/srv/irclogs.ubuntu.com/2020/06/01/#snappy.txt

=== benfrancis0 is now known as benfrancis
=== dennwc_ is now known as dennwc
=== Haxxa is now known as Guest67024
=== KindTwo is now known as KindOne
mupPR snapd#8773 opened: overlord/configstate: add sysctl option <Created by EthanHsieh> <https://github.com/snapcore/snapd/pull/8773>06:25
mborzeckimorning06:54
pstolowskimorning06:59
mborzeckipstolowski:hey07:02
mborzeckiwhat's the fuss about openttd disappearing from the store?07:02
zygagood morning07:03
zygamborzecki: ooh?07:03
zygawhat happened?07:03
mborzeckihttps://forum.snapcraft.io/t/openttd-is-gone/17860/207:04
zygainteresting07:06
mborzeckiheh too bad the tweet went out on 30.0507:07
* zyga returns to tool changes07:32
zygahttps://github.com/snapcore/snapd/pull/8691/files looks interesting!07:34
mupPR #8691: tests: plan to improve the naming and uniformity of utilities <Created by pedronis> <https://github.com/snapcore/snapd/pull/8691>07:34
zygapstolowski, mborzecki: do you guys want to have a look before I merge https://github.com/snapcore/snapd/pull/869107:52
mupPR #8691: tests: plan to improve the naming and uniformity of utilities <Created by pedronis> <https://github.com/snapcore/snapd/pull/8691>07:52
mborzeckizyga: it's 2 tabs away from me looking at it07:52
pstolowskizyga: i'm happy with it07:56
zygaok07:56
mupPR snapd#8691 closed: tests: plan to improve the naming and uniformity of utilities <Created by pedronis> <Merged by bboozzoo> <https://github.com/snapcore/snapd/pull/8691>08:20
zygamborzecki: replied on https://github.com/snapcore/snapd/pull/8770 but not going to change things until there's some more usage08:25
mupPR #8770: snap/naming: add ParseSecurityTag and friends <Created by zyga> <https://github.com/snapcore/snapd/pull/8770>08:25
zygamborzecki: my goal is to shring the 2K+ diff of refresh-app-awareness-v208:25
zygamborzecki: I'm open to renaming this later on though08:25
zygamborzecki: mainly for practicality but also because I'd like to see how we use security tags across the codebase08:26
zygapstolowski: ^08:26
zygapstolowski: not sure if you saw that PR yet08:26
zyga+ systemd-run --system --service-type=forking --unit=qemu-ndb-preseed.service '' --fork -c /dev/nbd0 /home/gopath/src/github.com/snapcore/snapd/tests/main/preseed/cloudimg.img08:28
zygaJob for qemu-ndb-preseed.service failed because the control process exited with error code.08:28
zygaSee "systemctl status qemu-ndb-preseed.service" and "journalctl -xe" for details.08:28
zygapstolowski: ^ should '' have had a value?08:28
zygato czesc https://github.com/snapcore/snapd/pull/876408:28
mupPR #8764: tests: add ubuntu 20.10 to spread tests <Created by sergiocazzolato> <https://github.com/snapcore/snapd/pull/8764>08:28
zygasorry, that's a part of https://github.com/snapcore/snapd/pull/8764 :)08:28
pstolowskizyga: it shouldn't fail, something is not ready for 20.10, i'll find out, thanks08:30
zygaok08:30
pstolowskizyga: yes, i've #8770 opened and started reviewving08:30
mupPR #8770: snap/naming: add ParseSecurityTag and friends <Created by zyga> <https://github.com/snapcore/snapd/pull/8770>08:30
zygamborzecki: https://github.com/snapcore/snapd/pull/8733 is fixed now08:30
mupPR #8733: tests: port document-portal-activation to session-tool <Test Robustness> <Created by zyga> <https://github.com/snapcore/snapd/pull/8733>08:30
zygamborzecki: I have 3 more patches (2 ready, one needs an hour to polish) and no more dbus-daemon leaks08:31
