[00:23] <mup> PR snapcraft#1547 closed: plugins: extract sitecustomize logic from python <Created by kyrofa> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/1547>
[05:52] <zyga-ubuntu> good morning
[06:05] <mvo> hey zyga-ubuntu!
[06:05] <mvo> zyga-ubuntu: do you want me to do the reshuffle for 3814? looks like gustavo reviewed it last night
[06:06] <zyga-ubuntu> mvo: looking
[06:08] <zyga-ubuntu> mvo: yeah, go for it, thank you
[06:09] <mvo> zyga-ubuntu: sure thing, happy to help
[06:18] <zyga-ubuntu> mvo: thank you, I'm trying to go over https://github.com/snapcore/snapd/pull/3621
[06:18] <mup> PR #3621: cmd/snap-{confine,update-ns}: apply mount profiles using snap-update-ns <Created by zyga> <https://github.com/snapcore/snapd/pull/3621>
[06:19] <mvo> zyga-ubuntu: ok
[07:59] <Chipaca> good morning all
[08:00] <zyga-ubuntu> hey Chipaca
[08:00] <Chipaca> today's going to be less Chipacaful than usual: i've got physio in about an hour, and then i need to hop on a train to london, and then race back for a school run. I'll be working on the train, but connectivity is very spotty (and nonexistent in the underground portions of it)
[08:01] <zyga-ubuntu> Chipaca: I made that package with %dir to ensure I understand what's going on
[08:01] <zyga-ubuntu> Chipaca: ouch, sounds like a hard-to-focus day :(
[08:01] <Chipaca> zyga-ubuntu: thank you!
[08:01] <Chipaca> zyga-ubuntu: how did it go?
[08:02] <zyga-ubuntu> Chipaca: I built it and didn't install it :') I was exhausted and collapsed sleeping :)
[08:02] <zyga-ubuntu> Chipaca: I can try it and share it now
[08:02] <Chipaca> heh
[08:02] <Chipaca> no worries
[08:02] <zyga-ubuntu> one sec
[08:02] <Chipaca> zyga-ubuntu: so the question is whether it gets cleaned up on removal
[08:02] <zyga-ubuntu> right
[08:02] <zyga-ubuntu> installing
[08:02] <Chipaca> and if not, how to make it do that
[08:03] <Chipaca> zyga-ubuntu: opensuse, right?
[08:03] <zyga-ubuntu> yes
[08:04] <zyga-ubuntu> Chipaca: ok, removing the package removes the directory
[08:04] <zyga-ubuntu> it was empty
[08:04] <Chipaca> zyga-ubuntu: but it won't be
[08:05] <Chipaca> zyga-ubuntu: what happens if you drop a file in it?
[08:05] <zyga-ubuntu> Chipaca: trying with some hand-made files
[08:05] <Chipaca> yeah
[08:05] <zyga-suse> Chipaca: it stays behind
[08:06] <zyga-suse> Chipaca: you need a %postun section then
[08:06] <zyga-suse> https://fedoraproject.org/wiki/Packaging:Scriptlets
[08:06] <zyga-ubuntu> Chipaca: note that there are two "variants", one on remove and one on purge
[08:06] <zyga-ubuntu> or actually, one on upgrade and one on uninstall
[08:07] <zyga-ubuntu> Chipaca: look at the "Syntax" section there, then at the table inside it
[08:07] <zyga-ubuntu> yes, the values are actually $1 == 0 or $1 == 1
[08:07] <zyga-ubuntu> because words are long and rpm is old
[08:08] <Chipaca> we already have a %postun
[08:08] <Chipaca> in fact it has a preun and a postun and both are the same (?)
[08:08] <zyga-ubuntu> so it needs to work on the cache dir
[08:08] <zyga-ubuntu> hmm? that's odd
[08:11] <Chipaca> zyga-ubuntu: it's 0, 1, or more because it's “the number of packages of this name which will be left on the system when the action completes”
[08:12] <zyga-ubuntu> Chipaca: aha, curious
[08:13] <Chipaca> zyga-ubuntu: sorry, i misread
[08:13] <Chipaca> they're very similar but different
[08:13] <Chipaca> service_del_preun and service_del_postun
[08:18] <Chipaca> zyga-ubuntu: do you know if service stopping happens in preun or postun?
