[04:58] morning [04:58] Good morning [04:59] Cold night [05:06] zyga: yeah, and hoar frost everywhere :/ [05:55] hm guess the store is offline atm [05:55] hope it's not as bad as it sounds [05:56] mborzecki: yeah, the store is offline right now, and we're working to get it back online [05:56] as always, status.snapcraft.io [05:58] sparkiegeek: yeah, read the note about hardware failure, ouch [06:29] good luck sparkiegeek [06:30] zyga: thanks! [06:33] mborzecki: I sent some patches about facts API yesterday [06:34] zyga: ack, i'm looking at libsnap-confine-private parts PR [06:34] thanks! [06:56] hey mvo [07:00] zyga: hey, good morning [07:06] morning! [07:12] pstolowski: mvo: hey [07:14] hey pstolowski and mborzecki [07:26] silly question, but how do you revert a revert? [07:28] mborzecki: refresh? [07:29] pstolowski: that would poke the store which is currently down :) [07:30] mborzecki: we're now stumbling back to life; trying to fend off the advances from all the snapds in the world :) [07:52] * zyga does reviews [07:52] mborzecki: you can refresh to a revision you have [07:52] I think that is even offline [07:53] pstolowski: doing hot plug helpers now [07:54] zyga: awesome, thanks! [08:05] Dang, store is dooown [08:12] uh indeed [08:18] Store read operations are on their way back up [08:18] PR snapcraft#2296 opened: part grammar processor: lazily capture attributes from plugin [08:18] Just will be sluggish for a while as all the snapds retry. [08:18] Write operations are a while off yet [08:19] wgrant, that's alright, read ops will get my tests passing again [08:19] wgrant, sounds like you had a fun night [08:19] Er. Day. Never can tell with you [08:19] :P [08:19] For once something actually went wrong during my day [08:19] I think it's the first time I've ever woken someone up rather than being woken up [08:19] Oh how lovely that must have been! [08:20] * sparkiegeek is now a full pot of tea in to his woken-from-slumber [08:20] kyrofa: woah, you're up early [08:20] zyga, I'm in spain for a week [08:20] .... [08:20] zyga.eyesfulloftears [08:21] have fun :) [08:21] in BCN? [08:22] zyga, a bit outside of madrid, a village called manzanares el real. I think I may come back next year with the family for a few months. I love this place [08:22] zyga, roscon is in madrid this weekend, so it didn't make sense for me to fly over the atlantic multiple times [08:23] woah, even smaller than the town I was in [08:23] lovely :) [08:23] zyga, yeah I was specifically looking for that [08:23] specifically looking to find where zyga was? [08:23] zyga: I think you might have a stalker [08:23] Yes. Don't be jealous [08:23] that's all right, I love kyrofa :) [08:23] Yeah, see? [08:23] Back at you zyga [08:24] No I just wanted to experience small town spain, grocery shopping, actually living here for a little while. It's been lovely [08:24] kyrofa: careful [08:25] the next thing you say is "oh my god, how many years has it been already?" - except you say that in Spanish to your little Spanish speaking kids :D [08:25] (but then it's all right because it's the best place to live) [08:25] Hahahaha [08:26] zyga, yeah if we come here for-- I dunno, three months-- maybe my spanish will flourish [08:26] kyrofa: when I stayed for 6 years I was only planning on ... max ... 6 months [08:26] Ha! [08:26] zyga, explain something to me though, I feel rude asking anyone here [08:27] Let me upload a picture first to help my question [08:28] ruh-roh [08:28] haha :-) [08:28] right? [08:30] PR snapcraft#2297 opened: [legacy] part grammar processor: lazily capture attributes from plugin [08:31] zyga, alright, got it. This town is small, quiet, and seemingly perfectly safe. Kids out on the street, etc. However, I swear every property resembles this one: https://pasteboard.co/HFEc6pt.jpg [08:31] hmmm [08:31] as in $$$ or as in fort knox? [08:31] Everyone's yard is behind a serious wall with a gate [08:32] not sure if that's the same reason [08:32] Nah, not all the properties look quite as nice [08:32] on costa brava there were lots of multi-million apartments and houses [08:32] for rich sport stars, actors and business men [08:32] typically behind a tall fence [08:32] so maybe that's that [08:32] just real madrid [08:33] but I don't know for sure [08:33] just ask though [08:33] I'm sure all the locals know [08:33] ask in a grocery store [08:33] I dunno, I need to take a pic of the neighborhood I'm in. It really just feels architectural, but there must be some interesting history behind it [08:33] kyrofa: one thing to note is that Spain forgot how to make houses [08:34] kyrofa: during franco, everything was apartments [08:34] kyrofa: do you speak Spanish? [08:34] * zyga realised his earlier question about $$$ _or_ Fort Knox was a bit silly [08:34] zyga, I speak enough spanish to explain that I don't speak it [08:34] haha [08:34] well, I hope you do better than I did [08:34] yeah, I'd really just ask [08:34] then translate the answer back [08:35] (tip: kids help) [08:38] hmm, new Vmware ! [08:39] shiny goodies [08:42] zyga: looking forward to a deal/sale (black friday maybe), let me know if you spot any [08:42] sergiusens: hey, but snapcraft changed the default pwd when at build stage? [08:42] pstolowski: yeah, same here [08:42] workstation 15 feels like the bigger win now [08:42] highdpi fixes and what not [08:43] fusion 11 is just "keep on rolling" from what it seems [08:43] but I haven't played with either [08:43] pstolowski: I _barely_ missed the free