[01:12] PR snapcraft#837 opened: kernel plugin: allow collecting the same mod deps [01:15] cwayne thanks! [01:15] cwayne I hope you notice it is a smaller snap than before too ;-) [01:23] ooh I like that :) [02:06] PR snapcraft#838 opened: pluginhandler: take the file encoding into account === bpierre_ is now known as bpierre [04:36] PR snapcraft#839 opened: Fixing 'sign-build' integration tests [04:37] otherwise if falls apart with the static name, https://travis-ci.org/snapcore/snapcraft/builds/163606844 [05:34] o/ [05:48] PR snapd#2003 closed: debian: adjust packaging for trusty/deputy systemd [06:41] good morning [07:43] hey dholbach [07:43] morphis_: hey [07:43] morphis_: I got things to work last night [07:44] hey zyga [07:46] can somebody respond to the comment in https://plus.google.com/u/0/+DanielHolbach/posts/Xiho2rVSmEg?cfem=1? [07:57] zyga: wow! [07:57] zyga: how? [08:12] ogra_: so it looks like my bug from yesterday is because of upgrades so I'm going to do a bunch of testing around that today and see if I can trigger it [08:14] hello [08:15] morphis_: re [08:15] morphis_: let me show you [08:16] morphis_: http://paste.ubuntu.com/23250152/ [08:16] try this script please [08:17] zyga: what should I do when I am in the bash shell [08:17] how to contact the community manager :? [08:18] help me please [08:18] morphis_: you can mount anything in /media [08:18] morphis_: and look at /proc/self/mountimfo [08:18] info* [08:18] morphis_: both inside and outside of the shell [08:18] deymon0077: hey, how can I help you? [08:19] zyga: wow [08:19] it works :-) [08:19] morphis_: I'll make the changes to snap-confine today [08:19] great [08:19] will read through the script later [08:20] I have a question: can I in the town of open courses circle to work with Linux such as Ubuntu? [08:21] deymon0077: hmm, can you re-phrase that question? [08:21] yes of course 1 minute [08:23] if I could, in our city open circle work on Linux Ubuntu in particular [08:23] is it possible to? or not? [08:24] I am waiting for an answer to the impatient [08:25] deymon0077: I don't see any problem with that [08:25] deymon0077: this is not the best forum to ask about things like that [08:25] deymon0077: I would recommend to post questions to ... [08:25] deymon0077: https://lists.ubuntu.com/#Ubuntu+Worldwide+LoCo+Teams [08:26] deymon0077: match the country that you are living in [08:26] deymon0077: you can connect to the local community of ubuntu users and developers this way [08:26] deymon0077: on that page you can also find many interesting mailing lists [08:27] Thank you very much! Good luck. I will try it =) [08:28] Have a nice day and all delicious cookies) [08:39] PR snapd#2005 closed: asserts,overlord,snap: add prepare-device hook for device registration [08:55] PR snapd#2028 opened: doc: added timezone-control to interfaces doc [09:07] PR snapd#1990 closed: many: allow use of the system user assertion with create-user [09:08] Hello, good advise Russian forum ubuntu? [09:08] I'm from Ukraine [09:09] or a good foreign forum [09:18] tell me a good forum ubuntu [09:18] ? [09:29] deymon0077: perhaps ask in #ubuntu [09:29] deymon0077: this channel is specific to snappy === davmor2_ is now known as davmor2 === ara_ is now known as ara [10:20] PR snapd#2029 opened: daemon, store: switch to new store APIs in snapd [10:28] PR snapd#2030 opened: coreconfig: nuke it. Also, ignore po/snappy.pot [10:45] when you connect a content plug and slot, is there a directory that gets mounted? [11:06] PR snapd#2017 closed: store: download deltas if explicitly enabled [11:19] mectors, wait for mhall119, i think he has some experience with the content interface [11:21] ogra, thanks === hikiko is now known as hikiko|ln [11:41] zyga: I think I actually lied to you yesterday. I had dont apt upgrade and snap was working indeed. But when I did install the next version of my snap (build with the same version of snapd) I ran into the same issue again. After new revision of snap got installed it once again complains about wrong arm