[08:19] <zyga-ubuntu> I suspect in preun, but I don't know
[08:19] <zyga-ubuntu> it's not automatic
[08:19] <zyga-ubuntu> it must be done by our code explicitly
[08:19]  * zyga-ubuntu notices he was off wifi all day
[08:19] <pedronis> this continues to happen relatively often: Error: Failed to synchronize cache for repo 'updates' in fedora prepare
[08:21] <Chipaca> zyga-ubuntu: the service_del_preun/_postun say they stop it (unless asked not to), but don't say _which_ stop it
[08:23] <zyga-ubu1tu> 10:19 < zyga-ubuntu> I suspect in preun, but I don't know
[08:23] <zyga-ubu1tu> 10:19 < zyga-ubuntu> it's not automatic
[08:23] <zyga-ubu1tu> 10:19 < zyga-ubuntu> it must be done by our code explicitly
[08:23] <zyga-ubu1tu> 10:19  * zyga-ubuntu notices he was off wifi all day
[08:23] <zyga-ubu1tu> 10:19 < pedronis> this continues to happen relatively often: Error: Failed to synchronize cache for repo 'updates' in fedora prepare
[08:23] <zyga-ubu1tu> 10:20 < zyga-ubuntu> pedronis: aha, thank you
[08:23] <zyga-ubu1tu> 10:20 < zyga-ubuntu> pedronis: maybe fedora has some auto-mirror system and we choose a bad mirror
[08:23] <zyga-ubu1tu> 10:20 < zyga-ubuntu> pedronis: I'll try to see what's the case there
 zyga-ubuntu: the service_del_preun/_postun say they stop it (unless asked not to), but don't say _which_ stop it
[08:25] <zyga-ubu1tu> which stop it?
[08:25]  * zyga-ubu1tu looks
[08:26] <zyga-ubu1tu> Chipaca: sorry, can you rephrase?
[08:27] <Chipaca> zyga-ubu1tu: in https://en.opensuse.org/openSUSE:Systemd_packaging_guidelines
[08:29] <Chipaca> zyga-ubu1tu: it says that using "these macros" (%service_add_pre/_add_post/_del_preun/_del_postun) the default is to restart on upgrade and stop on removal
[08:29] <Chipaca> zyga-ubu1tu: it doesn't say which macro actually does it
[08:30] <zyga-ubu1tu> ah, I see
[08:30] <zyga-ubu1tu> let me expand them
[08:30] <zyga-suse> service_add_pre: https://paste.gnome.org/prvbogtlc
[08:31] <zyga-suse> service_add_post: https://paste.gnome.org/pbub2ldyo
[08:31] <zyga-suse> service_del_preun: https://paste.gnome.org/p2aogqht0
[08:31] <zyga-suse> service_del_postun: https://paste.gnome.org/p6nylwgov
[08:32] <zyga-ubu1tu> Chipaca: I think it's clear from those macros now
[08:32] <zyga-ubu1tu> Chipaca: service_del_preun is stopping stuff
[08:33] <zyga-ubu1tu> (disable then optionally stop)
[08:33] <zyga-ubu1tu> (but optionally is actually on by default
[08:34] <zyga-ubu1tu> then service_del_postun just removes any sysv migrated service files and reloads systemd
[08:34] <mup> PR snapd#3914 closed: snap-seccomp: skip mknod syscall on arm64 <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/3914>
[08:34] <zyga-ubu1tu> Chipaca: note that this is tumbleweed, leap _may_ use different macros
[08:35] <zyga-ubu1tu> but I think the spirit is the same
[08:35] <Chipaca> 's fine. I remove the files in preun if $1=0, but after service_del_preun
[09:08] <mup> PR snapd#3908 closed: snap-seccomp, osutil: use osutil.AtomicFile in snap-seccomp <Created by mvo5> <Merged by pedronis> <https://github.com/snapcore/snapd/pull/3908>
[09:08]  * Chipaca made a mess of things, but won a half hour back (?)
[09:09] <zyga-ubu1tu> error: cannot refresh []: cannot refresh snap-declaration for "classic": Get https://assertions.ubuntu.com/v1/assertions/snap-declaration/16/QbSFwGGAgvG8zHl9nWLY7vEee8lhgFsp?max-format=2: x509: certificate has expired or is not yet valid
[09:09] <zyga-ubu1tu> this is from a dragonboard that is on core 1166
[09:10] <Chipaca> zyga-ubu1tu: check system time
[09:10] <zyga-ubu1tu> Chipaca: it's up-to-date now
[09:10] <zyga-ubu1tu> core syncs
[09:10] <zyga-ubu1tu> thanks!
[09:11] <zyga-ubu1tu> so, now I have a ppc box, an aarch64 box and an armhf box (with HDD) all available in case people want to log in
[09:11] <Chipaca> zyga-ubu1tu: Pharaoh_Atem: how do you say "/var/cache" in rpmspeak?
[09:12] <Chipaca> i'm guessing %{_sharedstatedir} is /var/lib
[09:12] <Chipaca> or is it /usr/lib
[09:13] <zyga-ubu1tu> Chipaca: one sec
[09:13] <Chipaca> definitely /var/lib. But what's /var/cache :-)
[09:14] <zyga-ubu1tu> juse use /var/cache
[09:14] <zyga-ubu1tu> or %{_localstatedir}/cache but YUCK
[09:15] <zyga-ubu1tu> when I see a distro that uses /smurfs for /var I'll regret
[09:15] <zyga-ubu1tu> but I don't regret not using those silly variables
[09:18] <Chipaca> zyga-ubu1tu: and how does arch handle uninstall?
