[05:01] good morning :) [06:20] morning [06:20] good morning mborzecki [06:23] mvo: so crng init struck core image now? [06:23] mborzecki: it seems to have done so yesterday - not so sure about today, there was a new kernel update [06:23] mborzecki: so maybe that fixed things I'm just looking [06:28] mborzecki: hm, the diff is gigantic, no idea [06:29] mvo: kernel diff? [06:30] mvo: do you have a link? [06:31] mborzecki: sure: https://launchpad.net/~canonical-kernel-team/+archive/ubuntu/ppa/+packages?field.name_filter=&field.status_filter=published&field.series_filter=bionic [06:32] mborzecki: the first one is linux-4.15… [06:49] PR snapd#6382 closed: tests: use pc-kernel from stable instead of edge to build core-16 and core-18 [06:52] PR snapd#6370 closed: interfaces/builtin/opengl: allow access to NVIDIA VDPAU library [06:53] PR snapd#6362 closed: cmd/snap-update-ns: explicitly check for return value from parse_arg_u [06:58] degville: could you please do a final look/approval on pr#6369 (the one about PATH)? feel free to merge if you are happy [07:18] Hey [07:26] hey zyga [07:42] zyga: hey [07:47] PR snapd#6372 closed: tests: define new "tests/smoke" suite and use that for autopkgtests [08:00] mvo: PRs are turning green ;) [08:00] PR snapd#6373 closed: overlord/ifacestate: helper API to obtain the state of connections [08:01] mvo: wdyt about this one? https://github.com/snapcore/snapd/pull/6383 [08:01] PR #6383: tests: provide a fake random device to the core images === pstolowski|afk is now known as pstolowski [08:01] mornings [08:01] pstolowski: hey [08:05] mborzecki: I was thinking about it this morning [08:05] hey pstolowski [08:05] mborzecki: I think its ok, we do the same on classic afterall [08:05] mvo: yes, we do [08:05] mborzecki: but it won't help with the entropy misisng on boot [08:06] mborzecki: so I think its ok to have it but we shouldn't expect much from it except that it speeds up key generation when we do gpg tests (which iirc we skip on core anyway because we have no gnupg snap) [08:07] mborzecki: wdyt? [08:07] mvo: yeah, gppg and maybe ssh would benefit [08:07] mborzecki: aha, ssh indeed [08:07] mborzecki: that might be a good reason, we generate ssh host keys on first boot [08:08] re [08:09] welcome back zyga [08:11] mvo: i'm looking at the prepare code, maybe w could generate the host keys while preparing the image and just copy them over to system-data/etc/ssh/ [08:11] this would speed up the first boot [08:11] hey, how are things on your side guys? [08:11] I'm looking at debian package update now [08:11] mborzecki: yeah, that would work [08:11] mvo: will do! [08:11] mborzecki: we should measure first though [08:12] mborzecki: I mean, it might be just so fast its not worth the work on our side :) [08:22] PR snapd#6375 closed: tests: fix enable-disable-unit-gpio test on external boards [08:33] PR snapd#6326 closed: interface: raw-usb: Adding ttyACM ttyACA permissions [08:56] mvo, hi, we have refreshed the pc-kernel snaps to add some fixes to the image. smb has more details about it, he's working on that [08:57] klebers: thanks for getting back to me. we had trouble with the previous kernel, it failed to boot in most tests but the most recent one is fine again. [08:58] mvo, we are planning to update soon the version on 18/stable, would that affect someone? [08:59] klebers: we don't have that many 18 users yet, but let me check [09:10] hey Chipaca ! good morning [09:11] Iä! Iä! mvo fhtagn! [09:11] haha [09:11] was that finnish ? [09:12] ogra: the long form is «Iä! Iä! mvo fhtagn! Ph'nglui mglw'nfah mvo R'lyeh wgah'nagl fhtagn!» if that gives you any more clues [09:12] cthulu language, rly'something [09:13] oh, chtulu ! [09:14] * Chipaca practices for the post-brexit world [09:14] heh [09:14] Chipaca: I think you were offline by the time I got back last night. My Atom looks like this :-( https://i.imgur.com/YVeUqls.png [09:15] Gargoyle: looks like a problem in the theme to me [09:16] Gargoyle: is that plain Ubuntu? [09:17] Yup. 18.10 with Yaru theme [09:17] I have been playing with builder a few months ago - wonder if some dev libraries are messing things up? [09:18] Gargoyle: is there a bug or forum post talking about the issue? [09:19] Not that I have found. There is one relating to file open/save dialogs which are all messed up too. [09:19] Chipaca: isn't that some ubuntu specific gtk extension that moved the menu bar somewhere to the top bar under unity? [09:20] mborzecki: in 18.10? not that I'm aware [09:21] Gargoyle: so when you asked "is this issue fixed yet" you were assuming we had some kind of telepathic