mupPR snapd#8713 closed: interfaces/avahi*: update avahi-daemon labelling to also allow "avahi-daemon" <Needs security review> <Squash-merge> <Created by jetpackdanger> <Merged by zyga> <https://github.com/snapcore/snapd/pull/8713>09:00
zygasigh09:08
zygawhy no os.ModeRegular09:08
mborzeckihaha09:17
mborzeckino other mode == regular09:18
mborzeckibtw. ModeIrregular is quite interesting tho09:18
zygamborzecki: yeah09:31
zygaaka "unix says something but we cannot map it"09:31
zygapstolowski, mborzecki: as for https://github.com/snapcore/snapd/pull/8770 -- I'd like to rename those later as they are unused now and the PR is green09:34
mupPR #8770: snap/naming: add ParseSecurityTag and friends <Created by zyga> <https://github.com/snapcore/snapd/pull/8770>09:34
zygaif you don't mind that09:34
mborzeckianyone using fish shell?09:36
zyganope09:37
mborzeckizyga: https://forum.snapcraft.io/t/how-to-use-default-commands-from-those-apps-which-are-installed-using-snap-without-having-to-use-snap-run-infront-of-original-command/17878/7 idk seems to work here quite fine09:41
mborzeckiheh i still find it funny when when the coolness factor is what makes the choice ;)09:42
zygamborzecki: I wonder if there's a difference if fish is your default shell09:43
zygawow09:43
zygaarguably fish looks awesome09:43
mborzeckizyga: idk, tried `fish -l` should give me a login shell, so a vanilla setup then?09:43
zygano09:43
zygadoubt it09:43
zygawell, maybe09:43
zygabut I doubt it :)09:43
mborzeckiok, let me switc the shell and try again09:45
zygaI just did09:45
mborzeckipffff09:46
mborzeckidoesn't work09:46
zygaIt matters09:49
zygaNo snaps in sight09:50
zygaheh09:50
mborzeckihmm09:53
mborzeckiok maybe we could drop somethign in /etc/fish/conf.d/snapd.fish ?09:53
mborzeckios better /usr/share/fish/vendor_conf.d09:55
ogradoes anyone knwo who creates the lxd user if i install the lxd snap ? is that lxd itself or snapd on behalf ?09:56
zygaogra: I think there was a hack of some kind09:59
zygamaybe snapd.deb does it09:59
zygaI don't recall09:59
zygabrb09:59
zygaback pain09:59
zygasuck a f*** annoying always-on pain :/09:59
ograzyga, well, it makes appliances with lxd preseeded unable to create a user09:59
ograi'm wondering against whom to file a bug 🙂10:00
ograoh man ... get well10:00
mborzeckizyga: hm looked at /usr/share/fish/config.fish but i can't tell where the PATH could get set10:01
zygawhen in doubt10:02
zygastrace and grep the source10:02
zygadoing that now10:02
zygabut really, fish is nice10:03
mborzeckizyga: config.fish seems to do soemthing, but it's not immediately obvious10:03
zygait handles path internally10:04
zygaat least a little10:04
zygalet me look at the source for a moment10:04
mborzeckianyways, fish env seems a bit off, even eshell includes the snap path, can't get more weird than that10:07
zygahttps://fishshell.com/docs/current/tutorial.html#path10:15
zygafish has some weird ideas though10:15
zygamaybe we should use that fish_user_paths thing10:16
zygamborzecki: check that out https://fishshell.com/docs/current/tutorial.html#universal-variables10:17
mborzeckizyga: hm interesting, stil it'd be nice to have it working, bu ti don't feel we need to fix it, i'm sure fish users have more experience as to what should be set where10:23
mborzeckior at least i hope so10:23
pstolowskizyga: +1 for the security tag PR10:36
* pstolowski lunch10:39
zygata10:40
zygafish_xdm_login_hack_hack_hack_hack10:47
zygathat's a function name:D10:47
mborzeckiprelude to a lot of fun i presume?10:47
zyganah, not this time10:49
zygabut funny10:49