[09:19] <zyga-suse> Chipaca: users do it
[09:19] <zyga-suse> Chipaca: it's all on the wiki typically
[09:20] <zyga-suse> Chipaca: there are scripts but they just mention "look at the wiki" I think
[09:20] <zyga-suse> though ...
[09:20] <zyga-suse> maybe not
[09:20] <zyga-suse> le me recheck
[09:27] <zyga-ubu1tu> Chipaca: confirmed, hand made
[09:28] <Chipaca> zyga-ubu1tu: can you edit the arch wiki so https://wiki.archlinux.org/index.php/Snapd doesn't talk about the ubuntu-core snap?
[09:28] <zyga-ubu1tu> checking
[09:31] <zyga-ubu1tu> Chipaca: ok, shall I just rename that to 'core'
[09:31] <Chipaca> :-)
[09:31] <zyga-ubu1tu> done
[09:31] <zyga-ubu1tu> Chipaca: note that anyone can sign up and edit
[09:32] <Chipaca> noted
[09:32] <zyga-ubu1tu> though it will ask you a pacman question when you try to make an account :D
[09:32] <zyga-ubu1tu> so :D
[09:32] <Chipaca> zyga-ubu1tu: but i'm racing to get a reasonable commit to #3866 in <5 minutes i have left before i need to run
[09:32] <mup> PR #3866: many: implement fetching sections and package names periodically <Blocked> <Created by chipaca> <https://github.com/snapcore/snapd/pull/3866>
[09:32] <zyga-ubu1tu> o/
[09:33] <Chipaca> there, that should do it
[09:34] <zyga-ubuntu> sigh
[09:34] <zyga-ubuntu> so gnome-shell sometimes crashes/logs-me-out when I exit my terminal
[09:34] <Chipaca> zyga-ubuntu: also, it says to use --devmode to sideload stuff
[09:35] <Chipaca> zyga-ubuntu: ouch
[09:39] <popey> ogra_:
[09:39] <popey> oops, ogra_ did you make any progress on nano pi ssh over usb?
[09:45] <ogra_> popey, not really
[09:46] <ogra_> i didnt forget about it :)
[09:46] <popey> ok. was hoping to take my nano pi to nyc, but not sure how to get it on the wifi there
[09:46] <ogra_> (there is no kernel driver fix yet )
[09:47] <ogra_> the only kernel were it wks is the 3.x BSP kernel
[09:47] <popey> cani re-run the network config thing over serial?
[09:47] <ogra_> sure
[09:48] <popey> what's the command?
[09:48] <ogra_> sudo consoole-conf
[09:48] <ogra_> -o
[09:49] <ogra_> alternative you can edit /etc/netplan/00-snapd-config.yaml
[09:49] <popey> perfecto! thanks
[09:49] <ogra_> and run "netplan generate" and "netplan apply"
[09:49] <popey> i love these console ui things. the conjure-up one is so pretty
[09:49] <ogra_> never tried it :)
[09:49] <popey> i tried it at ubucon paris after dustin did a quick talk of it
[09:50] <ogra_> is it of any use in non-cloud ?
[09:50] <popey> lulz
[09:50] <popey> Sure, if there's something complex you want to deploy locally
[09:50] <popey> (I tried kubernetes and killed my laptop - it's a bit of a monster)
[09:50] <ogra_> heh
[09:51] <popey> oddly console-conf has no exit option
[09:52] <ogra_> yeah, you need to run through it completely
[09:52] <popey> strange, should have a [quit] option imo
[09:52] <ogra_> (it skips the use creation if a user exists)
[09:52] <ogra_> well, if invoked from a tty it should
[09:52] <ogra_> if invked on first boot we want to avoid that
[09:53] <ogra_> file a feature request at mwhudson's desk ;)
[09:53] <popey> will do :)
[10:00] <ogra_> hrm
[10:01] <ogra_> running gradio on my desktop ends up in a constantly popping up notification
[10:01] <ogra_> and yuo cant actually stop it ... urgh
[10:02] <ogra_> (i mean the whole app ... it keeps playing when closing it)
[10:02] <popey> https://github.com/haecker-felix/gradio/issues
[10:02] <popey> best place to file issues
[10:02] <ogra_> well, i see edge is 4 versions ahead
[10:03]  * ogra_ tries that
[10:04] <ogra_> k, the notification only shows once with that
[10:05] <mup> PR snapd#3916 opened: add support for socket activation in apps <Created by albertodonato> <https://github.com/snapcore/snapd/pull/3916>
[10:22] <mwhudson> ogra_: i was looking at console-conf bugs today, they've been kinda neglected recently :(
[10:22] <mwhudson> popey: there is an old plan around using console-conf for fixing your network config only
[10:23] <ogra_> mwhudson, well, it works fine already
[10:23] <ogra_> properly skips user stuff if a user exists
[10:23] <ogra_> i use it all the time to reconfigure network stuff on my boards
[10:24] <ogra_> the prob is really that it has no exit option when invoked manually
[10:24] <ogra_> so in case there is an actual network issue you can never get back to your tty because it will not exit on its own nor can you stop it manually
[10:26] <mup> PR snapd#3898 closed: interfaces/network: allow using netcat as client <Created by alfonsosanchezbeato> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/3898>
[10:27]  * ogra_ glares at that PR ... huh ? i ave been using netcat in snaps since 15.04 without probs 
[10:28] <ogra_> ah, but abeato doesnt ship it in his snap ... k
[10:29] <abeato> ogra_, the MP is just to avoid having to do that... in the end nc is part of core, and that should be something stable
[10:29] <ogra_> yeah
[10:52] <ackk> where should I look for stuff that's failing in https://travis-ci.org/snapcore/snapd/builds/275400722 ?
