[05:14] <mborzecki> morning
[05:43] <zyga> o/
[05:44] <zyga> mborzecki: can you look at https://github.com/snapcore/snapd/pull/8726
[05:44] <zyga> it's a one liner and I have a follow-up it opens
[05:44] <mup> PR #8726: tests: silence stderr from dbus-monitor <Test Robustness> <Created by zyga> <https://github.com/snapcore/snapd/pull/8726>
[05:46] <mborzecki> zyga: ack
[05:47] <zyga> for some context, privately I've been using a version of https://github.com/snapcore/snapd/pull/8728
[05:47] <mup> PR #8728: tests: detect stray dbus-daemon <Test Robustness> <Created by zyga> <https://github.com/snapcore/snapd/pull/8728>
[05:48] <zyga> and it essentially prints a TODO list of tests to improve
[05:49] <zyga> mborzecki: I replied but let's chat here
[05:49] <zyga> right now there's nothing more on stderr, if it goes to a file it'd be rather short-lived as that directory goes away when session-tool quits
[06:13] <mborzecki> zyga: if something fails does is the directory removed too?
[06:16] <zyga> yes
[06:16] <zyga> I looked at the rest of the script and we log other things there
[06:16] <zyga> but it's a remnant of development when I just would comment out the cleanup logic
[06:21] <pedronis> fedora-32 is listed in the workflows but doesn't exist in spread.yaml, this is making it fail constantly
[06:24] <zyga> pedronis: oh?
[06:24] <zyga> perhaps mvo added that with the caching PR
[06:25] <pedronis> seems so
[06:25] <pedronis> zyga: see here:  https://github.com/snapcore/snapd/pull/8729
[06:25] <mup> PR #8729: snap,many: mv Open to snapfile pkg to support add'l options to Container methods <Squash-merge> <UC20> <Created by anonymouse64> <https://github.com/snapcore/snapd/pull/8729>
[06:27] <zyga> yeah, indeed, it's there in the patch
[06:28] <zyga> it's not marked as requires so it was masked by opensuse failing as well, I suspect
[06:29] <zyga> if we cannot get a working image today I'll drop it from the matrix with a comment
[06:29] <pedronis> it was marked as required
[06:29] <pedronis> or maybe I'm confused
[06:29] <pedronis> anyway
[06:30] <pedronis> #8729 needs a 2nd review
[06:30] <mup> PR #8729: snap,many: mv Open to snapfile pkg to support add'l options to Container methods <Squash-merge> <UC20> <Created by anonymouse64> <https://github.com/snapcore/snapd/pull/8729>
[06:37] <mborzecki> i'll fix the sanpd selinux policy today so we'll be able to merge cachio's fedora 32 pr
[06:39] <zyga> pedronis: https://github.com/snapcore/snapd/pull/8729#pullrequestreview-418005303
[06:39] <mup> PR #8729: snap,many: mv Open to snapfile pkg to support add'l options to Container methods <Squash-merge> <UC20> <Created by anonymouse64> <https://github.com/snapcore/snapd/pull/8729>
[06:39] <pedronis> mvo: mborzecki: hi, Pawel's #8697 needs reviews
[06:39] <zyga> mborzecki: cool
[06:39] <mup> PR #8697: packaging: build cmd/snap and cmd/snap-bootstrap with nomanagers tag <Created by stolowski> <https://github.com/snapcore/snapd/pull/8697>
[06:41] <mvo> pedronis: sure, looking
[06:42] <mup> PR snapd#8723 closed: github: remove workaround for bug 133 in actions/cache <Simple 😃> <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/8723>
[06:43] <mup> PR snapd#8721 closed: devicestate, sysconfig: revert support for cloud.cfg.d/ in the gadget <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/8721>
[06:55] <mup> PR snapd#8729 closed: snap,many: mv Open to snapfile pkg to support add'l options to Container methods <Squash-merge> <UC20> <Created by anonymouse64> <Merged by pedronis> <https://github.com/snapcore/snapd/pull/8729>
[07:03] <pstolowski> morning
[07:14] <mup> PR snapd#8730 opened: data/selinux: allow snapd to remove/create the its socket <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/8730>
[07:16] <mborzecki> quick drive by fix when testing package updates pushed by Eighth_Doctor ^^
[07:16] <pstolowski> mborzecki: hi, can you take another look at https://github.com/snapcore/snapd/pull/8697 ?
[07:16] <mup> PR #8697: packaging: build cmd/snap and cmd/snap-bootstrap with nomanagers tag <Created by stolowski> <https://github.com/snapcore/snapd/pull/8697>
[07:16] <mborzecki> sure
[07:22] <zyga> pstolowski: small comment https://github.com/snapcore/snapd/pull/8697/files#r430205753
[07:22] <mup> PR #8697: packaging: build cmd/snap and cmd/snap-bootstrap with nomanagers tag <Created by stolowski> <https://github.com/snapcore/snapd/pull/8697>
[07:23] <zyga> not incorrect, just spreading the awesomeness of make
[07:30] <pstolowski> zyga: that's nice, thank you
[07:31] <mvo> zyga: oh, nice!