zygait runs for login shells10:50
zygahmm11:01
zygamborzecki: this is unexpected11:01
zygasystemctl --user show-environment11:01
zygaPATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games11:01
zygano snap11:01
zygawhat the?!11:01
mborzeckihm?11:01
mborzeckiintersting11:01
zygafish is not the problem something happened when I switched shells11:01
zygaI strongly think it's systemd11:01
* zyga looks11:01
zygainterestingly XDG_DATA_DIRS=/usr/local/share/:/usr/share/:/var/lib/snapd/desktop11:02
zyga(from the same output)11:02
mborzeckizyga: i did chsh and the only change i had is missing snap from path, user session looked ok11:02
mborzeckii mean user session env11:02
mborzeckibut i did not reboot11:02
=== benfrancis8 is now known as benfrancis
zygaI rebooted11:03
zygabut this is good because nothing in fish itself seems broken11:03
zygaI wonder if setting a shell like "gulash" which is just a hardlink to bash would change something11:04
mborzeckizyga: idk, show-environment showed the right PATH, fish had differetn PATH, seems broken to me11:04
zygamborzecki: for me show-environment shows that11:05
zygahttps://www.irccloud.com/pastebin/4Y7rV8b6/11:05
mborzeckizyga: is snapd installed? are there any other user-session-env generators running?11:06
zygaeverything is installed11:06
zygathis is my regular systewm11:06
zygaI just changed shell and rebooted11:07
=== benfrancis9 is now known as benfrancis
mborzeckican't reboot this system, but i'll try a fedora vm11:08
zygaok,11:08
zygaI'll keep digging11:08
zygajust need to stretch11:08
zygagrab coffee11:08
zygaand take the dog out11:08
zygasee you in 1 ~ h11:08
mborzeckizyga: fish looks broken to me: https://paste.ubuntu.com/p/DMcVX2CBnW/11:09
zygaWhy did our results differ?11:22
zygain the office11:52
zygawhat a lousy day :/11:52
pstolowskire12:05
* ogra files #188158812:06
mupBug #1881588: pre-seeding lxd on Core appliances breaks snap create-user <snapd:New> <https://launchpad.net/bugs/1881588>12:06
mborzeckizyga: i have no clue, that's f32 with fish shell set as default12:07
zygamborzecki: mmm, I'll dig later12:09
mborzeckizyga: but guess what, on arch after a reboot there's no snap in PATH in show-environment12:10
zygamborzecki: oh?12:17
zygamborzecki: so arch and fedora differ12:17
mborzeckizyga: hah, check this out12:19
mborzeckizyga: this is on arch: https://paste.ubuntu.com/p/Btwnx85TM8/12:20
zygao_O12:20
zygapam talking to systemd?12:20
mborzeckizyga: something imported environment when i logged in via gdsm12:20
zygacheck out /etc/pam.d/12:21
zygafor both the login session getty12:21
zygaand the one for the display manager12:21
mborzeckizyga: should i be looking for anything in particular?12:22
zygayes, diff the two files used12:22
zygaone will import more things12:22
zygacan you find the shorter one first12:22
zygathe one from getty12:22
zygahow did you log in?12:22
zygavia getty or via ssh?12:22
mborzeckizyga: ssh12:22
zygassh is easy to find12:23
zygafind the other one12:23
mborzeckiwhen logged in via getty, the env is ok too12:24
mborzeckizyga: hmm there's like 4-5 gdm-* files though12:25
mborzeckii guess gdm-password is the one12:25
mborzeckizyga: https://paste.ubuntu.com/p/gT3dCBB5gJ/12:25
mborzeckizyga: seems pretty similar, only difference is gdm opens gnome-keyring12:28
zygahmmm12:31
zygahmm12:31
mborzeckithere's also gdm-launch-environment but no clue when that applies12:32
mborzeckiit clearly looks like something pulls in the environment from the shell and plugs it into the session12:32
zygare12:35