architecture. :( [11:42] jdstrand, hello, fyi, https://github.com/snapcore/snapd/pull/2028/files & apologies for forgetting about it in the earlier PR [11:42] PR snapd#2028: doc: added timezone-control to interfaces doc [11:43] hmm, whats up with the latest snapd ... [11:43] after my dragonboard auto-upgraded today i get: [11:43] ogra@dragon:~$ snap list [11:43] error: cannot list snaps: cannot communicate with server: Get http://localhost/v2/snaps: dial unix /run/snapd.socket: connect: connection refused [11:43] ogra@dragon:~$ [11:44] ogra@dragon:~$ systemctl status snapd.service [11:44] ● snapd.service - Snappy daemon [11:44] Loaded: loaded (/lib/systemd/system/snapd.service; enabled; vendor preset: enabled) [11:44] Active: inactive (dead) (Result: exit-code) since Sat 2016-09-24 11:24:12 UTC; 5 days ago [11:44] Process: 2106 ExecStart=/usr/lib/snapd/snapd (code=exited, status=1/FAILURE) [11:44] Main PID: 2106 (code=exited, status=1/FAILURE) [11:48] wow, thats really badly broken [11:52] mvo, zyga ^^^ seems i had a failed upgrade for some reason, now my dragonboard install is in a unrecoverably state telling me snapd is to old [11:52] ogra@dragon:~$ cat /proc/cmdline | tr ' ' '\n'|grep snap_core [11:52] snap_core=ubuntu-core_425.snap [11:52] ogra@dragon:~$ ls -l /var/lib/snapd/snaps/ubuntu-core_*|tail -1 [11:52] -rw------- 1 root root 64577536 Sep 26 06:17 /var/lib/snapd/snaps/ubuntu-core_745.snap [11:52] ogra@dragon:~$ [11:53] Sep 24 11:24:09 dragon systemd[1]: Reached target Bluetooth. [11:53] Sep 24 11:24:10 dragon snapd[2007]: error: cannot downgrade: snapd is too old for the current system state (patch level 4) [11:53] Sep 24 11:24:10 dragon systemd[1]: snapd.service: Main process exited, code=exited, status=1/FAILURE [11:53] Sep 24 11:24:10 dragon systemd[1]: snapd.service: Unit entered failed state. [11:53] Sep 24 11:24:10 dragon systemd[1]: snapd.service: Failed with result 'exit-code'. [11:53] Sep 24 11:24:10 dragon systemd[1]: snapd.service: Service hold-off time over, scheduling restart. [11:53] Sep 24 11:24:10 dragon systemd[1]: Stopped Snappy daemon. [11:53] thats what i see on boot [11:53] ogra_ well now you know, snapd is too old :-P [11:54] sergiusens, well ... i see three ubuntu-core snaps ... one seems to be the broken 745 one, the other the reliably working 738 one ... but for some reason there is also a 425 that got picked by the auto-rollback mechanism [11:55] (instead of rolling back to 738) [11:55] ogra_ thanks for clarifying conf/modules ;-) [11:55] :) [11:55] ogra_ surely a new feature/bug as I don't think we support that in 16 at all (use to in 15.04 through ubuntu-core config hook) [11:56] well, and we shouldnt ... [11:56] the kernel needs to auto-load whats needed ... [11:56] i suspect the modules.dep is wrong .... it needs to be re-generated for exactly the module set that is inside the initrd [11:57] before re-compressing [11:57] ogra_ it should load what it needs, but what if the module takes some specific options like disable_802.11n=true (or however it was formatted) [11:57] ah, yeah, that would need something in th emodules.d dir ... a config file [11:57] *in the modules.d dir [11:57] ogra_ it used to be correct as it used to work and the kernel plugin only got love from ppisati since its introduction and this bug fix now [11:58] errr [11:58] /etc/modprobe.d that is [11:59] i guess we need more info like a kernel log from serial during boot and such ... [11:59] sborovkov: hmm, interesting [11:59] to see why it wasnt loaded [11:59] sborovkov: can you strace both cases? [11:59] so back to my main prob ... [12:00] why do we keep three vervisions of ubuntu-core ... and why do we roll back to the obviously obsolete one [12:00] an end user would now have to re-flash [12:00] *revisions [12:01] i guess thats a serious blocker [12:05] pstolowski: thanks! [12:06] PR snapd#2028 closed: doc: added timezone-control to interfaces doc [12:06] eeek [12:06] ogra@dragon:~$ ls -l /var/lib/snapd/snaps/ubuntu-core_* [12:06] -rw------- 1 root root 57589760 Sep 29 10:20 /var/lib/snapd/snaps/ubuntu-core_425.snap [12:06] -rw------- 1 root root 64577536 