/srv/irclogs.ubuntu.com/2019/09/26/#snappy.txt

mupPR snapd#7512 closed: interfaces/docker-support,kubernetes-support: misc updates for strict k8s <Created by jdstrand> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/7512>04:37
mupPR snapd#7513 closed: interfaces/docker-support,kubernetes-support: misc updates for strict k8s - 2.42 <Created by jdstrand> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/7513>04:39
mborzeckimorning05:02
mupPR snapd#7514 opened: cmd: add `snap debug bootvars` that dumps the current bootvars <Simple 😃> <Created by mvo5> <https://github.com/snapcore/snapd/pull/7514>05:19
mvohey mborzecki05:19
mborzeckimvo: hey05:19
mborzeckimvo: should we allow dumping of all bootvars too?05:20
mborzeckimvo: since we've taken away all the tools to inspect them05:21
mvomborzecki: we don't have an interface in our bootloaders to say "give-me-all"05:22
mvomborzecki: plus uboot will have a whole lot of unrelated stuff05:23
mvomborzecki: but yeah, it could be useful but I am not sure how to do it?05:23
mborzeckimvo: i see, teh Bootloader interface would need to be extended05:26
mborzeckimvo: otoh, we build a map of all bootarg=value anyway, at least for grub and uboot05:27
mvomborzecki: yeah and its a bit unclear how this would apply to e.g. lk05:27
mborzeckimvo: i suppose for lk it'd be ok to return just snap_*, reboot_reason and bootimg_file_name, there does not seem to be a genereic environment mechanism like uboot has05:29
mvomborzecki: right05:30
mborzeckimvo: can you take a look at #7509? i still need to address the panics in the tests, we have helpers for packing snap files in the test but the test data is incomplet witht he checks i added05:32
mupPR #7509: gadget, snap/pack: perform extended validation of gadget metadata and contents <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/7509>05:32
mvomborzecki: sure05:33
mborzeckimvo: thanks!05:33
mvomborzecki: is the test failing for silly reasons in that pr, i.e. does it just need a restart?05:51
mvomborzecki: I had some green runs this morning :)05:51
mborzeckimvo: it's failing bc of the PR, i need to push one more fix there06:10
mvomborzecki: ok06:14
mborzeckimvo: the checks have failed in #7514, something about a cmd/snap-seccomp/s.c in the tree06:51
mupPR #7514: cmd: add `snap debug bootvars` that dumps the current bootvars <Simple 😃> <Created by mvo5> <https://github.com/snapcore/snapd/pull/7514>06:51
mvomborzecki: oh, woah06:51
mvomborzecki: I did not touch this, let me look at the failure06:51
mvomborzecki: this test failure is bizare06:53
mvomborzecki: looks like 7166 can almost land (just two small comments from samuele). thanks again for picking this one up!07:05
=== pstolowski|afk is now known as pstolowski
pstolowskimorning07:05
mvohey pstolowski07:05
mborzeckimvo: yup will address those soonish07:05
mborzeckipstolowski: hey07:05
mupPR snapd#7515 opened: daemon: return "snapname_rev.snap" style when using /v2/download <Simple 😃> <Created by mvo5> <https://github.com/snapcore/snapd/pull/7515>07:12
mvothanks for your review pedronis \o/07:15
zygaHey guys. I will file today off, if feel much worse than yesterday.07:18
zygaTime to really get better07:19
mvozyga: uh, get some rest07:20
pstolowskihi, get well zyga!07:20
pedronismvo: hi, made some initial comments asking for changes also in #751007:24
mupPR #7510: daemon: make /v2/download take snapDownloadOptions <Created by mvo5> <https://github.com/snapcore/snapd/pull/7510>07:24
mvopedronis: thanks07:25
pedroniszyga: get well!07:27
mupPR snapd#7434 closed: seed/seedwriter: implement WriteMeta and tree16 corresponding code <Created by pedronis> <Merged by pedronis> <https://github.com/snapcore/snapd/pull/7434>07:46
pedronis#7441 and #7467 are ready for review07:53
mupPR #7441: asserts,seed/seedwriter: follow snap type sorting in the model assertion snap listings <Simple 😃> <Created by pedronis> <https://github.com/snapcore/snapd/pull/7441>07:53
mupPR #7467: seed/seedwriter: resolve channels using channel.Resolve* for snaps <Created by pedronis> <https://github.com/snapcore/snapd/pull/7467>07:53
pedronismvo: couple more comment in 7514, I missed the Bootvars the first time around08:00
mvopedronis: thanks, fixing that now as well08:02
mborzeckimvo:  are you planning to pick bootvars into 2.42?08:19
mvomborzecki: not really, why?08:22
pedronismvo: is John swapping today?08:23
mborzeckimvo: #7508 is tagged for 2.42, wondering whether to land it now or wait for bootvars to land first, 7508 get updated and have both cherry-picked to 2.4208:23
mupPR #7508: tests: skip the ubuntu-core-upgrade on arm devices on core18 <Simple 😃> <Created by sergiocazzolato> <https://github.com/snapcore/snapd/pull/7508>08:23
mvopedronis: let me check, iirc no but I could be wrong08:24
mvomborzecki: yeah, should be ok to skip on 2.42 as long as we fix master08:24
mborzeckimvo:  ok, landing 2.42 then08:25
mborzeckiduh, landing 750808:25
mvomborzecki: heh, ok08:25
mvomborzecki: I was worried for a moment08:25
mvopedronis: nothing in the calendar08:25
mvopedronis: maybe just a bit late?08:25
