=== j12t_ is now known as j12t === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun === dvladar_ is now known as dvladar [07:47] good morning [08:07] good morning [08:40] good morning mvo! i have a quick question [08:40] fgimenez: sure [08:40] fgimenez: and good morning [08:41] mvo, in rolling/edge #287 there's no generic-amd64 snap installed, that's ok, right? [08:41] fgimenez: hm, that smeels like a bug [08:41] fgimenez: I think its because of the rename of oem and gadget [08:42] mvo, http://paste.ubuntu.com/14086622/ [08:46] mvo, this is the failed test http://162.213.35.179:8080/job/snappy-cloud-test/21/consoleFull (search for "FAIL:" in there), when there's a new image available a new snappy-cloud-test is triggered [08:50] fgimenez: thanks, I will prepare a fix [08:51] mvo, thanks, btw regarding the all-snaps test you mentioned yesterday i've added a card in the old trello board to keep track of it https://trello.com/b/4PQViyUQ/snappy-core-stakeholders-backlog [08:51] fgimenez: \o/ [08:51] mvo, we can't we begin with it with the current rolling/edge, right? [08:52] fgimenez: I will start thinking about how to port the teests to the new world [08:53] mvo, great, another thing, in case you didn't notice we have integration tests executions with each PR at last \o/ [08:53] fgimenez: rolling/edge is not yet all-snaps, but I guess we should change this RSN :) [08:53] fgimenez: and yes, I noticed and its super cool, thanks for this!!! [08:55] mvo, np, i really needed to see it happening :) the PRs targeted to master are already working, we need this branch landed for 15.04 to work https://github.com/ubuntu-core/snappy/pull/261 [09:02] fgimenez: I guess you need review for this :) ? [09:04] mvo, well yes, if you have the time it would be great :) [09:06] fgimenez: I will do my best, the oem snap thing has some complications, I need to think about it a bit more. we changed the directory from /oem to /gadget that is trivial to re-add but we also changed the type from oem to gadget so this involves two places [09:06] fgimenez: but once oem and sudo are under control I should have time to review [09:08] mvo, great thanks a lot, no rush at all [10:29] Good morning all; happy Friday, and happy Bake Cookies Day! 🍪 === chihchun is now known as chihchun_afk [11:55] ogra_: you know if there is a way to install a core image going to be 16.04 on the raspberry-pi2? [11:55] didn't found any images yet [11:55] you need to use ubuntu-device-flash and create your own ... [11:56] we do not provide any images for rolling [11:56] (for none of the arches) [11:58] ogra_: ah I see [11:59] so rolling is what 16.04 will be [11:59] ogra_: so I just pick a device tarball from http://people.canonical.com/~platform/snappy/raspberrypi2/ and generate an image for rolling, right? [11:59] no [11:59] you only use u-d-f [12:02] no device tarball anymore? [12:02] sudo ubuntu-device-flash core rolling --channel=edge --oem=pi2.canonical/stable --device=armhf_raspi2 --enable-ssh -o myimage.img [12:02] or oem snap? [12:02] I see [12:03] oem snap comes from the store, device tarball from the system-image server [12:03] nice [12:03] (and soon everything comes from the store as a snap and system-image will be gone) [12:03] hm, gives me: [12:03] Device armhf_raspi2 not found on server https://system-image.ubuntu.com channel ubuntu-core/rolling/edge [12:04] funny, it is building here [12:05] bah [12:05] if i would have pasted the correct line from ym terminal it would work at least [12:05] :P [12:05] sudo ubuntu-device-flash core rolling --channel=edge --oem=pi2.canonical/stable --device=raspi2_armhf -o myimage.img [12:05] try that one [12:05] :) [12:06] :-) [12:06] (and --enable-ssh if you want to log in remotely) [12:06] jupp, works [12:25] mvo, I ran into an issue yesterday I'd like to pass by you [12:42] ogra_: worked quite nicely :-) [12:42] :) [12:49] sergiusens, I've been asked about including #169 in 1.x. Sound okay to you? [12:50] kyrofa, yeah, we should get a list; but we need to cherry pick #169 and #162 [12:50] sergiusens, ah, indeed [12:51] kyrofa, want to take that or should I? They should be a direct cherry pick [12:51] kyrofa, btw, nice