[00:31] <netphreak> Evening, guys!
[00:42] <netphreak> Any news on the Snappy RaspberryPi 3 support
[00:42] <netphreak> ?
[00:48] <manik_> netphreak: https://lists.ubuntu.com/archives/snappy-devel/2016-March/001639.html
[00:49] <manik_> chiluk: we officially support plan to support the 96boards' dragonboard 410c for aarch64
[00:59] <netphreak> Great news :)
[00:59] <netphreak> I suppose bluetooth and wifi is supported?
[01:00] <manik> netphreak: i haven't played with it myself, no h/w
[01:00] <manik> please check with ogra tmrw and confirm
[01:00] <netphreak> ok..
[04:52] <beiriannydd> hello from snappy/docker
[04:52] <beiriannydd> weird thing, I had problems building on snappy, so I exported the home volume to another docker container and built in that and everything worked
[04:53] <beiriannydd> permissions being squashed?
[06:47] <dholbach> good morning
[08:58] <mvip> morning, didrocks :)
[08:58] <mvip> was working on other things yesterday
[09:18] <mvip> or asac maybe you know as well -- what's the deal with PPA builds for snaps?
[09:18] <mvip> I've spoken to a few guys and they said that we should look into that for building our snaps, but I cannot seem to find any docs about that (i.e. building ARM snaps)
[09:29] <asac> mvip: yes
[09:29] <asac> i think that service is available in beta
[09:30] <asac> let me check
[09:32] <asac> let me try something
[09:33] <asac> mvip: https://code.launchpad.net/~asac if you go to your code page (not asac)
[09:33] <asac> do you see something about snap packages on the rigth?
[09:34] <mvip> asac so hmm we haven't pushed any code there yet, so i'd have to first create a package. Give me a few
[09:35] <mvip> but yes, i see a 'snap package' link
[09:35] <asac> mvip: i just creawted a test package: https://code.launchpad.net/~asac/+snap/tomcat-webapp-demo
[09:35] <asac> so just push sopmething with snapcraft.yaml there
[09:35] <asac> you can also use git if you dont like bzr
[09:35] <mvip> ah thank god ;)
[09:35] <asac> then there is a link "request a snap package"
[09:36] <asac> you hit that
[09:36] <asac> and after you can navigate to that snap package and hit "request a build"
[09:36] <asac> where you can select amd64 and i386 right now
[09:36] <asac> at least thats the experience i get
[09:37] <mvip> will this be integrated in the app store?
[09:37] <asac> ah i can also request arm
[09:37] <asac> you hav to configure processor
[09:37] <asac> athen you have more acrchitecture
[09:37] <asac> mvip: for sure at some point :)
[09:37] <mvip> ;)
[09:37] <asac> for now you have to upload yourself and sign it for 16 i guess
[09:37] <mvip> cool alright let me play around with this for a bit
[09:38] <mvip> ah right, there's a signing process too
[09:38] <asac> but natural evolution of this is to allow you to directly opush to edge channel in store or so
[09:38] <mvip> exactly
[09:38] <asac> right. but surely this can be delgated to maybe  a builder key or something
[09:38] <mvip> yeah let me play with this for a bit and bug you more later ;)
[09:39] <didrocks> asac: mvip: hey! yeah, it's not in beta anymore, just need to push a bzr branch
[09:40] <mvip> didrocks alright i just installed bzr on my box ;)
[09:40] <mvip> going to test with some snappy sample code
[09:41] <asac> didrocks: where is the announce mail?
[09:41]  * asac tried to find it but the search DEITY wasnt on his side
[09:41] <didrocks> asac: I don't think cjwatson made a public one
[09:41] <asac> oh ok
[09:41] <didrocks> I just read it written on IRC and some ML :)
[09:41] <asac> didrocks: maybe  you wanna blocg about  it and update ML with the blog?
[09:41] <asac> yeah well, i thought i saw an annoucne somewhere too
[09:41] <asac> but couldnt find
[09:41] <asac> *shrug*
[09:42] <asac> wonder what happens if you build a snap for powerpc
[09:42] <asac> :)
[09:42] <asac> wonde why the tomcat example failed though
[09:43] <asac> err where are the logs>?