[10:54] <zyga-ubuntu>        lookup api.snapcraft.io on [::1]:53: read udp [::1]:39352->[::1]:53: looks like a store issue
[10:57] <ackk> zyga-ubuntu, where did you see that?
[11:01] <zyga-ubuntu> ackk: in the log you provided
[11:01] <zyga-ubuntu> ackk: there are "folds" that can be expanded
[11:01] <zyga-ubuntu> ackk: a qucik tip is to look for "error preparing" or "error executing"
[11:01] <ackk> zyga-ubuntu, ah yeah, found it in the raw log
[11:01] <ackk> zyga-ubuntu, gotcha, thanks
[11:02] <ackk> zyga-ubuntu, is it possible to kick the travis job again?
[11:03] <zyga-ubuntu> yes, let me do that for you
[11:03] <ackk> zyga-ubuntu, thanks
[11:07] <mup> PR snapd#3917 opened: cmd,dirs: treat manjaro the same as arch <Created by zyga> <https://github.com/snapcore/snapd/pull/3917>
[11:39]  * zyga-ubuntu thinks about lunch now
[12:52] <mup> PR snapd#3918 opened: hooks: substitute env vars when executing hooks <Created by stolowski> <https://github.com/snapcore/snapd/pull/3918>
[13:00] <niemeyer> Hey folks
[13:01] <niemeyer> Won't be on the stand up today, apparently.. power outage here for a while
[13:01] <zyga-ubuntu> niemeyer: ok
[13:02] <niemeyer> Actually, I can join from the phone.. still enough battery.. hold on
[13:03] <Dmitrii-Sh> ogra_: I'll try the /snap/bin fix with a wayland session
[13:04] <ogra_> cool
[13:06] <diddledan> ikey: you about?
[13:06] <ikey> ish
[13:06] <ikey> sup?
[13:06] <diddledan> :-)
[13:06] <diddledan> can I pm a sec?
[13:06] <ikey> sure
[13:06] <zyga-ubuntu> mvo: gustavo just joined the main one
[13:07] <mvo> zyga-ubuntu: ok
[13:07] <ogra_> Dmitrii-Sh, oh, wit ... switching core to edge wont help since the profile.d snippet from the PR actually needs to come from the deb (it needs to live outside of the core snap in your session login setup)
[13:07] <ogra_> s/wit/wait/
[13:08] <Dmitrii-Sh> ogra_: hmm, shall I revert back?
[13:08] <ogra_> but 2.28 will have it ... https://forum.snapcraft.io/t/2-28-release-cycle-started/2026
[13:08] <Dmitrii-Sh> ogra_: I haven't tried yet, multi-tasking at the moment
[13:08] <ogra_> Dmitrii-Sh, well, check the black window issue ... thats unrealted to the PATH stuff
[13:08] <bloodearnest> Heya folks. From within a strictly confined snap, I'd like to find out my system's default dns resolver, but I can't read /etc/resolv.conf. Any ideas?
[13:09] <ogra_> bloodearnest, add the network interface to your snap
[13:09] <niemeyer> Sorry.. nevermind.. my phone reboots every time I join the hangout /o\
[13:09] <ogra_> (i think network is enough, perhaps network-bind though, if network alone doesnt help)
[13:09] <niemeyer> mvo: Thanks, please move on without me
[13:10] <bloodearnest> ogra_: thx, snap already has network and network-bind. Should that allow it to read /etc/resolv.conf?
[13:11] <bloodearnest> ogra_: ah, the app doing the configuration doesn't have those plugs. Will try that, thanks
[13:11] <ogra_> bloodearnest, hmm, actually ... grepping for resolv in the source only reveals network-control ...