[07:31] <pedronis> mvo: the prereq has landed squash merged, now #8711 needs a 2nd review and hopefully to be (almost) green
[07:31] <mup> PR #8711: snap/squashfs: also symlink snap Install with uc20 seed snap dir layout <Needs Samuele review> <Squash-merge> <UC20> <Created by anonymouse64> <https://github.com/snapcore/snapd/pull/8711>
[07:31] <mvo> zyga: this should make some of the other stuff we do much more readable if we use this trick
[07:32] <mvo> pedronis: \o/ I have a look
[07:32] <zyga> mvo: cool :)
[07:33] <zyga> mvo: it's annoying we need to repeat this across packaging
[07:33] <zyga> I wish we'd use snaps.mk more
[07:34] <zyga> brb
[07:34] <mvo> zyga: unfortunately I don't think that's an option for debian/ because if we ever want someone to maintain this it has to be as clsoe as possible to a "standard"
[07:35] <zyga> mvo: I don't think so, it's not unlike dh_auto* and using a makefile in the upstream project
[07:35] <zyga> mvo: it's all about integrating dpkg-buildflags inside our native makefile
[07:43] <mup> PR snapd#8731 opened: snap,many: mv Open to snapfile pkg to support add'l options to Container methods (2.45) <Created by mvo5> <https://github.com/snapcore/snapd/pull/8731>
[07:43] <mvo> pedronis: I backported 8711 to 2.45, straight cherry-pick did unfortunately not work
[07:44] <pedronis> mvo: it's because of the prereq? #8729 ?
[07:44] <mup> PR #8729: snap,many: mv Open to snapfile pkg to support add'l options to Container methods <Squash-merge> <UC20> <Created by anonymouse64> <Merged by pedronis> <https://github.com/snapcore/snapd/pull/8729>
[07:44] <pedronis> or something in it itself?
[07:46] <pedronis> mvo: ah, you backported the prereq actually https://github.com/snapcore/snapd/pull/8731 ?
[07:46] <mup> PR #8731: snap,many: mv Open to snapfile pkg to support add'l options to Container methods (2.45) <Created by mvo5> <https://github.com/snapcore/snapd/pull/8731>
[07:47] <pedronis> mvo: there's a bit of extra tests in boot, do they pass?
[07:51] <pedronis> mvo: we should backport #8721
[07:51] <mup> PR #8721: devicestate, sysconfig: revert support for cloud.cfg.d/ in the gadget <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/8721>
[07:59] <zyga> pstolowski: with this trick we could actually reduce quite a lot of clutter in the code building commands
[07:59] <zyga> pstolowski: with this trick we could actually reduce quite a lot of clutter in the code building commands
[07:59] <pstolowski> zyga: ok, but perhaps as followups
[07:59] <zyga> yes, totally
[07:59] <pstolowski> zyga: i'd really like to get this landed to unblock stuff
[07:59] <zyga> let me do it after your bits land
[08:02] <mvo> pedronis: 8721> yes, looking at this too
[08:09] <mborzecki> pstolowski: can you take a look at https://github.com/snapcore/snapd/pull/8730 ?
[08:09] <mup> PR #8730: data/selinux: allow snapd to remove/create the its socket <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/8730>
[08:13] <zyga> mborzecki: https://github.com/snapcore/snapd/pull/8656#discussion_r430233548
[08:13] <mup> PR #8656: snap-mgmt: perform cleanup of user services <Created by zyga> <https://github.com/snapcore/snapd/pull/8656>
[08:16] <pstolowski> siure
[08:20] <mborzecki> i'll force push to cachio's f32 pr and drop the last commit he made
[08:20] <mup> PR snapd#8726 closed: tests: silence stderr from dbus-monitor <Test Robustness> <Created by zyga> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/8726>
[08:21] <zyga> ta mvo
[08:21] <zyga> and mborzecki :)
[08:21] <mup> PR snapd#8725 closed: testutil/exec.go: set PATH after running shellcheck <Simple 😃> <Test Robustness> <Created by anonymouse64> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/8725>
[08:24] <mup> PR snapd#8730 closed: data/selinux: allow snapd to remove/create the its socket <Created by bboozzoo> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/8730>
[08:25] <mup> PR snapd#8604 closed: interfaces/builtin/desktop: do not mount fonts cache on distros with quirks <Squash-merge> <Created by bboozzoo> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/8604>
[08:41] <mvo> 8351 needs a second review
[08:44] <mvo> pedronis: do you want to do a review of 8661? it has two +1 at this point (but none from you)
[08:47] <mup> PR snapd#8732 opened: data/selinux: update policy to allow forked processes to call getpw*() <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/8732>
[08:47] <mborzecki> pstolowski: another one for you ^^
[08:47] <pstolowski> mborzecki: siure
[08:47] <pstolowski> grr
[08:47] <pstolowski> *sure
[08:48] <pstolowski> zyga: can you remind me where in the spread suite do we check for leftovers from namespaces etc?
[08:48] <zyga> pstolowski: what do you mean by leftover from namespaces?
[08:49] <pstolowski> zyga: sorry, probably mount changes; anything that lxd can leave behind
[08:50] <zyga> I see
[08:50] <zyga> I don't think we do that regularly
[08:50] <zyga> we kind of do it in one spot in mount-ns test
[08:50] <zyga> a branch that I have long open tried to but it picked up too many failures so it was unable to land
[08:50] <zyga> we may get that, a bit, with invariant-tool
[08:50] <pstolowski> zyga: aah, mount-ns test, i think that was it
[08:50] <zyga> but it needs to be done still
[08:52] <pstolowski> thanks
[08:57] <zyga> brb
[09:00] <pedronis> mvo: ? I don't see the 2nd review
[09:01] <pedronis> mvo: as I remarked I think we should apply mborzecki idea of trying not to restart if nothing has changed
[09:01] <pedronis> mvo: I think it needs a review from pstolowski
[09:03] <pstolowski> pedronis, mvo i'll review it in a bit (and also the vitality PR)
[09:04] <zyga> re
[09:06] <mup> PR snapd#8711 closed: snap/squashfs: also symlink snap Install with uc20 seed snap dir layout <Needs Samuele review> <Squash-merge> <UC20> <Created by anonymouse64> <Merged by pedronis> <https://github.com/snapcore/snapd/pull/8711>
[09:06] <pedronis> mvo: ^ I squash-merged it
[09:08] <mvo> pedronis: \o/
[09:09] <mvo> pedronis: re 8661> sorry, ian did two +1 I misread that
[09:10] <pedronis> ah, was confused for a sec
[09:10] <pedronis> pstolowski: actualFsTypeAndMountOptions doesn't need to return an error anymore as well
[09:10] <pstolowski> pedronis: missed that, thanks
[09:11] <pedronis> pstolowski: np, but that's where I started thinking about this, because I had move the call to it and then wondered if it would change the failure modes but actually it can't error in itself
[09:12] <pedronis> *I had you
[09:13] <pedronis> pstolowski: mvo: do we need/want the preseed fixes in 2.45.1 ?
