15:01:14 <pwhalen> #startmeeting Fedora IoT Working Group Meeting
15:01:14 <zodbot> Meeting started Wed Feb 10 15:01:14 2021 UTC.
15:01:14 <zodbot> This meeting is logged and archived in a public location.
15:01:14 <zodbot> The chair is pwhalen. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:01:14 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:01:14 <zodbot> The meeting name has been set to 'fedora_iot_working_group_meeting'
15:01:14 <pwhalen> #chair pwhalen pbrobinson bcotton tdawson puiterwijk coremodule
15:01:14 <zodbot> Current chairs: bcotton coremodule pbrobinson puiterwijk pwhalen tdawson
15:01:14 <pwhalen> #topic roll call
15:01:30 <pwhalen> Good morning folks, who's here today?
15:01:34 <bcotton> .hello2
15:01:35 <zodbot> bcotton: bcotton 'Ben Cotton' <bcotton@redhat.com>
15:01:44 * coremodule is here, good morning, and I'm already loving this new meeting time :)
15:01:44 * pbrobinson is here, and elsewhere
15:02:33 <pwhalen> alright, lets get started
15:02:35 <pwhalen> #topic 1) ==== Working Group process and admin  ====
15:03:00 <pwhalen> Anything for admin?
15:03:57 <pbrobinson> nothing from me here
15:04:12 <pwhalen> #topic 2) ==== Fedora 33 status ====
15:04:58 <pwhalen> #info Latest - Fedora-IoT-33-20210128.0
15:04:58 <pwhalen> #link https://openqa.fedoraproject.org/tests/overview?distri=fedora&version=33&build=Fedora-IoT-33-20210128.0&groupid=1&groupid=5
15:05:08 <pwhalen> no change here, has anyone hit anything new in F33?
15:05:52 <pwhalen> #info No issues reported in F33.
15:06:03 <pwhalen> Do we plan on a new compose this week?
15:07:26 <pbrobinson> awaiting a fix for the ostree issue for f33
15:09:13 <pwhalen> right, thanks
15:09:30 <pwhalen> #info Upgrade removes all overrides - https://bugzilla.redhat.com/show_bug.cgi?id=1925717
15:11:22 <pwhalen> #topic 3) ==== Fedora 34 (Rawhide) status ====
15:11:31 <pbrobinson> branching happening today
15:12:18 <pwhalen> #info Fedora 34 IoT Edition branching to be completed today.
15:12:48 <pwhalen> #info Latest compose Fedora-IoT-34-20210208.0
15:12:48 <pwhalen> #link https://openqa.fedoraproject.org/tests/overview?distri=fedora&version=34&build=Fedora-IoT-34-20210208.0&groupid=1&groupid=5
15:13:06 <pwhalen> #info Podman failure fixed with kernel-5.11.0-0.rc7.149.fc34
15:13:35 <pwhalen> this one has a new service failure I was looking at before the meeting,
15:13:49 <pwhalen> systemd-oomd[1556]: Requires swap to operate
15:14:06 <pbrobinson> pwhalen: does it not work with zswap?
15:14:43 <pwhalen> I dont see any swap, so I need to look more
15:16:44 <pwhalen> the iot_rpmostree_overlay failure on aarch64 I need to look at as well
15:17:12 <pwhalen> looks like the test just died, will rerun it
15:17:49 <pwhalen> I think thats it for F34, has anyone else hit any issues not yet reported?
15:17:55 <pbrobinson> pwhalen: is that for overlaying packages?
15:18:20 <pwhalen> yes, looking at the screens it just seems to be waiting, not actually failing
15:19:01 <pwhalen> the other issue is dma spam on the rpi3, but thats not unique to iot. Need to try to find where it started
15:19:23 <pwhalen> "bcm2835-dma 3f007000.dma: vchan 00000000df0f59a0: txd 00000000d943061b[b14e]: submitted"
15:21:54 <pwhalen> Anything else for F34?
15:22:06 <pbrobinson> pwhalen: there's a regression in package layering due to bits between libsolv and rpm-ostree
15:22:10 <pbrobinson> I suspect that's the failure
15:22:51 <pwhalen> pbrobinson: I thought the same, but it works on x86. I'll verify and file a new bug if needed
15:23:37 <pbrobinson> pwhalen: https://bugzilla.redhat.com/show_bug.cgi?id=1925717
15:25:04 <pwhalen> thx
15:25:10 <pwhalen> #topic 4) ==== Open Floor ====
15:26:17 <pwhalen> Anything else for today?
15:26:17 <coremodule> I wonder if we could spitball some ideas for testdays?
15:26:50 <pbrobinson> coremodule: as in when, or what to test?
15:26:54 <coremodule> what are your thoughts on multiple test days, each after different milestones?
15:26:54 <pwhalen> no speaking moistly here
15:26:59 <coremodule> lol
15:27:10 <coremodule> both! Is there a specific focus area that would benefit us more than just general testing?
15:28:36 <pwhalen> I think testing on actual hardware
15:28:59 <coremodule> awesome, that's a good clarification
15:29:08 * jsmith shows up late, sorry
15:29:29 <pbrobinson> jsmith: we voted you can organise the test day, thanks!
15:29:40 <jsmith> Sure, that's fine :-)
15:29:42 <pwhalen> :D
15:29:52 <coremodule> haha yes!
15:29:56 <jsmith> It's about time I did something around here :-p
15:30:30 <coremodule> How about three test days total? one in two-three weeks?
15:30:49 <coremodule> and then one a week or two after beta and final each?
15:31:05 <jsmith> Seems fairly reasonable to me.
15:31:21 <pwhalen> Works for me too, coremodule: do you want to take the lead on that?
15:31:29 <pbrobinson> coremodule: the concern I have with 3 test days is will the amount of people that show up make the effort over just doing one worthwhile
15:31:42 <coremodule> sure, I'll spearhead this
15:32:04 <coremodule> pbrobinson, I get that, it's sometimes not worth it at all.
15:32:30 <coremodule> This time around though, we already have most of the testcases written, so the amount of work to coordinate it has gone down significantly
15:32:49 <coremodule> So, while the turnout might be small, the input to get there will also be small(er)
15:33:43 <coremodule> anything specific to test this go around? uefi?
15:33:57 <pwhalen> I think two might fit better, before beta, and before final
15:34:52 <coremodule> pbrobinson, does two sound better to you? I'd really like to host at minimum two, so that any bugs found in the first can be tested for fixes in the second. Not that we can't do that ourselves, but community support always helps
15:35:01 <coremodule> I'm good with two
15:35:32 <pwhalen> I would love to expand the tests. I never really worked out a good way to test parsec. Also the http provisioning with zezere. I've never got that one working and need to revisit
15:36:42 <pwhalen> and any others we should be looking at
15:37:01 <coremodule> okay, I've added those to my list of stuff to look at for the next test day
15:37:08 <pwhalen> if people can think that over, make some suggestions to discuss next week
15:37:23 <coremodule> that sounds great to me. we can pick dates and such next meeting
15:37:26 <pbrobinson> coremodule: you're organising it so it doesn't overly bother me TBH, it was just a concern
15:37:45 <coremodule> pbrobinson, sure, understood.
15:37:59 <coremodule> in that case, let's plan on having two test days; one before beta and one before final
15:38:17 <pwhalen> sold
15:38:36 <pwhalen> anything else?
15:40:00 <pwhalen> #endmeeting