[00:53] PR # closed: snapcraft#1649, snapcraft#1720, snapcraft#1875, snapcraft#1905, snapcraft#2020, snapcraft#2135, snapcraft#2176, snapcraft#2229, snapcraft#2239, snapcraft#2289, snapcraft#2354, snapcraft#2359, snapcraft#2369 [05:11] morning [06:20] good morning [06:20] mborzecki: hey :) [06:20] zyga: hey hey [06:20] I walked my daughter to school today [06:20] it's so misty I was afraid of letting her go by herself [06:20] how are you doing? [06:20] and it's cold too [06:21] glad i swapped the tyres in the car on friday [06:22] yeah, scarfs, hand gloves and caps [06:25] * zyga reads some news about the election yesterday [06:25] then back to patches [06:28] zyga: heh, elections, the situation in lodz is 'stable' :) [06:37] mborzecki: so nothing changed? [06:38] zyga: yeah, exit polls suggest the same president as before with 70+% votes [06:40] I signed up to Apple Music yesterday as a trial to show my dad how it works [06:40] I'm in love :) it's fantastic [06:41] zyga: btw. https://status.github.com/messages [06:41] DRM free music, everything in iTunes seems to be available, 19zł a month [06:41] oh [06:41] github is down? I'm reading reviews now and it's ok so far [06:41] woah [06:41] data corruption [06:41] well [06:41] monday :) [06:42] zyga: appl music? basically like spotify? [06:42] no [06:43] it's like iTunes (you buy music, it's DRM free on your disk) [06:43] all you can eat [06:43] you can stream but you can also just download and keep offline forever [06:43] I usually spend 20-50 a month per music [06:43] so this is much cheaper [06:44] there's also a family plan but I didn't look at that yet, for up to six people [06:44] zyga: download as some format and keep forever elsewhere or are you forced to use the app? [06:44] it's just mp4 [06:44] normal format [06:44] one sec [06:56] zyga: heh, have you tried doing reviews today? [06:56] Good morning everyone [06:56] well, I'm trying now [06:56] hey dot-tobias :-) [06:56] mborzecki: what are you observing? [06:56] zyga: ui getting stuc waiting for response when adding comemnts [07:00] zyga: and got a unicorn erorr page :P [07:00] haha and the review was not sent [07:00] omg [07:02] so much for reviews [07:06] re [07:06] that's not encouraging :/ [07:08] quite possibly a dumb question, but … can I assume that the “last modified” date for edge images (http://cdimage.ubuntu.com/ubuntu-core/16/edge/current/) can be completely ignored and the build is in fact “daily”? [07:08] hmmm, good question dot-tobias [07:08] it looks fishy to me [07:08] I think it's not the latest one [07:09] morning [07:10] pstolowski: hey [07:50] PR snapcraft#2378 opened: Release chagelog for 3.0 [07:51] mo'in [07:51] oi [07:52] mborzecki: https://pastebin.ubuntu.com/p/B6QY2h3gwB/ [07:52] Chipaca: nice [07:53] I wish we had a "can run things from such-n-such snap" interface [07:53] PR snapcraft#2378 closed: Release chagelog for 3.0 [07:53] then taht wouldn't need classic :-) [07:53] Chipaca: we could [07:53] we have the capability [07:53] mborzecki: if you want to push it feel free; otherwise i could [07:54] zyga: it's like a content interface but different [07:54] haha [07:54] yes [07:54] not content ;) [07:55] it'd be an interface where we'd have a set of snap commands that a given snap can run [07:55] enforced at snap-confine level, I suspect [07:55] mborzecki: otoh we could also package shellcheck into the snap ourselves [07:55] mborzecki: or offer spread-shellcheck upstream [07:55] mborzecki: the latter needs a bit of paperwork [07:56] Chipaca: shellcheck gets frequent updates, at least here on arch, it's probably upstream pushing new changes [07:58] on a completely unrelated subject: WTH https://i.imgur.com/YcjvITL.png [07:59] Chipaca: promoted! [07:59] WAT [07:59] exactly [07:59] somebody paid money to do that [07:59] * zyga puts on pants [07:59] must be important [08:01] zyga: maybe it's a US/UK thing [08:01] Chipaca: no pants? [08:01] mborzecki: https://en.wiktionary.org/wiki/pants [08:01] mborzecki: look at the pics [08:05] PR snapcraft#2378 opened: Release chagelog for 3.0 [08:06] hmmm [08:06] github is indeed wonky [08:06] oh, nice, github's databases are eventually inconsistent right now [08:06] chipaca's critical PR is marked as approved [08:06] but if you click on it, it is not, no comment history [08:07] https://mobile.twitter.com/TheRegister/status/1054267789232431104 [08:07] the pic is very