[05:05] morning [06:04] PR snapd#5116 closed: interfaces: move host font update-ns AppArmor rules to desktop interface [06:19] Thanks! [06:24] zyga: hey, pushed more test cases to 4504 https://github.com/snapcore/snapd/pull/4504#discussion_r187152683 [06:24] PR #4504: snap, wrappers: systemd WatchdogSec support [06:25] zyga: i think you missed the parenthesis in the if condition [06:37] Ack [06:37] I will check shortly [06:43] store hiccups continue today? [06:53] please tell me that isn't so :) === pstolowski|afk is now known as pstolowski [06:54] morning [06:56] cześć [06:56] jak tam? :) [06:56] I need to plan an upgrade from artful [06:56] I should probably do a clean bionic install from the final iso to see how that is like [06:57] and then move my data over [06:58] jamesh: hey, I merged master into #5115 [06:58] PR #5115: interfaces: add xdg-document-portal support to desktop interface [06:58] is it the final state of that PR, there are some pending comments [06:59] also I recall reading your report on ubuntu-devel, what's the state of the portal work, is it the last branch you expect in this area for now, is there anything buggy or missing from your pop? [06:59] pov [07:01] zyga: I was just updating it now, and running some of the spread tests locally (which failed due to a 503 error from the store) [07:01] ack, feel free to force push if it makes your life easier, just merge master to get the diff updated please :) [07:03] hmm. there's a new "checks" tab on pull requests [07:06] zyga: fwiw, I've been able to run the full file open/file save workflows using these snapd changes [07:06] and it's somewhat confusing, there's checks that are run on each PR listed just below the review statuses and a separate checks tab (github apps marketplace?) [07:06] jamesh: sweet! lots of lots of people will love that [07:07] jamesh: yes, it's a new GitHub feature apparently [07:08] checks sound like it could be useful for automated syntax checking and the like [07:09] or a compile test. [07:09] afk for a bit [07:09] The "status" feature is probably still what you want for test suites though [07:10] yes, but we may opt into the new thing [07:10] it seems travis now has a unified model where we could pay to get more test slots for FOSS projectsa [07:39] re [07:40] heh error: received an unexpected http response code (503) when trying to download https://api.snapcraft.io/api/v1/snaps/download/b8X2psL1ryVrPt5WEmpYiqfr5emixTd7_1797.snap [07:40] zyga: don't want to upset you.. but ^^ [07:40] sigh :) [07:40] well [07:40] weather is weather [07:40] and the store is the store [07:42] mborzecki: can you look at https://github.com/snapcore/snapd/pull/5142 again please [07:42] PR #5142: many: add "snap debug sandbox-features" and needed bits [07:42] I updated it and documented the behaviour on the forum [07:44] zyga: already reviewed it in the morning https://github.com/snapcore/snapd/pull/5142#pullrequestreview-118985554 [07:44] PR #5142: many: add "snap debug sandbox-features" and needed bits [07:44] oh, indeed, my tab was out of date [07:46] replied :) [07:49] hello [07:49] is there a way to fill the snap version at build time? [07:49] yes [07:49] version-script [07:50] AFAIR, google that there are examples around [07:50] version-script the keyword I was looking for, thanks zyga [07:50] ogra_: https://github.com/snapcore/pi3-gadget/pull/14/files [07:50] PR pi3-gadget#14: Update SPI slot definition [07:50] ogra_: can you please have a look at this [07:55] moin moin [07:57] hey hey [07:57] kjackal: if you're using git you can also pass 'git' as the version: field but it is slightly broken [07:59] thanks for the suggestion mcphail, I am packaging kubernetes so I would like the version in the snap to reflect the version of kubernetes binaries. I will take a look at git [08:01] kjackal: i hit this issue with underscores in the git tags, though: https://bugs.launchpad.net/snapcraft/+bug/1769519 [08:01] Bug #1769519: Setting the snapcraft.yaml version field to 'git' produces undocumented and broken magic [08:02] zyga: mvo is off today? [08:02] I think so [08:03] * zyga hears a lot of fire sirens [08:03] hmm [08:03] a *lot* [08:03] zyga: too early for warsaw uprising anniversary [08:03] feels like something major nearby [08:04] pedronis: zyga: is there anything useful we can do to not use a lot of cpu while not making progress when there's