[01:18] <mup> PR snapcraft#922 closed: On gradle unit tests, unset the proxies <Created by elopio> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/922>
[03:06] <tsimonq2> cwayne: Quick, propose that your snap be merged: https://github.com/atom/atom/pull/13293
[03:06] <mup> PR atom/atom#13293: Add Flatpak support <needs-review> <Created by mattdangerw> <https://github.com/atom/atom/pull/13293>
[03:06] <tsimonq2> :P
[05:27] <hades007> how to put an apparmor profile in complain mode with ubuntu core 15.04 ? sudo  "apparmor_parser -rC /var/lib/apparmor/profiles/click_lxd_lxc_2.0.5-1" doesn't to work
[07:12] <foxmask> bonjello
[07:48] <dholbach> hey hey
[09:39] <mup> Bug #1643816 opened: improve systemd configuration support <Snappy:New> <https://launchpad.net/bugs/1643816>
[10:04] <markusfluer> Good morning,
[10:04] <markusfluer> How would I set the prefixes for nginx in a snap?
[10:05] <markusfluer> Or should they be just untouched and let autotools manage them
[10:10] <zyga> markusfluer: hey, you can try to set the prefix to /snap/nginx/current/ and use organize to strip that path from installation
[10:10] <zyga> markusfluer: as snapd will ensure that at runtime the content of your snap is visible from that location
[10:12] <markusfluer> zyga: Thank you very much.
[10:13] <zyga> markusfluer: I don't remember how to use organize to do this though,
[10:13] <zyga> markusfluer: experiment on hello world or something that builds quickly
[10:13] <zyga> markusfluer: and when you do, please share
[10:14] <markusfluer> zyga: I will try it out and comment it in the snapcraft.yaml.
[10:14] <markusfluer> I will publish it on github anyway
[10:15] <zyga> thanks
[10:23] <didrocks> zyga: I thought we shouldn't hardcode /snap though?
[10:23] <didrocks> this is what niemeyer told multiple times
[10:23] <didrocks> (but we don't have any other answer AFAIK)
[10:24] <zyga> didrocks: well, once we get a better answer people can adapt
[10:25] <zyga> didrocks: walk, run, fly :)
[10:48] <mup> PR snapcraft#923 opened: WIP pluginhandler: source management moved to the core <Created by sergiusens> <https://github.com/snapcore/snapcraft/pull/923>
[11:43] <jon___> hi I've just followed the ubuntu core setup steps, all ok until it now asks for a login and password, any ideas what this is for an initial boot?
[13:37] <mup> Bug #1643885 opened: ubuntu core image has different content between archs <Snappy:New> <https://launchpad.net/bugs/1643885>
[13:43] <mup> Bug #1643888 opened: first snap try doesn't install core snap <Snappy:New> <https://launchpad.net/bugs/1643888>
[13:55] <jdstrand> tedg: hey, if I wanted to file a bug against the results of locking snaps to the unity7 launcher, where would I file it?
[13:55] <tsdgeos> maybe someone should try to approach Valve since they seem to be leaning to trying flatpack https://forums.xamarin.com/discussion/comment/232301/#Comment_232301 maybe they can be convinced to use snappy
[13:55] <mup> Bug #1643893 opened: snapd doesn't handle well flaky network <Snappy:New> <https://launchpad.net/bugs/1643893>
[13:56] <jdstrand> JamieBennett, dholbach: fyi tsdgeos comment ^
[13:56] <Trevinho> jdstrand: what's the bug?
[13:57] <Trevinho> jdstrand: anyway... unity project I guess, but it depends...
[13:58] <jdstrand> Trevinho: I'm trying to find a simple reproducer
[13:58] <Trevinho> jdstrand: ok, ping me once you've it
[13:59] <jdstrand> Trevinho: basically what I see happen sometimes is that when I lock to the launcher, /snap/<name>/path/to/exe is used instead of /snap/bin/name
[13:59] <jdstrand> Trevinho: which is a pretty serious security bug because /snap/<name>/path/to/exe runs outside of confinement
[14:00] <Trevinho> jdstrand: I guess that's because the app didn't match properly to something, right?
[14:00] <Trevinho> jdstrand: howver, that's BAMF..
[14:00] <Trevinho> jdstrand: like you're running a snap from a shell, then you try to lock it, right?
