuild for weird reasons" for now in the Qt stack document
[10:50] <sitter> Mirv: sounds good
[10:50] <Mirv> we haven't noticed such needs in the past with 5.3.0/5.3.2 migrations though
[10:51] <sitter> kf5.kded: Could not load KDED module "networkmanagement" : "Cannot load library /usr/lib/x86_64-linux-gnu/qt5/plugins/kded_networkmanagement.so: (/usr/lib/x86_64-linux-gnu/libKF5KDELibs4Support.so.5: undefined symbol: _Z20qMessageFormatString9QtMsgTypeRK18QMessageLogContextRK7QString)" (tried plugins named: ("networkmanagement", "kded_networkmanagement") )
[10:51] <sitter> that's somewaht more interesting
[10:51] <sitter> perhaps that ultimately is why ksmserver crashes as well
[10:53] <Mirv> notably though with 5.4.0 I haven't gotten to run unity8 yet so I wouldn't yet know if there's something evil lurking now with 5.4.0
[10:54] <sitter> oh actually
[10:55] <sitter> Mirv: check out the symbol retractions here https://launchpadlibrarian.net/192781773/buildlog_ubuntu-utopic-i386.plasma-framework_5.5.0%2Bgit20141216.1359%2B14.10-0ubuntu0_FAILEDTOBUILD.txt.gz
[10:55] <sitter> it seems a bit odd that these symbols would show up in plasma-framework at all
[10:59] <sitter> shadeslayer, Riddell: one of you available right now?
[11:00] <shadeslayer> yeah
[11:00] <shadeslayer> ssup
[11:01] <sitter> shadeslayer: please run for f in /usr/lib/x86_64-linux-gnu/libQt5*.so; do echo $f; objdump -tT $f | grep "_Z20qMessageFormatString9QtMsgTypeRK18QMessageLogContextRK7QString"; done
[11:01] <shadeslayer> http://paste.ubuntu.com/9549427/
[11:01] <sitter> thanks
[11:01] <sitter> Mirv: core lost a symbol :O
[11:01] <shadeslayer> yw
[11:02] <shadeslayer> fun
[11:02] <sitter> http://paste.ubuntu.com/9549433/ that's mine with 5.4
[11:02] <sitter> shadeslayer: see, there's another reason why we should CI qt :P
[11:03] <Mirv> sitter: right, I can see that. that's ugly indeed.
[11:04] <shadeslayer> jebus
[11:05] <Mirv> the 5.4 transition might get a bit ugly then
[11:05] <shadeslayer> sitter: one would think Qt has ABI breakage checking
[11:05] <sitter> one would assume KDE has it...
[11:05] <shadeslayer> in their own CI
[11:05] <shadeslayer> sitter: sure, they're called packagers
[11:10] <sitter> e968793e81ffcb439210e7eb422c0063834cfdc9
[11:12] <sitter> https://qt.gitorious.org/qt/qtbase/commit/e968793e81ffcb439210e7eb422c0063834cfdc9
[11:13] <sitter> the qustion is how or why a random exported symbol got to show up in kdelibs4support at all
[11:15] <sitter> ./src/kdecore/kdebug.cpp:extern Q_CORE_EXPORT QString qMessageFormatString(QtMsgType type, const QMessageLogContext & context,
[11:15] <sitter> ./src/kdecore/kdebug.cpp:    return qMessageFormatString(type, context, str);
[11:15] <sitter> WHAT THE FUCK
[11:15]  * sitter shakes head
[11:17] <sitter> Mirv: there's your reason for why it broke. it's a one-time 5.4.0 rebuild ... kdelibs4support with qt<=5.3 externs a function exported but not "headered" by qtcore thus doing implicit private api usage
[11:18] <Mirv> sitter: whee... anyhow, if this is now limited to kdelibs4support the world still has hope
[11:19] <Mirv> I'll keep the "weird" section of the needs-rebuilds just for that, though
[11:19] <sitter> weird sounds like an understatement now that I know why it breaks, but yeah ^^
[11:24] <sitter> mhhh, running startkde in fluxbox, the only real way to start a proper plasma instance \o/
[11:39] <Riddell> sitter: dare I ask how qca is getting on?
