[05:07] morning [07:09] morning [07:12] good morning [07:13] sorry for being late, not feeling well [07:13] oh, sorry to hear [07:14] pstolowski: zyga: hey [07:20] I need a review for https://github.com/snapcore/snapd/pull/8977 to make progress on raa [07:20] PR #8977: cmd/snap: track started apps and hooks [07:21] it's almost entirely spread tests + tiny new feature in snap-ru [07:21] run [07:23] quick errand, back in 30, and then i'll do a review of zyga's PR [07:23] thank you [07:30] * zyga tweaked his PR and goes to review top-to-bottom [07:31] pstolowski: https://github.com/snapcore/snapd/pull/8976 fails on preseed tests [07:31] PR #8976: snap-confine: don't die if a device from sysfs path cannot be found by udev (2.45) [07:31] do you need to put something into the release branch to unbreak it? [07:31] it seems unrelated to the change in the PR itself [07:37] zyga: oh... we have quite a bit of delta with master for these tests & lib/nested.sh :/ [07:37] hmm, making a point release could be difficult then [07:38] zyga: maybe we should disable them in release . might be painful if we update nested.sh [07:38] at the same time, the preseed test is pretty important [07:39] zyga: many changes in spread.yaml as well, it's all interconnected [07:40] do you know why the existing test may have stopped working? [07:43] zyga: i'll investigate in a moment, but from a quick glance it seems to be core -> snapd snap transition that we fixed before in master [07:43] afair it affected match rules in the tests & nested.sh helpers [07:52] re [08:06] mvo: I got around to testing out a scheme to share private fontconfig caches: https://forum.snapcraft.io/t/shared-fontconfig-cache-prototype/18660 [08:07] brb [08:07] jamesh: mvo is on a sprint this week [08:07] zyga: ah. [08:08] he's starting at around noon and has limited availability [08:08] zyga: you might be interested in the post too [08:08] (so roughly 3-4 hours later than usual to be in sync with US) [08:08] yeah, I just opened it :) [08:08] I'm very interested, I had to rm -rf cache on my groovy system the other day [08:09] brb, need to fetch glasses from the other room [08:09] zyga: in particular: what are the chances of relaxing the restrictions on where content interface can mount stuff? [08:09] jamesh: where do you need to mount stuff? [08:10] zyga: Ideally I'd like to do a content interface share mounted to /var/cache/fontconfig [08:10] I see [08:10] let me think, I think in theory content could behave more like layouts [08:10] but first, glasses [08:11] PR snapd#8974 closed: spread.yaml: remove tests/lib/tools from PATH [08:11] PR snapd#8976 closed: snap-confine: don't die if a device from sysfs path cannot be found by udev (2.45) [08:11] zyga, jamesh good morning [08:12] jamesh: oh, private caches! exciting! [08:12] Trying to combine content interface plus layouts didn't work because the layouts mount occurred first. I just ended up with an empty directory on /var/cache/fontconfig [08:13] hi mvo! [08:15] re [08:15] mvo: hey :) [08:15] hey pstolowski [08:15] anything I can help with that does not require too much brian power :) ? [08:15] mvo: who is brian? ;) [08:16] brain [08:16] see, that's my point :) [08:16] :) [08:16] mvo: I wanted to ask you about https://github.com/snapcore/snapd/pull/8974 but you just merged woot! [08:16] PR #8974: spread.yaml: remove tests/lib/tools from PATH [08:16] thank you :) [08:16] :) [08:33] pstolowski: brian, the man called brian? ;) [08:38] zyga: https://www.jeffgeerling.com/blog/2020/uasp-makes-raspberry-pi-4-disk-io-50-faster [08:38] via hackernews [08:38] woah [08:38] that's neat [08:39] once I'm okay again and can work from the office I will get something like this for my 8GB