[14:00] <seb128> tsdgeos, that's probably something worth mentioning on the snappy mailing list, IRC backlog info is likely to get lost
[14:00] <jdstrand> Trevinho: I guess?? tbh I've never, ever, ever understood how that matching was supposed to work, no matter how hard I tried
[14:01] <jdstrand> Trevinho: that is one scenario. I've also launched from the dash and seen the same thing
[14:01] <jdstrand> Trevinho: but not everything. eg, vlc is working fine
[14:01] <Trevinho> jdstrand: so..... I asked multiple times to include inside the wrapper script a BAMF env variable that points to the actual desktop file when possible...
[14:01] <mup> Bug #1642900 changed: libgcc_s.so.1 not found by app using ubuntu-app-platform content snap <Ubuntu App Platform:New> <https://launchpad.net/bugs/1642900>
[14:02] <Trevinho> jdstrand: so, in case that there's no other way to match the app, then we try to undestand what binary launched it... Thus the problem you mentioned. This actually worked well in non snap-world. But I guess we've to tune it too
[14:04] <Trevinho> jdstrand: have a look to this though: https://github.com/snapcore/snapd/pull/1580#issuecomment-234546220
[14:04] <mup> PR snapd#1580: wrappers: set BAMF_DESKTOP_FILE_HINT for unity <Created by mvo5> <Closed by chipaca> <https://github.com/snapcore/snapd/pull/1580>
[14:04] <Trevinho> jdstrand: if that was done in the wrapper script, then it would fix also the issue you're talking about
[14:05] <Trevinho> indeed we've to cover also the case where a desktop file is not provided, and avoid doing what you mentioned, but this would b already a good point
[14:05] <jdstrand> Trevinho: maybe I will file a bug against snapd then
[14:06] <jdstrand> Trevinho: then point you at it for comment?
[14:06] <Trevinho> jdstrand: ok
[14:07] <JamieBennett> Trevinho, jdstrand, After a quick search I see no confirmation of Valve using flatpak atm, just a bunch of very vocal non-Valve people talking about flatpak and discounting snaps for no apparent reason than it comes from Canonical (which is try but is also wrong as the snap format is governed outside of Canonical). ev_, dholbach, or popey may have more context though.
[14:07] <Trevinho> jdstrand: here there's more context https://trello.com/c/xP1hN3BF/152-improve-desktop-file-support-by-adding-a-new-bamf-desktop-file-hint-environment-hint
[14:07] <dholbach> I don't have much info on valve's use of snaps.
[14:11] <popey> me either
[14:14] <tsdgeos> seb128: i'm not on that list i think
[14:14] <tsdgeos> seb128: what's the list name/address/mailman/thing?
[14:15] <seb128> tsdgeos, seems JamieBennett is on the topic since he just mentioned it so I guess you can consider it as something the snappy team is aware of and probably no need to email the list
[14:16] <tsdgeos> ok :)
[14:22] <jdstrand> JamieBennett, dholbach: I think tsdgeos point was that perhaps we should be proactive rather the reactive, but I'll leave it in your hands
[14:22] <JamieBennett> jdstrand, understood
[14:26] <mup> Bug #1643910 opened: BAMF_DESKTOP_FILE_HINT not set in correct place for unity7 <Snappy:New> <https://launchpad.net/bugs/1643910>
[14:26] <jdstrand> Trevinho: ^
[14:26] <Trevinho> ack
[14:27] <jdstrand> mvo: fyi, I think bug #1643910 needs attention sooner than later. it can result in snaps running unconfined under certain circumstances under unity7
[14:27] <mup> Bug #1643910: BAMF_DESKTOP_FILE_HINT not set in correct place for unity7 <Snappy:New> <https://launchpad.net/bugs/1643910>
[14:30] <mhall119> jdstrand: panteon-mail needs #1590679, and so do most (or all?) KDE apps.
[14:30] <mup> Bug #1590679: Apps can't own session bus names (unity7 interface) <snap-desktop-issue> <snapd-interface> <Snappy:In Progress by jdstrand> <snapd (Ubuntu):Confirmed> <https://launchpad.net/bugs/1590679>
[14:31] <mhall119> sitter has said that it's the only blocker for most of his KDE snaps
[14:31] <mhall119> some of them have other blockers, but the simple ones just need that
[14:32] <jdstrand> mhall119: are there devmode apps for these things?