mupPR snapd#7508 closed: tests: skip the ubuntu-core-upgrade on arm devices on core18 <Simple 😃> <Created by sergiocazzolato> <Merged by bboozzoo> <https://github.com/snapcore/snapd/pull/7508>08:26
mborzeckimvo: on that note, can you cherry-pick bb963563d0a8a1ad406a512c539072ca24ba8336 from that PR into 2.42 ?08:26
mvomborzecki: sure thing08:27
popeyzyga: friend of mine noticed snaps seem a broken on 19.10 - and found this bug https://bugs.launchpad.net/snappy/+bug/1656340  - any ideas?09:01
mupBug #1656340: XDG_RUNTIME_DIR is not created on app startup <bionic> <cosmic> <eco-team> <xenial> <Snappy:Triaged by zyga> <snapd (Ubuntu):Confirmed> <https://launchpad.net/bugs/1656340>09:01
Chipacastgraber: is running nested snapped lxds supported? asking because #1777017 so either there are more steps needed, or we're doing something wrong09:04
mupBug #1777017: snap install lxd doesn't work within a container <AppArmor:Invalid> <snapd:New> <snapd (Ubuntu):New> <https://launchpad.net/bugs/1777017>09:04
mupBug #1622336 changed: [DOC] snappy on KVM recommends cmd line that warns about RAW devices <Snappy:Fix Released> <https://launchpad.net/bugs/1622336>09:32
mupBug #1621769 changed: impossible to create a model assertion compatible with both snapd 2.13 and 2.14 <Snappy:Won't Fix by mvo> <https://launchpad.net/bugs/1621769>09:35
mupPR snapd#7516 opened: daemon: allow /v2/assertions/{assertType} to query store <Created by mvo5> <https://github.com/snapcore/snapd/pull/7516>09:39
Chipacapedronis: #1662534 is interesting, passphrases can't end in whitespace because of how we use gpg10:14
mupBug #1662534: snap create-key misses space at end of passphrase <snapd (Ubuntu):Confirmed> <https://launchpad.net/bugs/1662534>10:14
Chipacaor because gpg :-)10:14
mborzeckiwow, whitespace in passwords10:14
pedronisChipaca: sounds strange, I don't think we are in charge of the i/o there10:17
pedronisbut maybe I'm missing something10:18
Chipacapedronis: as I say there, gpg strips whitespace from the batch mode thing10:18
Chipacait's documented under "unattended gpg key generation" in the info pages10:19
pedronisChipaca: ok, sounds low to me10:19
Chipacatotes :-)10:19
Chipacabut it's the sort of thing that can have you lose a lot of time if you don't know about it10:20
Chipacaand as we are in control of _reading_ the passphrase, maybe we could warn?10:20
pedronissounds still low10:20
pedronisChipaca: are you asking me if you should had a warning today10:21
pedronis?10:21
pedroniss/had/add/10:21
Chipacapedronis: no10:21
Chipacai agree it's low10:21
Chipacai'd say the low thing is to warn / block a passphrase we know is not going to work10:21
Chipacasomehow fixing it so it works would be a "nah" :)10:21
pedronisthat can be even medium10:22
pedronisChipaca: I'm just a bit missing what input you need from me right now?10:22
Chipacapedronis: nothing, just sharing an obscure bit of info10:22
Chipacaotherwise it's just me that knows this :)10:22
Chipacame and launchpad10:22
pedronisok, maybe I will retain it, or not :)10:23
Chipacaheh10:23
Chipacai forget not everybody's brain is a sponge for stupid trivia10:23
Chipacapstolowski: https://paste.ubuntu.com/p/qy4Y3mgm2h/10:25
Chipacapstolowski: you're probably the most comfortable in that bit of code (configstate), care to look and see if that panic could still happen today?10:26
pstolowskiChipaca: ah, you got me nervous for a second, it is a very old bug isn't it?10:27
Chipacapstolowski: yes :-)10:28
pstolowskiChipaca: bug# for more context? i'll check it10:28
Chipacapstolowski: #167050110:28
mupBug #1670501: invalid memory address or nil pointer dereference in transaction.go <snapd (Ubuntu):New> <https://launchpad.net/bugs/1670501>10:28
pstolowskiChipaca: ty10:29
Chipacapstolowski: if you answer, start by apologising for us being bad at bugs10:29
Chipacaplease :-)10:29
pstolowskiChipaca: i'll be very humble10:31
cjwatsonwhite space> TIL10:34
Chipacazyga: #1681068 is all yours10:34
mupBug #1681068: Unable to use content interface with read-write source paths bind mounted over read-only targets <snapd (Ubuntu):New> <https://launchpad.net/bugs/1681068>10:34
Chipacazyga: and #1681099 i guess10:35
mupBug #1681099: Bind mounting is not performed when a directory created in a config hook is used with the content interface <snapd (Ubuntu):New> <https://launchpad.net/bugs/1681099>10:35
Chipacapstolowski: another one, looks to be the same thing: #169976810:43
mupBug #1699768: "snap set" causes snapd crash <snapd:Fix Released by stolowski> <snapd (Ubuntu):New> <snapd (Ubuntu Xenial):New> <https://launchpad.net/bugs/1699768>10:43
pedronisChipaca: zygmunt is off sick today10:47
* Chipaca assigns all the bugs to him10:48
pstolowskiChipaca: thanks, i'll double check and mark as dupe if needed10:52
mupPR snapd#7514 closed: cmd: add `snap debug boot-vars` that dumps the current bootvars <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/7514>10:54
mupPR snapd#7517 opened: run-checks: allow overriding gofmt binary, show gofmt diff <Simple 😃> <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/7517>11:19