power? :-) [09:21] OK. You are correct Chipaca. Switched "Applications" theme to Adwaita and atom's menu and file dialogs look normal again. [09:22] Chipaca: It's related to the filesystem one [09:22] *filesystem dialogs one [09:22] Gargoyle: how so? [09:22] Chipaca: haha - I read a lot of lovecraft but forgot this one [09:24] Gargoyle: in any case I can confirm in 18.10 with the default theme in atom the menu line is black-on-black or something [09:24] also it takes almost a minute to start the first time [09:25] (but this might be the snapd version) [09:25] s/almost/what feels like/ [09:26] It's something to do with the GTK themes and the switch to 18.10 and Yaru. I asked a while ago and someone like popey confirmed there was an ongoing issue. Didn't get a link to a forum post though. [09:29] I can't find one now ether, so I'll grab some screenshots / video and post one up this morning. [09:34] Gargoyle: ta [10:15] PR snapd#6369 closed: Add check for snap binaries dir not being in path

[10:26] PR snapd#6384 opened: snapd: fix race in TestSanityFailGoesIntoDegradedMode test [10:36] mvo: left a comment ^^ [10:40] mborzecki: \o/ for that [10:40] mborzecki: excellent catch [11:02] PR snapd#6385 opened: debian: fix silly typo in the spread test invocation [11:26] Chipaca: Found the underlying issue - it's electron and not specific to the snap. (https://github.com/electron/electron/issues/15194) [11:52] dum di dum.... Electron api demo app works fine. Building atom from source... [11:59] PR snapcraft#2437 closed: repo,baseplugin: support trusting repo keys [12:09] mborzecki: hey, can you take a look at https://github.com/snapcore/snapd/pull/6379 ? [12:09] PR #6379: ifacestate/tests: extra test for hotplug-connect handler [12:09] PR snapd#6386 opened: tests: skip lp-1802591 on "official" images [12:10] (the test you suggested in previous hotplug PR) [12:10] pstolowski: ack [12:10] thanks [12:13] hi all, can anyone interpret why spread failed for me: https://travis-ci.org/snapcore/snapd/builds/479834032 ? [12:17] Chipaca: i've updated #6333 [12:17] PR #6333: daemon: introduce /v2/connections snapd API endpoint [12:18] greyback: sorry the link doesn't seem to work, which PR is this? [12:18] greyback: 6361? [12:18] mborzecki: https://github.com/snapcore/snapd/pull/6361 [12:18] PR #6361: kvm: load required kernel modules if necessary [12:18] yep [12:21] greyback: hmm, the link to travis job from the PR doesn't work either, wth? [12:21] mvo: hey, can we get the 2.37 release tarballs on the release page please [12:22] mborzecki: hmm, I just tried it now, broken here too ("We couldn't find repo snapcore/snapd") Worked ~10 mins ago [12:23] travis induced coffee break :) [12:23] :D the new https://www.xkcd.com/303/ [12:26] mvo: I'm working on the 2.37 debian update [12:26] finally got over the hurdles there [12:26] (in the early prep of how-do-I-even-start) === ricab is now known as ricab|lunch [12:33] greyback: i've restarted the spread job [12:47] mborzecki: thank you [12:51] or I guess I can do that too [12:52] mborzecki: is connectionJSON used outside of api_connections/ [12:52] ? [12:52] mborzecki: in the whole picture i mean [12:53] Chipaca: yes, there's corresponding struct in the client package [12:53] planned to push the client bits to #6016 [12:53] PR #6016: [RFC] move various name validation helpers to snap/name package [12:53] uh, #6079 [12:54] PR #6079: [RFC] `snap connections` command [12:59] * pstolowski lunch [13:03] mborzecki: is there a reason to have the structs twice? I've been trying to avoid that unless there's a reason [13:03] bah, maybe it's "it still uses the old json stuff that's in daemon and i didn't want to refactor", that's a reason for example :-) [13:16] cachio: hey, what's the state of 2.37 [13:16] I'm trying to determine if the debian update should go with 2.36.x or with 2.37 [13:16] mvo: ^ CC please comment as well if you can [13:16] zyga: 2.37 [13:17] mvo: hey :) [13:17] zyga: I'm working on it right now and we should (fingers crossed) have a 2.37 final today or tomorrow [13:17] I tried to get the tarballs and I was in a busy meeting [13:17] zyga: I really really want it today [13:17] is there a .0 release yet? [13:17] and you will go with .1 [13:17] or is the work now focused on 2.37.0 [13:17] zyga: its not out just yet, we are at 2.37~rc1 [13:17] zyga: two more PRs [13:17] ok [13:17] zyga: one race fix, one autopkgtest fix [13:18] I'll continue to assess the debian side, we should have some actions there [13:18] for one [13:18] zyga: and thanks for working on the update [13:18] I would like to go through all