PI4 [08:42] ok, on to the reviews [09:02] zyga: do you have a spread test for tracking where everything works and with v2? [09:02] mborzecki: tracking is already tested in v2 [09:02] cgroup-tracking shows this [09:02] raa is not tested here yet as I broke the patch so that overlord bits are not there yet [09:02] but the test for that exists in the other branc [09:03] zyga: the large branch right? [09:03] well, not so much anymore but yeah :) [09:03] the bigger branch [09:03] ok [09:03] after this lands it will be quite small actually, mostly removals and small changes to overlord, and one last spread test [09:03] there's more later, mainly test coverage improvements [09:03] for hooks in particular [09:20] zyga: https://github.com/snapcore/snapd/pull/8977#pullrequestreview-442883866 [09:20] PR #8977: cmd/snap: track started apps and hooks [09:20] zyga: some tweaks, are the unit tests in the other branch too? [09:20] mborzecki: lookig [09:20] mborzecki: there are no more unit tests in the other PR [09:22] mborzecki: yeah, about mocking, I wonder where those tests went, I think I wrote some but perhaps I didn't or perhaps they got lost somehow along the path [09:22] mborzecki: thanks for the review, I'll finish pawel's service PR and get back to it [09:30] zyga: https://forum.snapcraft.io/t/snapd-for-aarch64-opensuse/18663 [09:32] thanks [09:42] PR snapd#8979 opened: tests: more checks in core20 early config spread test [10:06] pstolowski: fwiw, 8960 has a conflict [10:07] pstolowski: https://github.com/snapcore/snapd/pull/8960#pullrequestreview-442914596 [10:07] PR #8960: o/snapstate,servicestate: use service-control task for service actions (9/9) [10:08] I will park reviews and go back to https://github.com/snapcore/snapd/pull/8977 [10:08] PR #8977: cmd/snap: track started apps and hooks [10:08] mvo: ah, on it, ty [10:20] zyga-x240: thanks for the review! [11:17] PR snapd#8980 opened: tests: backport preseed test fixes to 2.45 [11:20] zyga-x240: ^ this should fix the error you reported earlier [11:23] pstolowski: mborzecki: I updated #8906 [11:23] PR #8906: asserts: introduce SequenceMemberAfter in the asserts backstores [11:23] pedronis: thanks [11:23] mborzecki: I'll do a follow up about 001 [11:23] thanks, checking [11:23] and 010 [11:24] pstolowski: thanks [11:54] brb === ijohnson|pto is now known as ijohnson [12:00] hello folks [12:07] ijohnson: hey [12:07] hey zyga-x240 [12:19] hi ijohnson ! [12:20] o/ pstolowski === ijohnson is now known as ijohnson|sprint [12:39] mborzecki: I've added tests to https://github.com/snapcore/snapd/pull/8977 [12:39] PR #8977: cmd/snap: track started apps and hooks [12:39] mborzecki: I'll do a pass over spread tests and comment suggestions after lunch/standup [12:39] mborzecki: thanks, your comment helped to find a bug [12:43] zyga-x240: hm hm? where? [12:43] snap run? [12:45] yeah [12:45] mborzecki: it's a subtle bug that wasn't screaming broken [12:45] but testing uncovered it just ifne [12:45] *fine [12:52] cachio: hey, #8558 has conflicts [12:52] PR #8558: tests: make the nested library usable independently of spread [12:56] pstolowski, checking, thanks [12:59] hmm [12:59] no standups? [12:59] I don't see a standup in my calendar [12:59] no more standup link? [13:00] I guess we can skip [13:00] or do it here [13:00] how do you guys feel about doing an IRC standup/ [13:01] pstolowski, I'll need to apply some changes to this once #8942 is merged [13:01] PR #8942: tests: support different images on nested execution [13:01] zyga-x240: standup? [13:02] mborzecki: where? [13:02] mborzecki: there's no standup in the calendar [13:02] cachio: ah, ok, let's try to get #8942 in [13:16] cachio: thank you for all the tooling improvements [13:16] cachio: i made a few comments to your PRs [13:16] I know it's a lot of work but I really think those are important for the longevity of the testing code [13:20] cachio: how would you feel about adding a test helper for the defer.sh/tac trick? [13:20] zyga-x240, it is ok [13:20] I wanted to use it without repeating the code around [13:21] I'll add this to my todo list [13:21] tests.cleanup perhaps [13:21] so I can start it after the snaps.sh migration is proposed [13:21] zyga-x240, nice, yes, makes sense [13:51] cmatsuoka: https://github.com/snapcore/snapd/pull/8981 [13:51] PR #8981: boot, bootloader: query kernel command line of run mod and recovery mode systems [13:51] mborzecki: ack, will check [13:51] ijohnson|sprint: ^^ if have the time between the meetings [13:52] mborzecki: nice I will add it to my queue, let's see how much time I have :-) [13:53] PR snapd#8981 opened: boot, bootloader: query kernel command line of run mod and recovery mode systems === msalvatore_ is now known as msalvatore [13:58] cmatsuoka: ijohnson|sprint: thanks! [14:03] mborzecki: I had a prepare failure on centos 7 on my PR, is tht known? [14:03] pedronis: do you ahve the log? [14:04] pedronis: or which pr, and i can take a look [14:04] mborzecki: https://github.com/snapcore/snapd/pull/8906/checks?check_run_id=841248103 [14:04] PR #8906: asserts: introduce SequenceMemberAfter in the asserts backstores [14:05] pedronis: looks like somehting temporary [14:06] mborzecki: updated https://github.com/snapcore/snapd/pull/8977 [14:06] PR #8977: cmd/snap: track started apps and hooks [14:06] pedronis: other prepare jobs were successful, so maybe something with a particular mirror getting synced at the time of the request [14:06] * zyga-x240 needs a moment for painkillers to work [14:07] pedronis: restarted the spread jobs in that pr [14:22] time to run some errands [14:22] PR snapcraft#3202 opened: build providers: fix base change warning message [14:42] PR snapcraft#3155 closed: storeapi: specify Content-Type for icon [15:07] * cachio lunch [15:11] cachio: I have something for that now, [15:13] cachio: I'll open a pull request in a minute [15:39] Issue core18#161 closed: mknod: /home/ubuntu/core18/parts/boostrap/install/dev/null: Operation not permitted [15:44] Issue core18#162 opened: Beef up build instructions [15:51] mborzecki: the centos 7 problem seems more persistent my PR failed again [15:51] same way [15:53] pedronis: can you make it non-required for now? [15:53] I should be able to now [15:54] funny, how prepare failed on one now, but all the tests run on the other nodes [15:54] bbiab === the-mentor9011 is now known as the-mentor [15:57] mborzecki: done [15:58] PR snapd#8906 closed: asserts: introduce SequenceMemberAfter in the asserts backstores [16:18] PR snapd#8982 opened: snapshots: export of snapshots [16:27] PR snapcraft#3097 closed: colcon v2 plugin + ros2 extension [16:32] cachio: https://github.com/snapcore/snapd/pull/8983 [16:32] PR #8983: tests: add tests.cleanup helper [16:32] PR snapcraft#3203 opened: experimental ros2 extension & colcon v2 plugin [16:33] PR snapd#8958 closed: tests: nested test improvements from master (2.45) [16:33] PR snapd#8983 opened: tests: add tests.cleanup helper [16:43] zyga-x240, nice, I'll take a look [17:32] PR snapcraft#3097 opened: colcon v2 plugin + ros2 extension [17:36] hey, zyga, and anybody who mention ;-) how are you ? [17:36] zyga: is the merge ok enough? the unit-test stop with some pgp ... [17:37] mvo: how are you? [17:39] hi, alan_g - you generalist ... what do you today ? [17:40] sorry, i'm afk ... [17:46] zyga: look's like nobody talk to me. [17:57] sdhd-sascha, there is a virtual company sprint going on i guess the