[13:12] <ogra_> and network-manager ...
[13:12] <ogra_> but that kind of binds your app to desktop only
[13:25] <bloodearnest> ogra_: is read access to /etc/resolv.conf a security issue?
[13:25] <bloodearnest> feels like network plug ought to give read access, at least?
[13:36] <ogra_> bloodearnest, yeah, it should ... ask jdstrand :)
[13:37] <zyga-ubuntu> bloodearnest: let me check
[13:37] <zyga-ubuntu> bloodearnest: nothing allows it explicitly, let me check abstractions
[13:37] <__chip__> hey all
[13:38] <__chip__> sorry i dropped :-/
[13:38] <zyga-ubuntu> bloodearnest: it is provided through the nameservice abstraction that is used by a number of interfaces, including network
[13:38] <zyga-ubuntu> bloodearnest: but because /etc/ is shared from the host, cna you please tell me if /etc/resolv.conf is a symlink?
[13:39] <bloodearnest> zyga-ubuntu: it is a symlink
[13:39] <zyga-ubuntu> bloodearnest: and also tell me which distribution are you on ("snap version" output helps)
[13:39] <zyga-ubuntu> bloodearnest: right, what's the target?
[13:39] <bloodearnest> to ../run/resolvconf/resolv.conf
[13:40] <zyga-ubuntu> ok, let me check one thing
[13:40] <zyga-ubuntu> bloodearnest: ok
[13:40] <bloodearnest> https://www.irccloud.com/pastebin/Ru1Xgwu4
[13:40] <zyga-ubuntu> bloodearnest: it should still work because of the nameservice abstraction
[13:40] <zyga-ubuntu> aha, you are on 16.04
[13:41] <zyga-ubuntu> can you show me "dmesg | grep DENIED" please
[13:42] <bloodearnest> zyga-ubuntu: one possibly important detail: this inside a xenial lxd
[13:42]  * ogra_ thinks read access to /etc/resolv.conf should be a basic part of the network interface ... its not like that could be harmful or significantly security relevant
[13:43] <bloodearnest> zyga-ubuntu: audit: type=1400 audit(1505394898.461:17247): apparmor="DENIED" operation="create" namespace="root//lxd-estore_<var-lib-lxd>" profile="snap.snapstore.hook.install" pid=4122 comm="snap-exec" family="inet6" sock_type="stream" protocol=6 requested_mask="create" denied_mask="create"
[13:44] <ogra_> thats just some ipv6 noise
[13:45] <zyga-ubuntu> aha
[13:45] <zyga-ubuntu> ogra_: it is
[13:45] <ogra_> what ? security relevant ?
[13:45] <zyga-ubuntu> bloodearnest: in that case we may have a bug in lxd+snapd
[13:45] <zyga-ubuntu> ogra_: is is allowed
[13:45] <ogra_> ah
[13:45] <ogra_> where ?
[13:46] <ogra_> grepping for esolv in te interfaces subdir reveals nothing in network.go
[13:46] <ogra_> *resolv
[13:46] <ogra_> is it allowed internally somehow ?
[13:47] <__chip__> hhmmm, the opensuse linode tests block on a prompt?
[13:47]  * __chip__ goes to school run while pondering this
[13:53] <sergiusens> ogra_ problem is /etc/resolv.conf is a symlink, so you will need access to what it points to too
[13:55] <bloodearnest> zyga-ubuntu: I still get permission denied when installing the snap on my host, with both the network and network-bind plugs
[13:55] <bloodearnest> so lxd probably red herring?
[14:06] <mvo> zyga-ubuntu, ogra_: could you please check https://github.com/snapcore/core/pull/59 ?
[14:06] <mup> PR core#59: fix object path used by usr/bin/xdg-open in core <Created by mvo5> <https://github.com/snapcore/core/pull/59>
[14:08] <mup> PR snapd#3917 closed: cmd,dirs: treat manjaro the same as arch linux <Created by zyga> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/3917>
[14:18] <zyga-ubuntu> mvo: re, sorry
[14:18] <zyga-ubuntu> checking
[14:26] <matiasb_> sergiusens, o/ hey, just checking... any updates on the metadata thing?
[14:31] <mup> PR core#59 closed: fix object path used by usr/bin/xdg-open in core <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/core/pull/59>
[14:34] <jdstrand> bloodearnest, ogra_: the network interface includes the nameservice apparmor abstraction and it has this rule: /etc/resolv.conf r,
[14:35] <GNUtoo-lede> Hi, with "Debian GNU/Linux 9.1 (stretch)", after doing apt install snapd I get: "Setup snap "core" (2849) security profiles (cannot update mount namespace of snap "core": cannot update preserved namespace of snap "core": fork/exec /usr/lib/snapd/snap-update-ns: no such file or directory)" when doing snap install hello-world
[14:35] <jdstrand> bloodearnest: is /etc/resolv.conf a symlink to somewhere? do you have security denials in the logs?
