=== popey9 is now known as popey [06:28] morning [07:34] PR snapd#9812 closed: packaging/deb, tests/main/lxd-postrm-purge: fix purge inside containers [07:45] mvo: hey [07:46] good morning mborzecki [08:11] hello! [08:20] welcome back everyone [08:20] hello pawel! [08:21] how was your holiday break? [08:32] zyga: hi! was awesome, very relaxing [08:32] how about you? [08:33] pstolowski: hey [08:33] hey mborzecki ! [08:34] pstolowski my break was shorter but we made it very active [08:34] pstolowski lots of time spent outdoors [08:36] sounds awesome [08:40] good morning pstolowski ! nice to have you back. hope you had a great break [08:40] zyga: good morning, so nice to see you too :) === mborzecki_ is now known as mborzecki [08:55] PR snapd#9776 closed: gadget: add validation for "$kernel:ref" style content [08:55] PR snapd#9809 closed: snap: skip help output tests for go-flags v1.4.0 [09:00] PR snapd#9804 closed: interfaces/opengl: allow RPi MMAL video decoding [09:07] * ogra hugs mvo [09:13] * mvo hugs ogra [09:14] * ogra is on his way to replace his libreelec installs with UC20 and https://snapcraft.io/kodi-pi-standalone ... thats a big step towards it [10:04] mvo: can you cherry pick https://github.com/snapcore/snapd/pull/9812 to 2.48 in case we do a point release? [10:04] PR #9812: packaging/deb, tests/main/lxd-postrm-purge: fix purge inside containers [10:05] mborzecki: let me have a look [10:15] PR snapd#9778 closed: asserts/repair.go: add "bases" and "modes" support to the repair assertion [10:37] pedronis: I updated 9777 (just fyi, no real rush at this point) [10:43] mvo: now I can request myself as reviewer again (mystery), I just queued it [10:43] pedronis: nice [10:43] pedronis: and yes, mystery! [12:05] PR snapcraft#3401 closed: ci: uprev pyinstaller and switch timestamp server [12:10] PR snapcraft#3385 closed: yaml_utils: promote module to a package [14:27] mvo, hey there [14:28] zyga: hey, how are you? [14:28] mvo, we're sorting out project planning in the team but I wanted to drop you a link for something that's coming up this sprint (2 weeks) [14:28] https://git.ostc-eu.org/planning/testing/-/issues/2 [14:28] I will work on an OSTC fork but I will also upstream things [14:29] ostc is the open source thing I'm participating in [14:29] cachio, ^ FYI [14:30] cachio, where is your fork/branch of spread btw? [14:30] is it just your fork or did you rename it by any chance [14:32] zyga, hey [14:33] how are you? [14:33] cachio, hola, como estas [14:33] this is the fork https://github.com/sergiocazzolato/spread/tree/master2 [14:33] muy bien :) [14:33] ah, great, your github fork [14:33] yes [14:33] I wanted to keep you in the loop of spread patches I'm planning, if any of those would be useful to you or would be relevant to include you in the design phase [14:34] zyga, yes [14:34] nice [14:34] thanks [14:34] just ping me if you need anything [14:35] cachio, I was thinking that if spread did support booting, at least on qemu, with a specific system image, it might be useful to organize core tests to support that, e.g. take image, that's already prepared by a separate stage and just boot it to run tests [14:35] for us it will be the image produced in the previous stage in the pipeline [14:36] (and our images are small so passing them from one step to the next is not hard) [14:36] the ultimate goal will be to take an arm image and run it via spread muxpi backend [14:36] cachio, right now any feedback on the yaml example there is great [14:37] I don't see any of those as final, that will come when I try to upstream it [14:37] that's why I want to have a separate fork, to have unrelated timelines for downstream and upstream work [14:37] zyga, that feature makes sense for qemu [14:38] so far the qemu backend has many restrictions [14:38] which should be addressed [14:38] I'll take a look to the yaml [14:44] cachio, thanks, I'm happy to talk if you want to tweak it at this stage, I'll pick it up tomorrow [14:44] zyga: nice [14:44] cachio, (not