| === amurray` is now known as amurray | ||
| === pstolowski|afk is now known as pstolowski | ||
| pstolowski | morning | 07:17 |
|---|---|---|
| mvo | hey pstolowski, good morning | 07:18 |
| mvo | 07:24 | |
| Chipaca | moin moin | 07:56 |
| Chipaca | ...rude. | 07:58 |
| mvo | Chipaca: hey good morning! | 07:59 |
| mvo | Chipaca: rude? | 08:00 |
| Chipaca | mvo: joke about somebody leaving without a word just after I said hello | 08:01 |
| mvo | haha | 08:01 |
| === jacekn_ is now known as jacekn | ||
| Chipaca | mvo: do you ready z_yga's "+1 but" as an actual +1, i.e. can I merge #5524? | 08:31 |
| mup | PR #5524: cmd/snap: check for typographic dashes in command <Created by chipaca> <https://github.com/snapcore/snapd/pull/5524> | 08:31 |
| Chipaca | read* | 08:31 |
| Chipaca | mvo: also, did we reach a consensus on the flag to watch for "do not error if none found"? I know we agreed it wasn't --oknodo :-) | 08:32 |
| Chipaca | was thinking of --no-error-on-empty taking a page from xargs, but it does seem a bit verbose | 08:33 |
| Chipaca | alternatively, and perhaps more naturally given the flag is only for use with --last, we could add syntax to --last, e.g. «--last=foo?» vs «last=foo» | 08:35 |
| mvo | Chipaca: I'm not great with names but --no-error-on-empty seems to be ok (yes verbose but). alternatives might be "--empty-ok" | 08:35 |
| mvo | Chipaca: I like "foo?" | 08:35 |
| mvo | Chipaca: not sure about zygas +1, but he should be online soon(ish) | 08:36 |
| Chipaca | sorted :-) | 08:36 |
| mvo | Chipaca: ta | 08:36 |
| Chipaca | mvo: ah, i thought they were further west | 08:36 |
| mvo | Chipaca: its 6h from here, so I would expect (jetlag and all that) in ~2h | 08:40 |
| zyga | 4am-orning | 08:52 |
| * ogra_ grumbles about his kiosk app being rejected because of missing .desktop file ... how pointless | 08:53 | |
| Chipaca | hehe | 08:54 |
| Chipaca | zyga: morning | 08:54 |
| Chipaca | zyga: we were just talking about you and jetlag | 08:54 |
| Chipaca | ogra_: wat? | 08:54 |
| Chipaca | ogra_: what's rejecting your app because of that? | 08:55 |
| zyga | Store Review tools | 08:55 |
| zyga | How are you guys doing? | 08:55 |
| ogra_ | yeah ... it uses an x11 loopback interface to itself ... the review tools block if you dont have a .desktop file when this interface is found | 08:55 |
| Chipaca | zyga: in #5224, does your "+1 but" mean an actual +1? | 08:55 |
| Chipaca | wait | 08:55 |
| Chipaca | zyga: #5524 | 08:55 |
| * Chipaca kicks mup | 08:55 | |
| zyga | Mup is partially off | 08:56 |
| ogra_ | jetlag | 08:56 |
| zyga | Chipaca: that depends, if you agree then you can tweak the feature, if you don’t you can treat my review as +1 | 08:57 |
| Chipaca | zyga: I disagree (explained a bit why on the pr) | 08:57 |
| zyga | Ok | 08:57 |
| Chipaca | zyga: (also i brought this other approach up with Juanreal before implementing it) | 08:57 |
| zyga | I haven’t read that yet | 08:57 |
| mup | PR #5224: interfaces: remove Plug/Slot types <Created by stolowski> <Merged by stolowski> <https://github.com/snapcore/snapd/pull/5224> | 08:58 |
| mup | PR #5524: cmd/snap: check for typographic dashes in command <Created by chipaca> <https://github.com/snapcore/snapd/pull/5524> | 08:58 |
| zyga | mvo: how is core18? | 09:00 |
| mvo | zyga: mostly happy, the snap-stop-mode test was still racy though, looks like output related, I added marker files and with that I got no failures yet (running with -repeat 100) | 09:15 |
| mvo | zyga: the integration PR is failing with what looks like unrelated issues | 09:16 |
| zyga | Thank you! That is good news | 09:16 |
| mvo | zyga: I am discussing kernel track names right now, I added you to the mail thread and maybe you can ask Gustavo to have a look so that we can make a decision. its one of the blockers for the model assertion for core18 | 09:16 |