progress while I was gone [12:51] * sergiusens is happy [12:51] sergiusens, I got it :) [12:51] kyrofa, do you know if Leo is going to be here today? [12:51] sergiusens, should I make a PR for them? [12:52] kyrofa, I guess so, just to be sure the tests pass before pushing it [12:52] sergiusens, I'm not sure honestly. His power/internet has been a little flaky, we haven't connected the last few days [12:52] sergiusens, however, knowing his leave situation, I wouldn't be surprised if he was outta here [12:58] kyrofa: sure, what is it? [12:59] mvo, Snappy services device the user data path using %h/apps. However, that directory does not seem to be created for services like it is for binary wrappers [12:59] mvo, so anything that tries to use that variable ends up trying to create /root/apps, which obviously can't happen [13:00] kyrofa: aha, ok [13:00] mvo, how would you feel about remapping the user data path to the app data path for services? [13:00] mvo, or should we actually create the user data path? [13:01] kyrofa: in a meeting right now, will reply in a wee bit [13:01] mvo, no problem. I'm happy to take care of this, just want to come to an agreement on the proper fix :) [13:09] sergiusens, #170 and #171 [13:09] kyrofa, you can't create though, apparmor won't let you [13:09] sergiusens, I mean as a snappy fix [13:10] oh right, that seems logical; maybe it should be in the core launcher though; not sure where the dirs are created these days [13:10] sergiusens, well for binaries it's in the wrapper [13:10] sergiusens, but moving it to the launcher would solve it for both [13:10] sergiusens, not a bad idea [13:14] sergiusens, but it also makes sense to just have services logging to the data path rather than somewhere in /root [13:14] kyrofa: I think mapping the user-data-path to the app-data-path makes most sense, so +1 from me on that [13:14] kyrofa: and nice catch! [13:14] mvo, ah, I love the easiest solution [13:14] mvo, no problem-- you'll see a PR soon [13:14] \o/ [13:15] jdstrand, ^^ in case you were curious [13:16] * jdstrand happens to be lurking, but will wander off for sure [13:16] nice! [13:20] kyrofa, question, should I squash `new-cli` before proposing a merge into master? [13:20] sergiusens, nahh [13:20] sergiusens, special case, I'd say [13:20] sergiusens, that's usually to get rid of "oh and do this too" commits [13:20] sergiusens, which you've already done [13:21] kyrofa, k, so just rebase; the reason I ask about elopio is that he targetted the testing change to `new-cli` [13:21] sergiusens, maybe wait an hour and see if he comes to standup? [13:22] sounds good [13:23] sergiusens, how should we update bugs now that the code and bugs are separate? Fix committed when it gets merged, fix released when it's a .deb? [13:23] kyrofa, fix released when the milestone is released [13:23] sergiusens, also, how do we indicate that something is fixed for 2.x but not 1.x? [13:24] sergiusens, hmm... define milestone? [13:24] kyrofa, we need to create series [13:24] kyrofa, https://launchpad.net/snapcraft/+milestone/0.6 [13:24] Ah ha [13:24] kyrofa, or the thing on the far right in the bug [13:24] Okay so yeah, fix released when it's in the archives, essentially [13:25] kyrofa, yeah [13:25] Ah, and bugs target them, okay nice I gotcha (not a heavy launchpad user here) [13:27] sergiusens, is creating the series relatively easy? [13:28] kyrofa, already done [13:28] sergiusens, hey, look at you! [13:28] kyrofa, I'll do bug management for a bit [13:30] sergiusens, so if I have a bug where I'd like to track state for both 1.x and 2.x, do I "target to series"? [13:30] kyrofa, like this I guess https://bugs.launchpad.net/snapcraft/trunk/+bug/1523912 [13:30] Launchpad bug 1523912 in Snapcraft 1.x "Shebang write uses overly broad regex" [Critical,New] [13:34] sergiusens, ah, pretty painless [13:36] kyrofa, we need this merged first for tests to pass https://github.com/ubuntu-core/snapcraft/commit/3ec8328a144c79b158c62ec836c356281f3b080f [13:36] eh, cherry picked [13:36] sergiusens, oh oops. Fixing now [13:43] sergiusens, https://github.com/ubuntu-core/snapcraft/pull/172 [14:00] kyrofa, merged [14:02] mvo: hey, so, ubuntu-core-security 16.04.9 got all caught up in the perl transition and is still in -proposed. are you planning on making another os all snap image before eoy? [14:03] mvo: that has the new caps from the brazil sprint and would be nice if people are playing with snappy over the holidays if they were there [14:03] * jdstrand isn't really here btw [14:03] sergiusens, alright, rerunning PR tests [14:07] jdstrand, out until the new year? [14:12] jdstrand: hm, that sounds like an important reason - so far we build without -proposed but I can build one with -propsoed enabled [14:26] kyrofa: I am [14:26] mvo: well, I expect it to get through proposed sometime today [14:26] jdstrand: aha, nice. in this case its easy, I can do a new upload today [14:26] mvo, not sure thats clever ... [14:27] given there sits also a half transitioned perl in -proposed [14:27] (building with propoed i mean) === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk [15:01] lool, be tere in a bit [15:02] sergiusens: ok [15:04] lool, I'm in [15:08] hello all. I’m new to snappy and am running into some trouble getting a my snap (which is really just a docker image) to work. if anyone could help me out, I’d really appreciate it. [15:09] my snacraft.yaml: https://gist.github.com/skawaii/5c710b27eb9eac1cc5cc [15:09] my exe sh script: https://gist.github.com/skawaii/8f71b3d76498b29ca989 [15:10] the erro i’m seeing is this: [15:10] (amd64)ubuntu@ubuntu-snappy:/apps$ whaley.speak [15:10] (amd64)ubuntu@ubuntu-snappy:/apps$ whaley.speak [15:10] /speak.sh: 2: /apps/whaley.sideload/IIDeccTbeLdT/./bin/speak.sh: docker: Operation not permitted [15:13] i can run my docker image outside of the snap via “docker run skawaii/docker-whaley” without problem, so I’m a little lost here [15:26] mvo: https://github.com/ubuntu-core/snappy/pull/269 [15:27] kyrofa: sergiusens: hello. Just checking in... [15:27] Hey elopio! [15:27] sergiusens: what do you need me to do with the integration tests branch? [15:27] elopio, are you out? [15:27] kyrofa: well, I'm in bed, sleeping a lot :D [15:28] elopio, hahaha [15:28] Today I'm going to start disconecting. I'll send an email. [15:28] elopio: how's that different from normal ;) [15:29] kyrofa: do you need something from me before I start partying? Today is the maker space party, cryptobeers without crypto. [15:30] elopio, that sounds amazing. And no, go away! [15:30] elopio, heads up though if you have a minute: sergiusens said you were targeting the new-cli with some tests [15:30] elopio, but he needed to rebase it to master, so he pushed it to a new branch instead of messing up your tests [15:30] elopio, you might want to retarget that new branch [15:31] davmor2: holidays are the same as any day, but without alarm clocks. [15:32] kyrofa: ah, ok, let me see. [15:32] elopio, ah, yeah your py_integration branch I think [15:34] elopio, also, if you have a minute to give https://github.com/ubuntu-core/snapcraft/pull/146 one last look, it's been cleaned up and I think it's good to go, I just want one more pair of eyes before I merge it [15:34] I'll try a merge with his rebased-new-cli branch. [15:37] kyrofa: I had already merged it on master. I've just merged this one too. [15:37] now sergiusens needs to update a lot of docs with his new cli :) [15:37] elopio, excellent, thank you :) [15:37] Haha, yes indeed [15:40] kyrofa: how do I merge my branch with a branch from sergio? Do I need to add him as a remote? [15:40] elopio, yeah, git remote add sergio_stuff then git fetch --all [15:44] mvo: thanks! [15:44] kyrofa: got it, the merge seems ok. The commit message and the diff does not. I'm not sure how to turn my change in one single commit on top of sergio's. [15:45] elopio, use 'fixup' [15:45] ogra_ (and mvo): note, ubuntu-core-security doesn't care about perl. it has a Build-Depends on apparmor that needed a rebuild that delayed things [15:45] ogra_ (and mvo): but it doesn't need the rebuilt apparmor or perl [15:46] jdstrand, well, its all or nothing ... if you enable proposed you get all packages from there including all breakage too [15:46] elopio, so what you should really do is rebase onto the new cli branch, and you'd have the