[09:14] <mborzecki> quick errand, back in 30
[09:16] <mup> PR core20#65 closed: Do not attempt signing changes during snap build <Created by xnox> <Merged by xnox> <https://github.com/snapcore/core20/pull/65>
[09:16] <pedronis> mvo: this is the comment I was referring to in 8661: https://github.com/snapcore/snapd/pull/8661#discussion_r427920348
[09:16] <mup> PR #8661: configcore: add "service.console-conf.disable" config option <Created by mvo5> <https://github.com/snapcore/snapd/pull/8661>
[09:16] <pstolowski> pedronis: that would be nice to have
[09:17] <mup> Issue core20#63 closed: fix travis builds with new subiquity from git <Created by xnox> <Closed by xnox> <https://github.com/snapcore/core20/issue/63>
[09:17] <mup> Issue core20#66 opened: Use focal base in travis <Created by xnox> <https://github.com/snapcore/core20/issue/66>
[09:17] <pstolowski> otherwise it will take long time before it's easily usable with lxd
[09:19] <mvo> pedronis: preseed> it would be a good idea, yes
[09:20] <mvo> pedronis: 8661> aha, sorry, I will look at this
[09:24] <mup> PR core20#49 closed: fix broken symlinks in /etc/writable <Created by mvo5> <Merged by xnox> <https://github.com/snapcore/core20/pull/49>
[09:26] <zyga> mvo: I will most likely miss standup today
[09:26] <zyga> mvo: I'll send notes to the standup doc
[09:27] <zyga> I may join from a phone if I have the conditions to
[09:30] <mvo> zyga: ok
[09:30] <pedronis> pstolowski: mvo: #8697 failed on centos-8, seems unrelated but not sure
[09:30] <mup> PR #8697: packaging: build cmd/snap and cmd/snap-bootstrap with nomanagers tag <Created by stolowski> <https://github.com/snapcore/snapd/pull/8697>
[09:31] <pstolowski> pedronis: yes, i've restarted it
[09:35]  * zyga goes to look at 8697 restart not wasting money anymore
[09:47] <mborzecki> re
[10:07] <pedronis> mborzecki: Ian made a suggestion in #8718 then it should be good to go
[10:07] <mup> PR #8718: boot, many: require mode in modeenv <UC20> <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/8718>
[10:18] <mborzecki> pedronis ijohnson  thanks for the reviews
[10:18]  * zyga hugs mvo 
[10:18] <zyga> mvo: the caching behavior is BRILLIANT
[10:18] <zyga> 2 seconds
[10:18] <mborzecki> need to address some commentrs in 8716 too
[10:18] <zyga> 0.00 cost
[10:18] <mborzecki> zyga: hah nice :P
[10:18] <zyga> I wonder if this will show up in load charts on the spread runner
[10:19] <zyga> as we're usually bound by memory
[10:19] <ijohnson> morning folks
[10:20] <mborzecki> ijohnson: heya
[10:20] <ijohnson> zyga: re: your comment on 8729 about the header being one byte larger than magic, the magic bytes is just 4 bytes, so a header of 20 bytes already is 16 bytes larger ?
[10:20] <ijohnson> hey mborzecki
[10:20] <zyga> ijohnson: yeah but it's a constannt
[10:20] <mup> PR snapd#8733 opened: tests: port document-portal-activation to session-tool <Test Robustness> <Created by zyga> <https://github.com/snapcore/snapd/pull/8733>
[10:20] <zyga> it's okay but not for the right reasons :)
[10:20] <ijohnson> zyga: so you think it should be only 5 bytes ?
[10:21] <zyga> I think it should be exactly len(magic) + 1
[10:21] <zyga> :D
[10:21] <ijohnson> got it
[10:23] <mup> PR snapd#8734 opened: tests: port interfaces-wayland to session-tool <Test Robustness> <Created by zyga> <https://github.com/snapcore/snapd/pull/8734>
[10:23] <mup> PR snapd#8735 opened: snap/snapfile,squashfs: followups from 8729 <Simple 😃> <Created by anonymouse64> <https://github.com/snapcore/snapd/pull/8735>
[10:23] <ijohnson> zyga: applied in a followup ^
[10:23] <zyga> super
[10:23] <zyga> thanks!
[10:24] <zyga> I was going to do it myself but I was working on a streak of patches
[10:33] <zyga> mborzecki: https://github.com/snapcore/snapd/pull/8736
[10:33] <mup> PR #8736: tests: log stderr from dbus-monitor <Simple 😃> <Created by zyga> <https://github.com/snapcore/snapd/pull/8736>
[10:33] <pedronis> ijohnson: not a fan of len(magic)+1, it's not particularly better or worse than the existing especially now that its in squashfs itself
[10:33] <mup> PR snapd#8736 opened: tests: log stderr from dbus-monitor <Simple 😃> <Created by zyga> <https://github.com/snapcore/snapd/pull/8736>
[10:34]  * zyga returns to https://github.com/snapcore/snapd/pull/7825
[10:34] <mup> PR #7825: many: use transient scope for tracking apps and hooks <Security-High> <Created by zyga> <https://github.com/snapcore/snapd/pull/7825>
[10:34] <pedronis> ijohnson: I'll push something else there
[10:34] <ijohnson> pedronis: ok
[10:34] <pedronis> ijohnson: I mean a 5 bytes file is never a valid squashfs anyway
[10:34] <ijohnson> well that's not really for us to decide ?