Sep 25 08:19 /var/lib/snapd/snaps/ubuntu-core_738.snap [12:06] -rw------- 1 root root 64577536 Sep 26 06:17 /var/lib/snapd/snaps/ubuntu-core_745.snap [12:06] * ogra_ notices the timestamps [12:07] so obviously it didnt not roll back but *upgraded* to 425 ... which is the last stable ubuntu-core [12:07] PR snapd#2028 opened: doc: added timezone-control to interfaces doc [12:07] it did that automatically ! [12:08] smells like snapd has a bug in the channel handling here [12:08] PR snapd#2028 closed: doc: added timezone-control to interfaces doc [12:08] mvo, zyga did anything change in that regard recently ? [12:08] (with the new snapd) [12:10] jdstrand: hey [12:10] zyga: in an hour. Need to test my snap first. How do I strace what snapd does? [12:10] * ogra_ wonders if we have a GA blocker tag for bugs [12:10] sborovkov: hmm, good question, you can probably just strace the command and coupled with -f -o you should get the full output [12:10] hey zyga :) [12:10] ogra_: I don't know really [12:11] zyga, well, thats pretty disastrous [12:11] ogra_: I don't get notifications in irssi unless my name is the first one [12:11] ogra_: note that this is under the control of the store [12:11] PR snapd#2028 opened: doc: added timezone-control to interfaces doc [12:11] ogra_: it's not true that revisions can be compare [12:11] *compared [12:11] ogra_: the most recent revision is whatever the store is claming [12:11] claiming [12:11] * zyga needs that coffee [12:12] zyga, well, the image is created with "-c edge, it should never ever pull ubuntu-core from stable [12:12] unless i manually tell it to [12:12] PR snapd#2028 closed: doc: added timezone-control to interfaces doc [12:12] and i think thats set up inside the imge, not in the store [12:13] via seed.yaml or so [12:14] ogra_: that may be a bug in the image building tool [12:14] ogra_: but I don't know how this works either nowadays, [12:14] ogra_: you can peek at /var/lib/snapd/state.json [12:14] well, that image is from last week ... it sat there without me touching it and auto-upgraded on its own for the last days [12:14] ogra_: and check which channel you are tracking [12:14] i#'m actually logging in for the first time today [12:15] it did the auto-upgrade just fine until today it seems [12:15] zyga: re irssi-- is that your intended workflow? I use irssi and get proper highlighting. (I use '{ text = "jdstrand"; nick = "yes"; word = "yes"; },' in the 'hilights' section of config). I also use something called hilightwin.pl [12:16] (perhaps you are missing 'word = "yes"' for your nick) [12:18] hmm, interestingly the pi3 went properly to 766 on the edge channel [12:19] jdstrand: is there a way to reload config without quitting? :) [12:19] * zyga wishes for vim for irc [12:19] zyga: yes. /reload [12:19] * ogra_ tickles mvo [12:19] zyga: I think you may need to do /save [12:20] zyga: otherwise when you shutdown it might go away. I can't recall otoh [12:20] PR snapd#2028 opened: doc: added timezone-control to interfaces doc [12:21] PR snapd#2028 closed: doc: added timezone-control to interfaces doc [12:22] zyga: hilightwin.pl isn't needed for this, but I find it handy. you can put it in a window and all your highlights show up there. good for seeing what to tend to first in backscroll [12:22] mvo, did you do any channel fiddling for ubuntu-core today in the store ? [12:23] which could have caused the aboev ... [12:23] **above === hikiko|ln is now known as hikiko [12:27] mvo: hi! wondering if someone is assigned to snapd failing migration in yakkety-proposed for the last few weeks. I ask because bug #1626121, which has been fixed for a while in yakkety-proposed, keeps coming up in my circles (latest was running snapd in lxd) [12:27] Bug #1626121: strict mode snaps crash with Segmentation fault on 16.10

