[05:19] <mborzecki> morning
[05:40] <mborzecki> that's a new none - Download snap "test-snapd-openvswitch-support" (11) from channel "stable" (stream error: stream ID 1; PROTOCOL_ERROR)
[06:24] <zyga> good morning
[06:24] <zyga> mborzecki: how are you :)
[06:24] <mborzecki> zyga: hey
[06:24] <zyga> mborzecki: we saw some of that last week, cause unknown but not investigated either
[06:25] <mborzecki> zyga: https://copr.fedorainfracloud.org/coprs/bboozzoo/snapcraft/
[06:25] <mborzecki> packaged the damned thing eventually
[06:26] <zyga> excellent
[06:26] <zyga> what's your plan for flock?
[06:28] <mborzecki> zyga: we've updated the slide deck, snapcraft is kind of packaged, i got the snapcore/go-gettext package approved and need sponsoring (got some reviews to do)
[06:29] <mborzecki> zyga: other than that, cover the important bits and improvise
[06:29] <zyga> mborzecki: if someone asks, what's the outlook for using snapcraft on fedora using fedora bases
[06:29] <jamesh> zyga: hi.  Was there supposed to be some review comments with https://github.com/snapcore/snapd/pull/7197#issuecomment-517256824 ?
[06:29] <mup> PR #7197: usersession: track connections to session agent for exit on idle and peer credential checks <Created by jhenstridge> <https://github.com/snapcore/snapd/pull/7197>
[06:29] <zyga> jamesh: hey, I don't know from the top of my head
[06:29] <mborzecki> zyga: tbh, the question is what's the outlook for non-ubuntu bases
[06:30] <zyga> jamesh: last week was rather busy, and I had little chance to look at PRs
[06:30] <jamesh> zyga: it looks like you might have had unsaved review comments at some point and clicked send
[06:30] <zyga> jamesh: oh, it still shows pending here.
[06:30] <mborzecki> zyga: this topic seems to be progressing really slow https://forum.snapcraft.io/t/base-runtime-freedesktop-sdk-runtime-19-08/11153
[06:30] <zyga> jamesh: github UX snafu, I'm sorry about that
[06:31] <jamesh> zyga: fair enough.  I know you were busy last week
[06:31] <zyga> I just sent it, it is only one comment though
[06:31] <jamesh> thanks
[06:31] <zyga> jamesh: I didn't go deeper than that function
[07:09] <pstolowski> morning
[07:10] <pstolowski> hey zyga, welcome back!
[07:12] <mborzecki> pstolowski: hey
[07:28] <zyga> pstolowski: hey, it's good to be back :)
[07:28] <zyga> pstolowski: ETOOMANYMILES
[07:29] <pstolowski> :)
[07:29] <pstolowski> hey mborzecki
[07:33] <zyga> pstolowski: one idea from the sprint https://github.com/snapcore/snapd/pull/7205
[07:33] <mup> PR #7205: rfc: introduce confinement options failsafe flag <Created by zyga> <https://github.com/snapcore/snapd/pull/7205>
[07:33] <zyga> pstolowski: it's not fleshed out fully
[07:34] <zyga> but should help when we have bugs that we don't know about
[07:35] <pstolowski> zyga: right, i saw this PR, was wondering if there was any discusson behind it, good!
[07:35] <zyga> pstolowski: only at the sprint
[07:36] <zyga> pstolowski: one more aspect to explore is to allow a specific interface endpoint to fail
[07:36] <zyga> without failing all of the device setup
[07:36] <zyga> (in failsafe mode)
[07:36] <zyga> e.g. allow serial-port-abc to fail without breaking snapd/core
[07:47] <pstolowski> zyga: added a comment
[07:47] <jamesh> zyga: for the mount namespace case, it might just mean that problems in the non fail-safe never get reported because they appear to work
[07:55] <zyga> jamesh: indeed, we can expand it to talk to the error tracker if needed
[07:55] <zyga> jamesh: but we also need to be mindful of a single error down in the guts propagating to effective DOS on snapd
[09:16]  * dot-tobias says hi
[09:27] <dot-tobias> sergiusens: Can you give me a quick hint here please? (outdated snapcraft homebrew formula) https://forum.snapcraft.io/t/install-snapcraft-on-macos/9607/4
[09:45] <mborzecki> google:debian-sid-64:tests/main/sbuild keeps failing
[09:46] <mborzecki> the debug log isn't helpful :/
[09:46] <zyga> mborzecki: do you have a sample?
