[06:45] <cpaelzer> good morning
[07:40] <pitti> xnox, slangasek: Ian reverted my commit to Debian's sysvinit to keep the package in sync (https://anonscm.debian.org/cgit/collab-maint/sysvinit.git/commit/?id=6ac609340af1), so the current zesty-proposed version brings back all the old cruft
[07:41] <pitti> xnox, slangasek: can you please remove it? remove-package in chroot complains with "Unable to find the server at api.launchpad.net" for me right now
[07:41] <pitti> (I'm talking to Ian how to fix this more properly)
[11:03] <fossfreedom_> seb128: hi - when you have a mo' please can I ask you to review the merge request for the gnome-menu patch we discussed on Friday last.  Cheers.  https://bugs.launchpad.net/ubuntu/+source/gnome-menus/+bug/1631745
[11:50] <ilmaisin> can anyone do sru if one can find someone with upload access to sponsor it?
[11:56] <jbicha> ilmaisin: yes! https://wiki.ubuntu.com/SponsorshipProcess
[12:07] <cult-> xnox: do you have any planned time schedule for odb of xenial?
[12:13] <xnox> cult-, no. it is not that urgent. even though it does affect your systems =/
[12:13] <cult-> xnox: is there anything i can do for it?
[12:14] <xnox> not really, no change uploads need to be prepared and uploaded. and one needs upload rights for that.
[12:14] <cult-> you mentioned that there's paperwork too for lts?
[12:15] <xnox> correct, i believe i have already filled all the boiler plate in the bug report description.
[12:23] <melodie> hi
[12:25] <melodie> I am using a Xenial install and being on a light system, trying to find a way to disable at-spi2. I don't find. There is no service for it so I can't use systemctl, and update-rc.d does not work anymore. If I remove the package it wants to pull off Evolution, which I need. What would be the best move? Create a service? Or ask the packager of Evolution to mark at-spi2 as optional?
[12:42] <seb128> fossfreedom_, hey, yes it's on my list, glad that we converged toward a real fix and not a workaround
[12:43] <fossfreedom_> :) thanks seb128
[12:43] <seb128> melodie, hey, what's the issue with at-spi? you might be able to turn off some bits in the a11y u-c-c settings
[12:56] <ilmaisin> xnox: did you get till's mail about the cups sru?
[13:00] <ilmaisin> xnox: if you ae in hurry i can maybe help with the paperwork if someone will sponsor the update
[13:03] <ilmaisin> xnox: as it got little more important as it was noticed that on some systems cups will shut down even with shared printers
[13:21] <xnox> ilmaisin, cups did not affect anybody at all. It only affected systems running -proposed, which no human should be doing.
[13:21] <xnox> ilmaisin, the fix is in the sru queue, and pending review by ~ubuntu-sru team, which i'm not part of
[13:21] <xnox> ilmaisin, also cups sru is available from a ppa, linked on the bug reprot
[13:21] <xnox> ilmaisin, also cups sru is available from a ppa, linked on the bug report
[13:32] <ilmaisin> xnox: in my last message about cups shutdown i meant the bug 1598300, not the bug 1642966 that is in testing, but they are interwoven that the fix to former depends on the fix to the later one
[13:33] <xnox> ilmaisin, sure i understand that; but the next step for it to get published in proposed channels is to have ~ubuntu-sru team review it.
[13:33] <ilmaisin> xnox: ok
[14:02] <tjaalton> doko: mesa 17.0.0 was released, I'll switch it to use llvm-4.0 instead of 3.9
[14:03] <doko> tjaalton: for trusty as well?
[14:03] <tjaalton> doko: no, just zesty
[14:03] <doko> ahh, ok
[14:03] <tjaalton> it'll be backported to xenial later
[14:03] <tjaalton> skipped 3.9 there
[14:04] <doko> just noticed that llvm 3.9 requires cmake 3.4.3 :-/
[14:04] <tjaalton> and it failed to build on ppc64el on xenial
[14:27] <ChrisTownsend> seb128: Hi!  So, as we discussed last week, the libertine landing needs a little help in -proposed.  Would you have time to help out with that?
[14:27] <seb128> ChrisTownsend, hey, let me have a look
[14:27] <ChrisTownsend> seb128: Thanks!