[11:39] <sitter> backburner
[11:39] <sitter> I need to send a mail polling opinions though
[11:39] <sitter> actually let me do that right now before I forget again
[11:40] <sitter> Riddell: what's the new packager list?
[11:40] <Riddell> I dare say okteta will live without it
[11:40] <Riddell> sitter: what new packager list?
[11:40] <Riddell> hi Neo31!
[11:40] <sitter> kde-distro-packagers@kde.org
[11:40] <sitter>  :P
[11:40] <Riddell> oh, no idea what that thing is for
[11:40] <Riddell> we get rid of one silly list and replace it with another
[11:41] <Riddell> maybe you can give it purpose
[11:43] <Neo31> hi Riddell 
[11:43] <Neo31> :)
[11:43] <Neo31> how r u today
[11:46] <Riddell> Neo31: all good
[11:46] <Riddell> Neo31: are you able to do some testing for alpha 1 today?
[11:48] <sitter> Riddell: no, I decided that distro is not relevant enough :P
[11:48] <sitter> but yeah, the new list I found  a bit weird
[11:48] <sitter> last I followed the discussion the idea was to use release-team which seemed like a suitable place to discuss distro things
[11:48] <sitter> but oh well
[11:57] <Neo31> not sure for tonight but probably i will get one hour at least
[11:58] <Neo31> that's in 9 hours from now, when I am setup at home
[11:58] <Neo31> Riddell, i have vivid already installed, all i need to to is apt update & upgrade then do some testing ?
[12:00] <Riddell> Neo31: nope needs full installs (can be on virtual machines or real)
[12:00] <Riddell> hi kranzer 
[12:00] <kranzer> hi Riddell 
[12:00] <Neo31> so i need to start downloading the daily iso ?
[12:00] <kranzer> Riddell: I'm starting doing task
[12:00] <kranzer> Riddell: I downloaded all stuff yesterday
[12:00] <Riddell> Neo31: yep please :)
[12:01] <Riddell> kranzer: great
[12:01] <kranzer> Riddell: I'll use Oracle Virtual box
[12:01] <Neo31> i'm on it Riddell  ;)
[12:02] <Neo31> downloading from here Riddell http://cdimage.ubuntu.com/kubuntu/daily-live/current/ ?
[12:03] <kranzer> Riddell: so
[12:03] <kranzer> Riddell: I just need to install it and use, don't I?
[12:06] <Riddell> kranzer: yep, install in the various different scenarios
[12:06] <kranzer> Riddell: what scenarios?
[12:07] <Riddell> manual partitioning, full disk, resize disk, disk with lvm etc, listed on  http://iso.qa.ubuntu.com/qatracker/milestones/328/builds/85671/testcases
[12:07] <Riddell> don't be afraid of duplicating tests, my testing does count as I'm the release manager and testing your own stuff is always prone to failure
[12:08] <Riddell> and you can duplicate each other, the most tests the better
[12:08] <kranzer> Riddell: I can install only x86 version
[12:09] <Riddell> kranzer: no amd64 machine?
[12:10] <kranzer> Riddell: virtualbox allows only x86
[12:11] <Riddell> kranzer: that'll be the case if you have a 32 bit install yes
[12:11] <Riddell> then concentrate on i386 and Neo31 can concentrate on amd64
[12:12] <kranzer> Riddell: could you give me link of testcases for i386?