[09:46] <mborzecki> zyga: https://paste.ubuntu.com/p/yjjz4S9ndQ/
[09:47] <mborzecki> zyga: heh E: Build failure (dpkg-buildpackage died)
[09:47] <zyga> hmmm
[09:47] <zyga> is this on master or on your branch?
[09:52] <mborzecki> on my branches
[09:56] <mup> PR snapd#7209 opened: firstboot: queue service cmds before mark-seeded <Created by stolowski> <https://github.com/snapcore/snapd/pull/7209>
[10:02] <zyga> mborzecki: are you changing dependencies?
[10:06] <zyga> mborzecki: network failed?
[10:06] <mborzeck1> zyga: duh, graphics froze, cpu fan 100%
[10:06] <mborzeck1> zyga: i.e. the year of linux desktop is upon us
[10:06] <zyga> mborzeck1: laptop or desktop?
[10:07] <mborzeck1> zyga: laptop
[10:07] <zyga> x230?
[10:07] <zyga> or x250?
[10:08] <mborzeck1> zyga: 250
[10:34] <zyga> mborzecki: does ohmygiraffe work for you on arch?
[10:35] <zyga> mborzecki: it crashes on TW
[10:35]  * zyga needs to update to 2.40
[10:37] <mborzecki> zyga: used to work
[10:38] <mborzecki> zyga: yup works
[11:33]  * pstolowski lunch
[12:39] <mborzecki> zyga: https://github.com/snapcore/snapd/pull/7210
[12:40] <mup> PR #7210: packaging/debian-sid: set GOCACHE to a known writable location <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/7210>
[12:40] <mup> PR snapd#7210 opened: packaging/debian-sid: set GOCACHE to a known writable location <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/7210>
[12:59] <zyga> mborzecki: what is that -pkgdir?
[12:59] <zyga> mborzecki: I have slight preference for /tmp/ for GOCACHE
[12:59] <zyga> but not strong one
[13:00] <mborzecki> zyga: -pkgdir is set by dh-golang
[13:00] <mborzecki> zyga: and afaiu dh-golang sets GOCACHE to point to _build as well
[13:29]  * zyga goes for lunch
[13:37] <mborzecki> - Fetch and check assertions for snap "test-snapd-tools_6" (6) (cannot fetch assertion: got unexpected HTTP status code 408 via GET to "https://api.snapcraft.io/api/v1/snaps/assertions/snap-declaration/16/eFe8BTR5L5V9F7yHeMAPxkEr2NdUXMtw?max-format=3")
[13:37] <mborzecki> pff
[13:38] <mborzecki> btw. PROTOCOL_ERROR has this in debug log:  DEBUG: Not retrying: http.http2StreamError{StreamID:0x1, Code:0x1, Cause:error(nil)}
[13:38] <mborzecki> soo that does look like HTTP2 level issue
[13:39] <mborzecki> well, at least the sbuild fix PR didn't fail in sbuild :)
[13:43] <cachio> mborzecki, did you see that today?
[13:48] <mborzecki> cachio: yes, 3-4 times already
[13:49] <mborzecki> cachio: last time in https://github.com/snapcore/snapd/pull/7210
[13:49] <mup> PR #7210: packaging/debian-sid: set GOCACHE to a known writable location <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/7210>
[13:50] <cachio> mborzecki, good, I'll see that with the store team
[13:51] <cachio> mborzecki, thanks for the info
[13:53] <zyga> cachio: I'll get back to you on the 2.40 release plan soon
[13:56] <cachio> zyga, ok
[13:56] <cachio> zyga, I' have an app with my accountant in 30 minutes
[13:57] <cachio> zyga, just ping me and I'll ping you back once I am back
[13:57] <zyga> sure
[14:13] <sil2100> zyga: hey! Since we'll be preparing the 18.04.3 point-release ubuntu-core images soon (both for 16 and 18), just wanted to touch base if the current snapd in stable is good to go to the new official stable images
[14:14] <sil2100> zyga: is the current stable snapd good for refreshed image release? ;)
[14:14]  * cachio afk
[14:51] <pstolowski> ogra: hey, i've a task to look at failure scenarios where a device fails at first boot setup and is rebooted, and doesn't come back (doesn't boot anymroe); i've heard you experienced such issues before, do you have any concrete examples, reproducers, or logs?