[10:35] <ijohnson> unsquashfs is the ultimate decider of if it's a valid squashfs ?
[10:35] <pedronis> ijohnson: not but as a predictor this is worse than before
[10:35] <pedronis> ijohnson: if we didn't care we could use len(magic) and Equal
[10:35] <ijohnson> I mean for something that's called magic I wouldn't expect it to be perfect
[10:36] <pedronis> why the +1
[10:36] <pedronis> anyway I will do yet something else
[10:36] <ijohnson> and I think we are splitting hairs over a tiny thing
[10:36] <ijohnson> sure feel free
[10:36] <ijohnson> I just took zyga's suggestion
[10:36] <ijohnson> I have no strong feelings about this really
[10:36] <ijohnson> just feels a bit pointless
[10:36] <pedronis> ijohnson: I know, I disagree with the suggestion, doing nothing would have been fine
[10:37] <pedronis> ijohnson: also fn to f is not an improvement either
[10:37] <pedronis> ijohnson: anyway, it's not your problem anymore, I will push something there
[10:37] <mup> PR snapd#8737 opened: snap/squashfs: also symlink snap Install with uc20 seed snap dir layout (2.45) <Created by mvo5> <https://github.com/snapcore/snapd/pull/8737>
[10:41] <pstolowski> Cannot allocate google:opensuse-tumbleweed-64: cannot allocate new Google server for opensuse-tumbleweed-64: invalid value for field 'resource.shieldedInstanceConfig': ''. Shielded VM Config can only be set when using a UEFI-compatible disk
[10:49] <zyga> pstolowski: known
[10:49] <zyga> sergio needs to fix the images
[10:49] <pstolowski> ok
[11:07] <mup> PR snapd#8731 closed: snap,many: mv Open to snapfile pkg to support add'l options to Container methods (2.45) <Created by mvo5> <Closed by mvo5> <https://github.com/snapcore/snapd/pull/8731>
[11:12] <mup> PR snapd#8737 closed: snap/squashfs: also symlink snap Install with uc20 seed snap dir layout (2.45) <Created by mvo5> <Closed by mvo5> <https://github.com/snapcore/snapd/pull/8737>
[11:13] <mup> PR snapd#8738 opened:  snap,many: mv Open to snapfile pkg to support add'l options to Container methods (2.45) <Created by mvo5> <https://github.com/snapcore/snapd/pull/8738>
[11:25] <pedronis> ijohnson: once I was looking there, I added some tests to 8735 as well
[11:25] <ijohnson> pedronis: ok, seems the scope of the PR has greatly increased though
[11:27] <ijohnson> anyways approved it
[11:27]  * ijohnson afk for bit
[11:32] <pedronis> pstolowski: zyga: I tweaked/expanded a little bit #8729, it needs a re-review
[11:32] <mup> PR #8729: snap,many: mv Open to snapfile pkg to support add'l options to Container methods <Squash-merge> <UC20> <Created by anonymouse64> <Merged by pedronis> <https://github.com/snapcore/snapd/pull/8729>
[11:32] <zyga> yeah, reviewing now
[11:32] <pstolowski> okay
[11:32] <pstolowski> pedronis: i've commented on the tooling proposal
[11:35] <zyga> pedronis: reviewed
[11:37] <pstolowski> pedronis: hmm but 8729 was merged?
[11:37] <pedronis> heh sorry
[11:38] <pedronis> pstolowski: I meant #8735, but another review is not stricly needed anymore
[11:38] <pstolowski> ah 8735 i guess
[11:38] <mup> PR #8735: snap/snapfile,squashfs: followups from 8729 <Created by anonymouse64> <https://github.com/snapcore/snapd/pull/8735>
[11:38] <pstolowski> yep
[11:40] <pedronis> pstolowski: seems 8697 can be merged now
[11:41] <pedronis> which should unblock 8533
[11:42] <zyga> pstolowski: thanks for commenting there
[11:42] <zyga> pedronis: I agree with pawel's comments
[11:43] <zyga> pedronis: I don't think we need a distinction between suite/task in the name, a common prefix plus something relevant to the tool function is good for me
[11:43] <mup> PR snapd#8697 closed: packaging: build cmd/snap and cmd/snap-bootstrap with nomanagers tag <Created by stolowski> <Merged by stolowski> <https://github.com/snapcore/snapd/pull/8697>
[11:43] <zyga> pedronis: if you pick a name for session tool I'll do a test mass patch to see how it looks like
[11:44] <pedronis> zyga: I'm not against the prefix, my issue is that I really care about distinguishing things that affects state across tasks vs not
[11:45] <zyga> hmmm
[11:46] <mup> PR core20#67 opened: tests: use SNAPCRAFT_PRIME dir instead of hardcoding "prime" <Created by mvo5> <https://github.com/snapcore/core20/pull/67>
[11:46] <zyga> so things like retry-tool vs things like lxd cleanup?
[11:48] <zyga> pedronis: under this classification, session tool does not affect state across tasks, correct?
[11:51] <pedronis> zyga: well session can survive if don't clean them up, no?