[12:15] <Riddell> kranzer: http://iso.qa.ubuntu.com/
[12:17] <macstar> guys there is something wrong the ppa:kubuntu-ci/unstable .... when i manually add this to a kde4 based kubuntu 14.10 and then do a dist-upgrade and upgrade it installs only the libs but not the plasma desktop
[12:18] <Riddell> macstar: it's for vivid only
[12:18] <macstar> riddell since when? i remember a few weeks back i ran 14.10 in vbox with the plasma next daily updates
[12:18]  * Riddell lunches
[12:20] <sitter> you really should not add unstable
[12:20] <sitter> at all
[12:20] <macstar> but i am trying this in virtualbox and want to hape it as up2date as possible to do some bug reports maybe
[12:21] <sitter> unstable breaks every other day
[12:21] <sitter> so unless you know how to recover broken packages you really should use unstable-daily for testing
[12:22] <macstar> sitter so what would you recommand then?
[12:22] <sitter> what I just wrote
[12:23] <macstar> that i should use unstable-daily... but i am doing this with kubuntu-ci/unstable or am i wrong?
[12:23] <sitter> there's 3 ppas
[12:24] <macstar> https://community.kde.org/Plasma/InstallingNext
[12:24] <macstar> there for daily builds it only links to ci
[12:24]  * sitter sighs
[12:25] <sitter> https://community.kde.org/Kubuntu/PPAs#Kubuntu_Continuous_Integration_.28CI.29
[12:25] <macstar> now that makes sense ty sitter
[12:28] <BluesKaj_> 'Morning all
[12:32] <kranzer> Riddell: one done!)
=== rdieter_ is now known as rdieter
=== kbroulik is now known as kbroulik-lunch
[13:25] <kranzer> Riddell: everything's tested
[13:25] <kranzer> Riddell: no problems at all
[13:30] <kranzer> Riddell: what should I attach to melange?
[13:44] <Riddell> kranzer: Non-English Live Session & Installation ?
[13:51] <kranzer> Riddell: I did kubuntu-desktop
[13:51] <kranzer> Riddell: not kubuntu-desktop-extra
[13:51] <Riddell> kranzer: how do you mean?
[13:52] <kranzer> Riddell: type: mandatory
[13:54] <kranzer> Riddell: like this http://iso.qa.ubuntu.com/qatracker/milestones/328/builds/85672/testcases/1301/results
[13:55] <Riddell> kranzer: would be great if you can do Non-English Live Session & Installation too
[13:56] <Riddell> ignore Live (netbook) that's not valid any more
[13:57] <soee> can someone givce link to isos testcases ?
[13:57] <kranzer> So, I need to install it again, don't I?
=== kbroulik-lunch is now known as kbroulik
[13:58] <Riddell> soee: http://iso.qa.ubuntu.com/
[13:58] <Riddell> kranzer: yep
[13:58] <kranzer> Riddell: okay but first I need to do some maths
[14:00] <soee> valorie: hiho, did you had time to work on the content for "Contribute" page ?
[14:17] <kranzer> Riddell: okay, I have installed it, and everything is ok)
[14:21] <Riddell> shadeslayer: what did I break?
[14:21] <Riddell> this doesn't sound like me
[14:22] <shadeslayer> Riddell: kscreen branches et all
[14:22] <shadeslayer> in debian git repo
[14:22] <shadeslayer> can't merge master into kubuntu_unstable
[14:22] <shadeslayer> that's what sitter says
[14:22] <shadeslayer> Riddell: it's holding up Debian CI :(
[14:23] <shadeslayer> I could cherry-pick the relevant commit from master
[14:23] <shadeslayer> sitter: ^^
[14:24] <sitter> wtf
[14:24] <sitter> just merge it
[14:24] <shadeslayer> I have conflicts
[14:24] <shadeslayer> and shitty ones at that
[14:26] <shadeslayer> sitter: ok, fixed
[14:31] <kranzer> Riddell: are you here?
[14:33] <Riddell> hi kranzer 
[14:35] <kranzer> Riddell: I did non-english one
[14:36] <Riddell> kranzer: was evening in the right language after install?
[14:38] <kranzer> evening?
[14:38] <kranzer> maybe everything?
[14:38] <Riddell> kranzer: yes, was everything in the right language after install?