month update :/ [08:43] I got workstation 14 recently [08:43] er [08:43] zyga: ouch... [08:43] not barely missed but barely got :/ [08:43] (ie. no update) [08:43] very solid product still [08:45] pstolowski: typically Vmware desktop products are 25-30% off nearly all year [08:45] with peak around Black Friday [08:45] Store read APIs should be stable now [08:46] thank you wgrant! must have been a busy morning [08:46] pstolowski: so my best advice is to just wait [08:46] Trevinho, what do you mean? [08:47] zyga: yep. the standard price doesn't justify moving away from virtualbox for me, to steep [08:47] *too [08:47] pstolowski: yeah, that's understandable [08:47] pstolowski: once you buy it once you can refresh it every few releases [08:48] you can get the lower cost update to 15 from all the way back to 12 [08:50] pstolowski: you can also buy it without VAT [08:50] so that's a nice change [08:53] zyga: do you remember the kernel selftest? we have a bit of a challenge to make it work with our testsuite [08:53] yes [08:53] oh? tell me more please [08:53] zyga: when enabling snapd will go into read-only mode on trusty [08:53] zyga: until we reboot [08:53] mmm, yeah [08:53] zyga: we can not reboot in prepare [08:53] zyga: if vmware products are 25-30% off all year, why don't they just drop the price? :) [08:53] zyga: but we do install some snaps in prepare [08:54] zyga: so its a bit of a catch-22 [08:54] mborzecki: marketing, magic [08:54] zyga: :/ [08:54] mvo: can we just make the 14.04 machine prepared with the right kernel? [08:54] zyga: also "funny" because while we need a newer kernel apparently things worked just fine with the stock 14.04 for most of our tests [08:54] mvo: or is that unreasonable for testing? [08:54] zyga: yeah, I think we need to do that [08:54] stock 14.04 kernel doesn't work at all [08:54] zyga: its slightly annoying for people who want to run local spread tests [08:54] are you sure? [08:54] zyga: it does not? [08:55] I'm pretty sure not much works on 3.10 there [08:55] zyga: let me dive deeper, so far I only looked at logs, let me login [08:55] last time I tried apparmor was not playing [08:55] and you would get issues on any operation [08:55] maybe mounting worked but not much more [08:55] mvo: also complex, maybe we should allow re-exec at least [08:55] but ... not sure [08:55] k [08:55] zyga: looking now [08:56] zyga: maybe our version check is just busted [08:56] zyga: anyway, another fun morning puzzle [09:03] Trevinho: when in pull, to changes to the part's source, when in build to the part's build dir, when in stage to the stage dir and when in prime to the prime dir. It has always been like this since introducing the override- directive [09:04] sergiusens: weird since the tdesktop stopped building for some weeks now as it wasn't find a file it was before in build (with a cp) [09:06] Trevinho can I see the snapcraft.yaml? [09:06] And how did it break? [09:07] sergiusens: failing here https://github.com/telegramdesktop/tdesktop/blob/dev/snap/snapcraft.yaml#L97 [09:07] sergiusens: log with some more debugging https://launchpadlibrarian.net/390331135/buildlog_snap_ubuntu_xenial_amd64_tdesktop_BUILDING.txt.gz [09:07] (for sure it has always been building) [09:07] but I'm fixing the path, it's indeed better to point to the actual src, but... [09:08] just curious if something changed [09:09] mborzecki: updated https://github.com/snapcore/snapd/pull/5805 [09:09] PR #5805: cmd/snap-update-ns: enforce trespassing checks [09:09] Trevinho, are you sure there's a snap/gui? https://github.com/telegramdesktop/tdesktop/tree/dev/snap [09:09] Trevinho your problem lies two lines above.. That relative path worked through an implementation detail [09:10] Oh, you make it [09:10] Blech [09:10] yeah, that's why I whished for years to do a SNAPCRAFT_PART_SOURCE env :) [09:11] but too in hurry to do a pr for it [09:11] zyga: lgtm, thanks! [09:12] zyga: ok, fun - the kernel test is busted, debugging now [09:13] zyga: so all good, much less of an issue than I expected [09:14] Trevinho, SNAPCRAFT_PART_SRC is a thing, but it wouldn't help you in this case, the snap/ dir isn't considered part of the src [09:15] yeah sure, but to pull from lib/xdg/* [09:16] sergiusens: another thing i wanted to ask, is there something going on for some caching of parts? As in the past I was trying to use travis cache for parts which didn't change, and would be nice to have a mode where snapcraft tries to reuse the parts if they're updates, otherwise it auto-cleans and pull/build/stage the ones which need to. As right now it gives an error message about what to do, but having something that forces it to clean... [09:17] mvo: oh, version compare wrong? [09:18] zyga: yeah, kernel version is "4.4.0-112-generic" - two "-" are illegal [09:19] oh [09:19] zyga: anyway, fix is easy [09:19] this upsets the compare logic? [09:19] zyga: yeah, its using the debian version compare rules [09:19] zyga: which give "-" a strict meaning [09:19] zyga: but no worries, fix is simple [09:22] Mornings [09:22] hey hey [09:22] feeling better? [09:22] niemeyer: good morning! [09:23] niemeyer: oh, I missed that you felt unwell? a cold as well? it seems to be going around :( [09:24] niemeyer: o/ [09:24] Trevinho, not completely clear what you're asking, but perhaps you'd be interested in the "lifecycle" section of these release notes? https://github.com/snapcore/snapcraft/releases/tag/2.43 [09:26] brb, getting more warm tea [09:38] kyrofa: oh, yes, that's the