mupPR snapd#7499 closed: many: move AppArmor probing code under sandbox/apparmor <Created by bboozzoo> <Merged by bboozzoo> <https://github.com/snapcore/snapd/pull/7499>11:20
mupPR snapcraft#2731 closed: meta: take no command-chain being prepended into account <Created by sergiusens> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/2731>11:36
stgraberChipaca: that user is attempting to run a nested container without having set security.nesting to true, so that's not going to work12:10
Chipacastgraber: is that something we can detect to point them in the right direction?12:10
stgraberChipaca: not particularly easily as for security reasons one normally cannot inspect their own apparmor policy12:11
Chipacastgraber: thank you (for the explanation, and the update on the bug)12:12
pstolowskihmm anyone having issues with github not offering just pushed branch for a PR?12:13
pstolowskinvm, it suddenly decided to show them.. must have been a temporary hiccup12:38
Chipacayeah github was broken for a bit12:45
mupPR snapd#7518 opened: cmd/snap: sort tasks in snap debug timings output <Created by stolowski> <https://github.com/snapcore/snapd/pull/7518>12:54
mupPR snapd#7519 opened: tests: use `snap debug boot-vars` in "ubuntu-core-upgrade" test <Created by mvo5> <https://github.com/snapcore/snapd/pull/7519>12:59
Chipacahuh, getting "this meeting has been updated to use Hangouts Meet" from the standup13:00
pstolowskiChipaca: same13:00
degville^ me too!13:00
Chipacaand the organizer is probably mvo :-|13:00
mborzeckiduh13:00
ijohnsonit had to happen sooner or later13:01
mvoChipaca: meh13:01
mborzeckiam i the only one there?13:01
ijohnsonhangouts actually died13:01
Chipacaso, let's just use https://meet.google.com/rne-gzqm-kgg?authuser=1&pli=113:01
mborzeckihmmm, that's different link than i have13:01
Chipacamborzecki: you have a link?13:01
mvoChipaca: I updated the inviite13:01
mborzeckiChipaca: typed in snappy-devel like i always did13:01
Chipacacachio: https://meet.google.com/agi-vdpd-myw?authuser=113:02
mupPR snapd#7511 closed: tests: when the backend is external skip the loop waiting for snap version <Created by sergiocazzolato> <Merged by bboozzoo> <https://github.com/snapcore/snapd/pull/7511>13:05
mvo7515 needs a second review13:06
Chipacamborzecki: which is the one that i need to review?13:08
mborzeckiChipaca: https://github.com/snapcore/snapd/pull/716613:08
mupPR #7166: client: add doTimeout to http.Client{Timeout} <Created by mvo5> <https://github.com/snapcore/snapd/pull/7166>13:08
=== ricab is now known as ricab|lunch
mborzeckimuch fun when the schedule is crossing the month boundary and the current day is on the other month13:33
Chipacamborzecki: let's talk about ISO week numbers13:34
mborzeckihahah13:34
Chipacajust not today13:34
mvoChipaca: if you want to get something fun to start reviews with, pick 751513:35
mvoChipaca: shouldn't be more than 5min of work13:35
Chipacamvo: my bias detector just exploded in a cloud of sparks13:35
Chipacamvo: *but* i had that one open already so eh13:35
mborzeckiand gnome-shell spamming the logs every second, wtf13:41
Chipacamborzecki: but it has all this information you need to know about!13:52
mborzeckiChipaca: repeated every second :/13:53
mborzeckiobviously there's no way to have journald drop those messages, now i can enjoy all of my journal history filled with that crap13:54
mupPR snapd#7515 closed: daemon: return "snapname_rev.snap" style when using /v2/download <Simple 😃> <Created by mvo5> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/7515>13:58
mupPR snapd#7348 closed: snapstate,store: check assumes early before downloading the snap <⛔ Blocked> <Created by mvo5> <Closed by mvo5> <https://github.com/snapcore/snapd/pull/7348>14:00
asymptoticallyhi all, i am trying to install microk8s via snap and it explodes while trying to make some certificates, here is the log: http://paste.debian.net/1102701/14:06
asymptoticallyi have tried the stable and edge channels but both fail, and now i'm not sure what else to try14:06
asymptoticallyand sorry if this is the wrong place. i couldn't find a channel for microk8s and guessed that this would be the next best place :)14:07
Chipacaasymptotically: whoa, that's not good14:08
Chipacakjackal: is that something you know about? ^^14:09
asymptoticallywhoops, forgot to add that i am using snap 2.41 on debian 1014:09
ograasymptotically, did you try --stable instaed of --edge ?14:09
ogratypically the edge channel is for active development14:10
kjackalreading14:10
asymptoticallyyeah, it gets the same error :( - i had a quick search through github issues and google, but it doesn't seem like anything similar has been reported (or my google-fu is weak)14:10
kjackalasymptotically: can you try --channel=1.15/stable?14:11
kjackal18747114:12
kjackalsorry14:12
asymptoticallythanks kjackal, seems like 1.15 installs and runs fine14:13
kjackalI hope these library issues will go away when we go to strict confinement14:13
asymptoticallyshould i just report this as a bug on the microk8s issue page? or would it be a problem with my system14:13
kjackalasymptotically: can you please open an issue at: https://github.com/ubuntu/microk8s/issues14:13