[14:38] <kranzer> yes
[14:39] <Riddell> great, that's a pass
[14:39] <kranzer> Riddell: I tried russian
[14:40] <Riddell> kranzer: you can't do any amd64 tests?
[14:40] <kranzer> Riddell: no, I can't(
[14:41] <Riddell> kranzer: can you take a screenshot of the russian install? just something for melange
[14:42] <Riddell> ah you added a url, I'll close it
[14:42] <Riddell> great, thanks
[14:42] <kranzer> Riddell: thanks)
[14:43] <kranzer> Riddell: are there other tasks?
[14:45] <Riddell> not currently sorry
[14:45] <Riddell> make an alpha 1 page for vivid? https://wiki.ubuntu.com/UtopicUnicorn/Alpha1/Kubuntu
[14:46] <Riddell> https://wiki.ubuntu.com/TrustyTahr/Alpha1/Kubuntu
[14:46] <Riddell> equivalent
[14:46] <Riddell> but saying it's all plasma 5 now, woo
[14:46] <Riddell> complete with screenshots
[14:47] <Riddell> kbroulik: sorry I need to leave now, let me know if you want that one
[14:47] <kbroulik> I guess you meant kranzer? :D
[14:49] <Riddell> mm yes sorry
[14:50]  * Riddell out
[15:24] <Neo31> sorry internet went down
[15:24] <Neo31> Riddell, i have the iso already, should i use the iso tracker for testing or follow other stuff?
[15:36] <Riddell> Neo31: yes please report on iso tracker
[15:38] <Neo31> ok Riddell thx
[17:03]  * Riddell out for evening
=== amichair_ is now known as amichair
[20:48] <mparillo> Riddell: kranzer: I took a start at https://wiki.kubuntu.org/VividVervet/Alpha1/Kubuntu
[20:48] <Riddell> mparillo: ooh great
[20:57] <Sergobot> Riddell: can you add a task for updating packages in repositories?
[20:59] <Sergobot> Riddell: Hello :)
[21:12] <Riddell> Sergobot: any package in mind?
[21:12] <Sergobot> Riddell: no
[21:12] <Riddell> Sergobot: calligra 2.9 beta needs doing
[21:13] <Sergobot> Riddell: can you create a GCI task for it?
[21:16] <Riddell> Sergobot: done
[21:16] <Riddell> will need to wait for valorie or someone else to approve it
[21:16] <Riddell> https://www.google-melange.com/gci/task/edit/google/gci2014/5853249311604736?validated
[21:16] <Riddell> but you can start if you wish
[21:17] <Sergobot> Riddell: okay
[21:18] <Neo31> hello Riddell 
[21:18] <Riddell> hi Neo31  how did testing go?
[21:20] <Neo31> I started testing on two Kubuntu 14.04 machines using virtualbox and i got an error related to virtualbox/kubuntu
[21:21] <Neo31> it shows a broken screen of colors when booting, then i have to hit host+F2 then host+F1 to get the graphics running correctly and proceed with the install
[21:21] <Neo31> where can I report such a bug Riddell ?
[21:22] <Neo31> one VM in doing entire disk test with updates, the other doing the same without updates (still in progress)
[21:23] <mparillo> Neo31: Could you bug be similar to: https://bugs.kde.org/show_bug.cgi?id=340966
[21:23] <ubottu> KDE bug 340966 in general "Plasma 5 crashes maximizing VMware Player" [Crash,Resolved: fixed]
[21:23] <mparillo> That is my upstream version of: https://bugs.launchpad.net/ubuntu/+source/open-vm-tools/+bug/1362631
[21:23] <ubottu> Launchpad bug 1362631 in open-vm-tools (Ubuntu) "Enter Full Screen Mode crashes Plasma Next Desktop in Kubuntu 14.10 and Daily Build of 15.04" [Undecided,New]
[21:25] <Riddell> mparillo: 14.04lts should be 14.10 for upgrades on alpha 1 page
[21:27] <Riddell> mparillo: link to bugs is utopic one and list of bugs are utopic ones, grab the new vivid link at qa.kubuntu.co.uk
[21:27] <Riddell> and the new bugs
[21:27] <Riddell> although would be good to triage the utopic ones to see which apply to vivid
[21:29] <Neo31> mparillo, no i don't even have to touch anything on a virtualbox VM it crashes without maximizing anything, but if i do the workaround it shows up correctly
[21:29] <Neo31> after the workaround everything is fine on virtualbox, i can maximize and everything...