one! [09:38] kyrofa: that works also when doing `snapcraft prime ` and then if some dep on that needs cleanup does it? [09:40] * zyga shivers [09:41] Trevinho, yes, but it's not magic. It will catch changes made to the files for local parts, but not remote parts. It will clean/re-run steps that you change (e.g. if you alter override-build it knows build needs to run again) [09:41] It's also pretty new, we're still tweaking it, but would love for you to give it some mileage [09:42] zyga: I updated 5768, it needs a second review but that should be trivial :) [09:42] ack, I'll look in a sec [09:43] zyga: thanks and no rush, still need to wait for tests [09:46] mvo: Sorry, missed your question [09:47] mvo: No, I just felt a bit brainwashed after last week :) [09:57] mborzecki: updated, thanks [09:57] mvo: looking now [09:58] mvo: whee, [09:58] something for 2.36? [10:07] unexpected benefit from having macOS tests in travis [10:07] no stupid log by default [10:08] (the one that is super slow to load) [10:11] jdstrand: hey, could you please look at 5395 again? [10:21] if a configure hook fiddles with socket serviec files, is there a way to have snapctl trigger systemd reload? [10:21] fiddle? [10:22] zyga: yes, say i have socket activated serice and i change the port it listens on [10:22] how can you do that? [10:23] zyga: sed -e with socket unit file? i guess it onl works here becase i have no apparmor to stop me [10:23] yeah. [10:24] exactly that [10:24] but yeah, say i have socket activated service and want to change something there, there is no way to do that, is there? [10:24] that's correct [10:30] PR snapd#5863 opened: overlord/ifacestate: add hotplug slots with implicit slots [10:34] niemeyer: would be great to land the oldest hotplug PRs that need your re-review when you've a moment. stacking up these hotplug PRs is becoming painful [10:39] when is https://dashboard.snapcraft.io/ supposed to get back up ? [10:39] https://status.snapcraft.io/ [10:39] it doesn't say [10:41] No ETA at present, but we're working on it. [10:42] pstolowski: Ack [10:43] niemeyer: sorry for pushing, let me know when i become annoying ;} [10:43] pstolowski: No worries, thanks for pinging [10:44] zyga: we don't use glib in snap confine, only in its tests right? [10:44] pstolowski: correct [10:45] zyga: allright then :). for a moment i was wondering why did you implement key=val parser if glib has one; makes sense, thanks [10:45] yeah :-) [10:59] brb [11:05] hmm tests/main/login failing? i guess the store is not fully up yet [11:05] yeah [11:06] the developer bits are still down, and login touches those [11:06] * zyga dives into propagation settings [11:09] mvo: Q: should I just bump the revno of go-flags in vendor? [11:09] or however that particular dance went :-) [11:11] bah, maybe when the store is happy again :) [11:13] Chipaca: please do and we just try it [11:14] zyga: commented on 5861 [11:14] looking [11:15] mvo: looks like something big breaks; need to dig [11:15] mvo: $ snap help [11:15] error: unknown command "help", see 'snap help' [11:15] :-) [11:17] pstolowski: replied to all, updating [11:18] PR snapcraft#2298 opened: requirements.txt: pin click to v6.7 [11:19] apparmor="DENIED" operation="open" profile="snap.pi-kiosk.hook.prepare-device" name="/etc/writable/hostname" pid=1942 comm="prepare-device" requested_mask="wc" denied_mask="wc" fsuid=0 ouid=0 [11:19] mborzecki: FWIW, https://standards.freedesktop.org/menu-spec/latest/ar01s03.html says Environments (OnlyShowIn and NotShowIn) are case-sensitive, and https://specifications.freedesktop.org/desktop-entry-spec/latest/ar01s03.html says case is significant "everywhere in the file" [11:21] jdstrand, is hostname_control missing write permission to /etc/writable/hostname ? (i have a gadget that tries to set the hostanme in prepare-device but even thgouh the interface is connected i get the above error) [11:21] Chipaca: hmm menu spec [11:21] mborzecki: that's where OnlyShowIn and NotShowIn are defined (the second is the desktop spec) [11:22] second link* [11:34] mvo, hey [11:35] yesterday I saw this with the lxd snap https://paste.ubuntu.com/p/XD87JBKzxs/ [11:35] if I install/remove lxd snap in beta or stable many times it works properly === pstolowski is now known as pstolowski|lunch [11:36] but when I install/remove it with core snap in one channel and then refresh to another channel and install/remove lxd snap it fails removing [11:36] mvo, and it can't be installed anymore [11:38] git bisect is awesome sauce [11:39] heading for lunch and also need to take my daughter to the dentist for a small but urgent correction of her braces. hope to make it back for standup [11:51] zyga: ok [11:53] pstolowski|lunch: ttyl [11:53] jdstrand: thanks! [11:54] PR snapcraft#2299 opened: [legacy] requirements.txt: pin click to v6 [11:56] Chipaca, git bisect is one of the best git features [11:57] Especially when it points to someone ELSE's commit [12:00] well, in this case I think it points to something that worked for us by accident, but it might indeed be a bug [12:00] I need to dig a little more [12:00] (the fix if it's on our end is very simple) [12:10] * zyga feels like he has fever :/ [12:11] I need to put a heater in this room [12:15] mvo: hey there, question for you, is core18 production ready to use as a base? [12:21] niemeyer, any chance you've done anything about enabling nested kvm on gce for spread? [12:21] kyrofa: No, haven't touched htat [12:21] niemeyer, we're to the point where we