[09:43] <asac> didrocks: https://code.launchpad.net/~asac/+snap/tomcat-webapp-demo/+build/449
[09:43] <mvip> didrocks OT, but here's a bash script that you can use for the updated marketing page for flashing out the disk image. https://gist.github.com/vpetersson/6346d67c8466ed513c3adb716a054cdd
[09:43] <didrocks> asac: yeah, that would be great if some of us can blog about it, but already not enough time to do what I need to do…
[09:43] <mvip> it's a bit rough still but with some polish it could streamline the flashing a fair amount
[09:43] <didrocks> mvip: oh nice, thaks! Yeah, I'll have a look
[09:44] <didrocks> (this is on Mac, right? I see diskutils being used)
[09:44] <mvip> didrocks i can put it in a github repo if you want
[09:44] <mvip> didrocks yeah, but should be relatively easy to make it conditional to Linux too
[09:44] <didrocks> mvip: just a note, bs=32M
[09:44] <didrocks> that's better than 1Meg nowdays :)
[09:44] <mvip> didrocks haha yeah
[09:45] <mvip> didrocks let me throw it up in a github repo
[09:45] <didrocks> ok ;)
[09:46] <asac> ogra_: did yo guys find a real fix for the sudoers thing yesterday?
[09:51] <ogra_> asac, sure, for both bugs ... a) dont loop mount sudoers with a readonly file ... b) mount /dev/pts in the container
[09:52] <asac> ogra_: did you land the fix?
[09:52] <ogra_> land ?
[09:52] <asac> yes, is it fixed now>
[09:52] <asac> ?
[09:52] <ogra_> i have no idea where either of them would have to land
[09:52] <asac> so wee dont want to fix it?
[09:53] <ogra_> thats an mvo thing
[09:53] <ogra_> jamie wanted to talk to him
[09:53] <mvip> didrocks: https://github.com/vpetersson/sdflasher
[09:55]  * ogra_ sighs
[09:56] <ogra_> there must be a million of such scripts out there ... and every year someone writes a new one :P
[09:57] <ogra_> i should re-vive ubuntu-imagewriter ;)
[09:57] <didrocks> mvip: PR your way :)
[09:58] <mvip> ogra_ i'm sure, but are any of them platform independent (OS X + Linux) and supports multiple formats of image files? :)
[09:58] <ogra_> didrocks, make a note that widows users need windows 10 for bash scripts ;)
[09:58] <mvip> ogra_ haha
[09:58] <mvip> didrocks awesome
[09:59] <ogra_> mvip, https://help.ubuntu.com/community/Installation/FromImgFiles
[09:59] <didrocks> ogra_: good one :p
[09:59] <ogra_> add the script there too if you want to officially promote it
[10:00] <ogra_> thats where all other pages link to (for user documentation at least)
[10:00] <ogra_> hmm, and the imagewriter bit should be removed
[10:01] <mvip> ogra_ might as well add it there, yeah. Thanks
[10:03] <ogra_> (would also be nice if you could make that script consistent .... thats half bash and half POSIX shell )
[10:04] <mvip> ogra_: it says 'immutable page' for me
[10:04] <ogra_> logged in ?
[10:04] <mvip> ogra_ yup
[10:04] <mvip> ogra_: yeah, this was just a hackish script i had laying around that has grown over time. Need to clean it up a bit more
[10:05] <ogra_> just drop the [[]] from your if/elsif ....
[10:44] <mvip> asac i was able to build a snap using the PPA. Yay! :)
[10:54] <mvip> asac it's very slow however.
[10:54] <mvip> asac been waiting for 15 min now for building the 'py3-test' example and it says 23 minutes remaining.
[10:54] <mvip> (for arm build)
[10:57] <mvip> hmm now i got 'Failed to build' but no error description at all.
[11:30] <asac> mvip: yes i already flaggedt hat issue. folks say it happened during LP upgrade yesterday and said they will investigate asap
[11:30] <asac> i dont know why arm is slow... if its slow building its probably not beefy builders or even emulators
[11:31] <asac> if its just long wait before it gets build it just means long build queue
[11:31] <asac> usually temporary because someone is doing an archive rebuild or something on the bilders
[11:33] <mvip> asac got it. thanks
[11:33] <mvip> asac well at least i know how to build it now.
[11:34] <mvip> btw, can we create a (private) projects w/ multiple contributors somehow?
[11:34] <asac> please shoot me a mail for that question. thanks
[11:35] <mvip> asac sure thing
[11:41] <ogra_> yeah, arm builders are non-native by default for PPAs unless you explicitly ask for native
[11:42] <ogra_> (use the canonical-arm-dev PPA if you want native, i can add you to it if needed)
[11:42] <mvip> ogra_ how do I ask for native ones?
[11:42] <mvip> ah :)
[11:42] <mvip> ogra_ yes please ('vpetersson')
[11:45] <ogra_> mvip, oops,  sorry, that PPA is canonical only
[11:46]  * ogra_ forgot about that
[11:46] <mvip> ogra_ meh. ;(
[11:47] <ogra_> sorry for that
[12:53] <kyrofa> Good morning
[13:02] <josepht> good morning
[13:11] <kyrofa> jdstrand, I have a seccomp question for you when you get in
[13:22] <jdstrand> kyrofa: go ahead
[13:23] <kyrofa> jdstrand, good morning!