[11:51] <zyga> pedronis: sure but the point is to pair session prepare and restore
[11:52] <pedronis> so it affects state across task, unless you use it just right
[11:52] <pedronis> the classification is based on ideal usage, it's about what gets touched
[11:52] <zyga> I think virtually everything has this property
[11:52] <zyga> mkdir does
[11:53] <pedronis> zyga: mkdir is a tool
[11:53] <pedronis> sorry
[11:53] <pedronis> mkdir is not a tool
[11:54] <zyga> so what is it?
[11:57] <pedronis> zyga: for analogy, it's the difference between a helper function vs a methond on the suite struct in go
[11:57] <zyga> ok
[11:57] <pedronis> we have some stateless helpers attached to suite sometimes
[11:58] <zyga> I think attaching this to suite is misleading as it reflects a particular way we use it, I look at it more like a context manager in python, it's an enter/exit pair with user code in between
[11:58] <pedronis> zyga: we don't have a with statement
[11:58] <mborzecki> pedronis: ijohnson: https://github.com/snapcore/snapd/pull/8716 is updated now
[11:58] <zyga> as such it could be used with the same vailidity, at the level of a task or the level of a suite
[11:58] <mup> PR #8716: o/devicestate: refactor current system handling <Simple 😃> <UC20> <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/8716>
[11:59] <zyga> pedronis: sure but the pairing nature is similar to one, it's just an analogy
[11:59] <pedronis> zyga: anyway it seems you are objecting to suite. , not to the dinstiction
[11:59] <zyga> yes, and I understand your point better now
[12:00] <zyga> so that glancing at the tool name gives some context as to what is the possible impact
[12:00] <pedronis> yes
[12:00] <zyga> I don't know if I have a better name suggestion but I see the goal
[12:03] <pedronis> zyga: pstolowski: how much is short names a goal for you?
[12:04] <zyga> pedronis: I'm okay with a non-short name
[12:04] <mborzecki> time to revisit https://github.com/snapcore/snapd/pull/7414
[12:04] <zyga> I prefer test- to tst- or similar
[12:04] <mup> PR #7414: tests: keep track of installed packages and restore the state after the test <Test Robustness> <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/7414>
[12:04] <zyga> mborzecki: looking
[12:04] <zyga> mborzecki: could that be framed in the terms of invariant-tool?
[12:06] <zyga> I think we should be more about detecting issues than automatically fixing the issue as that is easy not to know about; while somewhat more verbose I'd prefer to see a test both install and remove a package rather than rely on another layer computing and removing everything
[12:06] <zyga> just my 0.02
[12:06] <zyga> ok, folks, I'm arriving at my mom's place, I will miss the standup but I'll write something in the document; remember it's mother's day :)
[12:07] <ijohnson> zyga: interesting mother's day in the US was like 2 weekends ago
[12:07] <zyga> oh really? :)
[12:07] <ijohnson> I didn't realize it was different in different countries
[12:07] <zyga> same here
[12:07] <zyga> wonder why :)
[12:07] <mborzecki> zyga: yeah, it's mother's day in pl and maybe some other countries
[12:07] <ijohnson> yeah it's may 10th in US
[12:07] <zyga> ok, ttyl
[12:07] <zyga> I'll be back online when we leave
[12:08] <mborzecki> wow, checked wikipedia, actually may 26 is mothers day only in PL
[12:08] <pstolowski> pedronis: not super important
[12:08] <pstolowski> mborzecki, ijohnson heh, interesting
[12:08] <ijohnson> haha yeah that's interesting
[12:12] <pedronis> mborzecki: 8716 looking good, couple small comments
[12:17] <pstolowski> cachio: hi, i've addressed you comments to #8710
[12:18] <mup> PR #8710: tests: spread test for preseeding in lxd container (3/3) <Preseeding 🍞> <Created by stolowski> <https://github.com/snapcore/snapd/pull/8710>
[12:22] <pedronis> mborzecki: ijohnson: seededSystem and System are internal vs external and have different content, in theory we can unify seededSystem and currentSystem by adding an optional omitempty actions to the former
[12:22] <pedronis> ijohnson: would that help?
[12:23] <pedronis> also those struct definitions should probably move to systems.go as well
[12:23] <ijohnson> yes that sounds like it would help but I don't want to be a bother so if y'all are ok with it as is then it's fine for me too
[12:23] <pstolowski> we're at the brink of having 4th page of PRs..
[12:24] <cachio> pstolowski, thanks
[12:27] <pstolowski> zyga: do you have a moment for https://github.com/snapcore/snapd/pull/8709 ?
[12:27] <mup> PR #8709: cmd/snap-preseed, systemd: fix handling of fuse.squashfuse when preseeding (2/3) <Preseeding 🍞> <Created by stolowski> <https://github.com/snapcore/snapd/pull/8709>
[12:39] <pedronis> ijohnson: mborzecki: let's land it as is for now, and I'll look into it
[12:39] <mborzecki> ack
[12:39] <ijohnson> fine with me thanks
[12:39] <pstolowski> pedronis: does https://github.com/snapcore/snapd/pull/8608 still need to be blocked? was it blocked because of 2.45?