could use it [12:25] niemeyer, and when I say "could use" I mean "kinda need" === pstolowski|lunch is now known as pstolowski [12:36] PR snapd#5680 closed: [RFC] hotplug: handling of simple add/remove scenario [12:41] PR snapd#5864 opened: make sure hostnamectl can be used on core too [12:45] sergiusens: in a meeting, I will get back to you [12:46] kyrofa: I don't recall where the conversation stands.. do we have a PR already? [12:47] niemeyer, you just need to enable it on gce, it's not spread-specific [12:47] kyrofa: Where? [12:47] niemeyer, https://cloud.google.com/compute/docs/instances/enable-nested-virtualization-vm-instances [12:48] niemeyer, mvo: I'll skip standup [12:48] I feel so-so [12:48] but the real reason is that there's a 20+ movie crew in my house [12:48] shooting an episode of some police para-documentary now [12:48] heh [12:48] so... [12:48] yeah [12:49] things that happen when you live with parents ;-) [12:49] zyga: whaat? [12:49] there's fake police [12:49] and fake gangster [12:49] and fake hostage [12:49] and lights and lots of staff [12:49] and people with makeup [12:49] did you take a wrong turn on the way home from the sprint... are you actually in LA? [12:49] and I learned that my dog can bark [12:49] call the real police ... tell them the fake ones hold you back from attenting an important meeting !!! [12:49] he was so stressed about all the people he started barking at them [12:49] kyrofa: "You can enable nested virtualization using the API or gcloud component. To enable nested virtualization, you must create a custom image with a special license key that enables VMX in the L1 or host VM instance and then use that image on an instance that meets the restrictions for nested virtualization. The license key does not incur additional charges." [12:50] (normally he doesn't because his jaw was damaged) [12:50] kyrofa: I suggest working with cachio on that [12:50] anyway [12:50] zyga, do we hav to call you zygmunt kardashian soon ? [12:50] "action" [12:50] they are shooting now [12:50] well [12:50] the film [12:50] no shots in the scene [12:50] ogra: we didn't, but now we do. [12:50] haha [12:51] he'll be rich and famous soon ... and give us all his makeup tips ! [12:51] tomwardill: in autocratic Poland, movie shoots you ;-) [12:51] niemeyer, alright. cachio are you around? [12:51] ogra: as long as we can skip the semi-nude photos on Instagram, I'm not sure the world is ready for zygdashian [12:51] sparkiegeek: my ass is ready for the photo op [12:52] \o/ [12:52] and here I was thinking that getting fever would mean the day is bad [12:52] uhh ooh ... there can be mind-pics that cant be unseen !!! [12:52] http://www.eyebleach.me/ [12:52] Well that's enough internet for today [12:52] kyrofa: You (or someone) will need to understand how those images are really different [12:52] I can also not leave my room now as the door is blocked by their filming equipment [12:53] kyrofa: As apparently the magic is in the image contnet, not in the project [12:53] sparkiegeek: I don't think zyga needs to skimp on clothes to break the internet [12:53] ogra, zyga this is all it takes: https://www.youtube.com/watch?v=35mQ29Cj-og [12:53] Chipaca: right? have you seen my patches? :-) [12:53] zyga: Ack.. I suggest actually taking the day off if you're not feeling well [12:53] niemeyer: I plan to, as soon as those ... people leave [12:53] and I can actually go upstairs [12:54] zyga: Otherwise it's neither here not there and the recovery doesn't work so well [12:54] I need to type more quietly [12:54] kyrofa, lol ! [12:54] they are shooting right next to my keyboard [12:54] * zyga breaks [12:54] the re-takes are fun to watch [12:54] PR snapcraft#2294 closed: snap: workaround the dirty tree [12:55] kyrofa: once you know how those magic images work I'd love to know [12:55] zyga, haha, I know zero about gce so hopefully cachio does [12:55] kyrofa: (in Yoda's voice) but you will (be) [12:56] ;-) [12:56] ok quiet now [12:56] authentic spanish chorizo... ah. Almost as good as mine [12:56] kyrofa: tell them the one from a town over is better [12:56] kyrofa: that'll go down well [12:57] Hahaha [12:57] my daughter will have a surprise once she returns home [12:57] police, cameras, tons of people [12:57] that won't be intimidating [12:58] I'll tell her that *next* time she needs to do homework! [12:58] login test fails [12:58] another take [12:58] zyga, convince them to put fake blood on you and lay on the floor for when she walks in. Too much? [12:58] kyrofa: you'll be a great parent [12:58] ;D [12:58] zyga, I dunno, how old is she... maybe she'll be happy? [12:59] :P [12:59] 10 [12:59] Ah, not old enough [12:59] "whee, someone killed daddy" [12:59] :D [12:59] * zyga must stop laughing [12:59] yeah, it'd have to be on the dog, not the zyga, for it to be bad at that age [12:59] you should a will right next to you when you do that [12:59] Hahahaha [12:59] I hear handcuffs [12:59] and fighting [12:59] "I leave everything to my son" [13:00] (it's a scene where a guy tries to feel the cops) [13:00] don't forget to add "but please leave the lab running for Chipaca" to your will [13:00] hahahaha [13:00] :D [13:00] I meant to say flee [13:01] gee, spell checkers are so awkward sometimes [13:01] who will clear your browser history? [13:01] zyga, you're missing an opportunity to get famous. Ask them "does your show have anything to do with the cyber?" [13:01] sparkiegeek: I personally trained my dog [13:01] sparkiegeek: he will eat the