representative of IS work [08:07] not [08:08] Chipaca: did you get my review of https://github.com/snapcore/snapd/pull/6026 ? when i go to the PR there's like 3 reviews pending (?) [08:08] well, it's the reg [08:08] PR #6026: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [08:08] mborzecki: i got yours [08:08] mborzecki: nobody else's yet [08:08] PR snapcraft#2378 closed: Release chagelog for 3.0 [08:08] Chipaca: well i got a unicorn in return :P [08:08] mborzecki: but as above, github's hard drive is full [08:09] unicorns [08:09] unicorns everywhere [08:10] zyga: https://www.youtube.com/watch?v=DX1iplQQJTo&feature=youtu.be&t=90 [08:11] https://imgflip.com/i/2knjpg [08:12] I guess the windows upgrade went well at GitHub today [08:24] GH reviews are still in read only mode [08:25] zyga: at least one got through :P [08:26] oh? I keep try to send a trivial comment to see what happens [08:33] haha got zyga's review email, but the reviews are not visible https://github.com/snapcore/snapd/pull/6026 [08:33] PR #6026: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [08:34] well I got mborzecki 's review twice [08:34] so there [08:34] hahah [08:34] zyga: and i just got a review from you as well [08:34] oh and in zyga's review the comment was doubled [08:34] yes [08:35] zyga: so, depends how you look at it :-) that was the thing that triggered the panic, so that's the immediate cause [08:35] zyga: but before that, cmdTry was different from other snap ops in that it didn't pass the error to errorToCmdMessage [08:35] https://blog.github.com/2018-10-21-october21-incident-report/ [08:35] zyga: so that's the deeper cause [08:36] zyga: because there _is_ an opts for this error, it's just not in main [08:36] zyga: which one of those causes is the crux, i guess it depends on how you look at it? [08:44] mmm, I see [08:45] I made exceptionally hot coffee [08:45] my hands are freezing here [08:49] * Chipaca hands zyga a "crux of the matter" trophy [08:49] (just got another few emails) [08:49] hahah [08:49] it's still being published? [08:50] I see them now [08:51] zyga: github liked you comment so much it keep on sending it ;) [08:51] haha [08:51] eventually it will stop ;-) [08:51] haha [09:22] is it just me or are none of the 2 reviews posted in https://github.com/snapcore/snapd/pull/6026 visible? [09:22] PR #6026: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [09:32] zyga: hey! Did you have a moment to check the dragonboard image? [09:32] yes! [09:32] let's do it :) [09:32] is the one from last week good? [09:32] or shall I pick lastest [09:33] http://cdimage.ubuntu.com/ubuntu-core/18/pending/ubuntu-core-18-arm64+snapdragon.img.xz <- latest is always the best ;) [09:33] Oh, no mvo today? [09:34] sil2100: no, sprinting [09:34] sil2100: flashing [09:34] zyga: do you maybe know when's the ETA for a stable-channel release of snapd? [09:34] no idea [09:34] maybe mid sprint? [09:35] but really no idea [09:36] sil2100: are you talking about 2.36? [09:37] Chipaca: I'm talking about the snapd snap [09:37] Chipaca: we don't have any version of it in stable from what I can see at least [09:37] ah [09:37] ok :-) [09:37] no idea [10:01] Chipaca: shall we try posting the reviews again? [10:02] mborzecki: zyga: please [10:02] * Chipaca afk for a bit [10:02] yep [10:02] bah [10:02] 10:47 British Summer TimeWe continue to monitor restores which are taking longer than anticipated. We estimate they will be caught up in an hour and a half. [10:02] so maybe let's check after lunch [10:02] zyga: mborzecki ^ [10:03] (that was 15 minutes ago) [10:03] too late, already clicked :P [10:03] aand unicorn! [10:09] unicorn days [10:10] I tried to fork a repo just now [10:10] 404 and errors [10:15] sil2100: it works [10:16] https://www.irccloud.com/pastebin/bPMblg1d/ [10:16] dragon board is operational :) [10:16] 11:47 CESTWe continue to monitor restores which are taking longer than anticipated. We estimate they will be caught up in an hour and a half. [10:16] from status.github.com [10:17] Issue # closed: classic-snap#7, classic-snap#8, classic-snap#14, classic-snap#15 [10:17] PR # closed: classic-snap#24, classic-snap#27, classic-snap#28, classic-snap#30 [10:19] Let's SHIP IT [10:20] PR core-build#11 closed: remove cruft from the writable-paths [10:20] PR core-build#22 closed: unit testing for sync_dir() [10:20] PR core-build#26 closed: move most of the customization into