[13:23] <kyrofa> jdstrand, simple question, I think: why do we use SCMP_ACT_KILL as opposed to SCMP_ACT_TRAP in the launcher?
[13:24] <jdstrand> niemeyer, zyga: I've been looking at our udev tagging and there are several bugs on 16.04 (that's ok, zyga's udev branch hasn't landed yet so we'll just fix them)
[13:24] <jdstrand> kyrofa: trap is racy
[13:24] <zyga> jdstrand: in my branch or in the code outside of snappy?
[13:24] <zyga> jdstrand: can you be more specific
[13:24] <jdstrand> zyga: everywhere :)
[13:24] <kyrofa> jdstrand, ah! It was indeed a simple question then, thank you :)
[13:24] <zyga> jdstrand: can you please tell me more (just enough)
[13:25] <jdstrand> zyga: on 15.04 we generate things like this:
[13:25] <jdstrand> zyga: I'm getting there, hold on :)
[13:25] <jdstrand> KERNEL=="kmsg", TAG:="snappy-assign", ENV{SNAPPY_APP}:="hello-world.canonical"
[13:25] <jdstrand> that's fine
[13:25] <jdstrand> in the current 16.04 code, we do something like:
[13:26] <jdstrand> KERNEL=="kmsg", TAG:="snappy-assign", ENV{SNAPPY_APP}:="hello-world.echo"
[13:26] <jdstrand> KERNEL=="kmsg", TAG:="snappy-assign", ENV{SNAPPY_APP}:="hello-world.env"
[13:26] <jdstrand> KERNEL=="kmsg", TAG:="snappy-assign", ENV{SNAPPY_APP}:="hello-world.evil"
[13:26] <jdstrand> ...
[13:26] <zyga> jdstrand: and those overwrite?
[13:26] <jdstrand> the idea was I'm assuming to allow each individual app access to the device
[13:26] <zyga> jdstrand: (I mentioned that yesterday)
[13:27] <jdstrand> yes, := overwrites. only one of those will work
[13:27] <zyga> jdstrand: anyway, the other meeting is about to start, if you want we can jump there and we can continue this in a second
[13:27] <jdstrand> ok
[13:27] <kyrofa> jdstrand, is ERRNO also racy?
[13:28] <jdstrand> kyrofa: no. I suggested errno as EPERM and was told that we prefer kill
[13:28] <zyga> jdstrand: can we merge those so that we just say ENV{SNAPPY_APP}="hello-world.echo,hello-world.env,hello-world.evil" ?
[13:28] <jdstrand> zyga: well, there seems to be some bugs on property match
[13:28] <jdstrand> I'm not sure yet
[13:28] <zyga> jdstrand: I see
[13:28] <kyrofa> jdstrand, oh, that's too bad!
[13:28] <zyga> jdstrand: thanks for mentioning this
[13:28] <zyga> jdstrand: it's not used by any interface yet
[13:29] <jdstrand> no
[13:29] <zyga> jdstrand: but it will be required for various hw-assign-like scenarios
[13:29] <jdstrand> I knew the review was coming and it is related to my 'remove needle detection from launcher' bug
[13:29] <jdstrand> zyga: of course
[13:29] <jdstrand> zyga: I wanted to talk to pitti about it, but still investigating
[13:30] <jdstrand> I actually have a method that works
[13:30] <zyga> jdstrand: let's join the hangout
[13:30] <jdstrand> zyga: is per app device assignment a requirement?
[13:30] <jdstrand> zyga: link?
[13:31] <zyga> jdstrand: did you get it, I'm not nickserv-registered
[13:31] <zyga> jdstrand: also sent on telegram
[13:31] <jdstrand> I got it. I joined what you pasted
[14:13] <zyga> morphis: hey, can we discuss the work you are doing here?
[17:07] <elopio> sergiusens: can I change the mqtt parts in the wiki to use https?
[17:07] <elopio> that won't break anything, afaics
[17:40] <sergiusens> elopio, sure
[18:15] <kyrofa> Is it possible to change the keyboard layout in snappy?
[18:17] <kyrofa> I assume dpkg-reconfigure isn't going to work
[18:25] <sergiusens> elopio, can you explain this http://162.213.35.179:8080/job/github-snapcraft-autopkgtest-cloud/350/console ?
[18:25] <sergiusens> kyrofa, it was possible in the past, not sure snappy config ubuntu-core still exposes it
[18:47] <kyrofa> sergiusens, alright thanks, doesn't look like it