computer [13:01] kyrofa: your show would be so much easier to do with snaps [13:01] ;-) [13:01] that did need training ?? [13:01] another take [13:02] last I checked the hollywood snap was broken though :( [13:02] volunteering to lend my dog to anyone that needs anything eating [13:02] he's got lots of experience [13:02] tomwardill: did you leave him in the server room this morning? [13:02] ha! busted [13:03] don't pee on the primary replica [13:03] $ hollywood [13:03] Illegal instruction (core dumped) [13:03] [13:03] yeah, still b0rked [13:03] kirkland: ^ can haz fix please [13:03] sparkiegeek: which cpu do you have? [13:03] zyga: core I7 [13:04] sparkiegeek: there would be a lot more mess to clean up if I'd done that [13:04] zyga: I reported this to dustin ages ago, but then he quit :) [13:04] That's about six generations of CPU :P [13:04] Which microarch? [13:04] Er nine generations I guess [13:04] wgrant: well, I'm 90% sure it would fail on all of them; go ahead, try it! [13:05] Intel(R) Core(TM) i7-4790K CPU @ 4.00GHz [13:05] for the record [13:05] * tomwardill can't install it due to 503 [13:07] sparkiegeek: which gen? [13:07] sparkiegeek: I can try [13:07] 4790 is haswell refresh [13:07] 4th gen, iirc [13:07] 5 was Broadwell, yeah [13:07] 6 is skylake [13:08] https://ark.intel.com/products/80807 [13:08] I think they will be shooting more scenes in the basement now [13:08] so I _may_ just go to bed [13:08] it crashes on Skylake here [13:08] so boo [13:08] (another take) [13:08] yeah, it's a bug in the snap :) [13:12] * zyga looks at the tests/main/login test [13:14] sparkiegeek: zyga hollywood might be broken because it never got to experience the magic of magic of patchelf in a rebuild [13:14] yeah, it crashes in ld.so [13:14] * sergiusens is failing to write down all the words that flow through his head [13:15] * zyga hugs sergiusens [13:15] don't worry [13:15] zyga: it is using plain old python from the archive and is classic, but unpatched [13:15] binary patched that is [13:15] * zyga nods [13:18] wgrant: store auth is still down [13:18] is that part of dashboard.s.io? [13:18] yes [13:18] Well partially [13:18] User auth is [13:19] error: cannot get snap access permission from store: store server returned [13:19] status 503 (see 'snap help login') [13:19] right [13:19] ok [13:26] zyga: today is a day to reflect and maybe write docs 🙂 [13:34] sergiusens: core18 should be fine, its a foundations team question now actually :) [13:35] niemeyer: can you take a look at https://forum.snapcraft.io/t/snap-refresh-over-metered-connections/5001/39 ? that's snap refresh on metered connections, seems like core devices may be a bit of a problem becuse some end up using systemd-networkd and NM is oblivious to interfaces that are not managed by it [13:46] https://github.com/jessevdk/go-flags/issues/275 in case anyone was interested [14:08] mborzecki: asked a question there [14:17] popey: ping [14:17] om26er: pong [14:17] popey: there is a new update of Android Studio available since Monday but it hasn't hit the snap store, when do snaps auto build ? [14:18] ah yes. [14:18] 1) when someone with access to build.snapcraft.io logs in and presses the "Manually build" button [14:18] 2) Whenever the git repo snapcrafts/android-studio changes [14:18] i put a build trigger file in every snapcrafters repo [14:19] ah, I though it checked everyday if a new update is available or something [14:19] if anyone creates a PR against that with a date in it, it'll trigger *travis* to test build, which will make sure it builds okay, and if we merge, then build.snapcraft.io will build it [14:19] sadly not [14:19] popey: so we need to do a manual build now [14:19] ok, wanna do a PR against .build-trigger? [14:19] you can do that yourself and travis will start work pretty quickly [14:20] popey: is a PR required ? [14:20] the changes will be automatically pulled [14:20] a pr will ensure everyone can see what's going on and it will be recorded as a change rather than ninja a file [14:20] i dont mind doing it now if you're busy [14:21] popey: I am not busy but I thought that will actually kill the purpose of auto build [14:21] https://github.com/snapcrafters/android-studio/blob/master/snap/snapcraft.yaml#L32 [14:21] Well, there are two things we're doing here [14:21] 1) build on each commit, so we can test it builds okay before we hit the store [14:21] 2) build once committed so we get a build in edge [14:21] travis does #1, build.snapcraft.io does #2 [14:22] s/commit/pr/ [14:24] daily builds (I believe) only happen for projects where it's source for the part is a git repo, not a deb or some other dumped package [14:25] popey: https://github.com/snapcrafters/android-studio/pull/35 does it look fine ? [14:25] PR snapcrafters/android-studio#35: Release Android Studio 3.2 [14:25] well it worked, travis is on it :) [14:25] popey: correct on limitations of daily builds [14:25] thanks sparkiegeek :) [14:26] I wonder if we should set the login test to manual until the store is fully back [14:26] this would allow us to land stuff again [14:27] * cachio lunch [14:27] mvo: gimme 5m to update statuses everywhere, we're basically back [14:27] sparkiegeek: yay [14:28] popey: since y'all are discussing this, is there anyway to setup LP/B.S.I to run tests against a snap that is built through a trigger? It seems like the only solution is to set up the tests externally on something like Travis as you have done [14:29] ijohnson: hiya. The only thing we have discussed