zygawife home, sorry12:35
mborzeckiheh, need to change shell to somethinf sane, the completion that fish does is super annoying12:35
zygahttps://twitter.com/zygoon/status/126642724978133401812:36
zygasmall distraction12:36
=== facundo__ is now known as facubatista
zygafood time :)13:36
pstolowskihmm debian-sid failing on nfs-support test? anyone seen this?13:48
pstolowskimount.nfs: an incorrect mount option was specified13:48
pstolowskiit's "+ mount -t nfs localhost:/home /home -o nfsvers=3,proto=udp"13:49
zygahmmm13:52
zygamaybe something changed13:52
pstolowskii'll find out13:55
ijohnsonzyga: yes I think qemu on arm64 rpi should work with "new" kernels, i.e. the one on 20.04+ I think14:17
ijohnsonzyga: as per https://bugs.launchpad.net/ubuntu/+source/linux-raspi2/+bug/1783961 should be in 5.3.0+ actually14:18
mupBug #1783961: CONFIG_KVM is disabled for linux-raspi2 (aarch64) <linux-raspi2 (Ubuntu):Fix Released> <linux-raspi2 (Ubuntu Eoan):Fix Released> <https://launchpad.net/bugs/1783961>14:18
zygaI’ll check that shortly14:24
zygaInstalling gobs of packages14:24
mupIssue pc-amd64-gadget#48 opened: console=ttyS0 is too slow and useless <Created by xnox> <https://github.com/snapcore/pc-amd64-gadget/issues/48>14:32
cachiopstolowski, hey, #8764 updated14:52
mupPR #8764: tests: add ubuntu 20.10 to spread tests <Created by sergiocazzolato> <https://github.com/snapcore/snapd/pull/8764>14:52
pstolowskicachio: thanks, will take a look14:53
cachiopstolowski, tx14:53
mupPR snapd#8774 opened: tests: disable test of nfs v3 with udp proto on debian-sid <âš  Critical> <Created by stolowski> <https://github.com/snapcore/snapd/pull/8774>15:17
pstolowski^ this should unblock master15:18
ijohnsonthanks pstolowski15:18
pstolowskicachio: there is a typo in you find command, qemu-nbd vs qemu-ndb, perhaps it is in qemu-utils after all?15:19
zygathanks pawel15:19
cachiopstolowski, ouch, let me check15:19
pstolowskicachio: it's confusing because we have this typo in a f ew places (e.g. unit names), but we call the right command at the end15:19
cachioah, I'll check15:20
oSoMoNogra, the zoom snap was refreshed yesterday on my wife's laptop and it stopped working (wouldn't start at all), I had to revert it15:21
oSoMoNknown issue?15:21
ograoSoMoN, nope, havent heard of it15:21
oSoMoNogra, the laptop is in use right now so I can't get details, but I will poke at it later15:22
ograoSoMoN, https://github.com/ogra1/zoom-snap/issues/new ... and attach ~/snap/zoom-client/current/.zoom/logs/zoom-terminal.log from a failing run please15:23
oSoMoNack, will do15:23
=== KindTwo is now known as KindOne
mborzeckiijohnson: would appreciate if you could later take a look at https://github.com/snapcore/snapd/pull/877515:31
mupPR #8775: [RFC] bootloader, boot: boot scripts, edition <Needs Samuele review> <UC20> <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/8775>15:31
mborzeckii can split it later to smaller bits if that helps land it sooner15:31
mupPR snapd#8775 opened: [RFC] bootloader, boot: boot scripts, edition <Needs Samuele review> <UC20> <Created by bboozzoo> <https://github.com/snapcore/snapd/pull/8775>15:32
cachiopstolowski, you were righ15:34
cachioright15:34
cachiotest paseed15:34
pstolowskicachio: nice!15:35
cachioI reverted part of the change and pushed again15:35
cachiothanks15:35
pstolowskiyw15:35
ijohnsonmborzecki: sure will take a look at that today15:39
mborzeckiijohnson: thanks!15:45
ijohnsonmborzecki: hmm from first glance were we going to install both ubuntu-seed and ubuntu-boot grub.conf's though?15:45