[14:32] <jdstrand> mhall119: I guess panteon-mail is devmode? are there any KDE apps that want to talk to each other?
[14:32] <jdstrand> sitter: ^
[14:33] <mhall119> jdstrand: I think they use dbus to talk to themselves (to prevent multiple instances) or with the plasma shell, but sitter can give better info
[14:34] <mhall119> jdstrand: http://distribute.kde.org/snappy-testing/ has all their CI builds of snaps
[14:35] <jdstrand> sitter: do any two snaps in http://distribute.kde.org/snappy-testing/ want to talk to each other?
[14:42] <ppisati> ogra_: have you ever happened to build a kernel snap from -proposed?
[14:43] <ogra_> ppisati, sure, all the time (for rpi) ... but it indeed only works if meta and the kernel are in sync (i'm on vacation btw ;) )
[14:44] <ppisati> ogra_: ah. stay in vcation then
[14:44] <ogra_> heh
[14:45] <ppisati> ogra_: because even if i ticked 'use proposed for the build' it picked only stuff from -updates
[14:45] <ppisati> ogra_: when you are back?
[14:45] <ogra_> next monday ... you need to add proposed in the tree too iirc
[14:48] <ogra_> ppisati, there is code to copy the sources.list from the outer build root to the inner one http://bazaar.launchpad.net/~snappy-dev/kernel-snap-makefile/trunk/view/head:/Makefile#L19 ... so you need to make sure our outer chroot has proposed
[14:50] <ppisati> ogra_: "cp /etc/apt/sources.list chroot/etc/apt/sources.list" - yeah, i saw that, and my build was against -proposed
[14:51] <ppisati> ogra_: so i expected it to be enabled
[14:51] <ppisati> Pocket for automatic builds:
[14:51] <ppisati>     Proposed
[14:51] <ppisati> ogra_: anyhow, you are on vacation so i won't bother anymore
[14:51] <ogra_> but the PPA isnt ... (whiczh you need additionally)
[14:51] <ogra_> here is some other code ... http://bazaar.launchpad.net/~ogra/+junk/kernel-snap-makefile-proposed/view/head:/Makefile
[14:51] <ogra_> (see line 24/25)
[14:52] <ppisati> ogra_: i see
[14:53] <ogra_> you coudl add a config file to the outer branch (the one with snapcraft.yaml) and then read a flag to make that dynamically switchable
[14:59] <gerry_> hi, i have succesfully uploaded my app to the snap store, what is the best way to edit its entry in "Software" ?
[15:00] <didrocks> gerry_: hey! I would say try to go to myapps.ubuntu.com and edit entries there (not sure you can edit them all those)
[15:01] <mvo> jdstrand: let me read the bug again
[15:02] <jdstrand> Trevinho: I think there may be a unity7 component to that. please see my comments and add the tasks if you feel it is appropriate
[15:03]  * Trevinho checks
[15:03] <Trevinho> jdstrand: ah, since I wanted to try your new interfaces... I tried to compile snapd but it seems it failed here... So, isn't there a daily ppa or something?
[15:03] <gerry_> didrocks:  thanks for your help its just at the moment all I have got is "version unknown, source unknown and nonfree!
[15:04] <didrocks> gerry_: oh right, you should edit them thus! :)
[15:04] <jdstrand> Trevinho: I think so, I don't know where it is
[15:04] <jdstrand> mvo: ^
[15:06] <jdstrand> Trevinho: refresh to see comment #5 as well
[15:06] <jdstrand> I feel like there are multiple things happening
[15:07] <jdstrand> like, maybe if things match (eg, vlc and gnome-easytag) things work, but if they don't, items are created in ~/.local/share/applications
[15:08] <mvo> Trevinho: please make sure you ran ./get-deps.sh to ensure your stuff is up-to-date
[15:10] <elopio> vila: ping. Have you seen this? bzr: ERROR: Connection error: while sending POST
[15:10] <jdstrand> Trevinho: ok, and comment #6
[15:10] <jdstrand> ah, you're notified, so I'll stop pinging you here
[15:11] <vila> elopio: that single line rings no bell, more context ?