kjackalasymptotically: I need to try it out14:14
asymptoticallyhttps://github.com/ubuntu/microk8s/issues/67914:16
kjackalgreat thank you14:17
ograjdstrand, did you forget about my "/sys/firmware/devicetree/base/soc/ranges" and "/sys/firmware/devicetree/base/system/linux,revision" addition to hardware-observe, or am i just to impatient (just making your it didnt get forgotten)14:23
jdstrandogra: not forgotten, just not done yet14:25
jdstrand(it is in trello with other updates that are accumulating)14:25
ograah, ok ... if there is a trello card i dont need to nag ... sorry then :)14:26
jdstrandogra: I added you to it14:27
ograthanks !14:27
cwayneNow ogras gonna write a nagbot that comments on a Trello card every hour :p14:28
ograhaha14:28
ograautomation FTW !!!14:28
cwayneWe literally have a nagbot for open MPs14:29
ograroadmr, yo ... i tried to register a snap with "core-*" in the name today and got a "this name is reserved" ... after dropping the prefix (and calling my snap just ldap-server which is rather a bit too generic ... ) it worked fine ... do we have a global catchall thigie for anything prefixed with "core" ?14:30
roadmrogra: we do ;)14:30
ograouch, k14:30
roadmrogra: if you really want core-whatevs we can grant it, you just need to request the name14:31
* ogra makes a mental note for next time14:31
roadmr(well if you really want it and have a good justification for it :) I want many things but...14:31
ograyeah, i was to lazy for buerocracy :P14:31
=== ricab|lunch is now known as ricab
roadmraskocracy works better :)14:31
ograhahaha14:31
ograanyway, its ldap-server for now ... if i get people using it oout of context i'll probably have to rename14:32
cwayneI prefer ircocracy14:32
ograbut i expect people to complain then14:32
Chipacahmm, i should probably have breakfast14:34
* Chipaca goes14:35
joedborghey all, could anyone let me know how to remove interfaces with the network-control plug (if possible)?14:44
joedborg`ip link delete` doesn't seem to work14:44
pedronis#7441 needs a second review (it's very small)14:47
mupPR #7441: asserts,seed/seedwriter: follow snap type sorting in the model assertion snap listings <Simple 😃> <Created by pedronis> <https://github.com/snapcore/snapd/pull/7441>14:47
pstolowskipedronis: done14:47
pedronispstolowski: thank you14:48
mupPR snapd#7441 closed: asserts,seed/seedwriter: follow snap type sorting in the model assertion snap listings <Simple 😃> <Created by pedronis> <Merged by pedronis> <https://github.com/snapcore/snapd/pull/7441>14:48
pstolowskifun fact, new spread test in my branch was failing on 14.04 on journalctl match, because without "-a" flag journalctl would cut off one of the critical log lines14:50
pstolowskipedronis: which of your PRs is next?14:52
pedronis#746714:52
mupPR #7467: seed/seedwriter: resolve channels using channel.Resolve* for snaps <Created by pedronis> <https://github.com/snapcore/snapd/pull/7467>14:53
pstolowskik14:53
mupPR snapd#7520 opened: tests: moving opensuse to unstable due to catastrofic issue on the repo <Created by sergiocazzolato> <https://github.com/snapcore/snapd/pull/7520>14:54
mupPR snapd#7517 closed: run-checks: allow overriding gofmt binary, show gofmt diff <Simple 😃> <Created by bboozzoo> <Merged by mvo5> <https://github.com/snapcore/snapd/pull/7517>14:56
Chipacajoedborg: why doesn't it work?15:02
ograor better "how" doesnt it work (whats the error ?)15:08
joedborgChipaca ogra15:09
joedborgso the issue is that it doesn't appear to be deleting the interfaces, not thowing any errors15:09
ogradid you monitor journalctl ?15:09
ograi cant imagine it just quietly fails15:10
pedronisChipaca: I asked for a review for https://github.com/snapcore/snapd/pull/7467 mostly because with default track/channel normalisation that code will need some care again, it has +2 though so I'm going to land it, you can look at it anyway15:10
mupPR #7467: seed/seedwriter: resolve channels using channel.Resolve* for snaps <Created by pedronis> <https://github.com/snapcore/snapd/pull/7467>15:10
Chipacapedronis: ta15:10
mupPR snapd#7467 closed: seed/seedwriter: resolve channels using channel.Resolve* for snaps <Created by pedronis> <Merged by pedronis> <https://github.com/snapcore/snapd/pull/7467>15:12
pedronismvo: pstolowski: I rebased the next one #7468 , this one is a bit more substantial given it add support for local snaps15:18
mupPR #7468: seed/seedwriter: support local snaps <Created by pedronis> <https://github.com/snapcore/snapd/pull/7468>15:19
pstolowskity15:19
mvopedronis: looking15:19
Chipacapedronis: given snap.MountFile(foo, 5) gives us "/path/to/foo_5.snap", what would be a good name for something that returned just foo_5.snap given those args?15:26
pedronisChipaca: we have also a lot of code that does filepath.Base(info.MountFile()) fwiw15:27
Chipacait's a silly thing but i'm itching to do this; we've got %s_%s.snap in two different non-test places, and filepath.Base(snap.MountFile(...)) in something like 1215:27
Chipacapedronis: quite15:27
pedronisChipaca: my worry here is that the surface of these things is already quite big15:27