| zyga | How about the core18 snapd hosting slots? | 09:17 |
| mvo | zyga: you mean 5527? or something else? | 09:17 |
| zyga | Yes | 09:19 |
| Chipaca | snap set system refresh.hold=+2h | 09:19 |
| zyga | I saw it was red | 09:19 |
| Chipaca | why can't we do that :-) | 09:19 |
| zyga | Didn’t check deeper yet | 09:19 |
| pedronis | Chipaca: the plan is to have a command to do that sort of stuff, was blocked on deciding how to call it and the fact we have too many commands | 09:33 |
| Chipaca | pedronis: I could add it to corecfg (if we ignore that the function is called 'validate' and not 'validate-and-normalise' | 09:34 |
| Chipaca | ) | 09:34 |
| pedronis | please don't | 09:35 |
| * Chipaca quietly does 'git reset --hard' | 09:35 | |
| mvo | zyga: aha, indeed, its red because settle is not converging | 09:36 |
| mvo | zyga: we need to either fake the production snapd id | 09:36 |
| mvo | zyga: or add the test snapd-id to the detection if its the snapd snap, we had this issue before I think | 09:36 |
| zyga | Ah, I remember now | 09:36 |
| zyga | Can you patch that in some acceptable way | 09:37 |
| zyga | I’d love to say “we can boot core18 and it works” | 09:37 |
| mvo | zyga: I can add a patch, we will see if that is "acceptable" ;) | 09:39 |
| mvo | zyga: just working on the sru for 2.34.1 will do in a little bit | 09:39 |
| zyga | Thank you :-) | 09:39 |
| zyga | Ok | 09:39 |
| * Chipaca steels himself to go to the dentist | 10:01 | |
| zyga | mvo: feedback from the field: we should have a workflow where developers can copy a new kernel, snap install it and reboot | 10:08 |
| mvo | zyga: to test new kernels? | 10:09 |
| zyga | currently that doesn't work and requires time-consuming workarounds or use of ubuntu classic | 10:09 |
| zyga | yes | 10:09 |
| zyga | to do kernel development | 10:09 |
| zyga | (driver etc) | 10:09 |
| mvo | makes sense, we have some restrictions here currently mostly to avoid users doing strange things | 10:09 |
| zyga | mhm | 10:10 |
| zyga | maybe just --dangerous --very | 10:10 |
| zyga | or something like this | 10:11 |
| pstolowski | --scary | 10:19 |
| zyga | --I-know-what-you-did-last-summer | 10:23 |
| pstolowski | zyga: how is the sprint going? | 10:24 |
| zyga | very good so far | 10:24 |
| zyga | we have one interesting thing that was discussed yesterday evening | 10:24 |
| zyga | we will change how refresh works | 10:24 |
| zyga | I will share all the details later today/tomorrow (as time permits) | 10:24 |
| zyga | it will involve a new hook as well | 10:24 |
| pstolowski | interesting | 10:25 |
| pstolowski | pedronis: RFC - https://github.com/snapcore/snapd/pull/5532 | 10:54 |
| mup | PR #5532: api/connect: report "nothing to do" if attempting to re-connect <Created by stolowski> <https://github.com/snapcore/snapd/pull/5532> | 10:54 |
| pedronis | pstolowski: mmh, that's a change of behavior | 10:56 |
| pedronis | I think the old code just did nothing, not an error | 10:56 |
| pstolowski | pedronis: indeed, the old code would just do nothing in the handler, but created the change etc. | 10:58 |
| pedronis | pstolowski: we have a similar case already | 10:58 |
| pedronis | pstolowski: we do this: https://github.com/snapcore/snapd/blob/master/daemon/api.go#L1436 but I think Connect shoudl return either an empty ts or some kind of error | 11:01 |
| pedronis | to trigger that behavior | 11:01 |
| pstolowski | pedronis: yes i considered empty ts, but wasn't sure a dummy change with 'done' status was a good idea (forgot we have a case already) | 11:03 |
| pstolowski | pedronis: okay, i will do it that way then | 11:05 |
| === pstolowski is now known as pstolowski|lunch | ||
| * Chipaca lunch | 11:46 | |