[14:57] <zul> mterry: can you have a look at LP: #1662215 please when you get a chance should be pretty striaght forward
[15:00] <rbasak> !dmb-ping
[15:09] <mterry> zul: on it
[15:09] <zul> mterry: thanks
[15:37] <hallyn> yay, another one of those updates that resets my xmodmap
[15:37] <hallyn> hadn't had one in awhile so i was perplexed why tmux was acting weird
[16:11] <slangasek> pitti: sysvinit/zesty-proposed removed
[16:12] <pitti> slangasek: cheers
[17:00] <xnox> tedg_, is cgmanager needed for systemd-ubuntu-app-launch?
[17:00] <tedg_> xnox: No it is not
[17:00] <tedg_> Which makes UAL a lot faster too :-)
[17:00] <xnox> tedg_, you need click to launch clicks?
[17:01] <tedg_> xnox: Yes, and for our click hooks in the package.
[17:27] <xnox> tedg_, is the C library libubuntu-app-launch upstart only?!
[17:27] <xnox> well the ubuntu-app-launch.cpp part of it?
[17:27]  * xnox guess it's not quite C library but C/C++
[17:28] <tedg_> xnox: Yeah, the C code is basically a wrapper on the C++ stuff so that current users don't have to switch.
[17:35] <xnox> tedg_, do you support launching clicks, under systemd? and in that case shouldn't one be launching (click?!) helpers with systemd?
[17:36] <tedg_> xnox: Yes, because they'll become snap helpers soon.
[17:36] <tedg_> xnox: Once all the interface hooks stuff lands in snapd.
[17:39] <xnox> tedg_, but i see that helpers are only launched with upstart, in libubuntu-app-launch/ubuntu-app-launch.cpp and ripping that out, appears to be breaking the ABI as well.
[17:39]  * xnox pretends that return FALSE; is not good enough here.
[17:40] <tedg_> xnox: Yes, that's part of what I'm doing to remove Upstart, is have helpers use the jobs objects, which take care of that.
[17:41] <xnox> tedg_, right, so i can't really rip out build dep on upstart just yet, as that will break things =(
[17:41] <xnox> bugger
[17:42] <tedg_> xnox: Yes, we were discussing plans in #ubuntu-desktop this morning. We're gonna work to remove Upstart.
[17:49] <juliank> jbicha: your packagekit merge just build successfully on amd64 :D
[17:51] <juliank> it was "fun"
[17:52] <juliank> Ah, I'm subscribed to PackageKit bugs because I'm a member of the PackageKit-team team
[17:53] <jbicha> thanks!
[17:54] <xnox> tedg_, nice.
[18:09] <nacc> tarpman: would you be able to look at some of the autopkgtest failures from the openldap merge in http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html#openldap ?
[18:23] <tarpman> nacc: looking
[18:26] <nacc> tarpman: thanks! i think libreoffice is unrelated (as the version in excuses is triggering the same failure on itself), dogtag-pki is probably also not related
[18:27] <tarpman> nacc: dovecot one looks like some kind of TLS error? ssl.SSLEOFError: EOF occurred in violation of protocol (_ssl.c:720)
[18:27] <nacc> tarpman: yeah
[18:27] <tarpman> no idea about libaws really but also doesn't look particularly ldap related :\
[18:28] <nacc> tarpman: yeah and it's odd that it is passing on a few and not on a few, i'll dig into that one
[18:53] <tarpman> nacc: do you happen to know whether the testbeds have anything installed a typical sbuild chroot wouldn't? e.g. ubuntu-minimal
[19:25] <nacc> tarpman: off the top of my head, sadly, no
[19:27] <nacc> tarpman: fwiw, the libaws regression is unrelated, i'm pretty sure -- it's also happening for libaws itself, as well on those archs
[19:27] <nacc> jgrimm: --^ FYI, openldap merge is done, but stuck in proposed while we figure out the tests
[19:27] <tarpman> nacc: but they need to be fixed for openldap to migrate, related or not, right?
[19:28] <nacc> tarpman: right, but the fix might not be in opendldap (e.g., i think the libaws ones needs rebuilds of specific packages)
[19:28] <jgrimm> nacc, ack
[19:32] <melodie> seb128 still here?