[14:53] <pstolowski> also ondra ^ ?
[14:55] <ondra> pstolowski yeah it's easy one
[14:55] <ondra> pstolowski when anything fails during first boot seeding, we do revert 'uninstall' all snaps
[14:56] <ondra> pstolowski this actually removes kernel and core snap sym links from 'vat/lib/snapd/snap'
[14:56] <ondra> pstolowski I think to make first boot more robust, we should do two changes, in snapd and initrd.
[14:57] <ondra> pstolowski when we look for core and kernel snap, we should consider two locations '/var/lib/snapd/snaps' '/var/lib/snapd/seed/snaps'
[14:57] <ondra> pstolowski so we have fallback when snaps are removed
[14:59] <pstolowski> ondra: ack, thanks, i need to look at the code
[15:03] <ondra> pstolowski easy way to reproduce is to make gadget snap with malformed gadget.yaml, and build image with it
[15:04] <ondra> pstolowski e.g. and non supported role:
[15:07] <ogra> pstolowski, ondra there is also the core18 case where snapd doesnt properly get seeded and you end up with an error from console-conf "snap command not found" or some such
[15:07] <ogra> coe18 is a lot more fragile in that due to the singling out of snapd
[15:19] <zyga> cachio: postponed till next week
[15:19] <zyga> sil2100: current as in 2.39.x?
[15:20] <zyga> sil2100: I think anything in stable is good, I don't know of any blockers that would affect images
[15:20] <zyga> sil2100: what's the time frame? we will publish 2.40 to stable next week
[15:22] <sil2100> zyga: it's all happening this week, with a ETA for the release this Thursday
[15:23] <ondra> ogra yes this is same case
[15:23] <sil2100> zyga: thanks for the heads up then o/
[15:23] <zyga> sil2100: In that case do use 2.39.x, 2.40 will bake and we can try next time :)
[15:23] <pstolowski> ondra, ogra thanks, i need to digest it and think about this
[15:47] <zyga> willcooke: help, I cannot create topics on the WSL category
[15:47] <zyga> I wrote the whole thing and saved it to clipboard
[15:47] <zyga> willcooke: when I attempt to save all I get is "You are not permitted to view the requested resource."
[15:49] <willcooke> zyga, zoiks!  One sec...
[15:49] <zyga> mborzecki: udev rewrite tests running, I think it will be green
[15:49] <zyga> mborzecki: I will need to sit down and think how to test it, I will start by writing simple unit tests tomorrow
[15:50] <willcooke> zyga, hm.  "everyone" can create/reply/see.  Investigating
[15:54] <willcooke> zyga, I think this might have been a "user trust level" thing.  I've bumped you up, can you try now?
[15:55] <zyga> willcooke: https://discourse.ubuntu.com/t/using-snapd-in-wsl2/12113
[15:55] <willcooke> zyga, \m/ \m/ thank you!
[15:55] <zyga> pleasure :)
[15:55] <zyga> diddledan: ^ I think you have more experience with this
[15:56] <zyga> perhaps you can expand on the topic
[16:00] <cachio> zyga, ok, thanks for the update
[16:08]  * zyga EODs
[16:08] <pstolowski> cachio: ping
[16:27] <cachio> pstolowski, yes
[17:55] <mup> PR snapd#7211 opened: tests: add more debug information to see the root cause of the protocol error <⛔ Blocked> <Created by sergiocazzolato> <https://github.com/snapcore/snapd/pull/7211>
[18:27] <ogra> jdstrand, pocketbeagle is my gadget ... no need to notify me about it ;)
[18:28] <jdstrand> ogra: ah, I missed that. can you request a manual review?
[18:28] <ogra> will do :)
[19:03] <diddledan> popey: you about?
[19:59] <popey> Sup?
[20:03] <diddledan> ooh, ello, pm a sec, popey ?
[20:04] <popey> Sho thang
[20:04]  * diddledan /msgs
[20:22] <mup> PR snapcraft#2654 opened: click: update to 7.0 <Created by cjp256> <https://github.com/snapcore/snapcraft/pull/2654>