| Chipaca | hmm, clearing refresh.hold with "" seems to be broken | 11:48 |
| Chipaca | 2018/07/18 12:48:39.556782 stateengine.go:101: state ensure error: internal error: cannot unmarshal snap "core" option "refresh.hold" into *time.Time: parsing time """" as ""2006-01-02T15:04:05Z07:00"": cannot parse """ as "2006", json: "" | 11:49 |
| Chipaca | hmm | 11:49 |
| * ogra_ curses loudly ... | 11:57 | |
| ogra_ | the damned review tools are making me mad today | 11:58 |
| popey | Don't get mad, get even. | 11:58 |
| ogra_ | i think i tried evyery possible way to ship a pointless .desktop file for my kiosk daemon app now | 11:58 |
| ogra_ | but they all got rejected | 11:58 |
| * ogra_ has no idea how to proceed with that | 11:59 | |
| ogra_ | snap/gui/ doesnt work, specifying it in snapcraft.yaml and shipping it doesnt ... not sure what else to try | 11:59 |
| popey | why not just put a valid desktop file? | 12:00 |
| ogra_ | ? | 12:00 |
| popey | youre talking about a "pointless" desktop file | 12:00 |
| popey | why don't you do what we do for every other snap and put a valid one in there | 12:00 |
| ogra_ | well, i use an IMHO valid file | 12:00 |
| ogra_ | the point is that this is a daemon | 12:00 |
| ogra_ | it doesnt need a desktop file since it cant be run from it | 12:00 |
| popey | can I see the project/yaml/desktop file? | 12:00 |
| ogra_ | https://github.com/ogra1/xdmcp-client | 12:01 |
| popey | i always put the desktop file in snap/gui/foo.desktop | 12:01 |
| ogra_ | (i originally shipped it in snap/gui, just moved it arouond to toplevel and defined it via deskto: ) | 12:01 |
| popey | and never mention it in the yaml | 12:01 |
| popey | it gets automatically found | 12:01 |
| ogra_ | yes | 12:01 |
| ogra_ | thats what did first | 12:01 |
| ogra_ | *I | 12:02 |
| popey | your icon path is invalid | 12:02 |
| popey | in the desktop file | 12:02 |
| popey | put .desktop and .png in snap/gui, and make sure they are both called <snapname>.* | 12:02 |
| ogra_ | yeah, now it is :( ... (wasnt when it was in snap/gui | 12:02 |
| ogra_ | 9 | 12:02 |
| ogra_ | thats what i had in the former revision ... | 12:02 |
| ogra_ | like in all other desktop snaps i create ... | 12:03 |
| ogra_ | but that just caused the very same review error | 12:03 |
| ogra_ | popey, https://github.com/ogra1/xdmcp-client/tree/c70eafedf3ea751730677a843782f567d7c2d816 was my last revision | 12:03 |
| popey | filename is wrong | 12:04 |
| popey | for the icon, make it xdmcp-client.png | 12:04 |
| ogra_ | <snapname>.desktop | 12:04 |
| popey | and refer to it as such in the desktop file | 12:04 |
| ogra_ | well, icon.png works for all other desktop snaps i have ... but happy to try | 12:04 |
| niemeyer | mvo, nessita: Heya | 12:05 |
| popey | you can run the snap review tools locally to test it | 12:05 |
| popey | rather than do the store turn around time | 12:05 |
| niemeyer | snap info on snaps published by canonical is showing that long email address instead of snaps@canonical.. | 12:05 |
| niemeyer | nessita: What do we need to do to fix that? | 12:06 |
| nessita | niemeyer, we should reindex the snaps, let me do that | 12:06 |
| niemeyer | nessita: Thanks! | 12:07 |
| nessita | niemeyer, have an example of a snap showing the wrong email address? | 12:09 |
| === abeato_ is now known as abeato | ||
| === pstolowski|lunch is now known as pstolowski | ||
| zyga | nessita: core | 12:25 |
| nessita | zyga, I see. That's the contact url of a snap which is set by the publisher and is not automatically changed when SSO profile data changes | 12:27 |
| popey | https://snapcraft.io/core could do with a metadata update too! as could https://snapcraft.io/canonical-livepatch | 12:28 |
| ogra_ | the core text came originally from mark FWIW | 12:31 |