something wrong with the seed? [08:04] can you give me an example [08:04] maybe #1767896 [08:04] Bug #1767896: Live 18.04 with persistence snapd high CPU usage [08:05] grah! why does 18.04 not have keyboard shortcuts for menu items >:-( [08:05] * Chipaca is glad he hasn't upgraded every time he runs it in a vm [08:06] Chipaca: because gnome decided they are not desired or something? [08:08] ogra_: do you know how the live cd dance happens? asking because i want to tweak /etc/environment before snapd gets a chance to run [08:10] hmmmm [08:10] Chipaca: should we refresh/seed on live CD? [08:10] zyga: yep [08:10] zyga: some work in to making it work, evne [08:10] even [08:11] pedronis said that was not expected to work [08:11] but the daily livecd broke [08:11] * zyga is getting confused [08:11] still [08:11] ? [08:11] 2018-05-09T06:45:35-04:00 ERROR rename /var/lib/snapd/snaps/gtk-common-themes_3.snap.partial /var/lib/snapd/snaps/gtk-common-themes_3.snap: no such file or directory [08:11] this is the thing to debug [08:11] it totally is expected to work [08:11] zyga: no [08:11] I asked pedronis yesterday and he said no [08:11] i mean [08:11] zyga: the snap not being there is the cause of this bug, yes [08:12] zyga: snapd trying again and again and again is less than smart [08:12] and that's what burns up cpu [08:12] yeah, no disagreement there [08:12] it should slow down exponentially [08:12] OTOH, if you boot the live cd _with network_, by the ~2nd try it just works [08:12] because the download succeeds [08:12] somehow? i think [08:12] oh I saw that [08:12] I have a live USB (not persistent) of bionic [08:12] and I installed a snap [08:12] and it failed [08:12] then again [08:12] and it worked [08:12] no idea [08:16] Chipaca: zyga: I'm not here, but in budapest live cd people said they had space problems with refreshes, asked how to turn them off, explained about hold, if they didn't do it, not sure whay [08:16] ah, I see [08:16] thanks for the clarification [08:17] pedronis: I think they have done it, at least i understood they have and i see something like it in this daily [08:18] pedronis: now go away [08:19] Chipaca: https://github.com/snapcore/snapd/pull/5091#issuecomment-385631992 you mean to show the log only then, i.e. one week in since the last refresh? right now there's a message logged each time refresh is held by being on a metered connection [08:19] PR #5091: many: hold refresh when on metered connections [08:33] mborzecki: I meant “in a week's time we're going to update *no matter what* so get to some cafe or something you nitwit” [08:34] Chipaca: ah ok [08:52] mborzecki: also the // TODO use warnings for this [08:52] Chipaca: pushed some updates to 5091, please rereview [08:52] mborzecki: will make it easier to remember / find them :-) [08:52] mborzecki: looking [08:54] Chipaca: hah totally forgot about timeutil.Human ;) [08:54] mborzecki: :=D [08:54] uh, that was meant to be a smile [08:54] not … anything that isn't a smile [08:55] hm one thing though, i can't drop the time in timeutil.Human right? [08:56] not with that attitude [08:56] for that you need timeutil.Puritan ;) [08:56] mborzecki: you're right [08:56] mborzecki: maybe leave it for now and then add a flag to Human (or find a name for the new thing) [08:57] HumanDays [08:58] we need Klingon package for i18n [08:59] ;-) [09:00] * Chipaca half-smiled half-groaned [09:00] * Chipaca ~> moar coffee [09:02] * Chipaca ~> really moar coffee [09:04] the review board does not auto update? [09:05] mborzecki: no, or if so it does so very infrequently [09:09] mborzecki: gustavo mentioned it was stuck [09:11] * Chipaca might have his snark levels dialed up too much today [09:11] a user said something about a bug being a issue *we're* having, and I almost replied "oh, we aren't having it; you are" [09:12] heh, someone is in a mood today :) [09:12] but that could be me, I have a food poisoning since last night [09:12] some cream was added to a sauce I didn't know about [09:12] zyga: have you upset your wfie again [09:12] sucks to be me today [09:12] :-) [09:12] Chipaca: my mother in law [09:14] zyga: I flagged you about something in the forum btw, if you want to pick that up === chihchun_afk is now known as chihchun [09:29] the weather outside is so nice today [09:31] meh, no fonts in Qt apps again [09:33] zyga, you've asked fro SNAP_CONFINE_DEBUG=1 snap