same commits as before [15:46] * jdstrand nods [15:46] commit messages anyway [15:46] elopio, then you can fixup as sergiusens suggests, or just squash manually if necessary [15:47] I propose we move to bzr. [15:53] elopio, hahaha [15:53] sergiusens: kyrofa: https://github.com/sergiusens/snapcraft/pull/1 [15:54] I used cherry-pick, which I learned yesterday. Please tell me if that works. [15:55] elopio, hey that works. The diff looks the same, although making a PR into the fork means the tests don't run on travis right? [15:55] kyrofa: right. But I suppose sergiusens will make a pr into master, and there they'll run. [15:55] I confirmed they still pass here. [15:56] elopio, yeah that seems reasonable === frecel_ is now known as frecel [16:01] elopio, kyrofa I can push the rebased-new-cli to the official repo, then tests will run [16:02] sergiusens, elopio that might be better just to have a record of that PR on the official project [16:02] But it's not super important to me anyway [16:02] kyrofa, hah, so elopio didn't squash :-P [16:02] kyrofa, https://github.com/sergiusens/snapcraft/pull/1 [16:03] Tsk tsk elopio [16:05] kyrofa, elopio ok, so PR to this so tests run [16:10] sergiusens, looks like those cherry-pick tests pass [16:10] elopio, you'll love git before we're done with you [16:25] kyrofa, ah, thought they were merged already.. [16:26] kyrofa: sergiusens: so I got lost. sergiusens PRs his changes to master. We land that, and then I PR my two commits against master? [16:26] sergiusens, nah, I wouldn't do that without at least one more eye [16:26] elopio, no sergiusens just pushed his branch into the official project rather than his fork [16:27] elopio, so make a PR on the official project to that branch instead of the one on his fork [16:27] elopio, and squash [16:28] elopio, which is the only resaon I didn't merge your branch earlier [16:28] sergiusens: I wanted to keep two commits. The first makes the tests pass in xenial. The second is a workaround for trusty that I want to remove the first day of next year. [16:28] elopio, if the is the case then it is fine [16:29] elopio, let me do the work then [16:29] elopio, meh, I'll leave it to you [16:29] yes, I got i [16:29] t [16:30] or not. Just a second... [16:32] https://github.com/ubuntu-core/snapcraft/pull/173 [16:35] elopio, great [16:35] kyrofa, elopio so maybe today I just update the docs all day long :-P [16:35] sergiusens, yay, fun day [16:53] green! [17:04] kyrofa, elopio ok, so this is now there https://github.com/ubuntu-core/snapcraft/pull/174 [17:04] * sergiusens needs to go out and be a nanny for a bit [17:05] Alright sergiusens, thanks! [17:05] sergiusens, will you do the docs after this merges, then? Or will you put it in this branch? [17:08] kyrofa, I was thinking new branch, this is getting too big; unless you want the docs here [17:09] sergiusens, works for me :) [17:58] sergiusens, is there no snapcraft all anymore? [18:02] sergiusens, or is that now just snapcraft snap? [18:03] sergiusens, oh man... this is already so much better [18:03] sergiusens, excellent work [18:16] sergiusens, is there a reason snaps seems so big? I have a student working on a nodejs snap and it's 10 mb's [18:20] balloons, because it includes all of nodejs? :) [18:21] beuno, yes, presumably so. It has not external depends [18:21] the app itself is like 1k, heh [18:21] right [18:21] so that's where the 10mb comes from [18:22] in exchange, once his app works it'll work forever* [18:22] *ish [18:23] balloons, lol 10MB [18:25] (Dragonboard)ubuntu@localhost:/mnt/home/linaro/xenial-chroot/upnp-server$ ls -lh upnp-server-arm64_0.1.0_arm64.snap [18:25] -rw-r--r-- 1 root root 42M Dec 15 18:43 upnp-server-arm64_0.1.0_arm64.snap [18:25] that only has minidlna and lighttpd inside :) [18:25] snaps are big by design [18:25] I suppose I should get used to that. [18:26] I'm seeing that :-) [18:26] i have an unconfined debootstrap snap here that has 150MB ... [18:26] simply pulling in a whole base system as deps [18:34] so it's interesting you can build an i386 snap, and push it to the store. But it can never be used, as there is no i386 image right? [18:35] money.rga is the