the core snap build [10:20] Issue # opened: classic-snap#7, classic-snap#8, classic-snap#14, classic-snap#15 [10:20] PR # opened: classic-snap#24, classic-snap#27, classic-snap#28, classic-snap#30 [10:21] PR core-build#11 opened: remove cruft from the writable-paths [10:21] PR core-build#22 opened: unit testing for sync_dir() [10:21] PR core-build#26 opened: move most of the customization into the core snap build [10:21] everything is broken :) [10:22] apparently ceph fails universally ;) [10:23] Chipaca finally figured Friday's issue greyback hit [10:23] PR # closed: snapcraft#1649, snapcraft#1720, snapcraft#1875, snapcraft#1905, snapcraft#2020, snapcraft#2135, snapcraft#2176, snapcraft#2229, snapcraft#2239, snapcraft#2289, snapcraft#2354, snapcraft#2359, snapcraft#2369 [10:24] pstolowski: oh nice! [10:24] pstolowski: what was it? [10:25] mborzecki: a commit i introduced on sep 7, to conflict & retry on discard-snap. I missed the fact that discard-snap is created sometimes by install as well to remove inactive revision. so, auto-refresh conflicts with itself [10:26] mborzecki: in this case it tries to refresh lxd and also schedules removal of old inactive revision of lxd [10:26] pstolowski: nice [10:26] PR # opened: snapcraft#1649, snapcraft#1720, snapcraft#1875, snapcraft#1905, snapcraft#2020, snapcraft#2135, snapcraft#2176, snapcraft#2229, snapcraft#2239, snapcraft#2289, snapcraft#2354, snapcraft#2359, snapcraft#2369 [10:27] pstolowski: sounds like a scenario we'd like to have a test for [10:27] mborzecki: yes, definately. i added a test for conflict, but no we have a case where we shouldn't conflict [10:27] *now we have [10:28] great [10:48] sil2100: there are some wifi errors on screen after a while [10:49] Oct 22 10:40:55 localhost kernel: wcn36xx: ERROR hal_remove_bsskey response failed err=6 [10:59] zyga: I think we need to poke the kernel team about those [11:00] where shall I report those? [11:06] any idea if it's possible to do spread -repeat= but directly in task.yaml? [11:10] heh after/before test is fiddly [11:18] mborzecki: no idea [11:18] count: 10? [11:21] zyga: doesn't appear to be anything like that [11:44] :/ [11:44] pstolowski: sounds like another critical 2.36 pr [11:44] how's our favourite unicorn factory? [11:44] Chipaca: sure, i'm in touch with mvo on that [11:44] and fix is in progress [11:45] pstolowski: 👍 [11:45] or maybe 🖒 [11:45] ¯\_(ツ)_/¯ [11:47] can i get 2nd reviews on https://github.com/snapcore/snapd/pull/6023 ? [11:47] PR #6023: overlord/snapstate, snap, wrappers: start services in the right order during install [11:50] off to pick up the kids [11:57] FYI, latest GitHub status [11:57] The majority of restore processes have completed. We anticipate all data stores will be fully consistent within the next hour. [11:57] I just pushed a patch without erorrs [12:01] I take that back [12:01] comments I added just disappeared [12:06] comment showed up and disappeared again [12:06] fun [12:11] zyga: were any of your reviews of 6026 a +1? [12:11] Chipaca: let me look [12:12] zyga: all I see are a bunch of "crux of the issue" comments :-) [12:12] was wondering if I should merge it [12:12] let me try to give it a +1 [12:12] my changes elsewhere on GH just go away after I reload [12:12] Hi, is there some interface that allows me to change the scaling_governor on the core? [12:13] sborovkov: let me look [12:13] Chipaca: green, merge it [12:13] zyga: ooh i see your +1 [12:13] sborovkov: no, I'm afraid not [12:13] would that be cpu_control, or do we need power_control? [12:14] zyga, is it something that's possible to add? basically we are running on rpis. we want to change the scaling governor to performance. Otherwise while it winds up video performance goes down [12:14] cpu-control sounds about right [12:14] and it goes down by a lot [12:14] yeah, I agree with ogra [12:14] PR snapd#6026 closed: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [12:14] feels like a reasonable tweak [12:14] sborovkov, thats weird, the ondemand governor sshould surely cope [12:15] PR snapd#6026 opened: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [12:15] ogra, it does work properly after some time. But when we have a fresh run - the times it takes us to render a frame goes upo from 32 to 64 ms for first 20 seconds and so [12:15] ogra: it takes about a second to ramp things up though, doesn't it? [12:15] (while we should add