[01:08] i'm seeing the local dnsmasq occasionally go screwy and not answering requests in trusty === sz0 is now known as sz0` [03:35] FYI, python-django security update introduced a regression. Have identified upsteam bug and fix and will release updates shortly. (LP: #1311433) [03:35] Launchpad bug 1311433 in python-django (Ubuntu Trusty) "REGRESSION: AttributeError: 'functools.partial' object has no attribute '__module__' " [Critical,In progress] https://launchpad.net/bugs/1311433 === sz0` is now known as sz0 === sz0 is now known as sz0` [04:49] slangasek: indeed; this just ships some additional binaries now like /lib/systemd/systemd, it doesn't change the default [04:49] slangasek: i. e. the PPA is meant to land as it is; I disabled systemd-sysv, as that's uninstallable as long as many things depend on upstart only [04:50] slangasek: so for the time being one has to manually boot systemd, as in my blog [04:50] slangasek: and ISTR that we discussed in the hangout (or in the TC bug) that the systemd-sysv approach wasn't what we wanted? [04:53] (I used a modified 10_linux to add it as an option to grub http://paste.openstack.org/show/K8PXsu8nhUxTz89yJg4M/ but then again I did several things differently than you did.) [06:45] good morning === MacSlow is now known as MacSlow|errand [07:54] pitti: Hello , are you here ? [07:54] hello NikTh [07:55] pitti: I just commented on your blog about systemd. It seems that there is no turning back to official packages ? e.g with ppa-purge I get an error about "Could not find package list..." [07:56] NikTh: if you don't boot with init=, it's not supposed to change anything; it seems this changed the NM behaviour somehow [07:56] NikTh: no idea why ppa-purge doesn't work; how did you call it? [07:56] it worked well enough some time ago [07:56] pitti: sudo ppa-purge ppa:pitti/systemd [07:57] pitti: ppa-purge is working on other ppas (I have tested it on other ppas just for sure) [07:57] pitti: Yes, the NM does not work on upstart if you have updated the packages with systemd repo. [07:58] NikTh: hm, NM works fine here [07:58] (with upstart) [07:58] NikTh: what's the output of "nmcli nm"? [07:58] and is NM running at all? (pidof NetworkManager) [07:59] Now I have booted in systemd , If I boot in upstart (now that I have updated through your repo) I don't have an active Internet connection. I can ping properly, but no application is able to resolve any host. [08:01] NikTh: right, so please give me the nmcli nm output under upstart; also, can you pastebin the complete ppa-purge output? this ought to work (and it's not specific to my PPA) [08:01] pitti: hey, very cool the work you did on bringing systemd to Ubuntu! :-) [08:01] zyga: it wasn't all that much really -- so far this is just what the Debian maintainers di :) [08:01] did [08:01] I suspect something with resolv.conf , because now (with init=/lib/systemd/systemd) I can see nameserver correctly in contents. When I boot without the parameter resolv.conf is empty. [08:02] zyga: I merely created a lightdm unit, and our NM doesn't have units, everything else pretty much just worked [08:02] NikTh: right, that means NM/dnsmasq isn't running properly [08:02] pitti: do you think it is realistic we'll see systemd in u series early on? [08:02] pitti: as a default? [08:03] zyga: not as default; we first need to provide systemd equivalents of all the /etc/init/* bits [08:03] zyga: but I do intend to upload the PPA contents to U as soon as it opens; so far this should be fairly harmless [08:03] pitti: ah, right, I recall the transition uds sessions, it's a difficult nut to crack [08:03] could someone who knows grub2 have a look at https://bugs.launchpad.net/ubuntu/+source/grub2/+bug/1311247 [08:03] Launchpad bug 1311247 in grub2 (Ubuntu) "error: malformed file, press any key to continue" [Undecided,Incomplete] [08:03] zyga: not particularly difficult, just some work (and probably merging with Debian and