recently was some automated testing that cwayne is setting up :) [14:29] * popey bats that over the wall [14:29] I see, thanks for confirming [14:30] om26er: I understand your frustration, I'd like everything autobuilding all the time too :) [14:30] om26er: I worked around this with a shell script which checks out projects, ninjas the date into .build-trigger and pushes, which forces a build [14:31] when we get security updates I run that script against 50 or so snaps and it goes "wheeeee" building them all :) [14:31] please tell me the output is literally 'wheeee' [14:31] because do want [14:31] popey: I might write a bot that: 1. checks every day if a new version of Android Studio is available 2. If available, creates a pull request and notifies us through email or sends a push notification to my desktop to look at the PR. [14:32] that way we can ensure our software is always the latest. [14:32] tomwardill: actually it prints an emoji "shrug" because it triggers a build and doesn't really know if it worked or not :) [14:32] popey: that'll do :) [14:40] ijohnson: It seems like the sort of feature that snapcraft itself might want to have [14:40] Rather than LP defining an LP-specific interface for testing a snap [15:01] snapd translations: where are they? https://forum.snapcraft.io/t/weird-error-message-whenever-i-run-the-snap-command/6418/6 [15:02] mvo: you appear to have access to those [15:02] wgrant: agreed having a way to specify tests to run inside a snapcraft.yaml that are run as part of snapcraft building a snap would be nice [15:03] diddledan: replied on the thread, but https://translations.launchpad.net/snappy/trunk for completeness [15:03] aha, thankyou :-) [15:05] diddledan: I think I do, let me look [15:07] PR snapcraft#2298 closed: requirements.txt: pin click to v6 [15:07] PR snapcraft#2299 closed: [legacy] requirements.txt: pin click to v6 [15:07] niemeyer: hi! I'm looking at https://forum.snapcraft.io/t/interfaces/6154. Is there a way (like there was in the old github wiki) to have links for each particular interface? eg, https://forum.snapcraft.io/t/interfaces/6154#dbus or https://forum.snapcraft.io/t/interfaces/6154#desktop [15:09] jdstrand: I think each interface (or small groups of interfaces) should be its own document [15:09] jdstrand: So that we can go deep into them, without creating a never-ending document which is very tiresome [15:09] sparkiegeek: diddledan: but I think that's not the right one [15:09] sparkiegeek: diddledan: the Dutch translation is empty, there [15:09] Chipaca: oh? *shrug* [15:10] jdstrand: The main interface page would be just a very shallow summary of what the interfaces do, linking into the detailed doc [15:10] Chipaca: probably findable from there though? [15:10] probably, but I'm bad at it [15:10] (CC degville) [15:10] diddledan: do you know what language that is? [15:10] jdstrand: That was my thinking around the issue at least [15:11] jdstrand: Still needs to pass through the real world test [15:11] looks ~dutch to me [15:11] https://bugs.launchpad.net/ubuntu/+source/snapd/+bug/1773174 [15:11] sigh [15:11] Bug #1773174: Dutch lowercase translation warnings [15:11] niemeyer: is there a format for the name of the topic you'd like to see? eg, https://forum.snapcraft.io/t/dbus-interface/... ? [15:12] jdstrand: Yeah, I think we already have a couple there [15:12] jdstrand: Content interface? [15:12] sparkiegeek: yeah, I had the same hunch but it seems its mostly untranslated inhttps://translations.launchpad.net/snappy/trunk/+pots/snappy/nl/+translate [15:12] that is one [15:12] https://forum.snapcraft.io/t/the-content-interface/1074 [15:12] mvo: it reads like dutch to me. Could be afrikaans. [15:13] that has 'the-' prefixed [15:13] mvo: but that's also untranslated there [15:13] so the translations are coming from somewhere else [15:13] we had this same issue before [15:13] where the .no translations had the same issue [15:13] and it took me a while to find them... [15:13] i should've taken notes [15:14] om26er: https://travis-ci.com/snapcrafters/android-studio/builds/86015873 [15:14] 3.2.0.26 - look good? [15:14] https://translations.launchpad.net/ubuntu/cosmic/+source/snapd/+pots/snappy/nl/+translate [15:14] mvo: diddledan: sparkiegeek: ^^ [15:14] Chipaca: ta [15:14] Chipaca: right, was about to go look at cosmic :) [15:15] aha, well found [15:15] number 169 for the first message [15:15] Chipaca: actually I'm not sure we should enforce this rule on translations about the upper/lower [15:16] mvo: why? [15:16] Chipaca: I mean, we basicly have no idea about what rules these langugaes have (at least I don't have any) [15:16] case doesn't mean the same thing in all languages? [15:16] yeah, exactly [15:17] mvo: i'm going to call a bridge-crossed-if-and-when thing [15:17] I know some languages uppercase more than English does [15:17] niemeyer: thinking of brand store snap decls, they would like links out to descriptions. it probably makes sense for me to do that [15:17] I don't know of a language that has cases that uppercases less [15:17] german certainly does :) [15:17] I mean, uppercase a lot more [15:17] yep [15:18] and 143 for the other [15:18] and this is 'uppercase because it's the start of a sentence', so I'd expect it to be pretty universal [15:18] niemeyer: do you want 'the-dbus-interface' or 'dbus-interface'? or something else less likely for someone to accidentally squat? [15:18] eg, snapd-interfaces-dbus [15:19] jdstrand: It has to read nicely.. the URL comes from