| ogra_ | (i had a way longer text for the initial uploads) | 12:31 |
| === ondra_ is now known as ondra | ||
| popey | That was when we only had "snap info" though, the web store needs more text to make it look compelling. | 12:37 |
| Chipaca | not sure core needs to be compelling :-) | 12:38 |
| nessita | zyga, niemeyer I filed an RT to have the contact_url changed for every snap | 12:38 |
| ogra_ | popey, true | 12:40 |
| popey | people link to it, it needs more than one line | 12:41 |
| ogra_ | popey, so ... whatever i try i still get the same denial even with local review-tools | 12:41 |
| ogra_ | ogra@acheron:~/Devel/xdmcp-client:master$ review-tools.snap-review xdmcp-client_0.1_amd64.snap | 12:41 |
| ogra_ | Errors | 12:41 |
| ogra_ | ------ | 12:41 |
| ogra_ | - declaration-snap-v2:slots_deny-connection:xephyr:x11 | 12:41 |
| ogra_ | human review required due to 'deny-connection' constraint for 'on-classic' from base declaration | 12:41 |
| popey | thats not the same error | 12:41 |
| popey | that's not a desktop file issue | 12:41 |
| ogra_ | (actually it doesnt look like the desktop file thing at all that i get by mail) | 12:41 |
| ogra_ | no | 12:41 |
| popey | one for jamies | 12:42 |
| ogra_ | the desktop file is what i got by mail from the review | 12:42 |
| ogra_ | popey, hmm | 12:43 |
| ogra_ | ogra@acheron:~/Devel/xdmcp-client:master$ review-tools.snap-review xdmcp-client_0.1_amd64.snap | 12:43 |
| ogra_ | Errors | 12:43 |
| ogra_ | ------ | 12:43 |
| ogra_ | - declaration-snap-v2:slots_deny-connection:xephyr:x11 | 12:43 |
| ogra_ | human review required due to 'deny-connection' constraint for 'on-classic' from base declaration | 12:43 |
| ogra_ | err | 12:43 |
| ogra_ | one sec, wrong paste | 12:43 |
| ogra_ | The reviewer provided the following feedback: | 12:43 |
| ogra_ | desktop interfaces (x11) specified without a corresponding meta/gui/*.desktop file. If using snapcraft, please see https://snapcraft.io/docs/build-snaps/metadata#fixed-assets. Otherwise, please provide a desktop file in meta/gui/*.desktop (it should reference one of the 'apps' from your snapcraft/snap.yaml) | 12:43 |
| ogra_ | thats what i get by mail ... | 12:43 |
| mup | PR #39: Login integration test closes #38 <Created by fgimenez> <Merged by chipaca> <https://github.com/snapcore/snapd/pull/39> | 12:43 |
| ogra_ | seems slightly contradicting to what the review tools tell | 12:44 |
| ogra_ | popey, is there a way to find out who did the manual review ? | 12:46 |
| ogra_ | https://dashboard.snapcraft.io/snaps/xdmcp-client/revisions/12/review/ would be one that caused the mail | 12:46 |
| popey | Jamie | 12:47 |
| ogra_ | oh, yeah, i didnt know i could open that link even ! | 12:47 |
| ogra_ | heh | 12:48 |
| ogra_ | well, the tools definitely reject it because the app needs to declare a " slots: [ x11 ]" in the apps: section | 12:49 |
| ogra_ | (to provide an x11 loopback interface for using XWayland under mir-kiosk) | 12:49 |
| ogra_ | geez .... i didnt expect that to be this complicated .... | 12:49 |
| ogra_ | yay ... new technology :) | 12:49 |
| * ogra_ updates https://forum.snapcraft.io/t/review-tools-forcing-desktop-file-for-kiosk-apps/6436 | 12:57 | |
| === chihchun_afk is now known as chihchun | ||
| === chihchun is now known as chihchun_afk | ||
| Chipaca | ogra_: "i didnt expect that to be this complicated" --napoleon, somewhere around sepember 1812 | 13:28 |
| kjackal | hi everyone. I am a bit confused. Is it possible to have the launchpad builders build and release snaps on a few different tracks? Can we have lp builders build from branch A and release to track A and at the same time have other(?) builders build from branch B and release to track B? | 13:28 |