maybe stealing some units from Fedora) [08:04] pitti: does systemd upstream support running as a user session? [08:04] a friend of mine hit that bug [08:04] after lts->lts upgrade [08:04] pitti: http://pastebin.com/raw.php?i=fLqniTEy [08:04] zyga: yes, but we'll look at that after settling the systemd side; one step at a time [08:04] pitti: well the transition decisions were hard (do we keep upstart to run unconverted jobs or do we break compatibility) [08:04] zyga: first we need to make that rock solid, and also make this work on the server and phone [08:04] zyga: see the BP; upstart should run as a "deputy init" for some time [08:05] pitti: has there been a decision to switch on phablet now? [08:05] * zyga wonders how that will affect our initial product launches [08:05] NikTh: what do you have in /etc/apt/sources.list.d/ ? is there a .list for my PPA at all? [08:05] NikTh: maybe that's already gone [08:06] zyga: we of course can't do anything which breaks the phone; hence I just want to make this opt-in for now, so that interested developers can play with it, but we don't break anything [08:07] pitti: nice, I indend to give it a try as soon as U opens [08:07] * zyga wonders when U will get a name and when the archive will be open [08:08] pitti: is there, something wrecked I suspect with servers ? with launchpad ? I will test it again later. [08:08] zyga: there's the PPA for now [08:08] zyga, pitti will be very carefull, if he breaks the phone he knows i can whine in his ear all day for the rest of the cycle if needed ;) [08:08] oh yes; and ogra_ lives near enough that I can hear him whining from here! [08:08] pitti, could we get armhf builds in your PPA ? [08:08] :) [08:08] heulsuse [08:08] lol [08:09] ogra_: my layman PPA doesn't have ARM, but I can do a manual build and put them on people or so [08:09] ogra_: but U will be open soon enough, then we'll just get them from the archive :) [08:09] pitti, just ask the LP team to enable armhf [08:09] ogra_: heh :-) [08:09] soon enough ... right [08:09] dear sabdfl, please give us an adjective [08:09] for the Undeniable Unicorn [08:10] or Untitled Ubuntu [08:10] * ogra_ thinks this is a conspiracy between mark and randall ross to make planet more popular again [08:10] the "unnamed ubuntu" :) [08:10] hehe right [08:10] pitti: What is your understanding on this ? http://pastebin.com/raw.php?i=6EbQSq8B [08:10] suddenly everyone has it open again and hits F5 all day [08:10] NikTh: oh, suddenly it works? [08:11] uber Uguisu [08:11] uncanny underdog [08:11] yeah, http://www.list-of-animals.com/u.html is hopelessly small :/ [08:11] yep [08:11] * mvo like uncanny underdog [08:12] pitti: I don't think it worked properly. Did not downgrade the packages properly. [08:12] actually, maybe ogra's head will look like http://www.list-of-animals.com/details-uakari after landing systemd :) [08:12] lol :) [08:12] LOL ! [08:13] NikTh: hm, right; so it seems ppa-purge is broken :( [08:14] NikTh: what happens if you try: sudo apt-get install network-manager/trusty ? [08:14] NikTh: oh, you need to remove /etc/apt/sources.list.d/pitti-systemd-trusty.list first and run sudo apt-get update [08:15] pitti: I will try something else first.. wait. I have added the ppa again and I will try to remove it again, because ppa-purge worked as it should on other ppas. Only this one seems to have some difficulties. [08:16] NikTh: otherwise, try [08:16] sudo dpkg --P --force-depends systemd [08:16] sudo apt-get install {udev,network-manager,libgudev-1.0-0,lightdm,systemd-services,libpam-systemd,gir1.2-gudev-1.0,libnm-glib-vpn1,libnm-util2,gir1.2-networkmanager-1.0,libnm-glib4,liblightdm-gobject-1-0,libsystemd-login0,libsystemd-daemon0,libsystemd-journal0,libudev1,libudev1:i386}/trusty [08:17] NikTh: but yes, re-adding and re-removing sounds worth a try, too [08:18] pitti: what would you answer on this ? http://pastebin.com/raw.php?i=xAxT8vXH . I would answer yes, but as