[15:18] <mup> PR snapcraft#924 opened: New plugin: jhbuild <Created by attente> <https://github.com/snapcore/snapcraft/pull/924>
[15:21] <mup> PR snapcraft#812 closed: New plugin: jhbuild <Created by attente> <Closed by attente> <https://github.com/snapcore/snapcraft/pull/812>
[15:25] <elopio> vila: http://paste.ubuntu.com/23517256/ All our bzr test in arm are failing in a similar way.
[15:29] <vila> elopio: bzr: ERROR: Connection error: while sending POST /~joetalbott/%2Bjunk/testpart/.bzr/smart: [Errno 110] Connection timed out
[15:29] <vila> elopio: firewall issue ?
[15:31] <jdstrand> willcooke, JamieBennett: fyi, bug #1643910. both mvo and Trevinho are aware of it so fyi only. this is a pretty significant usability and security bug when it hits. Please see my comments as well as the description
[15:31] <mup> Bug #1643910: BAMF_DESKTOP_FILE_HINT not set in correct place for unity7 <Snappy:Triaged> <bamf (Ubuntu):Triaged by 3v1n0> <https://launchpad.net/bugs/1643910>
[15:31] <jdstrand> ratliff, tyhicks: also, fyi ^
[15:32] <jdstrand> willcooke, JamieBennett: not saying anything needs to be done more than what they are already working on, but wanted you guys to be aware of the unpredictable behavior and security implications
[15:33] <jdstrand> so that the bug doesn't get deprioritized
[15:33] <jdstrand> ok, is done with that bug for now :)
[15:34] <lof_mt> hi flexiondotorg can you spare some minutes right now?
[15:34] <jdstrand> mhall119: fyi, panteon-mail is not in the store
[15:35] <flexiondotorg> lof_mt, In a meeting. Give me 30mins...
[15:38] <jdstrand> mhall119: do you know which of the snaps in http://distribute.kde.org/snappy-testing/ work? kmahjongg for example does not
[15:39] <mhall119> jdstrand: try kblocks
[15:39] <mhall119> or ktuberling
[15:39] <jdstrand> neither does ktuberling
[15:40] <jdstrand> $ ktuberling
[15:40] <jdstrand> /snap/ktuberling/x1/bin/qt5-launch: line 75: /snap/ktuberling/x1/usr/games/ktuberling: No such file or directory
[15:40] <jdstrand> ok, I'll try kblocks
[15:43] <elopio> vila: perhaps. But in theory, the rules should be the same for all these machines, and it doesn't happen in amd64. I will ask IS anyway.
[15:43] <elopio> thanks!
[15:44] <mhall119> jdstrand: kblocks and ktuberling are both also in the edge channel of the store
[15:44] <jdstrand> mhall119: kblocks fails
[15:44] <mhall119> ok, maybe it's an issue with those builds
[15:44] <jdstrand> kalgebra launched
[15:46] <jdstrand> mhall119: what is the name of the platform snap again?
[15:47] <jdstrand> mhall119: for kde. ktuberling is 5M and is complaining about not finding kde libs
[15:47] <jdstrand> snap find is not helping
[15:48] <Elleo> jdstrand: heya, would you mind giving this a review when you have a chance? https://github.com/snapcore/snapd/pull/2328
[15:48] <mup> PR snapd#2328: Add download manager interface <Created by Elleo> <https://github.com/snapcore/snapd/pull/2328>
[15:48] <jdstrand> kde-frameworks-5?
[15:48]  * jdstrand tries
[15:48] <jdstrand> Elleo: ok, I'll add it to the list
[15:48] <Elleo> jdstrand: thanks :)
[15:49] <mhall119> jdstrand: yes, from --edge
[15:50] <mhall119> snap find won't help,  because they're not in stable
[15:51]  * jdstrand nods
[15:51] <jdstrand> seems like kde-frameworks-5 could've been in stable...
[15:53] <gerry_> I have filled in all the forms at myapp.ubuntu but for some reason it does not appear in my entry in "software" do I have to change how its listed in my yaml somehow?
[15:53] <mhall119> jdstrand: maybe, but without stable app snaps there wouldn't be much point
[15:54] <jdstrand> mhall119: fyi, installing that and making the interface connection results in the same issue
[15:54] <jdstrand> mhall119: I have to /usr/lib/snapd/snap-discard-ns ktuberling for it to work
[15:55] <gerry_> or do I have to find a way to re-publish my app?