[12:27] I wonder if this causes a highlight (zyga) [12:27] hmmm [12:27] the zyga ! [12:27] nope :/ [12:28] let me restart [12:28] zyga the [12:28] what client ? [12:28] hi zyga! [12:28] irssi [12:28] grr [12:28] hmm [12:28] weird [12:28] nope [12:28] probably a typo [12:28] le me google [12:29] my config has "hilights = ( { text = "cjwatson"; nick = "yes"; word = "yes"; } );" and that works fine [12:29] sudo snap install hexchat ;) [12:29] probably /hilight -nick -word zyga or some such (possibly as two separate commands) [12:30] ogra_: one day :) [12:30] popey: I saw your game snapping video. It's really good. [12:30] haha [12:30] old habits you know [12:30] yeah, same here ... [12:30] (other way around though) [12:30] zyga: it could be a typo... I wonder if maybe in another part of your 'hilights' section [12:31] ogra_: new habits die fast? [12:31] who had the brilliant idea to call the option "hilights" [12:31] ah [12:31] that screams for typos ;) [12:31] sigh [12:31] yes [12:31] it does :) [12:31] re hilights> I know, right? [12:32] the fact that the config file is perl is nice too [12:32] hit me! [12:32] hey zyga [12:32] gaah [12:32] screw this ... [12:32] all work and no play makes zyga a dull boy [12:32] I'll get back to coding [12:32] is there a joke about writing an IRC client because irssi has a lousy config [12:33] zyga: I promise, it can be made to work :) (perhaps another time) [12:36] I've generally found that the path to wisdom in irssi is to forget about writing the config file by hand and to maintain it exclusively by running commands in irssi and then /save [12:36] far easier and you can test stuff on the fly [12:37] and as far as I'm aware you can do everything interactively like that anyway [12:38] cjwatson: that is probably very wise [12:38] it is really easy to mess up your config [12:38] hee, let's try this: snap-confine [12:38] hmm [12:38] hi zyga! [12:38] yes [12:38] it worked [12:38] \o/ [12:38] congrats [12:38] now I will catch all the people talking about snap-confine [12:38] thanks jdstrand :) [12:39] that is a mixed blessing ;) [12:39] zyga: you're welcome! :) [12:49] qengho: thanks! [12:53] zyga: so I see ther eis a bunch of snapd services. Which one I need to run with snapd? Anything I need to pass to it? can you give me the command line. Just upgraded everything for that [12:53] Bug #1628914 opened: ubuntu-core edge image switched to stable channel unpredictably and became unusable === chihchun is now known as chihchun_afk [12:56] sborovkov: it's not snapd, just run the program you care about [12:56] sborovkov: snapd doesn't participate in that [12:56] mvo, niemeyer, bug #1628914 seems pretty serious ... [12:56] Bug #1628914: ubuntu-core edge image switched to stable channel unpredictably and became unusable [13:00] ogra_: I promoted the ubuntu-core snaps from yesterday to beta, that was the only change I did [13:00] mvo, well, any idea how the above could happen ? [13:00] jdstrand: yaketty-proposed> sort of, autopkgtest is driving me nuts, there is a failure on i386 right now but its not happening in my adt VM that I use for testing [13:01] mvo: oh I hate that :( [13:01] ogra_: let me look [13:01] jdstrand: yeah, exactly, each adt run takes ~45min, so debugging that is a pain [13:01] the bug should have all info aggregated now [13:01] jdstrand: and its *only* failing on i386 and only on the actual adt runner on LP/whatever-stack, not locally [13:07] ogra_: can I get the uboot.bin too? or the output of fw_prrintenv? [13:07] sure [13:08] mvo, atteched on the bug [13:08] zyga: well... is not that arm version issue in snapd... I can't run my app easily. It's a bit complicated as it's buildrooted. [13:10] ogra_: so the ubuntu-core switched to "stable" which is strange [13:10] mvo, yeah [13:11] out of the blue [13:11] ogra_: the state has it set to "channel: stable" [13:11] ogra_: so somethng switched it, anything in history? [13:11] mvo, but beyond that switch ... the fact that snapd cant run when an older ubuntu-core is used is way more worrying [13:12] mvo, nothing ... i sideloaded kgunn's unity8-session snap and installed classic before to have wget available, thats all [13:13] beyond that i havent even logged in since i installed the board initially [13:15] ogra_: in a meeting right now, but it is definitely worrying [13:15] ogra_: channels should never switch except