whole snapd team is busy with that [17:57] ogra: thank you ;-) i didn't know [17:58] well, its an internal thing, usually not promoted to the outside world a lot [17:59] ogra: i had understood. But, you helped me. That is the cause, why i said "thank you", to you ;-) [17:59] 🙂 [17:59] :-) [18:03] sdhd-sascha: yeah, what ogra said, busy in meetings :/ but nice to "see" you :) [18:04] me too, i feel the same ;-) (don't know why...) [18:06] i listen: https://www.youtube.com/watch?v=QRQwZDWz1Pw&t=4051s [18:13] that's my educational friendly to youngsters bit done for today ;-) https://github.com/dotnet/runtime/issues/1634 [18:14] hah, teaching LD_PRELOAD to the masses ! [18:14] IKR! [18:15] obviously by "very experienced developers" I'm referring to myself, because until I got going on Snapping stuff I was oblivious [18:16] in my defence, I'm supposed to be a PHP dev :-p [18:16] as in using PHP to dev, not writing the PHP language [18:17] mvo: you, didn't believe it! some guy want to programm a webserver from me. I know, i can do low-level with sockets... and i can do the other way.... (But really, i feel sucked, but that question.... I learned all my way ;-) ) [18:18] sdhd-sascha: netcat into bash? ;-p [18:19] diddledan: ;-) nice ... sometime i didn't know "nc" ;-) [18:19] I've seen router firmwares running their webserver scripts as bash [18:20] as in using bash for the application code [18:20] sdhd-sascha: if you need a webserver, https://bazaar.launchpad.net/~snappy-dev/snappy-hub/go-example-webserver/revision/1 could be a starting point, [18:20] diddledan: you mean, that bash as parent process is a problem? i only see such routers ? ... [18:21] mvo: no... i mean, some person, ask me to write a socket "daemon" .... (me?) [18:22] ogra, about fabrica, could it possibly be extended to use containers a la docker, you think? I've got snapcrafty building into images for all the supported architectures, which work with qemu-user-static on Githug actions, so I wonder if we could add them to fabrica too [18:22] ... they also spin up a real snapd and systemd inside them [18:22] i surely wont reject patches 😉 [18:23] the downside is you need to start the containers in --privileged mode [18:23] well, fabrica is a daemon [18:24] though it would require a lot of code to make it use docker or some other container system ... the current code is only lxd go stuff [18:24] aye [18:25] but i indeed wouldnt refuse it 🙂 [18:25] * diddledan pops it on the toedoe list [18:25] heh [18:27] sdhd-sascha, I meant that the scripts generating the HTML for the browser are BASH. you open your browser, go to the router page, the webserver executees some bash and returns the output [18:27] insert-exploit-here [18:28] who remembers shellshock? :-o those were fun times [18:29] diddledan: ??? ok ??? i read alot how to exploit my router... but it isn't mine.... the owner is the telephonee company... so i do nothing!!!! [18:29] i have a few toshiba flash-air cards here ... you can actually flash the OS on them through such an exploit and make them run a proper linux 😄 [18:29] diddledan: thank you ;-) i respect you ;-) [18:30] diddledan: just soom boot"s ... i know ;-) [18:33] re [18:33] ogra: i really want to know more of you knowledge ;-) [18:35] allo zyga-x240 [18:35] zyga-x240: why did you change your nick? And... .... [18:35] I stopped using irccloud so I use separate nicknames for each machine [18:36] zyga-x240: really? what test at cannocial was it now ? and why, do you think? [18:37] I don't understand your question, sorry [18:37] here, at my place... i often use "rke" kubernetes installer... it works goood... but today, it says "no" ... [18:38] zyga-x240: no, problem... [18:39] zyga-x240: just, want say. That i respect you ;-) [18:40] zyga-x240, the #8973 is updated [18:40] PR #8973: tests: moving journalctl.sh to a new journal-state tool [18:40] cachio: thanks, I'll check it out tomorrow though [18:40] zyga-x240, thanks [18:40] I came to check the weird failures in some tests [18:44] zyga-x240: how are you weather today? we have small rain at 7. And the rest strong wind.... (But everybody is happy about the sun...) [18:44] sdhd-sascha: it was rather cold most of the day with some rain starting in the late afternoon [18:45] zyga-x240: cold ? really ? can't beleave ... it was 20 degreee here... [18:46] all the day... [18:47] zyga-x240: you are a good guy/people, for me at least ;-) [18:47] thanks, I'm just trying to do my best :) [18:47] that's why i say it ;-) [18:48] cachio: something very wrong and weird happend [18:48] i just listen (or read) ;-) [18:48] cachio: I think prepare/restore across the new "tools" suite is broken [18:48] zyga-x240: i know [18:48] cachio: all of the failures can be summarized as "tests/bin" is either empty or not on PATH [18:49] cachio: all of the tests pass in isolation, I'm trying a larger run now [18:50] cachio: is there no zyga left? [18:52] zyga-x240, are you talking about master? [18:52] cachio: I don't know yet, I cannot understand how those things are failing [18:52] I just ran a few of those tests one-by-one and they did not fail [18:52] zyga-x240, do you have a url with the errror? [18:52] do you see any typos there? [18:54] hmmm [18:54] zyga-x240, I'll try to reproduce [18:54] cachio: wait a sec [18:54] spread.yaml seems stale, maybe I haven't rebased this on recent master [18:54] I see tests/lib/tools in PATH still [18:54] maybe that's why? [18:55] i see typos... but i .... hmm. ... i wonder onllllllllly .... hmm .... i'm behind you and my strenght is for you ;-) [18:55] trying again [18:56] +49 15 15 5789 0 752 [18:57] what, try again ? [18:57] sdhd-sascha: hmm? what are you saying [18:58] zyga-x240: where are you from? [18:59] sdhd-sascha: from Poland, why? [19:01] zyga-x240: sorry, for my question. (again) my grandparents was from slowakiya ... I'm love earth. [19:04] zyga-x240: can i visit you? maybe my skepticial wife ... [19:04] cachio: I force-pushed without changes just after rebasing [19:04] sdhd-sascha: I don't think so [19:04] zyga-x240, good [19:05] sdhd-sascha: I'm totally not in shape right now and I don't know you that much to begin with [19:08] PR snapcraft#3202 closed: build providers: fix base change warning message [19:09] if cachio did you harm... than someone should stop him!! [19:10] that, was not the last, what i send [19:10] sdhd-sascha: ?!? [19:10] what are you talking about [19:11] zyga-x240: and you? are you ok ? [19:11] sdhd-sascha: I'm not okay but my colleague cachio has nothing do with that, what made you think that? [19:12] why, did you mention "cachio" ? i',m try to be fair to anyone... [19:13] PR snapcraft#3204 opened: Improvements to the flutter plugin [19:13] sdhd-sascha: I'm talking to cachio about test failures in a pull request [19:17] zyga-x240: thank you. Your mention. Is better, then others!!!! [19:24] yesterday, on german tv, there was titanic. [19:26] I'm stuck in bed with serious back problems, no TV for me [19:26] what back probs? [19:28] zyga-x240: if you give me a probLEM .... i solve it... maybe 3 weeks later.... [19:28] spinal cord herniation [19:28] yeah, that's not gonna get fixed remotely ;-p [19:29] oooh ;-) i try to translate... [19:29] hmmm... [19:29] diddledan: I wish it could [19:30] * diddledan files a spinal PR [19:30] i too:-) (Hope i understand) [19:30] heheheh [19:30] you need a rock band to give it a prod.. a spinal tap, if you will ;-p [19:30]