[12:39] <mup> PR #8608: configcore: issue a warning on core16 when journal.persistent option is set <Squash-merge> <⛔ Blocked> <Created by stolowski> <https://github.com/snapcore/snapd/pull/8608>
[12:40] <pedronis> pstolowski: it's blocked becausse mvo made the good point that we should auto-remove the warning if there's a reboot
[12:40] <pedronis> pstolowski: that needs a bit of infra work I fear
[12:41] <pstolowski> pedronis: ah.. good point indeed. i think the comment was made elsewhere, i wasn't aware
[12:42] <pedronis> anyway, it's a bit unclear if it's a good time to work on that
[12:42] <pedronis> let me put this in it at least
[12:42] <cachio> pstolowski, I left few comments
[12:42] <cachio> it is almost ready
[12:42] <pedronis> pstolowski: ah, you did
[12:42] <cachio> just 1 thng to fix
[12:42] <pedronis> pstolowski: thx
[12:45] <pstolowski> pedronis: ok, i can pick it up when time permits and there is nothing more pressing
[12:45] <pstolowski> cachio: thanks
[12:45] <pedronis> pstolowski: exactly
[12:45] <pedronis> pstolowski: I think early config stuff, services stuff, and probably after that snapshot stuff is higher priority
[12:46] <pedronis> and making sure preseed works as well
[12:47] <pstolowski> ack
[12:48] <pstolowski> pedronis: what's the conclusion re preseed fixes for lxd & 2.45.1?
[12:48] <pedronis> pstolowski: I think mvo said yes, but he can reconfirm
[12:55] <mup> PR snapd#8736 closed: tests: log stderr from dbus-monitor <Simple 😃> <Created by zyga> <Merged by zyga> <https://github.com/snapcore/snapd/pull/8736>
[12:58] <mvo> pstolowski: yes please unless it's a ton of work
[13:14] <pstolowski> mvo: ok, will do once the remaining 2 PRs land i nmaster
[13:15] <pstolowski> *in master
[13:15] <mup> PR snapd#8732 closed: data/selinux: update policy to allow forked processes to call getpw*() <Created by bboozzoo> <Merged by bboozzoo> <https://github.com/snapcore/snapd/pull/8732>
[13:23] <mup> PR snapcraft#3143 closed: cli: migrate close to use the new channel map <Created by sergiusens> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/3143>
[13:23] <mup> PR snapd#8718 closed: boot, many: require mode in modeenv <UC20> <Created by bboozzoo> <Merged by bboozzoo> <https://github.com/snapcore/snapd/pull/8718>
[14:00] <ackk> sergiusens, jdstrand, hi, FTR it seems even the new review-tools still complain about the $SNAP/bin/python symlink which snapcraft creates. I workaround'd it by excluding it via "prime" as it's not really needed, but it seems something either snapcraft or review-tools should handle?
[14:03] <mborzecki> afk for some time
[14:14] <zyga> running out of battery, more work back home
[14:14] <zyga> o/
[14:21] <mup> Bug #1880698 opened:  /etc/writable is double mounted <uc20> <Snappy:New> <https://launchpad.net/bugs/1880698>
[14:24] <mup> PR snapd#8739 opened: tests: port interfaces-desktop-* to session-tool <Test Robustness> <Created by zyga> <https://github.com/snapcore/snapd/pull/8739>
[14:30] <mup> PR core20#67 closed: tests: use SNAPCRAFT_PRIME dir instead of hardcoding "prime" <Created by mvo5> <Merged by xnox> <https://github.com/snapcore/core20/pull/67>
[14:36] <mup> PR snapd#8740 opened: spread.yaml: apply yaml formatter/linter <Simple 😃> <Created by anonymouse64> <https://github.com/snapcore/snapd/pull/8740>
[14:37] <mvo> cachio: 8693 has a conflict
[14:38] <cachio> mvo, thanks
[14:39] <cachio> I'll fix it
[14:40] <mvo> cachio: thanks!
[14:57] <jdstrand> ackk: hey, which snap was that again? I don't think the new review-tools is in prod yet
[15:10] <roadmr> jdstrand: it is, went to prod yesterday
[15:10] <mup> PR snapd#8735 closed: snap/snapfile,squashfs: followups from 8729 <Created by anonymouse64> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/8735>
[15:15] <mup> PR snapd#8656 closed: snap-mgmt: perform cleanup of user services <Created by zyga> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/8656>
[15:16] <mup> PR snapd#8738 closed:  snap,many: mv Open to snapfile pkg to support add'l options to Container methods (2.45) <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/8738>
[15:19] <pedronis> mvo: ^ great, now we can port the fix?
[15:19] <pedronis> *backport
[15:19] <mup> PR snapd#8741 opened: snap/squashfs: also symlink snap Install with uc20 seed snap dir layout (2.45) <Created by mvo5> <https://github.com/snapcore/snapd/pull/8741>
[15:20] <pedronis> ah :)
[15:20] <pedronis> thx
[15:20] <mvo> pedronis: you want 8656 for 2.45 ?
[15:21] <pedronis> mvo: the feaure is experimental, but yes, it would make sense
[15:21] <pedronis> if it's not too disruptive
[15:21] <mvo> pedronis: sure, let me double check
[15:25] <mup> PR snapd#8742 opened: snap-mgmt: perform cleanup of user services (2.45) <Created by mvo5> <https://github.com/snapcore/snapd/pull/8742>
[15:25] <mvo> 8351 needs a second review
[15:26] <ijohnson> mvo is https://github.com/snapcore/snapd/pull/8661 ready out of curiosity ?
[15:26] <mup> PR #8661: configcore: add "service.console-conf.disable" config option <Created by mvo5> <https://github.com/snapcore/snapd/pull/8661>
[15:27] <ijohnson> also mvo did you see my comment in the SU doc about needing to port https://github.com/snapcore/snapd/pull/8706 to 2.45 ?