pedronisthe next person will likely do "%s_%s.snap" again15:28
Chipacanot for the first time i wish go had some sort of a path interface that would hide this sort of stuff :-)15:28
Chipacapedronis: yeah15:28
Chipacait's already hard to figure out what you need, there15:28
Chipacafair point15:28
pedronisChipaca: which is the other place doing %s_%s.snap, I'm not finding it15:31
pedronisI just see one in info15:31
pedronis.go15:31
Chipacahmm, didn't m vo just add one?15:40
mvocyphermox: hey, I'm curious about the status of the netplan.io SRU. when do you think it will hit xenial-updates? mostly wondering because we prepare a new core stable release currently and it would be great if this fix would be part of it (cc ijohnson)15:41
pstolowskicachio: have you seen ijohnson's question to #7520?15:43
mupPR #7520: tests: moving opensuse to unstable due to catastrofic issue on the repo <Created by sergiocazzolato> <https://github.com/snapcore/snapd/pull/7520>15:43
cyphermoxmvo: gah, I wish I could tell you better news but tbh it slipped my mind -- the SRU was uploaded to the unapproved queue but never reviewed15:44
cyphermoxmvo: I'll try to get that in -proposed today and then we can see what we can do about this15:45
cyphermoxmvo: ETA for your stable release?15:45
Chipacapedronis: looks like i broke something in the latest policy change :-) i'm digging into that15:45
mvocyphermox: in a meeting right now will get back to you. rough eta is early next week so probably a bit too early for you15:47
pedronisChipaca: as in spread test break but not unit tests?15:56
Chipacapedronis: yeap15:56
Chipacapedronis: broke google:ubuntu-core-16-64:tests/main/ubuntu-core-upgrade15:56
Chipacacan't remove core:x3 despite it being not current and not in use15:57
pedronisok15:57
pedronisseems we are missing unit tests then15:57
Chipacayeap :-)15:57
ograis there any particular reason why on many systems snap version doesn't list the kernel version ?16:03
ograit feels like every second/third time i ask someone on the forum to provide snap version i get no kernel line returned16:03
Chipacaogra: people remove it because they consider it sensitive information16:04
Chipacaogra: or, maybe it's too old?16:04
Chipacaogra: but it has been there since forever16:05
ograyes, i know16:05
ograit just feels like it is missing more often recently16:06
Chipacaogra: since 2.23 at least16:06
Chipacaogra: but this is 2.21 :-/16:06
ograyeah16:06
ograsince he cant install core he has the deb version16:07
pstolowskiyay, #7355 finally green \o/16:07
mupPR #7355: interfaces/apparmor: load multiple profiles in a single batch <Complex> <Created by stolowski> <https://github.com/snapcore/snapd/pull/7355>16:07
Chipacaogra: looks like no squashfs supportfs16:08
Chipacaffs16:08
ijohnsonpstolowski: yay do you need a review on that ?16:09
ograChipaca, yep, like all pixelbooks/chromebooks16:09
ograseems there are more and more users of these devices though :(16:09
ograand the archive all these debian VMs use doesnt have squashfuse either16:10
mvocyphermox: meeting over. just read backlog. thanks for the update, would be great to have it in -proposed soon(ish). I will ponder a bit over this, we *might* upload it to our image ppa to get it even if its not in -updates yet but that would be a bit of a last-resort kind of thing16:10
mvocyphermox: please keep me updated on this one (and I will poke you again gently early next week :)16:11
pstolowskiijohnson: yes please, thanks!16:11
ijohnsonack will do today, probably after your EOD16:11
* Chipaca adds 220V to mvo's stick-of-poking16:11
* ijohnson adds US compatible 120V inverter on top16:13
* mvo hugs Chipaca and his enchanted +5 stick-of-management16:13
Chipacamvo: you forgot *glowing*16:14
cachiopstolowski, answered16:14
cachiowas in a meeting16:14
cachiothanks for the heads up16:14
* cachio out to get kids from school16:15
cachioI'll be back in 2016:15
pstolowskicachio: ty. fwtw i just had green run of spread on one of my PRs16:15
cachiopstolowski, the error is not happenning in 100% of the runs but is still happening16:16
cachioperhaps we need to wait today16:16
cachiodoes it makes sense?16:16
jdstrandzyga: hey, afaics, opensuse still has 2.39.1. I can never remember, it doesn't reexec, correct?16:17
ijohnsonjdstrand: zyga is off sick today16:17
jdstrandzyga: nm. feel better16:17
Chipacasiiigh16:17
Chipacai'm going to go for a run, maybe everything will be better when i get back16:18
jdstranddoes anyone remember if opensuse reexecs?16:18
ijohnsoncachio: yes that makes sense, I'll check in with you before EOD to see if opensuse repos are still not working16:18
ograChipaca, like ... bojo the clown stepped down ?16:18
jdstrandfedora is still on 2.39.216:18
jdstrandI know it doesn't reexec16:18
Chipacajdstrand: https://github.com/snapcore/snapd/blob/master/cmd/cmd_linux.go#L6916:19
stgraberyeah, suse & fedora is what's holding LXD on core1616:20
stgraber(we need 2.40 for the fix of base handling)16:20
jdstrandChipaca: I thought manjaro reexec'd...16:20
* jdstrand shrugs16:20
jdstrandChipaca: thanks :)16:20