example [18:35] yeah [18:35] no i386 images atm ... [18:36] (they might come though, there are IoT platforms that are i386 only ... once we support them we will have to have i386 images) [18:36] but til then you have to build for amd64 to get something usable [18:36] (or armhf ... and soon arm64) [18:38] balloons, you can build an image, there just isn't one promoted as avail [18:39] with snaps we want everyone to feel like compiling go with gc-go :-P [18:40] sergiusens, https://github.com/ubuntu-core/snapcraft/pull/175 . Sort of a "cleaning up nasty things" PR, a tad random I'm afraid [18:41] sergiusens, mostly tests [18:45] kyrofa, got it; too bad you merged the new cli thing [18:45] sergiusens, uh oh, how come? [18:45] * sergiusens mentions he likes to joke and troll, more so on Fridays ;-) [18:45] argh.. does --allow-unauthenticated fail for anyone else? Trying to sideload a snap and I get a signature verification failed. [18:45] * balloons tries with a different package [18:46] balloons, 15.04 or rolling? [18:46] 15.04 [18:46] kyrofa, so, I was just joking (just in case) [18:46] well, I'm running ubuntu-core/15.04/edge [18:46] sergiusens, you suck too [18:46] balloons, it should just install [18:46] kyrofa, glad to know [18:47] * sergiusens is probably on Santa's naughty list [18:47] Probably [18:47] sergiusens: do you every get off that list? [18:48] You were always that one kid that people actually DID buy that stupid wrapped coal present for [18:48] davmor2, haven't kept track; I just pay attention now as I don't want to splash the naughtiness onto my son [18:48] Does anyone ever buy those? [18:49] sergiusens: :) [18:49] kyrofa, I was once gifted a brick for my birthday in this huge box; it was a prank from a friend [18:49] my life forever changed [18:49] Haha, "this is heavy... must be expensive" [18:49] sergiusens: yeah only cause you used the brick to smash the windows to get the things you wanted right ;) [18:49] sergiusens, ok, so I believe I know what the issue is. It's displaying that error when I install a snap with a bad arch. Aka, an arm snap on amd64 [18:50] balloons, I don't think there will be user message polish on 15.04, but may be good to know if this is the case on rolling too [18:50] a preemptive bug would be good [18:50] sergiusens: now I understand why you are on Santas naughty list, I mean if friends treat you like that :D [18:51] if someone wishes to do a quick check that would be nice [18:51] try installing http://people.canonical.com/~nskaggs/money.snap for instance. It should yell at you nicely [18:51] balloons, us snapcrafters aren't really into having the latest and greatest images anymore [18:52] cuz those are broken [18:52] balloons: you have a snap that makes money what printer do I need ;) [18:52] lol.. so we just get elopio to test right? he's qa! [18:52] balloons, there you go! [18:53] balloons, hah, elopio is headed to the jungle now and has no solar powered laptop, much less any internetz ;-) [18:53] ahh right.. Friday already [18:54] you guys into github issues now, or lp bugs? [18:55] balloons, lp [18:55] balloons: no same as usual bungee cords and post-it notes [18:55] I guess I'll still give you a bug report. Even on lp [18:55] sergiusens, why lp, by the way? [18:56] sergiusens, just because it's always been there? [18:56] kyrofa, our bdfl suggested it [18:56] sergiusens, ah [18:57] balloons, if you go to snapcraft's issues at least you get pointed to lp; not sure how it is in snappy [18:57] * balloons spins it on it's head and asks why github if lp has git support [18:57] :p [18:58] balloons, because github contributions/streaks are way cooler than lp karma [18:58] balloons, mentions in reviews :-) [18:58] balloons, a nicer landing page [18:58] balloons, a bigger community [18:58] sergiusens, I know why you would choose github. It's just simpler and cleaner [18:59] balloons, everyone looks on github for projects before lp, too [18:59] balloons, right [18:59] just playing DA [18:59] mostly to point out having half the project one place and half the other is a bit annoying [18:59] balloons, no argument here [19:00] balloons, I'd like to be able to close issues from