[15:27] <mup> PR #8706: interfaces/serial-port: add NXP SC16IS7xx (ttySCX) to allowed devices <Created by tsunghanliu> <Merged by stolowski> <https://github.com/snapcore/snapd/pull/8706>
[15:27] <ijohnson> it was a few days ago
[15:40]  * cachio lunch
[15:41] <zyga> I’m home eating lunch
[15:41] <zyga> Back in some time
[15:44] <mup> PR snapd#8743 opened: tests: cherry-pick test fixes from master <Created by mvo5> <https://github.com/snapcore/snapd/pull/8743>
[16:01] <cmatsuoka> ijohnson: I created a model for the pi but my boot is failing because the assertion is signed with an expired key, I think it's caused by the system date being completely wrong there
[16:02] <cmatsuoka> ijohnson: how do you fix the date in order to be able to boot?
[16:05] <ijohnson> cmatsuoka: you mean that it's failing to do the first boot of install mode when the seed is verified during the initramfs?
[16:05] <cmatsuoka> ijohnson: it's failing in install mode
[16:06] <cmatsuoka> the-tool[285]: [ - assertion is signed with expired public key ...
[16:06] <ijohnson> cmatsuoka: but to be clear the error is in the initramfs correct?
[16:06] <cmatsuoka> yes, it's in the initramfs
[16:09] <ijohnson> cmatsuoka: do you know how far back the time is in the initramfs? I.e. what time the pi thinks it is?
[16:10] <cmatsuoka> ijohnson: I don't know, can I read that from uboot?
[16:10] <ogra> its the birthday of the granny of the founder of broadcom ...
[16:11] <ogra> you cant read it from u-boot
[16:12] <pedronis> ijohnson: cmatsuoka: we should be able to apply fixes like the one we do from the real systems from snap-bootstrap I suppose
[16:12] <ogra> (you could patch u-bot to read an RTC if there was a RTC ... but there isnt one on the pi)
[16:12] <ogra> *u-boot
[16:12] <cmatsuoka> ogra: how do you deal with this situation? I'm sure you went through this before
[16:13] <ogra> i wrote the fixrtc script 🙂
[16:13] <ijohnson> cmatsuoka: I can show you how to flash a new boot.scr when I'm back from lunch
[16:13] <ogra> but i think thats been dropped for some new thing that xnox worked on
[16:13] <ijohnson> That way you can add dangerous to kernel cmdline  and get an initramfs debug shell when the-tool fails
[16:14] <cmatsuoka> ijohnson: ok, I'll have lunch too and will be back in 30 min or so
[16:14] <ijohnson> Okay
[16:18] <mup> PR snapd#8716 closed: o/devicestate: refactor current system handling <UC20> <Created by bboozzoo> <Merged by pedronis> <https://github.com/snapcore/snapd/pull/8716>
[16:19] <mup> PR core20#68 opened: static: make /etc/writable mode "transition" in writable-paths <Created by mvo5> <https://github.com/snapcore/core20/pull/68>
[16:20] <xnox> cmatsuoka:  why did you create a new model for the pi? we have all the models one may need avaiable. including dangerous ones.
[16:20] <xnox> cmatsuoka:  or is it not good somehow?
[16:20] <xnox> cmatsuoka:  can you share the model please?
[16:27] <jdstrand> ackk: I found the snap
[16:28] <ackk> jdstrand, ah sorry, missed your earlier message
[16:31] <jdstrand> ackk: ok, I see the issue
[16:31] <jdstrand> cc sergiusens ^
[16:31] <ackk> jdstrand, it's canonical-rbac. https://dashboard.snapcraft.io/snaps/canonical-rbac/revisions/241/ is an example of one failure
[16:31] <jdstrand> yep
[16:31] <jdstrand> thanks
[16:32] <ijohnson> mvo: approved #8741 to 2.45 for you
[16:32] <mup> PR #8741: snap/squashfs: also symlink snap Install with uc20 seed snap dir layout (2.45) <Created by mvo5> <https://github.com/snapcore/snapd/pull/8741>
[16:33] <mvo> ijohnson: \o/
[16:33] <ijohnson> mvo: also did you see my earlier ping re: the serial-port regex PR for 2.45 ?
[16:33] <ijohnson> mvo: i.e. https://github.com/snapcore/snapd/pull/8706
[16:33] <mup> PR #8706: interfaces/serial-port: add NXP SC16IS7xx (ttySCX) to allowed devices <Created by tsunghanliu> <Merged by stolowski> <https://github.com/snapcore/snapd/pull/8706>
[16:34] <mvo> ijohnson: I did not see that, cherry-pick should be fine
[16:34] <ijohnson> mvo: ack just an FYI that I requested jdstrand review it, but it was merged before he could review it, not sure if that affects your cherry-pick decision or not
[16:34] <mvo> ijohnson: done
[16:34] <mvo> ijohnson: well, now it would have - oh well
[16:34] <ijohnson> sorry
[16:34] <ijohnson> I should have made that more clear
[16:35] <mvo> ijohnson: it's okay, it looks harmless enough
[16:35] <ijohnson> agreed, I think in the future I will mark as blocked PR's that require jdstrand's review so that this doesn't happen again
[16:42] <mvo> ijohnson: I added a needs-security label now
[16:42] <ijohnson> mvo: ok, sounds good
[16:45] <ijohnson> cmatsuoka: so when you're back, install u-boot-tools via apt, then take this script, https://pastebin.ubuntu.com/p/Nrd5N8n8BV/ and run:
[16:45] <ijohnson> cmatsuoka: `mkimage -A arm64 -O linux -T script -C none -n "boot script" -d bootscript.rpi /media/your-sd-card/ubuntu-seed/boot.scr`
[16:45] <ijohnson> cmatsuoka: that script should be named bootscript.rpi for example