when the user explicitely sets it [13:15] ogra_: could you add the "history" ? [13:15] mvo, right ... and i assume snapd should stay executable even if i roll back [13:16] ogra_: that isa separate problem :/ [13:16] mvo, which history exactly ? i added state.json [13:16] ogra_: and a pretty big one actually [13:16] ogra_: shell history [13:16] ah [13:16] sure [13:16] ogra_: what commands might have caused this misbehavior [13:19] jdstrand: hey, I need a review on a patch for snap-confine that gustavo asked me to release quickly, I will add a few spread tests for it but can you have a quickly initial look at https://github.com/snapcore/snap-confine/pull/161 [13:19] PR snap-confine#161: Prefer the "core" snap is one is available [13:21] mvo, added [13:22] mectors: you might need to talk to zyga about the 'write' property of the content interface, as well as ways to share snap data files between snaps [13:24] sergiusens: for this failure: https://travis-ci.org/snapcore/snapcraft/jobs/163709462 is that I need to add snap as a dependency, or that I am not faking enough? [13:25] ralsina before going crazy, https://github.com/snapcore/snapcraft/pull/839 [13:25] PR snapcraft#839: Fixing 'sign-build' integration tests [13:25] sergiusens: phew [13:25] it's not that failure tho :-) [13:26] ogra_: thank you [13:26] np [13:26] ralsina doesn't seem related though [13:26] yeah [13:27] I am using "snap sign" on an assertion I generate that is predictable and the integration test passes locally [13:27] But on staging it's not finding the snap command [13:27] on travis, I mean [13:27] ralsina we don't install snapd for our unit tests [13:27] sergiusens: hmmm [13:27] sergiusens: ok, I can fake it [13:28] ralsina yeah, that would be better; signing on unit tests can get tricky [13:28] mectors: the write property cannot be used at this time, we should expand (and design this) the interface to allow to refer to wriable data as well as to readable data/ code [13:30] zyga, it should definitely be reflected in the documentation as it takes a lot of trial and errors to figure out that it doesn't work at all [13:33] oparoz: yes, I know, there are many things that need improvements [13:33] oparoz: if you want please improve the documentation straight away, it takes a moment to do this and we merge lots of stuff daily [13:33] I'm happy to submit a PR [13:34] Just not sure about the direction. Just remove that attribute? [13:35] zyga: done [13:36] oparoz: document what works today perhaps [13:36] oparoz: start with someting and let's see what we get in the feedback [13:36] jdstrand: thank you [13:39] oh, so that's how it sounds when a kernel doesn't boot [13:39] the beep [13:39] man [13:39] don't run your kernel while listening to meeting/hangout [13:45] Hi All, is still possible to create snaps for 15.04? If so, where could I find the correct snapcraft version? [13:46] clobrano: I believe the answer is yes, you will need an older version of ubuntu as a base though (perhaps 15.04, I don't recall) [13:58] zyga: I see, thanks! === pbek_ is now known as pbek [14:13] PR snapcraft#839 closed: Fixing 'sign-build' integration tests [14:16] PR snapd#2031 opened: store: apply deltas if explicitly enabled [14:21] PR snapd#2032 opened: tests: fix create-key/snap-sign test isolation [14:25] PR snapcraft#826 closed: Do not depend on Content-Length when Content-Encoding is gzip [14:28] ogra_ so snapcraft#838 solves it for you, right? [14:28] PR snapcraft#838: pluginhandler: take the file encoding into account [14:28] PR snapcraft#837 closed: kernel plugin: allow collecting the same mod deps [14:28] sergiusens, thats what i added to the PPA yesterday afaik [14:28] sergiusens, though with completely turned off tests during package build [14:28] ogra_ yeah, this doesn't have the prints and has unit tests ;-) [14:29] then it should be fine [14:29] ogra_ did you install the image onto your phone though? that is the important question [14:29] i'll happily tests it from -proposed once there is a package [14:29] i rarely install snappy images onto my phone ... [14:29]