[16:01] <kalikiana_> gerry_: doesn't appear meaning what? it's not listed at all?
[16:01] <kalikiana_> how did you upload it?
[16:02] <kalikiana_> can you see it in myapps in the list?
[16:06] <icey> how long does stuff usually take to get from master on github to Launchpad builders?
[16:07] <kyrofa> icey, you mean with git-to-git imports?
[16:07] <icey> kyrofa: I mean: once a patch to snapcraft lands, how long does it usually take before the patch is in the launchpad snappy builders?
[16:08] <kyrofa> icey, ah ha
[16:08] <kyrofa> icey, well, the release process of snapcraft involves an SRU. Are you familiar with that?
[16:08] <kyrofa> icey, long story short, about a week or so
[16:08] <icey> kyrofa: yea
[16:09] <icey> kyrofa: that's a nice timeline ;-) any chance for a review on my PR? https://github.com/snapcore/snapcraft/pull/908
[16:09] <mup> PR snapcraft#908: Let Rust plugin fetch dependencies in pull <Created by ChrisMacNaughton> <https://github.com/snapcore/snapcraft/pull/908>
[16:09] <kyrofa> icey, snapcraft releases weekly, with a little slosh for SRU
[16:10] <kyrofa> icey, don't worry, we'll go back through it!
[16:10] <icey> heh kyrofa, it should have everything asked for now, and it's passing tests :)
[16:10] <kyrofa> Excellent
[16:10] <icey> I'm hoping to get it in soonish, since I wrote about it and started talking about it to the Rust community :)
[16:11] <icey> kyrofa: and apparently people are interested: https://www.reddit.com/r/rust/comments/5e754t/making_rust_packages_for_linux_with_snappy/
[16:12] <kyrofa> icey, good to hear!
[16:23] <lof_mt> ping flexiondotorg
[16:24] <flexiondotorg> lof_mt, Hi there :-)
[16:24] <flexiondotorg> What's up?
[16:24] <lof_mt> do you have a bit of time to help me with the gsettings problem?
[16:24] <gerry_> kalikana: sorry phone call
[16:27] <flexiondotorg> lof_mt, I can try.
[16:27] <flexiondotorg> What is the issue?
[16:27] <lof_mt> flexiondotorg: This is the current snapcraft file: https://paste.ubuntu.com/23513341/
[16:27] <lof_mt> builds well under 16.10 (maybe on 16.04 too not sure about gtk).
[16:27] <mvo> jdstrand, Trevinho: I was a bit busy before but now I looked at the BAMF_DESKTOP_FILE_HINT issue. I'm not sure how what the behavior should be, what if the snap installs multiple desktop files that launch the same binary (say libreoffice with libreoffice --writer). trying to match what desktop file to pick sounds fragile. I think I could support the easy case but it seems far from ideal
[16:28] <lof_mt> But the application crashes when run, because it lacks gsettings schemas. And i could not find documentation about how to include it.
[16:28] <didrocks> lof_mt: you are using the desktop-gtk3 cloud parts
[16:28] <flexiondotorg> lof_mt, The first thing I notice is that you don't have any stage packages.
[16:28] <gerry_> it is listed there is just no summary or details all it has is "version unknown source unknown"
[16:29] <didrocks> lof_mt: all doc is on snapcraft define desktop-gtk3
[16:29] <mvo> jdstrand, Trevinho: e.g. "Exec=libreoffice --writer " and the user uses "libreoffice ~/some/doc --writer" , the parser would have to understand about reordering of args,  etc and probably stll get it wrong because for some apps reodering is legal for some its not etc
[16:29] <didrocks> basically you need:
[16:29] <didrocks> - the home plug for gsettings
[16:29] <didrocks> - and prepend your command with desktop-launcher
[16:29] <didrocks> desktop-launch*
[16:30] <didrocks> flexiondotorg: he doesn't need all stage-packages, the desktop part provides most of them (but yeah, maybe more will be needed)
[16:30] <flexiondotorg> Yep.
[16:37] <lof_mt> didrocks: What do you mean with "the home plug for gsettings"? I have both the home plug and the gsettings plug for the application.