the topic title [15:19] jdstrand: Squat is also a non-issue in the forum.. we can always rename [15:20] mvo, Chipaca, sparkiegeek : I can make the change now I've found them if you'd like me to? [15:20] diddledan: are you fluent in Dutch? [15:20] niemeyer: The interface? [15:20] diddledan: go for it [15:20] no, it'll only be swapping the case of the first letter :-p [15:20] jdstrand: Yeah, or "The interfaces" in some cases [15:20] popey: so whats the next step for https://github.com/snapcrafters/android-studio/pull/35 ? [15:20] I'm not fluent, and the little bit I had has rusted to heck [15:20] PR snapcrafters/android-studio#35: Release Android Studio 3.2 [15:21] Do we just merge it ? [15:21] Chipaca, diddledan: I think this needs notes for TRANSLATORS so that it can be found before run-time in snapd [15:21] Chipaca: ok, I checked wikipedia and it says that all european languages capitalize the first word [15:21] sparkiegeek: it should have a TRANSLATORS note already (if not, I'll add it) [15:21] jdstrand: But we should sync with degville to see what's his take on the whole sisue [15:21] diddledan: please make the change [15:21] roger that [15:21] Chipaca: https://translations.launchpad.net/ubuntu/cosmic/+source/snapd/+pots/snappy/nl/143/+translate it does not [15:22] niemeyer: degville, can you comment? this is somehting I'd like to have done in the next few weeks [15:22] at least, not one that LP understands, and it got others :) [15:22] sparkiegeek: remind me, what was the original english? [15:22] Chipaca: so I think we are good until these !europeans come and complain :) [15:22] Chipaca: it's there :) [15:22] err, that was for degville only [15:22] done [15:22] niemeyer: the concept idea is fine. is there one otoh that you would like to see as a concept? [15:22] mvo: from the code: // note IsLower != !IsUpper for runes with no upper/lower. [15:23] or more than one [15:23] mvo: so we're covered for a bunch of 'em already :-) [15:23] niemeyer, jdstrand: I'm just getting to that part of the docs. We can definitely update the interfaces asap. [15:23] says it'll be held for review by someone who knows what they're doing :-p [15:23] Chipaca: yay, must have been a smart guy writing this code ;) [15:23] nah, i know that person, i think they just have moments of lucidity and then it's all chaos all the time [15:24] thanks diddledan, lets hope this review will happen quickly, if not I guess we need to poke people (i.e. write to the translations team) [15:24] Chipaca: heh [15:24] * mvo hugs Chipaca [15:24] mvo: as I said in the forum, i'm glad we changed it from a panic :-) [15:24] * diddledan prepares a feather quill for the handwritten note to the translations team [15:24] mvo: also FWIW this has been going on for Dutch people since at least https://bugs.launchpad.net/ubuntu/+source/snapd/+bug/1773174 [15:24] Bug #1773174: Dutch lowercase translation warnings [15:25] diddledan: in esperanto, right? [15:25] yeah, it's quite a while for them to have been suffering with it - I guess most folk just ignored it [15:25] degville: we've not met. I'm Jamie and I work on interfaces a bit and have written a lot of the text in that page [15:25] degville: that started as a wiki page in the github wiki and was migrated over [15:25] jdstrand: Not sure of a specific one.. I just half-expect some interfaces to be too simple and boring, which might be good candidates for some small aggregation [15:25] (I didn't do the migration) [15:25] niemeyer: ack [15:26] jdstrand: You've probably half-met last week :) [15:26] Chipaca: hm, thats month :/ oh well [15:26] mvo: yeh [15:26] degville: are you working on documentation? are you saying this change niemeyer and I discussed is something you would do or you're happy if it happens by someone else sooner [15:27] Chipaca: makes me wonder if we should sync the translations into our git tree, this way we could trivially analyze them [15:27] mvo: translations thinks it's syncing them somewhere [15:27] Chipaca: shall I assign the bug #1773174 about the dutch translations to you and mark it as in progress? [15:27] Bug #1773174: Dutch lowercase translation warnings [15:27] * jdstrand looks in canonical directory and sees that debis a Technical Writer (great!) [15:27] but I dunno :-) [15:27] mvo: I'd like to get them synced, then other distros could get them easier? [15:27] Chipaca: yeah, the upstream translations but those are the distro translations that are not quite right [15:27] Chipaca: yeah, that makes a lot of sense [15:27] sigh [15:28] anyway, I'll add all those TRANSLATOR notes [15:28] there's a bunch of 'em :-( [15:28] joy [15:28] documentation. who'd bother?! :-p [15:28] diddledan: the url above had "cosmic" in it, the bugreport is about "bionic", so we may need to update it in two place :/ would you be so kind? [15:28] okie dokie, I'll go do that [15:29] om26er: I merge it, it builds, success [15:29] 👍 [15:29] diddledan: thank you! [15:30] sparkiegeek: do you happen to know who can approve po/pot in LP these days? we have a pending https://translations.launchpad.net/snapd/trunk/+imports?field.filter_status=all&field.filter_extension=pot [15:30] sparkiegeek: and a bunch of pos [15:31] sparkiegeek: aiui we could enable translation sharing [15:31] mvo: I do not [15:31] sparkiegeek: ok, thanks, I will try to find something in the LP help [15:31] mvo: I assume it's https://translations.launchpad.net/+groups/ubuntu-translators ? but just a guess [15:31] Chipaca: ^- I think with the above (approved