[19:33] <nacc> tarpman: e.g., looking through excuses, i think all of the ada related packages are in 'regression' status for the same reason
[19:33] <melodie> seb128 my issue is about having the hand on the management of services and processes, which I don't have, because systemd-ui does not provide any kind of management
[19:35] <melodie> seb128 in Trusty it has been possible to disable the launch of at-spi2 and therefore a11y which is launched by dbus, using update-rc.d, but in Xenial with systemd there is no service written for this feature, so it's started for no reason as I don't  need it, and I can't remove it because it's tied to Evolution, which I do need.
[19:36] <melodie> I thought maybe I could somehow tweak the behavior in apt conf as the Arch users do with the pacman.conf file for their package manager, but I found no clue about that in the man of apt or apt.conf …
[19:36] <tarpman> nacc: ack. currently looking at dogtag-pki, but only here for a little longer
[19:37] <nacc> tarpman: if I absolutely had to guess, that one is also unrelated
[19:38] <nacc> tarpman: as we have very similar regressions across other source packages
[19:38] <tarpman> sure, but a better chance of being within my power to do something about :)
[19:41] <jbicha> melodie: what's the problem with at-spi2 running?
[19:50] <melodie> jbicha resource
[19:51] <melodie> jbicha I tend to eliminate any process which is of no use in my system, so I'd like to be able to manage the processes the way I feel fit
[19:52] <jbicha> you could try cp /etc/xdg/autostart/at-spi-dbus-bus.desktop ~/.config/autostart/  and then add Hidden=true to that file in ~/.config/autostart
[19:52] <melodie> jbicha I will try that, thank you
[19:53] <jbicha> but Ubuntu intentionally does not make it easy to disable autostart files in the default install since disabling them often causes more harm than good
[19:54] <melodie> should I comment "NoDisplay=true" before using the Hidden feature?
[19:55] <melodie> jbicha perhaps the person who manages the package Evolution could mark at-spi2 as "recommands" and not as hard dependency?
[19:57] <melodie> quitting the session and coming back to test
[19:57] <sarnold> melodie: if you just want to uninstall the thing, there's a package that lets you build a package with nothing in it, that serves only to make apt happy
[19:57] <sarnold> sigh
[20:01] <jbicha> melodie: from what I can see, it's not a hard dependency of Evolution; it's a hard dependency of gtk3
[20:02] <sarnold> melodie: if you just want to uninstall the thing, there's a package that lets you build a package with nothing in it, that serves only to make apt happy
[20:02] <sarnold> sadly I still haven't figured out the name of the package
[20:02] <melodie> jbicha strange
[20:02] <Laney> sarnold: equivs
[20:02] <melodie> jbicha because when I try to remove the package it wants to remove Evolution specifically
[20:03] <sarnold> Laney: that's the one! thanks :D
[20:03] <sarnold> melodie: maybe the 'equivs' package can help you
[20:03] <jbicha> melodie: it probably wants to remove everything gtk3 on your system!
[20:03] <melodie> jbicha I don't think so
[20:04] <jbicha> the package that has that autostart is at-spi2-core: https://packages.debian.org/sid/amd64/at-spi2-core/filelist
[20:05] <melodie> jbicha and?
[20:05] <jbicha> never mind, it's a different related pkg that gtk3 depends on
[20:06] <melodie> tried what you suggested it didn't work
[20:06] <Laney> You probably want to discuss this in a support channel
[20:07] <melodie> Laney it would be better if someone in the dev team could tell me if a service file would help have the hand on it
[20:07] <melodie> I'm not used to write any of these service files, but if I have to to be able to manage the service I might try to give it a go
[20:08] <melodie> then I could share it with other people
[20:08] <Laney> XDG autostart files can be overridden in your home directory
[20:08] <melodie> Laney that's what jbicha suggested but it didn't work
[20:09] <melodie> http://paste.ubuntu.com/23990496/
[20:09] <dobey> or by using an environment which doesn't do xdg autostart
[20:09] <melodie> dobey a
[20:09] <melodie> what about having a service file which allows people to manage their services the way they like it?
[20:10] <jbicha> melodie: there's also systemd user services, at-spi-dbus-bus.service and at-spi-registryd.service
[20:10] <melodie> jbicha in this case I try systemctl commands now
[20:10]  * Laney requests that you go to a support channel for 'tweaker' discussions