| roadmr | hello folks, there's currently an outage in the snap store, we're working to solve it, I'll update when it's fixed | 13:29 |
| kjackal | I may be missing some button | 13:29 |
| Chipaca | kjackal: I thought the builders only released to [foo?]/edge | 13:29 |
| Chipaca | kjackal: but https://docs.snapcraft.io/build-snaps/ci-integration seems to imply you should be able to choose branches | 13:30 |
| Chipaca | oh wait | 13:31 |
| Chipaca | those are lp branches :-) | 13:31 |
| Chipaca | kjackal: but in that same page, step 11 | 13:31 |
| kjackal | I think I understand it now. I have to start with the branch on LP and then create the builders. let me try that | 13:33 |
| roadmr | store outage should now be resolved, thanks for your patience | 13:34 |
| pstolowski | pedronis: if you have some time for reviews, https://github.com/snapcore/snapd/pull/4940 needs re-reviewing; and i'm waiting for spread test run of #5532 | 13:41 |
| mup | PR #4940: overlord: added UDevMonitor for future hotplug support <Created by stolowski> <https://github.com/snapcore/snapd/pull/4940> | 13:41 |
| mup | PR #5532: api/connect: ignore connect if already connected <Created by stolowski> <https://github.com/snapcore/snapd/pull/5532> | 13:41 |
| ogra_ | Chipaca, lol | 13:45 |
| === chihchun_afk is now known as chihchun | ||
| pedronis | Chipaca: hi, could you look at #5515 | 13:58 |
| mup | PR #5515: daemon: make sure most change generating handlers can produce errors with kinds <Created by pedronis> <https://github.com/snapcore/snapd/pull/5515> | 13:58 |
| Chipaca | pedronis: looking | 14:09 |
| zyga | how are things going? | 14:09 |
| zyga | or doing actually | 14:10 |
| oSoMoN | I've added plugs for gtk-3-themes, icon-themes and sound-themes to the libreoffice snap like was done for a bunch of gnome app snaps, but I'm getting the following error and don't know what's causing it: | 14:11 |
| oSoMoN | main.go:192: cannot change mount namespace of snap "libreoffice" according to change mount (/snap/gtk-common-themes/319/share/icons/Adwaita /snap/libreoffice/x1/data-dir/icons/Adwaita none bind,ro 0 0): cannot create writable mimic over "/snap/libreoffice/x1": permission denied | 14:11 |
| oSoMoN | can anyone enlighten me? | 14:11 |
| Chipaca | pedronis: youch | 14:12 |
| oSoMoN | (this is what I added to snapcraft.yaml: https://git.launchpad.net/~libreoffice/+git/libreoffice-snap/commit/?id=0e58538d71af93eef00c6d73b7767a548208e30d) | 14:12 |
| oSoMoN | and the bug report with all the details is https://github.com/ubuntu/gtk-communitheme/issues/350 | 14:12 |
| zyga | oSoMoN: hey | 14:12 |
| zyga | oSoMoN: can you share the snap yaml's of the two snaps and the output of "dmesg | grep DENIED" please | 14:13 |
| zyga | oSoMoN: also snap version but I suspect you are just on the latest stable | 14:13 |
| oSoMoN | zyga, snapcraft.yaml: https://git.launchpad.net/~libreoffice/+git/libreoffice-snap/tree/snap/snapcraft.yaml?h=6.0 | 14:15 |
| oSoMoN | $ snap version | 14:15 |
| oSoMoN | snap 2.33.1 | 14:15 |
| oSoMoN | snapd 2.33.1 | 14:15 |
| oSoMoN | series 16 | 14:15 |
| oSoMoN | ubuntu 18.04 | 14:15 |
| oSoMoN | kernel 4.15.0-23-generic | 14:15 |
| zyga | ok | 14:15 |
| zyga | oSoMoN: can you tell me if $SNAP/data-dir/{themes,icons,sounds} all exist in the snap? | 14:16 |
| zyga | I assume they don't | 14:16 |
| oSoMoN | zyga, they don't | 14:16 |
| zyga | ok | 14:16 |
| zyga | please share the denial | 14:16 |
| zyga | and /var/lib/snapd/apparmor/profiles/snap-update-ns.libreoffice from the affected machine | 14:16 |
| oSoMoN | zyga, there are multiple denials, most of them are similar to this: | 14:17 |
| oSoMoN | juil. 18 16:17:04 bribon audit[4274]: AVC apparmor="DENIED" operation="mount" info="failed mntpnt match" error=-13 profile="snap-update-ns.libreoffice" name="/tmp/.snap/snap/libreoffice/69/" pid=4274 comm="3" srcname="/snap/libreoffice/69/" flags="rw, rbind" | 14:17 |