translations in LP for snapd upstream) we could enable translation syncs [15:31] mvo: it looks like they automatically got the suggestions Chipaca and I already added [15:32] there aren't any translations for xenial or trusty (the list is empty) [15:32] diddledan: great, thanks for double checking [15:33] oh wow, that pending pot is old [15:33] back from april? [15:37] niemeyer: hey, do you have access to https://translations.launchpad.net/snapd/trunk/+imports?field.filter_status=all&field.filter_extension=all and can click on "approve" on the "needs review" link? it seems I can't but you own the snapd project on LP so you may [15:38] mvo: Grayed out here as well: https://usercontent.irccloud-cdn.com/file/D9bETlXs/image.png [15:40] niemeyer: thank you [15:40] oh dear. who stole the permission bits?! :-p [15:40] niemeyer: I will ask the store for help [15:40] diddledan: heh :) [15:42] mvo: We need to decide on the kernel names.. I think the idea of using tracks + limiting tracks displayed depending on what the model says sounds interesting. [15:42] mvo: Do we have any new concerns around this, or should we just go ahead? [15:43] I vote Wilma for one name! [15:44] what do you mean you want sensible suggestions?? [15:44] pfft [15:48] niemeyer: I think this is good, lets do it [15:56] mvo: Cool, replied [16:04] diddledan: all (or nearly all) the positional arguments had the right TRANSLATORS note, but none (or nearly none) of the options did :-( [16:19] PR snapd#5865 opened: cmd/snap: add a bunch of TRANSLATORS notes (and a little more i18n) [16:19] mvo: diddledan: ^ [16:29] reviewed :-) [16:29] * diddledan requested changes because I'm bossy :-p [16:31] * Chipaca deletes the branch and goes off in a huff, takes up farming [16:31] haha. now _that_ is a quality ragequit [16:31] heh [16:31] I had a client actually do that, once [16:32] still paid his bills, so that was alright, but still [16:32] yikes [16:32] a colleague quit his job as a support engineer to take up organic farming, far up north - so he pretty much only works from May to September now [16:33] also a friend, a dutch web dev, went off to become a farmer in Jujuy [16:33] which was a bit extreme, but he seems a lot happier these days ¯\_(ツ)_/¯ [16:33] roadmr: see, support people I can understand ragequitting the world [16:34] jujuy? is that the month that we do deployments from the commandline? :-p [16:34] diddledan: https://goo.gl/maps/9jAp64cfVh52 [16:34] reminds me of the "I'd rather shovel pig sh*t" in https://www.youtube.com/watch?v=b2F-DItXtZs [16:34] aha, a placee [16:35] I figured it was a mistype of "July" :-p [16:36] I love the emphasis that the tts puts on "web scale" [16:38] om26er: released to stable,beta! Thanks! [16:38] (I tested it first) :D [16:38] popey: Thanks a lot. [16:39] popey: did you release to beta to help people transition to latest ? [16:41] om26er: ah force of habit. For all the other snapcrafters ones the script detects the version in the beta channel to decide when to build the next release [17:25] Chipaca: nice, thank you! [17:26] PR snapd#5768 closed: selftest: actually run the kernel version selftest [17:31] mborzecki: just fyi (for tomorrow), I updated 5771 with your suggestion [19:16] hmm what's this about «Connect docker:home to core:home (connection denied by slot rule of interface "home")»? [19:16] I wanted to give snapped docker access to my home folder to share volumes with, but snapd denies that connection¿? [19:17] Saviq: how did you get that error? [19:20] Chipaca: `snap connect docker:home` [19:21] Saviq: does the snap list the home interface? [19:21] https://pastebin.ubuntu.com/p/ShTwKGh5bS/ [19:23] Saviq: the home interface auto-connects [19:24] well, yeah, but it's not connected... [19:24] that is weird [19:24] - docker:home [19:24] and: [19:24]  snap connect docker:home [19:24] error: cannot perform the following tasks: [19:24] - Connect docker:home to core:home (connection denied by slot rule of interface "home") [19:25] ah [19:25] Saviq: I'm not sure exactly, so I'd defer to zyga for details [19:25] Saviq: but [19:25] Saviq: that snap defines its _own_ home [19:26] Saviq: it defines docker:home, as opposed to core:home [19:26] Saviq: what channel are you on with the docker snap? [19:26] Saviq: I'm not at all sure that works the way it's meant to work [19:26] Saviq: (i might even have it all wrong -- this is a use i've never seen before) [19:27] I imagine this is from https://forum.snapcraft.io/t/auto-connection-for-home-interface-for-docker-snap/7502 [19:32] I had pushed a revert that undoes the read:all for the docker snap until I can get it approved by the store [19:36] Saviq: yes I think I had pushed a revert to edge but hadn't put it on beta yet, I am releasing that now. Sorry for the confusion [19:38] ijohnson: right, I'm on beta [19:38] so probably just hit that then [19:39] Yes I think so [19:41] ijohnson: is there a way to avoid `sudo` for all the docker commands? [19:41] If you add your user to the docker group, then you shouldn't have to use sudo with docker [19:42] right, but need to create that group manually [19:42] Yes [19:43] ack thanks! [19:44] The reason I was adding the read:all to the home plug was so that if you didn't want to create the docker group and add yourself to it and still use sudo docker, then you could at least build files outside of root's $HOME [19:44] But apparently this needs store approval, which I didn't realize (it connected fine locally) === pbek_ is now known as pbek