[16:38] <didrocks> lof_mt: oh sorry, as they were not in ascii ordered, I missed it, yeah, you are good for that one :)
[16:38] <didrocks> you only miss desktop-launch prepending thus
[16:38] <flexiondotorg> lof_mt, I tried to build in lxd. You're missing dependencies.
[16:39] <flexiondotorg> Check the Cmake output.
[16:39] <lof_mt> Do you use the cloud or container image? I only checked with desktop and server.
[16:41] <mvo> jdstrand, Trevinho: followed up in the bug, happy to implement naive matching in snap run if that helps but see the bugreport for my questions
[16:43] <kyrofa> zyga, ping
[16:44] <flexiondotorg> lof_mt, I used 'snapcraft cleanbuild' which requires you have lxd installed.
[16:45] <flexiondotorg> It provisions a temporary lxc container to build the snap in.
[16:45] <lof_mt> will test that
[16:49] <lof_mt> When can you omit the bin/ in command:? I saw it in the leafpad example. https://github.com/ubuntu/snappy-playpen/blob/master/leafpad/snapcraft.yaml
[17:00] <rvr> ogra_: fgimenez: Are you aware of any problem with edge and pi2/pi3? I can't boot any image (Waiting for ethernet connection... unable to connect)
[17:02] <fgimenez> rvr, i haven't worked with it in a while, i'll give it a try and let you know how it goes
[17:14] <mhall119> jdstrand: I'm afraid you'll have to get sitter or sgclark to help you here, they know the apps better than I do
[17:18] <willcooke> jdstrand, ack re: #1643910.  thanks
[17:18] <mup> Bug #1643910: BAMF_DESKTOP_FILE_HINT not set in correct place for unity7 <Snappy:Triaged> <bamf (Ubuntu):Triaged by 3v1n0> <https://launchpad.net/bugs/1643910>
[17:53] <lof_mt> the lacking dependency was "gettext".
[17:54] <dslul> Hello, may I ask you, is the raspberry pi3 core16 image still 32bit only?
[17:59] <ogra_> dslul, yes
[18:01] <ogra_> (the 64bit one eats a lot more ram and you might be missing driver support for binary stuff ... there will be a 64bit image at some point, but thats rather low prio)
[18:05] <mattyw> cprov, ping?
[18:06] <cprov> mattyw: hi there
[18:06] <jdstrand> niemeyer: no rush, just fyi that I responded to your question in the pr
[18:06] <jdstrand> niemeyer: (and hi! :)
[18:50] <niemeyer> jdstrand: Heya
[18:50] <niemeyer> jdstrand: Looking
[18:56] <niemeyer> jdstrand: Your idea of allowing the alternation seems like the best option
[18:57] <niemeyer> jdstrand: So we have name: org.foo.bar, and it only matches against another org.foo.bar, but it's okay to use an interface with (-[0-9]*)?
[18:58] <niemeyer> actually, (-[0-9]+)?
[19:00] <stgraber> jdstrand: heya, sorry to nag but have you heard back from the store deploy (or whatever was needed) to have the "lxd" slot work on the lxd snap? had a few people ask about that interface lately as it was mentioned in some recent blog post...
[19:01] <plars> elopio: is there a built-in kill timeout on spread? I seem to be hitting that while it's still installing packages on dragonboard, so it can't even get to the first tests because it kills it too soon
[19:05] <lof_mt> Is there something built in to deploy gsettings schemas? Or do i need to add it to the launcher like in the plank example? → https://github.com/ubuntu/snappy-playpen/blob/af7154cc9cac61d55b17c4aa9517c4d7c003a1a0/plank/launcher
[19:10] <icey> kyrofa: I'm working on cutting out build-essential from the Rust plugin now
[19:11] <icey> kyrofa: looks like it's generally suggested to do build-essential though: https://users.rust-lang.org/t/installation-problem-on-debian-cc-linker/6192
[19:17] <jdstrand> niemeyer: yes. I committed some changes to show how it works
[19:17] <jdstrand> roadmr: hey, what revision is the store at now?
[19:18] <icey> kyrofa: pushed change without build-essential; does t he build environment for Snapcraft's CI look like the Launchpad builders?
[19:21] <jdstrand> roadmr: nm
[19:22] <jdstrand> stgraber: it looks like the store now has r798 of the review tools, so I'll update the lxd snap declaration.