the build deps [13:18] zyga: \o/ [13:18] (which are really runtime deps) [13:18] zyga, hey [13:18] and then check if we are comfortable with them being as old as they are [13:18] hey cachio, how are you doing? [13:18] zyga: there is one potential PR I might pull in (6378) [13:18] zyga, initial validation for pre~1 is completed [13:18] mvo: and if no, actually go to salsa, work on updates (despite most likely missing this freeze) [13:19] zyga, mvo is gonna generate a new release soon [13:19] mvo: and I would honestly put this on us each release [13:19] mvo: where if we pay the painful price now once [13:19] mvo: we should be in a good place for each release (it should be cheap) [13:19] mvo, next friday 25 is my last day before vacations [13:20] mvo, I mentioned that in the standup but I think you were already on vacations [13:20] it was on december [13:21] cachio: ok [13:21] cachio: should be enough time for the validation [13:21] zyga: yeah, maybe we need to do it [13:21] zyga: not great but better than the alternatives it seems [13:21] mvo, I hope so [13:34] PR snapd#6387 opened: client: introduce helper for querying snapd API for the list of slot/plug connections <⛔ Blocked> [13:34] Chipaca: opened the client bits ^^ [13:35] * Chipaca shakes fist at the internet at large, and goes for coffee [13:36] ☕ [13:40] pstolowski, hey [13:41] pstolowski, when you have time, could you please take a look to this one? #5887 [13:41] PR #5887: tests: moving core-snap-refresh-on-core test from main to nested suite [13:41] pstolowski, it is related to nested suite [13:44] cachio: will do [13:44] How do we find out who is currently doing the packaging for Atom if it currently just says "snapcrafters" on the store page? [13:44] pstolowski, thanks [13:45] Gargoyle: that's the snapcrafters, which means anyone on the snapcrafters org lead by popey and wimpress [13:49] PR snapd#6385 closed: debian: fix silly typo in the spread test invocation [13:49] diddledan: OK. I've got a "working" build of atom. I say "working" because the menu and file dialogs are rendering properly but atom's own test suite fails horribly. But I can't get my head rount the branching on atom's github page! :/ [13:50] PR snapd#6359 closed: tests: fix listing tests to match "snap list --unicode=never" [13:51] PR snapd#6386 closed: tests: skip lp-1802591 on "official" images [13:52] 6384 needs a second review === ricab|lunch is now known as ricab [13:56] PR snapcraft#2440 opened: meta: make hooks executable instead of complaining they're not [14:18] Gargoyle: the snap of atom is created from https://github.com/snapcrafters/atom [14:36] cachio: btw 5887 has conflicts [14:50] pstolowski, ohh, sorry [14:52] pstolowski, fixed [14:52] thanks [14:57] mvo: i think we could land #6368 as it is now, even with the typo, given the spread job on master has failed because of this problem [14:57] PR #6368: tests: fix daemon-notify test checking denials considering all the log lines [14:58] mborzecki: if you are happy with it: +1 [14:59] PR snapd#6368 closed: tests: fix daemon-notify test checking denials considering all the log lines [15:00] mborzecki: thank you! [15:17] PR snapd#6378 closed: cmd: fix snap-device-helper to deal correctly with hooks [15:53] 6384 needs a second review [15:55] mvo: lgtm [15:55] ta! [15:56] PR snapd#6384 closed: snapd: fix race in TestSanityFailGoesIntoDegradedMode test [16:19] Seems I'm trying all the broken snaps today! :/ [16:19] * ogra hugs Saviq for https://bugs.launchpad.net/bugs/1812003 [16:19] Bug #1812003: hooks should be made executable instead of erroring out [16:20] Jenkins can't use https (probably any SSL) due to a trust store error. [16:20] PR snapd#6388 opened: tests: fix install-snaps test by changing the snap info regex [16:20] Does it need access to etc/ssl ? [16:23] * cachio lunch [16:33] * zyga resumes work on debian packaging [16:35] eh, quilt [16:35] how I love the [16:35] (not) [16:35] thee [16:45] created report here:- https://forum.snapcraft.io/t/jenkins-cannot-access-update-server-via-https/9499 [16:51] PR snapd#6389 opened: cmd/snap: small refactor of cmd_info's channel handling === pstolowski is now known as pstolowski|afk [17:07] PR snapd#5887 closed: tests: moving core-snap-refresh-on-core test from main to nested suite [18:01] ok, EOD from me [18:01] ttfn [18:48] PR snapd#6390 opened: release: 2.37 [19:23] cachio: all arches have 2.37 in beta now, just fyi [19:35] mvo, great [19:36] starting right now [19:47] cachio: \o/ [19:52] PR snapd#6390 closed: release: 2.37 [21:08] PR snapd#6391 opened: tests: simplify interfaces-contacts-service test