| oSoMoN | zyga, https://paste.ubuntu.com/p/HhXgw52vzY/ | 14:18 |
| zyga | ok | 14:18 |
| zyga | this is https://github.com/snapcore/snapd/pull/5395 | 14:19 |
| mup | PR #5395: interfaces: generalize writable mimic profile <Core18> <Created by zyga> <https://github.com/snapcore/snapd/pull/5395> | 14:19 |
| zyga | I will try to land this branch | 14:19 |
| zyga | as a workaround please create, inside the libreoffice snap, the data-dir/ | 14:19 |
| zyga | then the rest should work | 14:19 |
| zyga | it's the extra level of nesting that is breaking this | 14:20 |
| Chipaca | pedronis: is sideloadCheck just moved? | 14:20 |
| oSoMoN | zyga, thanks! I'll test creating the data-dir/ inside the snap | 14:21 |
| oSoMoN | I wonder why this is working for other snaps, like evince or gedit, though | 14:21 |
| oSoMoN | those don't have a data-dir/ directory either | 14:21 |
| zyga | I don't know | 14:24 |
| zyga | I will study the rules you gave me to see if this is the same thing I think for sure | 14:24 |
| oSoMoN | thanks! | 14:24 |
| zyga | oSoMoN: please let me know if this works for you | 14:28 |
| oSoMoN | zyga, yes, I'm repacking the snap with an additional data-dir/ directory, will let you know | 14:28 |
| zyga | oSoMoN: thank you | 14:28 |
| zyga | you can just unsquash it, mkdir and snap pack it | 14:28 |
| oSoMoN | zyga, yep, that's what I did | 14:30 |
| oSoMoN | but same error | 14:30 |
| oSoMoN | I verified that $SNAP/data-dir exists after installing | 14:30 |
| zyga | can you be more specific please, what is the error you are getting now | 14:30 |
| oSoMoN | cannot create writable mimic over "/snap/libreoffice/x1/data-dir": permission denied | 14:30 |
| zyga | that is more curious | 14:31 |
| zyga | hold on | 14:31 |
| zyga | well, add the extra three directories and that will go away | 14:31 |
| zyga | but I really need to get to the bottom of this | 14:31 |
| oSoMoN | ok, adding those | 14:31 |
| zyga | and please dmesg the denial that you saw now | 14:31 |
| oSoMoN | AVC apparmor="DENIED" operation="mount" info="failed mntpnt match" error=-13 profile="snap-update-ns.libreoffice" name="/tmp/.snap/snap/libreoffice/x1/data-dir/" pid=8248 comm="3" srcname="/snap/libreoffice/x1/data-dir/" flags="rw, rbind" | 14:32 |
| zyga | thank you | 14:33 |
| pedronis | Chipaca: yes | 14:33 |
| oSoMoN | repacking with all three directories now, let's see if that works | 14:33 |
| Chipaca | pedronis: ah yes i edited locally to confirm :-) | 14:33 |
| Chipaca | pedronis: before +1'ing it (without the edit it would've taken me a lot longer to review) | 14:34 |
| pedronis | Chipaca: was moved far for some reason from either it's first or last usage | 14:34 |
| pedronis | so I moved it | 14:34 |
| zyga | ah | 14:34 |
| zyga | I know why it didn't help | 14:34 |
| pedronis | Chipaca: that test file is too long (tm) :( | 14:34 |
| zyga | but it's still that bug | 14:34 |
| Chipaca | pedronis: as is api.go itself | 14:34 |
| zyga | oSoMoN: the reason is that gtk-common-themes is actually exportng a number of items that are nested themselves | 14:35 |
| zyga | the branch I referenced fixes this but, as I said, it's on me to land | 14:36 |
| === chihchun is now known as chihchun_afk | ||
| oSoMoN | zyga, I can confirm that creating the missing dirs works around the issue | 14:42 |
| abeato | jdstrand, mind taking a look at https://github.com/snapcore/snapd/pull/5469 when you can? | 15:18 |
| mup | PR #5469: interfaces/apparmor: (un)load profiles in one apparmor_parser call <Created by alfonsosanchezbeato> <https://github.com/snapcore/snapd/pull/5469> | 15:18 |
| zyga | oSoMoN: thank you | 15:23 |
| === pstolowski is now known as pstolowski|afk | ||