run ... when this happens, so here goes: https://gist.github.com/thresheek/05e0343b19cb1e16b745ebf6162d6599 [09:33] also, good morning === chihchun is now known as chihchun_afk [09:59] and again, it's fine after a reboot [10:03] thresh: `/usr/lib/snapd/snap-discard-ns ` should be enough [10:04] mborzecki, ok, I'll try next time this happens. What could be the cause? [10:07] thresh: maybe it's something with how the fonts are made available to snaps [10:11] thresh: do you have an nvidia prime thing? [10:12] (no, it's not a subscription service from nvidia to get free shipping) [10:12] (and no it's not condoms either) [10:13] Chipaca, yes I do [10:13] bumblbee or whatever [10:13] thresh: bumblebee was the old thing that supported per-window, right? [10:13] anyway, that confuses snapd [10:13] possibly [10:14] every time i switch from one to the other i need to snap-discard-ns [10:15] not sure if the fix is to tell snapd about this, or to tell whatever it is that does the switching about snapd :-) [10:20] Moin [10:21] who's the publisher of the gtk-common-themes snap? [10:21] i mean, who are the devs [10:22] zyga: is there any concern still about #5075 ? [10:22] PR #5075: snap/env: fix env duplication logic [10:22] hello niemeyer! [10:22] popey: do you know who pushes gtk-common-themes? [10:23] pstolowski: Yo o/ [10:23] Chipaca: desktop team own that. kenvandine probably [10:23] kenvandine: O HI [10:23] 6am for him, probably too early [10:24] popey: thanks [10:42] sergiusens: https://bugs.launchpad.net/snapcraft/+bug/1770364 is the issue I was struggling with yesterday [10:42] Bug #1770364: Using 2 remote parts, file not being copied from stage to prime [11:09] PR snapd#5124 closed: many: add wait command and seeded target [11:18] cprov: ping for when you're around [11:22] niemeyer: I am in a dentist appt. and will be back in 30m. Urgent ? [11:24] Chipaca, to change livecd defaults you wantz ro change casper and re-build the image [11:25] *want to [11:27] ogra_: I want many things, but that is not one of them [11:27] ogra_: :-) [11:28] Chipaca, well, thats the only sane way to change /etc/environment that i know [11:28] ogra_: then i guess i won't [11:28] casper creates all defaults in the livefs overlay [11:30] ogra_: it was to debug a bug that i've now marked as won't fix, so ¯\_(ツ)_/¯ [11:30] (alternative you could hack it in on the fly ... by adding "break=bottom" to the kernel cmdline, injecting it into the "nearly booted" system from the initrd shell and then move on with the boot by exiting the shell [11:30] ) [11:30] ah, k [11:37] * zyga is unsuited for work today :/ [11:37] is there a way to set environment variable for an installed snap, maybe using snapctl ? [11:48] * zyga files a sick day, sorry folks, I cannot help it today [11:49] om26er: no [11:58] for testing you can copy the wrapper script to ~/ ... edit it and bind-mount it over the one in the snap ... [11:58] but thats indeed no permanent solution or in any way sane :) [12:02] niemeyer: ping, I am back home [12:12] om26er: what are you needing to do? [12:13] Chipaca: my snap requires an environment variable exported for a specific feature to work, though I opted for a system-wide export and that did the trick. [12:13] om26er: it sounds like your snap should be using a config hook :-) [12:13] So not blocked on anything, just wanted to know if that was possible [12:13] I'll look into that [12:13] om26er: is is a fixed flag? [12:14] ogra_: like FOO_ENABLE=1 [12:14] Chipaca, that's my team, jamesh most specifically [12:14] zyga: yeah [12:14] er om26er ^ [12:14] kenvandine: O HI [12:14] kenvandine: :-) [12:14] om26er: can you juse add that to environment: in snapcraft.yaml? [12:14] Chipaca, but didrocks does the regular pushes [12:14] kenvandine: #1767896 is why i was wondering [12:14] Bug #1767896: Live 18.04 with broken seed causes snapd high CPU usage [12:14] zyga: don't want to make it permanent. On some systems exporting that variable is not needed, on other it is. [12:15] zyga, or CRAZY_ENABLE=1, yeah :) [12:15] is it harmful on the first group? [12:15] kenvandine: daily live cd broken because one of the snaps now depends on something that isn't seeded, and can't be seeded because it's not in the channel [12:15] if it is a configuration then yes, use configure hook [12:15] Chipaca, ugh [12:15] zyga: its untested, so can't really say. Would probably