I said something weird is happening with the packages. [08:19] NikTh: yes, that looks right [08:19] NikTh: i. e. "y" [08:20] pitti: :-) [08:20] NikTh: the PPA removes systemd-services and adds systemd, so ppa-purge should revert that [08:21] pitti: Here is another funny part : http://pastebin.com/raw.php?i=Ln2CwpWk [08:23] NikTh: ah, yes; better do that when running under upstart; the trusty version has no idea about systemd, it assumes upstart is there [08:26] pitti: the funny thing is that NM (as we said before) does not work under upstart, so it cannot read any packages list. [08:26] NikTh: it already downloaded everything, so it should [08:27] NikTh: but as I said, I have no idea why NM doesn't work for you, it does here; I need "nmcli nm" as a first step for debugging that [08:28] NikTh: but you should run apt-get install -f first; it seems with all the ppa-purge stuff you wrecked your packages quite a bit :( [08:28] pitti: quite a bit ? quite a LOT I would say. :P [08:32] I will reboot in upstart now.. what logs you may want to keep for the NM/dnsmasq problem ? [08:32] pitti: [08:32] NikTh: not sure yet; could be that NM doesn't start up at all (nmcli will fail then), or doesn't see the connections, etc. [08:33] NikTh: I kind of hope it fails to start completey (although I can't reproduce that), as everything else would be rather unrelated to upstart vs. systemd [08:33] Ok, I'll see. See you later. [08:33] NikTh: /var/log/syslog is interesting in most cases [08:53] mvo: with python-apt, I can only call mark_install() on an apt.package.Package object; but I have already picked a (non-default) version from its .versions array, can I somehow mark *that* for install instead of the most current version? [08:54] mvo: or do I have to call fetch_binary on the apt.package.Version object, and drop the usage of apt.apt_pkg.Acquire? [08:56] pitti: yes, just change the candidate version and then call mark install [08:57] pitti: i.e. pkg.candidate = some_different_ver should work [08:57] mvo: ooh! thanks! [08:57] pitti: if not let me know and point me to the code please so that I can poke around [08:59] mvo: it works wonderfully, thanks! [08:59] pitti: I'm back :) [09:00] pitti: great to hear [09:00] I have resolved(short of) the problem with resolv.conf in upstart mode. I "hacked" manually the /etc/resolvconf/resolv.conf.d/head and added a custom nameserver like 8.8.8.8 [09:01] NikTh: so is NM running at all? [09:01] But when I ran resolvconf -u guess what the message was ? [09:01] Yes, NM was running, nmcli nm returned enabled active running.. [09:02] As I told before that I can ping, but it cannot resolve any host. [09:02] so what was the error message? [09:02] /etc/resolvconf/update.d/libc: Warning: /etc/resolv.conf is not a symbolic link to /run/resolvconf/resolv.conf [09:04] ok, that sounds rather unrelated to what boots your system :) [09:04] So I made the symbolic link properly and now I am in upstart mode with NM full working. nslookup return the appropriate messages [09:04] nice [09:05] NikTh: well, it could certainly be related to /etc/init/resolvconf.conf not having a systemd equivalent [09:05] (nor an init.d one) [09:05] The thing here is that before I upgrade from your repo I didn't have any problem, but I will not insist, because I had one other testing repo except yours. :) [09:06] Anyway, I will try to fix the other errors now, with ppa-purge [09:06] NikTh: it could certainly be; in my VM /etc/resolv.conf isn't a symlink either [09:06] NikTh: so thanks for pointing that out! [09:08] pitti: I will still keep testing your repo/packages/systemd.. you can consider me as a tester on this migration :-) I use systemd in Arch Linux and Fedora.. so I think I can help a little bit :) [09:10] NikTh: nice! yes, I just confirmed that booting with systemd replaces the symlink with a file; I'll fix that [09:11] NikTh: so, thanks for reporting that! [09:11] pitti: Ok. Glad I helped :-) === dholbach_ is now known as dholbach [09:15] asac, HAPPY BIRTHDAY !!!! [09:15] oh, alles Gute asac! [09:15] pitti: ppa-pure now worked as it should. The user must purge the ppa in upstart mode (as you indicated earlier) or it will not work. Everything back to normal :) [09:16] yay [09:16] ohhh, asac HAPPY BIRTHDAY :) [09:18] omg [09:18] now the public reminder :P [09:18] :) [09:18] ogra_: pitti: mvo: thx [09:18] i feel honoured to be with you guys on my birthday :P [09:19] even if only remotely [09:19] hehe [09:19] heh [09:20] asac, happy birthday! [09:21] seb128: thanks! [09:25] asac: haha [09:26] mvo: at least you are ahead of me :) (i hope i remember correctly) [09:29] asac: I think so, yes - and ogra_ as well ;) [09:29] yeah, i'm an old fart [09:29] you old guys... [10:11] Laney: =))) happy two years! [10:11] still no name? [10:11] NoNameYet_xnox: !!! [10:17] Laney: ohhh, congrats [10:18] mvo: thanks [10:18] xnox too (we're twins) [10:18] (ish) [10:18] =) [10:21] NoNameYet_xnox, is there for as long as Laney is? [10:21] happy anniversary guys ;-) [10:22] seb128: yeah =) on parole that is, Laney was in the community for longer than I have been. [10:22] *payroll [10:26] NoNameYet_xnox, k, that's probably why it feels like Laney has been there for ever ;-) [10:27] doesn't to me :P [10:29] asac, happy birthday! :) [10:34] dholbach: o/ [10:34] thanks! [10:37] asac: happy birthday :-) === doko_ is now known as doko [10:40] asac, wußte ich doch daß asac für "Alter S***" steht ;-P Viel Spaß beim Zählen der Jahresringe =) [10:41] doko: :P [10:41] immer diese typisch deutschen witze [10:41] hehe [10:41] zyga: thanks! === MacSlow is now known as MacSlow|lunch [11:06] doko: wouldn't it be helpful for gcc-multilib to ship e.g. i386-linux-gnu-gcc (et.al.) binaries/symlinks? [11:07] (for symmetry with gcc:i386 on i386) [11:11] can't be symlinks, and needs some thoughts what the defaults should be ... [11:29] There's a translateable string in gedit that sets the preferred encoding for the current locale. It was translated wrong for the "el" locale until today, when I notified the gnome-gr team and we fixed it upstream. [11:29] I want to backport that translation for previous versions of Ubuntu, e.g. 12.04, 14.04 etc. If I go to translations.launchpad.net and edit it (I have the required rights), it'll show up in the next "language-pack-el", right? Or do I need to file a bug report/SRU about it? === work_alkisg is now known as alkisg [11:35] alkisg: yes, just edit it on translations.lauchpad.net for all affected releases. a lang-pack refresh will pick it up. [11:36] Thank you NoNameYet_xnox, done: https://translations.launchpad.net/ubuntu/trusty/+source/gedit/+pots/gedit/el/+translate?batch=10&show=all&search=ISO-8859-15 [11:36] alkisg: i think we do langpack refresh at point releases so 14.04.1 and 12.04.5 will have it fixed. [11:36] Cool === tvoss is now known as dvoss [12:15] Laney: I have a question regarding https://bugs.launchpad.net/ubuntu/+source/sosreport/+bug/1296755 [12:15] Launchpad bug 1296755 in sosreport (Ubuntu Saucy) "sosreport archive /var/lib/maas by default" [Medium,In progress] [12:16] Laney: you added a task for Quantal-backport, any reason for that ? this release is EOL [12:18] caribou: It's not, not for a few weeks yet [12:18] Laney: is it worth going through the backport trouble for that ? I wouldn't think so [12:18] Laney: sosreport is mainly used by the support engineers [12:19] it's up to you [12:19] Laney: ok, I will mark the Quantal backport as "Wont fix" and concentrate on Precise [12:19] If you want to update it in P then that's what is needed === dvoss is now known as tvoss [12:20] So sez the rules of backports [12:21] Laney: you mean I just have to wait a week and I'll only have to worry about precise ? [12:21] Probably more like 4 weeks, but there's no announcement about that yet [12:21] Laney: well, both P & Q sosreport versions are the same; I'll