[19:23] <jdstrand> stgraber: ok done. feel free to upload
[19:24] <stgraber> jdstrand: thanks
[19:26] <kkimdev> Q: is there a way to install a snappy package without root?
[19:30] <niemeyer> jdstrand: This looks super verbose: +    name=###DBUS_NAME###-{[1-9],[1-9][0-9],[1-9][0-9][0-9],[1-9][0-9][0-9][0-9],[1-9][0-9][0-9][0-9][0-9],[1-9][0-9][0-9][0-9][0-9][0-9]},
[19:30] <niemeyer> jdstrand: Is that the best we can have with apparmor?
[19:31] <niemeyer> jdstrand: What's the actually supported syntax there?
[19:32] <jdstrand> niemeyer: it is very precise to match pids. the aare is not normal regex
[19:32] <jdstrand> niemeyer: so, yes, that is the best
[19:32] <jdstrand> niemeyer: I mean, we could be less precise, but since we are trying to support pids, I went with very precise
[19:33] <niemeyer> jdstrand: That's good. The question was really if there wasn't richer syntax we could (ab)use
[19:34] <jdstrand> niemeyer: unfortunately, no. the aare is quite flexible but it has its rough spots
[19:36] <niemeyer> jdstrand: Can we nest {}?
[19:36] <niemeyer> Actually, we don't even have to
[19:37] <niemeyer> jdstrand: -[1-9]{,[0-9]}{,[0-9]}{,[0-9]}{,[0-9]}{,[0-9]}
[19:38] <jdstrand> I don't think that is going to eveluate right. note that I stole this from the apparmor upstream tunable for @{pid}
[19:38] <jdstrand> yes, that won't evaluate right
[19:38] <jdstrand> well, maybe
[19:38]  * jdstrand thinks
[19:39] <jdstrand> actually, it looks like it would. let me test something
[19:42] <jdstrand> yeah, that's fine
[19:42]  * jdstrand adjusts
[19:43] <niemeyer> Nice, thanks
[19:43] <niemeyer> jdstrand: Why the deny-connection in the base-declaration?
[19:43] <mup> Bug #1618095 changed: [SRU] 2.14.2 <verification-done> <Snappy:Fix Released> <snapd (Ubuntu):Fix Released> <snapd (Ubuntu Xenial):Fix Released> <https://launchpad.net/bugs/1618095>
[19:50] <dslul> hello, I installed ubuntu core 16 on my raspberry pi3, is it possible to configure it only via ssh? what are the login credentials?
[19:54] <roadmr> jdstrand: ... yes, 798.
[19:54] <roadmr> jdstrand: are you now using snap-declaration plugs/slots for lxd? \o/ Let me know how it works :)
[19:54] <roadmr> jdstrand: I used it for a test snap from the checkbox folks, seems to be working nicely
[19:59] <jdstrand> roadmr: yes, I just added a snap declaration for the lxd snap
[19:59] <roadmr> awesome :)
[19:59] <jdstrand> well, for the slots side. it already had the plugs side
[20:11] <niemeyer> jdstrand: Sent a review
[20:27] <dslul> Is it possible to do the initial configuration of Ubuntu Core using ssh and not with a usb keyboard and a screen?
[20:28] <dslul> I see that the ssh daemon is running, but I don't know the credentials
[20:43] <stgraber> jdstrand: lxd snap updated in all channels for the new slot
[20:43] <jdstrand> stgraber: nice! :)
[21:01] <dslul> nobody knows the answer?
[21:02] <mwhudson> dslul: no,  it is not
[21:02] <mwhudson> dslul: you can do it using serial as well
[21:03] <dslul> how can I do it using serial?
[21:03] <dslul> I mean, I know how to connect to a serial, but what then?
[21:05] <mwhudson> dslul: press enter when connected and it should work the same way as a the tty
[21:06] <dslul> ok thank you very much, I'll try this
[21:08] <kyrofa> icey, sorry for the delay, I was finally getting my couches delivered after nearly three months!
[21:09] <kyrofa> icey, I wouldn't rely on the snapcraft test infra being as vanilla as the LP builders-- you might want to double check there. But honestly, just make a clean lxc or run cleanbuild
[21:10] <kyrofa> And you should be able to determine if you've satisfied dependencies
[21:54] <wililupy> Question: Is there a specific reason why in the grub.cfg we use net.ifname=0?