[21:32] <Neo31> so mparillo Riddell on which platform i need to search for the bug please? bugs.kde.org or where exactly ?
[21:32] <Neo31> I will add the details on reproducing it
[21:34] <genii> Hm, updates to sddm aren't seeing there has been manual changes to /usr/share/sddm/scripts/Xsetup and just wipe it out.
[21:34] <Riddell> Neo31: I really don't know it could be anywhere of X, virtualbox, kde, plasma , qt :(
[21:34] <Sergobot> Riddell: should I write in changelog file vivid or utopic?
[21:36] <yofel> genii: needs to be marked as conffile for it to do that
[21:37] <yofel> why's that in /usr/share anyway -.-
[21:38] <genii> yofel: It's the only place I could reasonably find in sddm to turn off extra monitors before starting ( with xrandr)
[21:44] <genii> It would be easier if it honoured xinitrc or xsession settings in it's own home dir but it doesn't.
[21:46] <yofel> it's own home dir? You mean the users, or...?
[21:49] <genii> yofel: It's own, yes. The idea here is I want to turn off screen spanning for sddm, but not for users
[21:50] <genii> yofel: If I put the xrandr commands needed into /var/lib/sddm  .xinitrc or xsession, they are not executed
[21:50] <Neo31> Riddell, I don't know on what platform to report bugs for KDE5 and plasma
[21:57] <yofel> Neo31: plasma bugs go to bugs.kde.org
[21:57] <Riddell> Sergobot: vivid
[21:59] <yofel> lets see if I can get vivid+ci to crash
[22:01] <yofel> no, but that VM also has the vbox x11 driver installed
[22:02] <Neo31> thx yofel 
[22:02] <yofel> FWIW, the crash that I can reproduce is the one from kde 340966
[22:02] <ubottu> KDE bug 340966 in general "Plasma 5 crashes maximizing VMware Player" [Crash,Resolved: fixed] http://bugs.kde.org/show_bug.cgi?id=340966
[22:06] <mparillo> yofel: Thank you. There was some confusion on the notes to that bug, but it seems to be resolved in 5.1.2 (not 5.3.2). Is that an easy way to see when that lands in Vivid?
[22:11] <yofel> give me a minute to check whether it's the driver that matters or the fact that I have updated plasma and kscreen snapshots from CI
[22:13] <yofel> ok, the driver was already installed, so it's indeed fixed in the CI snapshots
[22:25] <yofel> mparillo: should be fixed in 5.1.2 from what I see, and that's in vivid-proposed
[22:29] <Neo31> Riddell, when i'm not sure the bug is related to which product exactly should I still report it and put in on iso tracker? I think this is a major bug because we cannot proceed with vivid installation if we don't do the workaround
[22:35] <soee_> if we try OEM install and pick some custom language the live session loads with english but ubiquity (when we click to install system) uses our custom language - this should work like that ?
[22:37] <mparillo> yofel: Thank you.
[22:43] <shadeslayer> option parsing sure is shit in all languages
[22:43] <shadeslayer> python had shit optparsing
[22:43] <shadeslayer> as does ruby
[22:43] <shadeslayer> -r '' -w foo gets parsed as : option : r , value : -w
[22:44] <yofel> fun
[22:45] <Neo31> hey guys, other than dmesg output what else do i need to report on a system startup crash GUI ?
[22:46] <yofel> hm, when exactly does it crash again?