* jdstrand is curious about when audio-playback will be everywhere16:21
jdstrandWimpress, popey: hey, do you have an in with electron? could you poke someone on https://github.com/electron-userland/electron-builder/issues/4234?16:22
jdstrandkenvandine: would you mind looking at https://github.com/ubuntu/snapcraft-desktop-helpers/pull/197 ?16:23
mupPR ubuntu/snapcraft-desktop-helpers#197: snapcraft.yaml: add audio-playback for audio <Created by jdstrand> <https://github.com/ubuntu/snapcraft-desktop-helpers/pull/197>16:23
jdstrandChipaca: btw, thanks for https://forum.snapcraft.io/t/show-interfaces-of-snap-before-installation/13252/316:25
jdstrandChipaca: that is quite handy for the time being16:25
Chipacajdstrand: note for really old snaps it won't work (they'll have an empty snap-yaml)16:26
Chipacabut, yeah16:26
jdstrandChipaca: it was a nice advertisement for the 'http' snap as well ;)16:27
jdstrandChipaca: cunning16:27
Chipacajdstrand: that's probably the only big caveat (the other is to note that it is per revision and the code i posted just grabs the first one)16:27
Chipacajdstrand: nah, regular http (via apt install httpie) can do that as well; if I plug the http snap it's using snapd:///... :-p16:27
Chipacabut yes :-D16:28
roadmrjdstrand: review-tools  20190924-2239UTC is now in production \o/16:30
ijohnsonpstolowski, any idea why systemd.Stop uses s.systemctl("stop",...) but everything else seems to use s.systemctl("--root",s.rootDir,...) ?16:33
ijohnsonshould stop (and also start I guess) use the --root opt as well16:34
pstolowskiijohnson: let me check the code.. i probably touched this long time ago is that right?16:35
ijohnsonpstolowski: well not sure if you touched this particular bit of code, but I think you had to do the same thing for snapctl stop , etc.16:37
ijohnsonjust not sure if there's ever a reason we need to _not_ use --root when we talk to systemctl16:38
ijohnsonseems like we should always use --root16:38
pstolowskiijohnson: i definately touched services wrt snapctl, yes, was long time ago16:38
om26erDid I report the bug at the correct place https://bugs.launchpad.net/snapd/+bug/1845533 ? Or should that be reported for the distro package ?16:38
mupBug #1845533: snap info gets confused if there is a directory of the same name <snapd:New> <https://launchpad.net/bugs/1845533>16:38
pstolowskiijohnson: yes i agree it looks suspicious, should probably be consistent, i'm not sure if there is a reason for that16:39
pstolowskiprobably an omission16:39
pstolowskieod for me16:40
pstolowskio/16:40
=== pstolowski is now known as pstolowski|afk
ijohnsonack, thanks pstolowski16:40
om26erIs snapcraft autobuild trigger broken ? I made a code change to the master of a project that has autobuilds enabled but it didn't trigger even after 15 minutes, I now triggered the build manually.16:45
kenvandinejdstrand, done16:56
pedronisChipaca: I think you fixed this recently https://launchpad.net/bugs/1845533 ?17:03
mupBug #1845533: snap info gets confused if there is a directory of the same name <snapd:New> <https://launchpad.net/bugs/1845533>17:03
jdstrandkenvandine: thanks!17:10
cwaynekenvandine: yo, your jobs should be fixed and running now17:15
kenvandinecwayne: thanks17:16
Chipacapedronis: yep, fixed in master18:22
Chipacapedronis: actually, no18:29
Chipacapedronis: the bug that's fixed in master is a different one18:29
Chipacapedronis: the one in that report is not a bug at all18:29
Chipaca"when I do snap info to a directory that isn't a snap, snap info gets confused: it says the directory isn't a snap!"18:31
Chipacaer...18:31
Chipaca...what's the bug again?18:31
Chipacaom26er: ^^ that's you i'm talking about :-)18:32
om26erChipaca you only need to run snap info for a snap that is not installed currently and have a directory of that name created18:33
om26ercd ~; mkdir slow-snap; snap info slow-snap18:34
Chipacaom26er: is 'slow-snap' a snap that exists18:34
om26erslow-snap is a snap that I publish but you will get the complain that `error: no snap found for "slow-snap/"`18:34
om26erChipaca, yes it exists and you can repeat that issue with any snap that is currently not installed18:35
Chipacaholy regressions batman18:35
Chipacawhen did this break18:35
Chipacaom26er: thank you. I need to figure out what's going on. I'll update the bug with this information.18:36
om26ergreat, you are welcome :-)18:36
om26erShall I change the bug to confirmed ?18:37
om26erjdstrand this got rejected in a funny way (after first getting manual approval) https://dashboard.snapcraft.io/snaps/deskconnd/revisions/49/18:47
om26ercontext: https://forum.snapcraft.io/t/auto-connection-request-for-deskconnd/1336418:47
Chipacahmmmmmmmmm18:55
Chipacaom26er: could you confirm for me please that if you 'snap refresh core --beta' (or --edge), such that 'snap version' tells you you are on 2.42 pre, then the issue goes away?18:57
Chipacai think i might have fixed this without meaning to :-)18:58
* Chipaca will still create a PR with a test for this18:58
om26erI assume you meant core18 instead18:58
Chipacayou assume wrong :-)18:59
om26erhah, ok, core it is then18:59