ijohnsonI thought we were only going to do ubuntu-boot15:45
mborzeckiijohnson: it's called for boot only now, but eventually we'd do both15:47
ijohnsonhmm15:47
ijohnsonmborzecki: but how would that work for i.e. custom u-boot in a gadget ?15:47
* cachio lunch15:48
mborzeckiijohnson: that's tbd, it will likely be limited to gadgets we control15:49
ijohnsonmborzecki: mmm I see15:49
mborzeckiijohnson: iow grub only for the time being ;)15:49
ijohnsonright :-)15:49
mborzeckianywyas, eod time, going to take the kids out for some ice cream15:50
mupPR snapcraft#3150 opened: cli: error/warn when using sudo <Created by cjp256> <https://github.com/snapcore/snapcraft/pull/3150>15:53
zygadiddledan: openttd --edge crashes on pi416:04
abeatoI'm seeing this problem with the docker snap, both in my laptop and in an arm64 machine: docker: Error response from daemon: OCI runtime create failed: container_linux.go:345: starting container process caused "exec: \"docker\": executable file not found in $PATH": unknown.16:04
abeatohttps://paste.ubuntu.com/p/4MthKrcd3w/16:04
abeatoreally weird16:04
zygaabeato: are you seeing any denials?16:07
abeatozyga, some, but I do not think they are related: https://paste.ubuntu.com/p/RHpfd7fRQ8/16:08
mupPR snapcraft#3151 opened: cli: disable --target-arch support on core20 <Created by cjp256> <https://github.com/snapcore/snapcraft/pull/3151>16:08
mupPR snapcraft#3152 opened: cli: disable --target-arch for multipass/lxd <Created by cjp256> <https://github.com/snapcore/snapcraft/pull/3152>16:08
zygahmm16:09
zygaI think it is trying to execute /snap/bin/docker16:09
zygaabeato: does this make sense to you? I think PATH is set up wrong and it searches and finds itself instead of the embedded binary16:10
pstolowskioh fun, debian-sid now failed on google:debian-sid-64:tests/main/command-chain:reexec116:11
zygapstolowski: how?16:11
pstolowskizyga:  https://pipelines.actions.githubusercontent.com/xS8oSnypZkPEQZqiZgDaRp2kdvQJKbOY08TesHp7E8vn7g4hYR/_apis/pipelines/1/runs/6044/signedlogcontent/59?urlExpires=2020-06-01T16%3A11%3A12.8468116Z&urlSigningMethod=HMACV1&urlSignature=Ip%2BQIctB93sO65jma2G8J0FSRPdRG2xYwo2iB6MJdY4%3D16:12
abeatozyga, not sure - but it looks like a problem with the docker snap: I saw this on an arm64 machine, then I tried and saw the same in my laptop. I'll post in the forum anyway16:12
zygapstolowski: expired URL16:12
pstolowskiweird.. copy: https://paste.ubuntu.com/p/R74RtVYXkd/16:13
pstolowskizyga: ^16:13
zygahmmm16:15
zygano idea16:15
zygabut look a that uber long line16:15
zygawith recent test history16:15
mupBug #1611424 changed: Additional "lts" channel or support for upstream series <landscape> <lxd> <nova-lxd> <openstack> <Snappy:Fix Released> <https://launchpad.net/bugs/1611424>16:15
mupBug #1611424 opened: Additional "lts" channel or support for upstream series <landscape> <lxd> <nova-lxd> <openstack> <Snappy:Fix Released> <https://launchpad.net/bugs/1611424>16:21
mupBug #1611424 changed: Additional "lts" channel or support for upstream series <landscape> <lxd> <nova-lxd> <openstack> <Snappy:Fix Released> <https://launchpad.net/bugs/1611424>16:24
zygaijohnson: hey, not sure if you have a moment, I'd like to land https://github.com/snapcore/snapd/pull/8770 to open some follow-ups18:23
mupPR #8770: snap/naming: add ParseSecurityTag and friends <Created by zyga> <https://github.com/snapcore/snapd/pull/8770>18:23
ijohnsonzyga: hey18:23
ijohnsonsure I can take a look this PM18:23