thiss to the interface, you should definitely not see much differenec in system behaviour when switching governors ... just that it doesnt clock down after it is done with processing) [12:15] that looks so bad on the customer screens [12:15] oh wow that's more than i'd expect [12:15] basically the timings were 2 times worse [12:16] once we enabled performance governor we never saw anything like that [12:16] sborovkov: can you add this to your snap [12:16] wow, is that with accelerated graphics or plain framebuffer? [12:16] and show me the --devmode denial [12:16] I can make a modification to the interface [12:16] PR snapd#6026 closed: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [12:16] i definitely dont see that with the kms driver [12:17] aaaaah now my pr is half-merged [12:17] non-kms indeed means you fully utilize the CPU and completely leave the GPU alone [12:17] Chipaca: what? [12:17] zyga, github fun today :) [12:17] wow, [12:17] yeah [12:17] I think today is not a good GH day [12:17] i actually learned their 404 pick is interactive ! [12:17] *pic [12:18] i never stayed long enough on it to even notice [12:18] but if you wiggle the mouse it makes nice layer shifting effects [12:18] PR snapd#6026 opened: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [12:18] alright I will need to build a devmode snap though or try it from some other one guys, I will write back when I do that. ogra what do you mean accelerated graphics or plain framebuffer? we do use h/w decoder. and yeah showing stuff with qt's qml, so it's using framebuffer [12:19] PR snapd#6026 closed: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [12:20] sborovkov, i mean the kms drm driver when you set "dtoverlay=vc4-kms-v3d" in config.txt (this turns on accelerated GLES ) [12:21] that's not enabled at the moment [12:21] it was breaking our qt app previously on the core [12:21] I should try to see if it works now actually [12:22] well, probably not without a compositor ... not sure [12:23] * ogra curses about this silly x11 restriction that forces you to add a .desktop file for a kiok app [12:23] *kiosk [12:23] re [12:23] * zyga turned on the heater [12:26] PR snapd#6026 opened: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [12:27] PR snapd#6026 closed: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [12:28] PR snapd#6026 opened: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [12:28] * Chipaca hugs mup [12:29] Pushing stuff to GH feels pointless now [12:29] the patches show up [12:29] the diff stays stale [12:29] PR snapd#6026 closed: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [12:29] zyga: let's move back to launchpad [12:29] mborzecki: anyway, I pushed trivial fixes to https://github.com/snapcore/snapd/pull/6010 [12:29] PR #6010: cmd/snap-discard-ns: add support for per-user mount namespaces [12:32] zyga: … or did you? [12:35] heh gh is the new single point of failure of modern software dev [12:36] pretty much that what we try to avoid when designing stuff ;) [12:37] mborzecki: wait until Microsoft moves GitHub to azure ;-) [12:38] PR snapd#6026 opened: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [12:38] 13h+ of outage [12:39] PR snapd#6026 closed: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [12:41] PR snapd#6026 opened: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [12:42] PR snapd#6026 closed: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [12:43] PR snapd#6026 opened: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [12:44] so does it make any sense to push stuff now or not? [12:47] PR snapd#6026 closed: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [12:49] mborzecki: probably not [12:49] PR snapd#6026 opened: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [12:49] ehh [12:49] damn [12:50] hmm 2018-10-22 14:50:23 Cannot allocate google:ubuntu-18.10-64: cannot find any Google image matching "ubuntu-os-cloud-devel/daily-ubuntu-1810-cosmic-v20181002" on project "ubuntu-os-cloud-devel" [12:51] PR snapd#6026 closed: cmd/snap: try not to panic on error from "snap try" <⚠ Critical> [12:51] mborzecki: get master [12:51] mborzecki: that was fixed by the cachio late friday [12:51] ah ok [12:51] #6024 [12:51] PR #6024: tests: new cosmic image for spread tests on gce [12:52] you'd call that "saturday" [12:52] :-) [12:59] PR snapd#6026 opened: cmd/snap: try not to panic on error from "snap try" <⚠ Critical>