Chipacaom26er: unless you have the 'snapd' snap, in which case that is the one to refresh18:59
Chipacabut you'd have to turn that on with an experimental flag afaik18:59
Chipacaso probably not18:59
Chipaca(i think the change to it not being experimental is in .42)18:59
om26erChipaca yes, seems to be fixed with core from beta channel19:02
Chipacahuzzah19:03
Chipacaom26er: that's going out to candidate and then stable as soon as a couple of integration ducks get in line, hopefully it's not inconveniencing you too much until then19:07
om26eryeah, its a v.small issue, compared to others that my softwares are facing right now ;-)19:08
* Chipaca hugs om26er 19:09
* om26er hugs Chipaca back19:09
om26erits been a while since my last IRC hug19:11
om26erroadmr are you around ? I am kind of blocked as all my snaps got auto-rejected (piconn, deskconn and deskconnd) :-)19:16
roadmrom26er: omg that's horrible :(19:16
roadmrom26er: let's have a look19:16
om26er```declaration malformed (unknown constraint key 'plug-publisher-id') declaration-snap-v2_valid_plugs (content, allow-auto-connection_plug-publisher-id)```19:17
roadmrright - it was jdstrand who wrangled those declarations19:17
roadmrlet's see if I can figure out what the issue is19:17
roadmrremember I said cross-publisher autoconnects were tricky? :)19:18
om26erroadmr indeed, you said that ;-)19:19
roadmrom26er: I think we'll need jdstrand to have a look - I can't spot anything wrong :/ he should be around in a bit19:30
om26erroadmr no problem, thanks for looking into it. I'll be on for a while so I hope I'll catch Jamie then19:31
roadmrsure; I see you posted on the forum so worst case, he'll see that later19:31
jdstrandom26er, roadmr: looking19:41
jdstrandI know the problem19:43
om26erI guess that's a good thing ?19:43
jdstrandyes19:44
jdstrandgimme a sec19:44
jdstrandom26er: ok r49 is fixed. let me fix the other revisions and snaps19:46
om26erwohoo! you are a life saver19:46
jdstrandroadmr: there was a typo in the wiki (https://wiki.canonical.com/AppStore/Reviews?action=diff&rev2=247&rev1=246)19:47
jdstrand(that's an internal url, sorry)19:47
roadmrjdstrand: oh! let's see!19:48
roadmrjdstrand: I checked that but I guess if the typo was on the *reference* document I was never going to spot that :)19:48
roadmrah, that makes sense :) thanks19:48
jdstrandom26er: ok, deskconnd is done. let me adjust the others19:51
jdstrandom26er: deskconn fixed and r133 approved19:53
jdstrandom26er: piconn fixed and r4 approved19:54
om26erjdstrand is it normal that auto-connection no longer works ?19:55
jdstrandom26er: gpiod fixed (nothing to approve, but next upload will auto-approve)19:56
jdstrandom26er: the bug in the snap decl would've caused that19:56
jdstrandom26er: with what I just fixed, it should work19:56
jdstrandom26er: sorry for the hiccup. these snap declarations can be rather delicate19:56
jdstrandom26er: please test to verify it is working as expected19:57
om26erjdstrand my snaps are not auto connecting to crossbar snap now19:58
om26erto reporduce `snap install crossbar --edge; snap install deskconnd --edge`19:59
* jdstrand is looking20:02
jdstrandom26er: there is an interplay here since you are wanting two different content slots auto-connected from two different publishers while the snap is providing a content interface itself20:05
jdstrandgimme a sec20:05
om26eryeah, here is the outlook20:06
om26ercrossbar: provides runtime environmentdeskconnd: connects to crossbar, provides its own unix domain socket as welldeskconn: connects to crossbar and uses unix domain socket provided by deskconndpicon: connects to crossbar and uses unix domain socket provided by deskconnd20:07
jdstrandom26er: I'm going to be fiddling with the snap declaration for deskconnd for a minute. if you get emails, don't worry about it20:07
om26erjdstrand sure, no problem20:08
om26ersorry for those pastes above, I am a little too used to Slack and didn't realize, IRC doesn't support much formatting20:08
jdstrandoh20:09
jdstrandom26er: slots:20:09
jdstrand  runtime-dir:20:09
jdstrand    content: executables20:09
mupBug #1845555 opened: snap refresh leaves behind old .mount files in /etc/systemd/system/ <Snappy:New> <https://launchpad.net/bugs/1845555>20:09
jdstrandom26er: 'executables' is what is needed, not 'runtime-dir'20:09
jdstrandom26er: did you change that? I didn't notice it before20:09
om26erno, didn't change that, its been like that from the beginning20:10
jdstrandom26er: ok, adjusting20:10
jdstrandcrossbar:runtime-dir                       deskconnd:crossbar-runtime-dir20:11
jdstrandlet me adjust the others20:11
Chipacacould somebody on 18.04 check the 'wallstreet' snap (and deb) and confirm whether it works there?20:14
jdstrandom26er: should be fixed: https://paste.ubuntu.com/p/zPqQ4cvz6P/20:15
jdstrandom26er: I adjusted piconn and gpiod too, but didn't test as they are armhf20:15
om26erjdstrand thank you, I tested deskconnd and deskconn they work great now. Will test piconn on the raspberry Pi20:15
jdstrandom26er: cool, you're welcome and sorry it wasn't a totally smooth experience. the docs are updated for next time20:16