| mvo | zyga: pr#5533 allows creating images with the new kernel-track feature of the model assertion. if Gustavo has a free minute it would be great to get his opinion on what kernel track name to pick. having one soon would be really great, then I can start writing an integration test | 16:18 |
| mvo | hm, no mup currently? | 16:23 |
| cachio | zyga, https://paste.ubuntu.com/p/BQHWShC7Zv/ | 17:54 |
| cachio | I see these denials executing a test for daemon-notify interface | 17:55 |
| cachio | zyga, any idea why it could be happening | 17:55 |
| cachio | = | 17:55 |
| zyga | cachio: yes, it is not auto-connected | 18:35 |
| zyga | you have a slot but it's not connected | 18:35 |
| cachio | zyga, I manually connected it | 18:35 |
| zyga | I wrote a thread about this on the forum | 18:35 |
| zyga | it's too late | 18:35 |
| zyga | you cannot connect it fast enough | 18:35 |
| zyga | well | 18:35 |
| zyga | I'd have to see your test service | 18:36 |
| zyga | but in general you cannot use that from non-service started by systemd | 18:36 |
| zyga | and you cannot connect it fast enough | 18:36 |
| zyga | so in practice it's not really usable | 18:36 |
| zyga | please read my forum post about this | 18:36 |
| zyga | https://forum.snapcraft.io/t/its-a-little-bit-hard-to-use-daemon-notify-for-sd-notify/6366 | 18:37 |
| zyga | mvo: the github part of mup is off because of the github availability issue | 18:37 |
| cachio | zyga, what I did is to connect it and the install it again | 18:37 |
| cachio | zyga, but same issue | 18:38 |
| zyga | that doesn't help | 18:38 |
| zyga | it's not auto-connected | 18:38 |
| zyga | because the install would likely fail | 18:38 |
| zyga | because the daemon: notify cannot be installed without successfully talking to systemd | 18:38 |
| zyga | can you show me the test please | 18:38 |
| cachio | zyga, https://paste.ubuntu.com/p/rkf8VXVg5K/ | 18:39 |
| cachio | https://paste.ubuntu.com/p/wrmrZz2rNY/ | 18:39 |
| zyga | ok, and what did you expect to happen? | 18:39 |
| cachio | zyga, I am checking logs to see I there is any denial | 18:40 |
| cachio | zyga, still doing it manually | 18:40 |
| zyga | the interface is not auto-connected so as soon as you start the service it is attempting to use a program it cannot use | 18:41 |
| zyga | you can connect it | 18:41 |
| zyga | restart the service | 18:41 |
| zyga | but then it may not work _anyway_ depending on what systemd does | 18:41 |
| cachio | ok | 18:41 |
| zyga | if it allows you to talk over the socket (separately from our permission) | 18:41 |
| cachio | zyga, I'll try that, thanks | 18:45 |
| thiras | hello | 19:43 |
| thiras | discord and spotify are broken | 19:44 |
| thiras | $ snap run discord | 19:44 |
| thiras | ln: failed to create symbolic link '/home/thiras/snap/discord/69/.config/gtk-2.0/gtkfilechooser.ini': File exists | 19:44 |
| thiras | Gtk-Message: Failed to load module "gail" | 19:44 |
| thiras | Gtk-Message: Failed to load module "atk-bridge" | 19:44 |
| thiras | ubuntu 18.04 | 19:45 |
| jdstrand | abeato: yes. it is on my list. I looked at it and am ok with the approach, but I wanted to check a few things before giving my ack. I apologize for the delay. I've been meaning to get to it and am sprinting this week | 19:45 |
| thiras | spotify has quite similar error log | 19:47 |
| cachio | zyga, I have updated the PR #5535 | 20:04 |
| mup | PR #5535: tests: fix tests expecting old email address <Critical> <Simple> <Created by zyga> <https://github.com/snapcore/snapd/pull/5535> | 20:04 |
| zyga | cachio: I saw just now, thank you! | 20:13 |
| cachio | zyga, np | 20:13 |
| === devil is now known as Guest74443 | ||
| === Guest74443 is now known as devil_ | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!