commit messages [19:01] ok, snappy playtime is done. Thanks for the responses guys. And enjoy your holidays! [19:01] balloons, that was easy :-P [19:01] /whois holidays [19:02] sergiusens, curl.curl is my favorite snappy package now btw. An IoT device with no cli network tool feels dead [19:03] * ogra_ has a wget-unconfined snap for that :) [19:03] ogra_, holidays are those things other people do sometimes. I know, I know. It's weird, but the silence is refreshing [19:03] for old-schoolers :) [19:03] * balloons is a wget man [19:05] ogra_, wait a minute... does that mean if I like wget I'm old? [19:06] not as old as me i guess :) [19:07] ogra_: oh common, you're old school enough to know how to use sed in some archaic manner to get urls and you know it ;) [19:07] hah [19:07] someone knows me :) [19:07] shhhh [19:09] yay ... my new kbd just got into delivery ! [19:10] ogra_: I just have deep seated respect for any developer who lived through the buttons moving wars ;) I mean I'll still treat you with disdain, sarcasm and indifference but know it comes from this deep seeated respect :D [19:10] -e [19:10] (the christmas present i made myself ) [19:10] ha [19:11] "the button moving wars" [19:11] you should write a book ;) [19:12] it wasnt actually a war ... that kind of needs two parties ... [19:12] after all it was just me whining :) [19:12] (and design not reacting) [19:12] hahaha [19:13] olga paid me a beer as compensation in austin though :) [19:13] (in silent agreement) [19:14] ogra_: that's a lot of beer I mean austin isn't a little place :D [19:14] :D [19:15] sergiusens, yeah I'm all over the place regarding "" versus `` huh [19:15] sergiusens, what do you prefer? [19:17] kyrofa, I was more inclined to use `` just because "markdown" [19:17] sergiusens, agreed. Fixing along with the last `find` [19:17] are that double backticks ? [19:18] ogra_, yeah [19:18] uuh [19:42] kyrofa, done with the review [19:47] sergiusens, thanks :) [20:22] sergiusens, alright, fixed up [20:36] hey! I messed up, and forgot my password.... [20:36] I've been looking for solutions and the only one's I find are for raspbian [20:37] I have a rpi2 [20:37] has anybody come across something like this before? [20:37] hathatwhereshat, what are you running? [20:37] snappy [20:37] hathatwhereshat, and you changed the default password of ubuntu? [20:37] cheers for showing interest sergiusens [20:38] yes... I wrote the password on a note, but I wrote it down wrong [20:38] hathatwhereshat, take the sdcard out; go into /system-data/var/lib/.../extrausers/shadow and edit the password there [20:38] that might be a way, not sure about the exact path to the file as I don't have a system running currently [20:39] that's fantastic news sergiusens! thank you very much [20:39] I'll let you know if I can get in [20:39] kyrofa, no tests have been harmed in this writing https://github.com/ubuntu-core/snapcraft/pull/178 [20:40] hathatwhereshat, you jst need to know the format of a shadow file `man shadow` [20:43] sergiusens for some reason I don't have these folders [20:44] hathatwhereshat, it should be in the partition named writable [20:44] oh ok. thank you so much sergiusens === retrack is now known as Guest59273 [20:45] kyrofa, I'm curious to know if you followed the link from __init__ before I made the comment it was a link [20:46] seemed sneaky after doing it :-P [20:56] hey sergiusens , I've been trying to find this partition (I'm just starting with programming) all over the place but can't find it.. I'm on a mac and I've tried "df -h" to find the volumes, I've also tried "ls -la /Volumes" but I don't find this writable partition [20:57] If you can't help I'm still stupidly grateful for what you've done til now!! [20:57] hathatwhereshat, oh, I have no idea how mountpoints are exposed on a mac, on Ubuntu if you plug the sdcard into a laptop it will show up on /media//writable [20:57] beowulf, maybe you can help ^ ? [20:57] if around of course [20:58] ok... I'll try a live version of ubuntu then! === retrack is now known as Guest5625 [21:37] kyrofa, feel free to ping me about whatever, but I am mostly EOY now :-) [23:57] any new url for 'snappy gui anyone?'