need a wide testing [12:15] if it can be globally enabled then just do so [12:15] Chipaca, i'll open and close the branch [12:16] So it basically tells Android Studio to use system-wide libstd instead of the one that's shipped with its source. [12:19] om26er: you could always have a wrapper check for e.g. [ -e $SNAP_USER_DATA/.do_the_crazy_thing ] [12:20] kenvandine: will you update the bug when you do? [12:22] Chipaca, done and done [12:25] kenvandine: awe, and some [12:26] :) [12:35] PR snapd#5117 closed: interfaces/apparmor: enable apparmor, even if partial [12:37] zyga: there are conflicts on #5090 [12:37] PR #5090: cmd/snap-update-ns: poke holes when creating source paths for layouts [12:37] oh [12:38] thanks === devil is now known as Guest68883 [13:25] PR snapd#5145 closed: boot: clear "snap_mode" when needed [14:12] kenvandine: hey, got a sec to look at https://github.com/ubuntu/snapcraft-desktop-helpers/pull/115 ? [14:12] PR ubuntu/snapcraft-desktop-helpers#115: Fix FONTCONFIG_PATH to point to where we generate fonts.conf [14:12] niemeyer: https://lists.fedoraproject.org/archives/list/server@lists.fedoraproject.org/message/XWJPNR2BIT6ZTA5M7OKSW7U6GY5EKQWH/ [14:14] greyback, saw that, it's in my queue to look at today [14:14] kenvandine: ack, thanks! [14:28] o/ [14:34] Wimpress: Hey! Who owns the name 'axel' on the snap store ? Can we reclaim that ? [14:34] * Wimpress goes looking... [14:34] I have proposed packaging upstream but they seem to be on a long vacation or something ;) [14:35] https://github.com/axel-download-accelerator/axel/pull/156 [14:35] PR axel-download-accelerator/axel#156: Add snap packaging [14:37] elopio: Hey! [14:38] elopio: do you know what will it take to get the Keybase Desktop app as snap ? The packaging you proposed upstream looks to be the CLI client. [14:38] om26er: That name is not taken as far as I can tell [14:39] Wimpress: hmm, says "The name 'axel' is already taken." [14:40] sparkiegeek: ^ help. why does that snap not show in search but snapcraft thinks it's registered? [14:52] zyga: ping [14:54] Yes? [14:55] niemeyer: pong [14:55] zyga: Hey [14:59] Wimpress: is it fine to create a new repository under snapcrafters ? [14:59] apparently I can click the 'New' button [15:17] om26er: Do you want to repo to be called `axel`? [15:17] Wimpress: yes. [15:18] I can create it and you can submit a PR :-) [15:18] sure [15:26] om26er: https://github.com/snapcrafters/axel [15:30] Pharaoh_Atem: matthew miller supports fedora base snap :) [15:55] om26er: https://news.ycombinator.com/item?id=17038734 - thank you! [15:56] \o/ [15:56] :) [15:56] Thought you might like that :) [15:57] popey: Definitely great to know and fyi android studio neared 10k installs :) [15:59] Nice! [15:59] One of them is me :) [16:00] Its exciting how snaps are fixing real issues, making my life a breeze every day. [16:00] That's great to hear. Looking forward to the next developer tool you need to snap ;) [16:02] I have my eyes on https://hub.github.com/ === pstolowski is now known as pstolowski|afk [16:24] Hey niemeyer are you planning on updating blender to 2.79b? [16:27] kyrofa: Yeah, definitely, and also need to update 2.8 === blackboxsw is now known as blackboxs === blackboxs is now known as blackboxsw === prime is now known as Guest32215 [17:17] PR snapd#5148 opened: boot: clear "snap_mode" when needed (2.32) [17:21] PR snapd#5149 opened: many: add wait command and seeded target [17:25] PR snapd#5150 opened: snapd.core-fixup.sh: add workaround for corrupted uboot.env [19:10] Wimpress: https://github.com/snapcrafters/axel/pull/1 [19:10] PR snapcrafters/axel#1: Add initial packaging [19:11] So I think you will need to register/own that name first to be actually able to publish that to the store. [19:38] om26er: i am trying, getting a store problem. [20:05] om26er: ok, axel is now registered to snapcrafters [20:41] PR snapd#5151 opened: tests: use journalctl cursors instead rotating logs [20:54] popey: lets merge that PR and enable builds for it :) [21:02] I can't enable builds yet. https://github.com/canonical-websites/build.snapcraft.io/issues/1091 [21:09] FYI https://forum.snapcraft.io/t/new-cdn-for-snap-downloads/5363 - new CDN online, 50% mix at the moment, please post on that forum thread is you see any download issues [22:32] PR snapcraft#2129 opened: Swap FROM in Dockerfiles from 'xenial' to 'bionic'