build debdiffs for both [12:22] Laney: thanks for the details [12:22] Don't think it'll be hard to boot a Quantal VM and see if it runs [12:22] the standard required isn't very high [12:22] indeed === MacSlow|lunch is now known as MacSlow === _salem is now known as salem_ [13:38] Trevinho: hi, I'm looking at bug 723167, and notice that this issue was partially fixed. Would you like me to open a new bug, or mark this bug back to 'New' status. Thanks [13:38] bug 723167 in cairo (Ubuntu) "Fuzzy fonts caused by Cairo antialiasing artifacts with Radiance theme." [Undecided,Confirmed] https://launchpad.net/bugs/723167 === ming is now known as Guest7729 === alkisg is now known as work_alkisg [14:31] pitti: I certainly expect that systemd will be providing /sbin/init eventually, and we're not going to have to manually boot, if that's what you mean [14:32] slangasek: ah, so we don't want a separate systemd-sysv, just move /sbin/init, poweroff, etc. from upstart to systemd? [14:32] slangasek: (as we only support one system, not multiple as Debian) [14:32] pitti: I think that would be the eventual goal; I'm not sure we need to diverge in the other direction from the Debian packaging [14:33] slangasek: ok; well, let's figure that out when the time comes :) [14:33] debian delta wise, it might just be easiest to have systemd-sysv after all [14:37] Riddell: yeah, everything seems to be in limbo due to not having a codename [14:37] we can't even update distro-info-data properly yet [14:39] almost one week after the release :/ [14:41] seems most unprofessional [14:44] Riddell: yes, it's really blocked; you wouldn't know what to put into debian/changelog and the .changes file, for starters [14:45] pitti: "unstable" :) [14:45] dobey: I still call dibs on "Untitled Ubuntu" [14:46] sabdfl is letting us down [14:47] PPAs FTW.. [14:47] pitti: well, you can upload to "devel" but that would still at the moment end up in trusty's queue. [14:47] yeah, wrong target [14:47] nice nick, Dimitri :P [14:48] Logan_, dont you have to call him John now ? [14:48] :) [14:48] pitti: at one point there was a plan to use devel alias all the time (like in debian) but apt complaints about something (can't remember i think if one uses devel in sources.list) [14:49] NoNameYet_xnox: I had used devel in apt sources for quite some time; but I reverted it as I ran into hash sum mismatches way more often than with "trusty" [14:49] pitti: it would be nice if we could keep devel always open, similar to how fedora's raw-hide is always open, and stable release is simply branched at FF or some such. [14:49] NoNameYet_xnox: I mainly suspect that the symlinks interact badly with apt-cacher-ng [14:49] NoNameYet_xnox: rolling release! *cough* [14:49] NoNameYet_xnox: so... Debian? [14:50] Logan_: well that's how i've been doing my merges yesterday - > by forwarding patches to debian and tricking maintainers to upload them =) [14:50] that's the best way to do them :) [14:50] i guess i should start mass NMUs.... [14:51] i just want my tab completion to stop getting confused on "tru" when i have both trusty and trunk directories [14:52] could we maybe make "u-series" an alias for whatever adjective he comes up with? [14:53] barry: well i have "trusty" and "u-series" [14:55] NoNameYet_xnox: yeah. i always branch into a dir named after the series. it was a bit painful last cycle ;) [15:09] jamesh_, ping === sz0_ is now known as sz0` === sz0` is now known as sz0_ [15:55] Is there a PPA for the latest Qt and Qt Creator on 13.10? [15:59] ryanprior: looks like https://launchpad.net/~alexey-ivanov/+archive/qtcreator has the latest Qt Creator [16:03] Logan_: thanks. Looks like that's still Qt4. I'm also interested in using the latest Qt but I haven't been able to find a repo with both the latest Qt and the latest Qt Creator. [16:03] build from source? [16:04] Logan_: I can sure do that. I just figured that since Ubuntu has been doing so much Qt