sure if URL is final) https://git.ostc-eu.org/planning/testing/-/boards [14:45] zyga: I like this [14:46] mvo, hehe, that's just me doing my usual convincing to use what's in the forge already, and not to use exteral systems [14:46] too bad free gitlab is so limited, paid versions have really nice features [14:47] zyga, thanks for sharing [14:49] zyga, I think the image flag could be usefull also for other backends [14:49] cachio, right, it doesn't have to be local image [14:49] so it is not needed to update the image in the spread.yaml in case you need to test a test image in gce for example [14:49] not sure how the google api looks like but it's plausible to build an image that's compatible in one phase, upload it to google in another and then use that image in the third phase [14:49] yeah [14:50] I think the bigger new item is the syntax for paramteres, the definition side is I think okay but the use side is confusing [14:50] or at least non-obvious [14:50] perhaps it should just be string interpolation with ${image} or $image or something like that [14:51] I'll sleep on this, it's not something I'm convinced about yet [14:52] zyga, agree [14:52] need to think which is the best way to implement that [14:56] * cachio afk [14:56] cachio, if we want to use parameters "anywhere" it may be hard, as it implies that the yaml parser needs to somehow understand that [14:56] cachio, ok, let's chat later [14:57] zyga, thanks [14:57] I'll be back in 20 [15:08] mvo, offtopic, we've set up mattermost IRC bridge lately and it works fairly well [15:09] I'm mentioning this because it lets people use (arguably) more comfortable mattermost while still being visible on IRC [15:09] it's per channel obviously [15:15] zyga, is that another server that you have to run, or is it like just a plugin that is enabled server side ? [15:16] ijohnson, it's a self-hosted service that runs but I honestly don't know the details as my colleague deployed it yesterday [15:16] it was generally easy with the exception of some confusing instructions early on during credentials management [15:16] ijohnson, it's live in #ostc if you want to look [15:17] zyga: ah I see, yeah I meant to set something like that up for myself to use only mattermost for this IRC channel, but I never got around to it yet [15:17] also happy new year zyga! [15:17] ijohnson, happy new year :) [15:20] :-) [15:57] can a devmode snap (ie: microstack) be updated, or do I need to remove it and lose everything in microstack and reinstall to get the latest version? [15:58] you can manually refresh devmode snaps [15:58] perfect, thanks ijohnson [16:35] pedronis: I reviewed #9803 [16:35] PR #9803: many: clarify gadget role-usage consistency checks for UC16/18 vs UC20 [16:37] ijohnson: thx, I'll work on the feedback in my morning [16:37] the feedback is pretty minimal fine for a followup if you prefer [16:43] ijohnson: about supporting multi-volume setups, they would pass the checks but we still don't support gadget updates with them atm [16:44] but you could still build an image with a multi-volume gadget and not do gadget updates though right ? [16:44] yes [16:45] admittedly, I didn't try to do so with your branch so perhaps there are other things missing in the stack that prevent using multi-volumes, but having just a unit test about validating a multiple volume gadget with all the ubuntu-* roles on a single volume seems useful to me [17:11] PR snapcraft#3402 opened: cli: add missing quote key creation hint [17:21] PR snapcraft#3403 opened: storeapi: remove unused MissingSnapdError === ijohnson is now known as ijohnson|lunch [18:25] * cachio afk === King_InuYasha is now known as Conan_Kudo === Conan_Kudo is now known as King_InuYasha === ijohnson|lunch is now known as ijohnson [20:38] PR snapd#9815 opened: {,sec}boot: pass "key-name" to the FDE hooks [21:03] PR snapd#9816 opened: interfaces, wrappers: misc comment fixes, etc [22:51] PR snapcraft#3404 opened: build(deps): bump lxml from 4.5.0 to 4.6.2