[16:46] <mup> PR snapd#8744 opened: interfaces-ssh-keys: Support reading /etc/ssh/ssh_config.d/ <Created by andrewsomething> <https://github.com/snapcore/snapd/pull/8744>
[16:47] <jdstrand> ijohnson, mvo: PR reviews start today. I'll do that one first
[16:48] <ijohnson> thanks jdstrand
[16:48] <mvo> ta
[17:00] <sergiusens> jdstrand: thanks
[17:04] <cmatsuoka> ijohnson: thanks! in this case it seems that the cause was a key that was new and not valid yet :(
[17:05] <cmatsuoka> ijohnson: I'm rebuilding the image with the dangerous model and overriding channels, that should work
[17:05] <cmatsuoka> ijohnson: but the store is incredibly slow right now
[17:05] <ijohnson> :-(
[17:47] <mup> PR snapcraft#3145 opened: cli: migrate upload and release to new channel-map <Created by sergiusens> <https://github.com/snapcore/snapcraft/pull/3145>
[17:53] <mup> PR snapcraft#3131 closed: plugin handler: debug build commands if developer debug enabled <enhancement> <Created by cjp256> <Closed by sergiusens> <https://github.com/snapcore/snapcraft/pull/3131>
[18:01] <mup> PR core20#68 closed: static: make /etc/writable mode "transition" in writable-paths <Created by mvo5> <Merged by xnox> <https://github.com/snapcore/core20/pull/68>
[18:34] <mup> PR snapd#8741 closed: snap/squashfs: also symlink snap Install with uc20 seed snap dir layout (2.45) <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/8741>
[19:05] <zyga> re
[19:05] <zyga> cachio: hey
[19:05] <cachio> zyga, hey
[19:06] <zyga> cachio: hey, I missed the standup today
[19:06] <zyga> I wanted to ask about opensuse images
[19:06] <zyga> did you notice those are not working now?
[19:06] <cachio> zyga, I am updating those
[19:06] <cachio> but takes a bit of time
[19:06] <zyga> ah, great, thank you :)
[19:06] <cachio> it should be ready in 15 minutes
[19:06] <cachio> both
[19:06] <cachio> zyga, np
[19:06] <zyga> sounds good
[19:07] <zyga> and another question, is fedora 32 operational?
[19:07] <zyga> if not perhaps we should disable it form the test matrix
[19:07] <cachio> zyga, https://github.com/snapcore/snapd/pull/8693/checks?check_run_id=709950381
[19:07] <cachio> last tests passed
[19:07] <mup> PR #8693: tests: add fedora 32 to spread.yaml <Created by sergiocazzolato> <https://github.com/snapcore/snapd/pull/8693>
[19:07] <zyga> oh that's great
[19:07] <cachio> I need to fix other images and then we can merge the pr
[19:07] <zyga> ok, once those are rolled out we should be green
[19:07] <zyga> ok
[19:08] <cachio> today I applied the changes suggested in the pr and those worked
[19:09] <mup> PR snapd#8734 closed: tests: port interfaces-wayland to session-tool <Test Robustness> <Created by zyga> <Merged by zyga> <https://github.com/snapcore/snapd/pull/8734>
[19:53] <mvo> a "review" for 8743 would be great, all just cherry picks from master to fix various test failures in 2.45 (probably more to come but this is a good start)
[19:53] <mup> PR snapd#8740 closed: spread.yaml: apply yaml formatter/linter <Simple 😃> <Created by anonymouse64> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/8740>
[20:00] <jdstrand> sil2100: hey, fyi, you may have noticed that the core snap was blocked on manual review due to tzdata updates in xenial-updates shuffling around files. wondering if you feel like that was useful or if it would be better to skip checking usr/share/zoneinfo
[20:00]  * jdstrand approved all of those btw
[20:04] <zyga> jdstrand: o/
[20:06] <mup> PR snapd#8521 closed: snap-bootstrap,secboot: clear tpm if provisioning fails <UC20> <Created by cmatsuoka> <Closed by cmatsuoka> <https://github.com/snapcore/snapd/pull/8521>
[20:12] <mup> PR snapcraft#3146 opened: build providers: snap sw to channels if injecting <Created by sergiusens> <https://github.com/snapcore/snapcraft/pull/3146>
[20:25] <zyga> cachio: o/ are the suse images out?
[20:26] <cachio> zyga, yes
[20:26] <zyga> thanks, I'll re-trigger a few of my tests
[20:26] <zyga> this job caching makes running tests just magical
[20:34] <cachio> zyga, indeed
[20:45] <jdstrand> hey zyga :)
[20:51]  * cachio afk
[20:57] <mup> PR snapcraft#3147 opened: Use `pylxd` instead of `lxc exec` <Created by abitrolly> <https://github.com/snapcore/snapcraft/pull/3147>
[21:10] <mup> PR snapd#8739 closed: tests: port interfaces-desktop-* to session-tool <Test Robustness> <Created by zyga> <Merged by zyga> <https://github.com/snapcore/snapd/pull/8739>
[22:52] <mup> PR snapd#8745 opened: tests: port xdg-open-compat to session-tool <Test Robustness> <Created by zyga> <https://github.com/snapcore/snapd/pull/8745>
[22:58] <mup> PR snapd#8746 opened: tests: port xdg-open to session-tool <Test Robustness> <Created by zyga> <https://github.com/snapcore/snapd/pull/8746>
[22:58]  * zyga wraps up for the night
[22:58] <zyga> cachio: ^ if you have energy left :)
[23:05]  * zyga really EODs
[23:05] <zyga> (I had one more branch ready)
[23:05] <mup> PR snapd#8747 opened: tests: port snap-routine-portal-info to session-tool <Test Robustness> <Created by zyga> <https://github.com/snapcore/snapd/pull/8747>
[23:10] <cachio> zyga, I'll take a look