[14:35] <GNUtoo-lede> *snap install hello
[14:37] <bloodearnest> jdstrand: /etc/resolvf.conf is a symlink to ../run/resolvconf/resolv.conf
[14:38] <jdstrand> bloodearnest: that is also allowed by said abstraction. do you have a security denial in the logs? 'grep audit /var/log/syslog'
[14:39] <ogra_> jdstrand, aha, thanks !
[14:40] <mup> PR snapcraft#1542 closed: tests: add integration tests for build snaps <Created by elopio> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/1542>
[14:42] <jdstrand> ogra_: fyi, you might be interested in 'apparmor_parser -p /path/to/profile' to see if a particular rule is in a profile (the -p is a preprocessor and will dump all rules, including those that are added by abstractions, to stdout)
[14:42] <bloodearnest> jdstrand: just rebuilding the snap in strict mode to try again
[14:42] <jdstrand> handy for grepping
[14:43]  * ogra_ notes down 
[14:43] <ogra_> thanks once more !
[14:43] <jdstrand> np :)
[14:44] <mup> PR core-build#20 closed: ubuntu-core-rootfs: deal with (broken) symlink in sync_dirs <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/core-build/pull/20>
[14:44] <Pharaoh_Atem> zyga-ubuntu: %{_var}/cache works too :)
[14:44] <Pharaoh_Atem> %_var points to %_localstatedir
[14:44] <Pharaoh_Atem> which, of course, points to /var
[14:45] <Pharaoh_Atem> likewise, %_usr points to %_prefix, which points to /usr
[14:46] <bloodearnest> https://www.irccloud.com/pastebin/U6OrCnvV
[14:46] <bloodearnest> jdstrand: ^^
[14:47] <bloodearnest> jdstrand: hmm, it's the configure hook that's trying to read /etc/resolv.conf
[14:48] <bloodearnest> can I give that network plugs?
[14:48] <bloodearnest> I currently grant them on each app as needed
[14:51] <bloodearnest> or, from the configure hook, can I invoke an app that has network plug?
[14:52] <mup> PR core-build#21 opened: Release 0.7.43+ppa23 <Created by mvo5> <https://github.com/snapcore/core-build/pull/21>
[14:55] <__chip__> Pharaoh_Atem: tell me about %{_var}
[14:55] <__chip__> Pharaoh_Atem: or better: tell me where i can read about all things %{}
[14:56] <__chip__> Pharaoh_Atem: ah, i see _var in https://fedoraproject.org/wiki/Packaging:RPMMacros?rd=Packaging/RPMMacros
[14:57] <mvo> ogra_: so I pushed the initramfs fix to the ppa, I will build a new core now. who should I ping to update the kernel snap so that it includes the updated initramfs? or can I just trigger the pc-kenrel snap build?
[14:57] <ogra_> mvo, talk to ppisati
[14:57] <ogra_> just a no-change rebuild
[14:57] <ogra_> preferably for all of pc-kernel, pi2-kernel and dragonboard-kernel
[14:57] <mvo> ogra_: yeah, we need the core in edge published first, right?
[14:58] <ogra_> not for the official kenels
[14:58] <ogra_> they use the deb directly during build
[14:58] <mvo> ogra_: using our ppa:snappy-dev/image ?
[14:58] <ogra_> all kernels that use the snapcraft kernel plugin need the fix in core though
[14:58] <ogra_> mvo, yep
[14:59] <mvo> ogra_: great
[15:00] <ogra_> after new-york split-initrd should be ready, then we wont need that mess anymore
[15:00] <ogra_> :)
[15:06] <jdstrand> bloodearnest: it is possibly to specify plugs for configure hooks iirc. let me look that up
[15:07] <jdstrand> bloodearnest: yeah, https://snapcraft.io/docs/build-snaps/hooks
[15:07] <davidcalle> jdstrand: bloodearnest you might want this version: https://github.com/CanonicalLtd/snappy-docs/pull/98
[15:07] <mup> PR CanonicalLtd/snappy-docs#98: Refresh the whole page, add install and remove hooks <Created by caldav> <https://github.com/CanonicalLtd/snappy-docs/pull/98>
[15:09] <pstolowski> niemeyer, hey, added https://forum.snapcraft.io/t/install-and-remove-hooks/2120 ; can you proof-read?
[15:10] <mup> PR snapcraft#1548 opened: meta: ensure main keys are ordered in snap.yaml <Created by sergiusens> <https://github.com/snapcore/snapcraft/pull/1548>
[15:10] <jdstrand> davidcalle: thanks
[15:11] <bloodearnest> jdstrand: davidcalle: \o/ great success. Thanks!