zygait's an extension of the existing validator to a parser for security tags18:23
ijohnsonthat was broken out of the big refresh app awareness PR right?18:23
zygayes18:23
zygaI will tweak naming in follow ups18:23
ijohnsoncool, yeah I remember the code as it was in that PR so hopefully shouldn't be too alien, my memory allowing :-)18:23
zygaI'm trying to land pawel's PR to unbreak master so I'd rather not push more here until that lands :)18:24
ijohnsonof course18:24
zygaijohnson: it's not that much new code, thanks!18:24
ijohnsonbtw, how does the github caching work when a PR is closed / re-opened ?18:24
ijohnsonI assume some keys set per the job are different so nothing gets cached18:25
zygaijohnson: I think it works the same way but the key is in the hash key18:25
zygaand IIRC we use the github job ID18:25
zygaso I think this gives us a new job18:25
ijohnsonhmm yeah that's kinda what I would expect18:25
ijohnsonnot sure if it would be desirable to have jobs cached across closing/re-opening18:25
ograzyga, in case you are bored with your pi4 ... here is something to play with https://people.canonical.com/~ogra/snappy/appliances/fabrica/18:57
cmatsuokaijohnson: adding a new group to extragroups with the same gid solved the audio access problem19:28
ijohnsoncmatsuoka: nice, did you have to remove the one from /etc/group too ?19:28
ijohnsonor did everything just work with the duplicated definitions ?19:28
cmatsuokaijohnson: no, I just added a new one. Probably the name doesn't matter as long as it has the same id19:28
cmatsuoka(in my case I added another audio group)19:29
zygaeh, mouse stopped working19:29
zygathinkpads with linux :|19:29
ijohnsonmmmm19:29
* ijohnson -> short break19:29
mupPR snapcraft#3149 closed: elf: search dynamic tags within sections, not segment <Created by cjp256> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/3149>19:54
mupBug #1875493 changed: [core] log rotation doesn't properly restart rsyslogd <Snappy:Invalid> <rsyslog (Ubuntu):Invalid> <https://launchpad.net/bugs/1875493>20:06
ijohnsonthanks for the review jdstrand!20:07
jdstrandnp20:08
mupBug #1875493 opened: [core] log rotation doesn't properly restart rsyslogd <Snappy:Invalid> <rsyslog (Ubuntu):Invalid> <https://launchpad.net/bugs/1875493>20:09
mupBug #1875493 changed: [core] log rotation doesn't properly restart rsyslogd <Snappy:Invalid> <rsyslog (Ubuntu):Invalid> <https://launchpad.net/bugs/1875493>20:12
mupPR snapcraft#3152 closed: cli: disable --target-arch for multipass/lxd <Created by cjp256> <Merged by sergiusens> <https://github.com/snapcore/snapcraft/pull/3152>20:14
mupBug #1875493 opened: [core] log rotation doesn't properly restart rsyslogd <Snappy:Invalid> <rsyslog (Ubuntu):Invalid> <https://launchpad.net/bugs/1875493>20:18
mupBug #1875493 changed: [core] log rotation doesn't properly restart rsyslogd <Snappy:Invalid> <rsyslog (Ubuntu):Invalid> <https://launchpad.net/bugs/1875493>20:21
ijohnsonzyga: #8770 is approved, shall I merge it for you?20:21
mupPR #8770: snap/naming: add ParseSecurityTag and friends <Created by zyga> <https://github.com/snapcore/snapd/pull/8770>20:21
mupBug #1875493 opened: [core] log rotation doesn't properly restart rsyslogd <Snappy:Invalid> <rsyslog (Ubuntu):Invalid> <https://launchpad.net/bugs/1875493>20:27
mupBug #1875493 changed: [core] log rotation doesn't properly restart rsyslogd <Snappy:Invalid> <rsyslog (Ubuntu):Invalid> <https://launchpad.net/bugs/1875493>20:30
=== KindTwo is now known as KindOne

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