work already, somebody might already be tracking the latest stable releases so that I wouldn't have to. [16:06] Logan_: if I start building from source at every new stable release, I might as well maintain a PPA with the latest Qt and Qt Creator packages. But I figured somebody here might already be doing that. [16:06] ryanprior: what do you define as "latest"? [16:07] ryanprior: qt5 in ubuntu is at 5.2.1 [16:08] ryanprior: which is latest stable release at this point. [16:08] NoNameYet_xnox: in Saucy it's 5.0.2 [16:09] ryanprior: why not upgrade? [16:09] ryanprior: upgrade to 14.04 LTS, it's out now. [16:10] NoNameYet_xnox: because I'm in the middle of development and don't have a second machine I can get all set up and migrate to. If I screw up my only machine during upgrade then I'll have to sink a day restoring my data and configuration from backups. [16:10] ryanprior: or start 14.04 in lxc container / chroot to launch/use newer qt. [16:10] or use a VM. [16:11] ryanprior: trying to retro-fit 5.2 onto 13.10 is the surest way to break your machine, since it's a large transition which is only correctly compatible with 14.04. [16:11] NoNameYet_xnox: okay, I'll look into the chroot option. That sounds pretty sensible. === bfiller is now known as bfiller_afk [17:28] guess who just got a text about a post by Mark :D [17:28] UTOPIC UNICORN LEGGOOOO [17:28] (I love it) === Logan_ is now known as UtopicUnicorn [17:35] NoNameYet_xnox: time to change your nick :P [17:35] ++ === roadmr is now known as roadmr_afk [17:40] ohhh yes [17:40] unicorns it is [17:40] * zyga looks forward to natty narwhal videos for unicorns [17:42] I'm going to need the t-shirt for this release === tkamppeter_ is now known as tkamppeter [17:55] https://www.youtube.com/watch?v=9auOCbH5Ns4 [17:55] dobey: but is that a *utopic* unocirn ;) ? === bfiller_afk is now known as bfiller [18:01] zyga: well it's not a breezy badger, waving its arms around [18:01] * zyga loves this job [18:03] Well [18:03] I guess it's time for some rainbows in default wallpapers [18:03] heh [18:05] hehe === sz0 is now known as sz0` === sz0` is now known as sz0 === roadmr_afk is now known as roadmr [18:30] pitti, i seem to remember us talking (maybe at a real uds so this could be old) that mountall should but did not fix premounted filesystem's options to match mtab, for things like /proc and /sys; can you recall the discussion? [18:30] pitti, and, ugg, does this flow over into systemd at all [18:31] apw, doesnt systemd require that mounting happens in the initrd ? [18:31] * ogra_ heard that somewhere ... might be a myth [18:32] * apw looks clueless about systemd (pretty impressive over irc i recon) [18:34] heh === ChrisTownsend1 is now known as ChrisTownsend === roadmr is now known as roadmr_afk === roadmr_afk is now known as roadmr === bfiller is now known as bfiller_afk === bschaefer_ is now known as bschaefer [20:59] mpt: Do you a suggestion for a color for the retracers graph for failed retraces for 14.10? https://bazaar.launchpad.net/~daisy-pluckers/errors/trunk/view/head:/errors/static/js/retracers.js [21:01] bdmurray: Pink. With unicorns surfing the peaks. [21:25] well, that's just special [21:26] I proposed a merge for a branch that I can push to [21:27] and? [21:27] i do that daily :) [21:28] it was a minor change that didn't really need approval, and I didn't realize I had access to the branch :P [21:39] tumbleweed: http://bazaar.launchpad.net/~ubuntu-dev/ubuntu-dev-tools/trunk/revision/1420 [21:39] fixed that from your change in 2011: http://bazaar.launchpad.net/~ubuntu-dev/ubuntu-dev-tools/trunk/revision/1201.1.33 :P [21:39] it was quitting if people said "yes" and continuing if "no" [21:39] and it was asking if people wanted to continue :P [21:43] Logan_: thanks [21:44] no problem === salem_ is now known as _salem === sarnold_ is now known as sarnold === NoNameYet_xnox is now known as xnox === wedgwood is now known as Guest68564