[15:11] <jdstrand> cool
[15:12] <pstolowski> davidcalle, ^ might also be interesting for you, although I believe you captured all the important aspects already
[15:13] <davidcalle> pstolowski: looking in a min, evilnick raised a fair point on the PR review: install hook only during initial install or subsequent revisions as well?
[15:14]  * ogra_ finds the forum text pretty clear " it gets executed ... when the snap is installed for the first time "
[15:14] <davidcalle> ogra_: ty ogra :)
[15:14] <pstolowski> davidcalle, only initial. but you had that before: "The install hook is called upon initial install.", did you remove it since?
[15:15] <davidcalle> pstolowski: was missing from the top summary
[15:15] <pstolowski> ah
[15:16] <sergiusens> davidcalle hey, mind updating the deprecations part of the site?
[15:17] <sergiusens> ogra_ split initrd would be grand!
[15:17] <sergiusens> my two year wait will be over ;-)
[15:17] <ogra_> sergiusens, i'm already using it for the splash screens ;) for actual modules and te like there are a few minor snapd changes missing that we'll solve in NY
[15:17] <davidcalle> sergiusens: deploying in a moment, when ^ gets merged (pstolowski, please give a final ack)
[15:17] <niemeyer> pstolowski: It's a good base, thanks! I will tune it a bit further once I have a keyboard again
[15:19] <pstolowski> davidcalle, I suggest tweaking the example yaml.. it's confusing with the extra plugs which will not work anyway
[15:20] <davidcalle> pstolowski: it's fixed, I think, I switched to upower-observe
[15:24] <pstolowski> davidcalle, ah, right, missed that, thanks
[15:35] <pstolowski> davidcalle, lgtm, approved
[15:35] <pstolowski> davidcalle, + a note about more advanced config got future consideration/updates
[15:39] <pstolowski> s/got/for/
[15:39] <oSoMoN> when will the wayland interface be available for consumption?
[15:41] <ogra_> oSoMoN, https://forum.snapcraft.io/t/the-desktop-interfaces/2042 says 2.28
[15:41] <davidcalle> pstolowski: noted, thanks
[15:49] <oSoMoN> ogra_, indeed, I read that post on Monday but my brain is a colander…
[15:49] <oSoMoN> thanks!
[15:49] <ogra_> oSoMoN, just wait til you are my age !
[15:49]  * ogra_ waves with his cane
[16:04] <pstolowski> davidcalle, when will your update go live? also, is it going to be https://github.com/snapcore/snapd/wiki/hooks ?
[16:05] <davidcalle> pstolowski: here https://snapcraft.io/docs/build-snaps/hooks , when Juju let me... Deployment of snapcraft.io is stuck right now because of some juju workers being down
[16:05] <davidcalle> So, at some point in the next hour :)
[16:05] <ogra_> can you still call them workers if they dont wok ?
[16:05] <ogra_> *work
[16:05] <pstolowski> davidcalle, cool. and I confused the wiki pages again ;)
[16:09] <pstolowski> ogra_, you don't stop calling someone a worker if he goes on vacation ;)
[16:09] <ogra_> yeah, but they went on strike !
[16:10] <ogra_> :)
[16:11] <pstolowski> ogra_, still a worker! union laws!
[16:11] <ogra_> haha
[16:22] <ogra_> hmm
[16:23] <ogra_> snaps arent really happy if you call pulseaudio -k to have pulse restart
[16:25] <sergiusens> ogra_ we should propose the removal of `-k` from pulseaudio's cli options ;-)
[16:25] <ogra_> well, i can also killall o pkill it :)
[17:22] <mup> PR snapcraft#1549 opened: plugins: extract pip from python plugin <Created by kyrofa> <https://github.com/snapcore/snapcraft/pull/1549>
[18:01] <mup> PR snapcraft#1548 closed: meta: ensure main keys are ordered in snap.yaml <Created by sergiusens> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/1548>
[18:20] <mup> PR core-build#22 opened: Simple tests <Created by mvo5> <https://github.com/snapcore/core-build/pull/22>
[19:05] <cachio> niemeyer, is it possible to exclude 1 test from the spread execution'
[19:05] <cachio> ?
[19:06] <cachio> I mean to run a whole suite but skip 1 test?
[19:06] <niemeyer> cachio: Only if you mark it as manual at the moment
[19:07] <cachio> niemeyer, ok, that could work, thanks
[19:13] <ryebot> Is there a problem with the core snap on s390x? I keep getting this: https://gist.github.com/wwwtyro/a37bb84db21e84723eeb70d3d2c6719e
[19:13] <ryebot> fwiw, this is under lxd.