jdstrandChipaca: I happened to be ssh'd into a bionic vm, so snap installed wallstreet (eek, --classic!) and it just hangs20:18
Chipacajdstrand: does it then OOM?20:18
jdstrandI don't know what it is supposed to do, but it seems that probably isn't it20:18
Chipacajdstrand: https://mobile.twitter.com/DustinKirkland/status/117731132281340314720:19
jdstrandjeez, first --classic, now twitter?20:19
jdstrand:)20:19
Chipacanot sure why that needs classic TBcmfH20:19
jdstrandChipaca: the deb does things that look like the tweet20:20
* jdstrand removes that and tries the snap again20:20
jdstrandok, the snap continues to not do anything20:21
jdstrandit doesn't seem to be leaking memory20:22
Chipacathank you20:22
ChipacaI wonder if that can be strictly confined20:23
ChipacaI don't see why not20:23
ChipacaI mean, it ships tmux20:23
jdstrandit seems to be using screen and/or tmux20:23
Chipacabut that's not necessarily classic, right?20:23
jdstrandso, 'no'20:23
jdstrandit is classic20:23
Chipacajdstrand: right, but it has tmux inside of itself20:24
jdstrandI mean, it must today be classic20:24
Chipacajdstrand: and uses them to run things from itself20:24
jdstrandyes but tmux does stuff that the sandbox disallows20:24
Chipacajdstrand: that works, surely?20:24
Chipacaahh20:24
Chipacaohhh20:24
jdstrandyou cannot run tumx in a snap20:24
Chipacabooooh tmux boooooh20:24
jdstrandor screen20:24
om26erjdstrand tested on the pi, things are fixed. Thank you again :)20:24
jdstrandthere are setuid/setgid bits and other things20:24
jdstrandom26er: woohoo! :)20:24
* om26er has a small home automation project that has UbuntuCore18 running and controls GPIO pins on the RPi for a few months20:26
jdstrandom26er: neat :)20:28
jdstrandChipaca: super terse notes detailing my recent exploration of a 'tmux-support' interface: https://paste.ubuntu.com/p/Rnf6Pfz85y/20:28
Chipacajdstrand: :-(20:29
jdstrandChipaca: I time-boxed it and found that there were a number of issues with the terminal in that env that would have to be debugged20:29
jdstrandChipaca: I was able to get some stuff going though. if I looked really hard at 'script' and learned some pty/tty fu, maybe...20:32
Chipaca:)20:33
* jdstrand maintains that screen and/or tmux should be in the core snap (which doesn't help wallstreet one bit)20:33
Chipacajdstrand: i thought we'd agreed on that already?20:33
mupBug #1845555 changed: snap refresh leaves behind old .mount files in /etc/systemd/system/ <snapd:Invalid> <https://launchpad.net/bugs/1845555>20:33
jdstrandChipaca: I thought the 'agreement' was 'no' :P20:33
jdstrandif it was 'yes', then yay!20:34
Chipacait wouldn't be the first time i got those two words mixed up20:34
Chipacapedronis will know more20:34
Chipacai'll try to bug him about it20:34
jdstrandcool, thanks :)20:34
jdstrandoh I left wallstreet running this whole time. still no OOM20:35
* jdstrand kills it and tears down the vm20:35
Chipacajdstrand: you commie you20:37
Chipacatearing down wallstreet20:37
* roadmr is now known as GordonGekko20:38
Chipacaroadmr: i expect you to update your profile picture accordingly20:39
jdstrandhehe20:39
* roadmr visits gravatar.com20:40
roadmrhttps://bit.ly/1KwjXmW <- whatcha think about this? not sure about the suspenders20:41
Chipacaroadmr: https://www.mirror.co.uk/news/uk-news/boris-johnson-vs-gordon-gekko-286080120:46
roadmroh hahaha what20:46
Chipacaroadmr: also https://wallstreet.fandom.com/wiki/Gordon_Gekko?file=Gordon_Gekko.jpg20:47
Chipacaanyway, i'm out of here20:48
Chipaca👋20:48
roadmro/ enjoy20:48
* roadmr scored 9/10 in the quiz \o/20:49
jdstrandroadmr: thanks for pushing 20190924-2239UTC :)21:05
jdstrandjoedborg: fyi, ^ has the override you need for your snap to pass automated review21:05
joedborgah awesome, thanks jdstrand roadmr21:06
joedborgI'm just removing all the unnecessary layout rules and I'll push again21:06
roadmr👍21:06
jdstrandjoedborg: yep, cool :)21:06
mupPR snapd#7521 opened: [RFC] many: use --root as first arg to systemctl everywhere <Created by anonymouse64> <https://github.com/snapcore/snapd/pull/7521>21:33
joedborgroadmr: jdstrand: managed to clear out alot of the layout https://github.com/charmed-kubernetes/snap-kubernetes-worker/commit/0d9548f866ae5114c64e99f039640553d60d804a21:38
roadmr\o/21:38
joedborgthe snap is uploading rn21:39
joedborgroadmr jdstrand rev 2 pushed21:53
jdstrandjoedborg: that's excellent!22:14
jdstrandjoedborg: I see it passed automated review too :)22:15
joedborgjdstrand: yeah :D22:20
joedborgJust need Snapcraft io to work now so I can schedule automated builds22:20
mupPR snapcraft#2732 opened: project: use os.sched_getaffinity instead of multiprocessing.cpu_count <Created by kyrofa> <https://github.com/snapcore/snapcraft/pull/2732>22:28
mupPR snapcraft#2733 opened: cmake plugin: support disable-parallel option <Created by kyrofa> <https://github.com/snapcore/snapcraft/pull/2733>22:52
mupPR snapd#7522 opened: httputil: set user agent for CONNECT <Created by chipaca> <https://github.com/snapcore/snapd/pull/7522>23:02

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!