[19:26] <ryebot> snap download works just fine
[19:27] <ryebot> er - `snap download core`
[19:32] <mup> PR snapcraft#1550 opened: demos: remove the py?-project demos <Created by sergiusens> <https://github.com/snapcore/snapcraft/pull/1550>
[19:44] <Dmitrii-Sh> ogra_: irccloud-desktop runs fine http://paste.ubuntu.com/25535844/ - no black windows (--edge core + apt update as of
[19:44] <Dmitrii-Sh>  date +%s
[19:44] <Dmitrii-Sh> 1505418240)
[19:45] <Dmitrii-Sh> couldn't check brave though:
[19:45] <Dmitrii-Sh> ➜  ~ type brave
[19:45] <Dmitrii-Sh> brave is /snap/bin/brave
[19:45] <Dmitrii-Sh> ➜  ~ brave
[19:45] <Dmitrii-Sh> ln: failed to create symbolic link '/home/dima/snap/brave/5/.config/gtk-2.0/gtkfilechooser.ini': File exists
[19:46] <Dmitrii-Sh> that's wayland btw
[19:50] <bschaefer> ogra_, hey, soo if /dev/dri didnt mount on my raspi3 ... i would assume somethings wrong with the vc4 driver?
[19:51] <bschaefer> though the strange this is this was happening the other day, but downloading the new image fixed it (seems like w/e is upgrading snap core seems to be causing issues?)
[19:52] <bschaefer> ... hoping i dont need to download the edge image each day to avoid the core upgrading and causing vc4 issues ... though not sure why else dri wouldnt be there :)
[20:15] <mup> PR snapd#3919 opened: interfaces: updates for default, browser-support, desktop, opengl, upower and stub-resolv.conf <Created by jdstrand> <https://github.com/snapcore/snapd/pull/3919>
[21:01] <ogra_> bschaefer, very weird, this doesnt happen for me at all, i can have my pi3 running for days (with daily refreshes of edge) and always have mir-kiosk on screen
[21:07] <bschaefer> happened 2 days ago ... and today but i just downloaded the new iage
[21:07] <bschaefer> where core doesnt upgrade
[21:07] <bschaefer> all i see is strange binding errors for vc4 but that seems common
[21:07] <bschaefer> (looks like it tries to bind until it can bind)
[21:08] <mup> PR snapcraft#1528 closed: recording: record the machine information collected by uname <Created by elopio> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/1528>
[21:27] <bschaefer> ogra_, hmm its happened again
[21:27] <bschaefer> on the newest edge image
[21:27] <bschaefer> core            16-2.27.6+git374.6f26821  2935  canonical  core
[21:27] <bschaefer> and no dri folder
[21:29] <ogra_> very weird ...
[21:30]  * ogra_ will monitor thaat the next days ... but i usually have my pi running in the office 24/7 and the mouse cursor is always up there ... 
[21:30] <ogra_> do you have HDMI constantly plugged in ?
[21:31] <ogra_> (perhaps the driver falls over when not getting EDID data or some such)
[21:33] <ogra_> (i'm also sure screenly would have told me if they ever observed something like this, we worked together a lot the last two/three weeks)
[21:36] <bschaefer> ogra_, yeah its plugged in, ill give you a pastebin of syslog if you feel like digging through it
[21:37] <ogra_> well, not at this time, but yeah, i'll dig into it tomorrow (my pi is in the office, i'm not near it atm)
[21:37] <bschaefer> ogra_, http://paste.ubuntu.com/25536380/
[21:37] <ogra_> dmesg would be good
[21:37] <bschaefer> alright! Yeah its only 2:37pm here
[21:37] <bschaefer> soo still pretty early in the day
[21:38] <bschaefer> ogra_, and dmesg for tomorrow: http://paste.ubuntu.com/25536382/
[21:38] <ogra_> heh, 11:37pm  here ...
[21:38] <bschaefer> it seems multiple reboots messes it up
[21:38] <bschaefer> o geez haha, have a good night!
[21:39] <ogra_> grepping for vc4 i dont see any unusual messages
[21:40] <bschaefer> yeah ... the only thing that looked strange was the bind = -517 but seems to happen wehn dri is mapped anyway
[22:16] <xnox> ryebot, it should work i don't think there is one in the stable channel though.
[22:16] <xnox> ogra_, is core snap available in the stable channel for s390x yet?
[22:16] <ryebot> I'm getting different behavior across different machines, so whatever the problem is, I think it's on my end right now.
[22:22] <mup> Bug #1717375 opened: snap find behaviour? <Snappy:New> <https://launchpad.net/bugs/1717375>
[22:35] <mup> PR snapcraft#1551 opened: dirs: set plugin, schema, and library dir for snap <Created by kyrofa> <https://github.com/snapcore/snapcraft/pull/1551>
[23:20] <mup> PR snapcraft#1550 closed: demos: remove the py?-project demos <Created by sergiusens> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/1550>