Meeting started at 2026-09-16 15:30:22 UTC

The Meeting name is 'fedora_coreos_meeting'

Meeting already in progress

!hi

Angel Cervera Roldan: Angel Cervera Roldan (acervera)

!hi

marmijo: Michael Armijo (marmijo)

!hi

Tiago Bueno: Tiago Bueno (tlbueno) - he / him / his

!hi

Hristo Marinov: Hristo Marinov (hricky) - he / him / his

!hi

jbtrystram: Jean-Baptiste Trystram (jbtrystram) - he / him / his

!hi

nimbinatus: Laura Santamaria (nimbinatus) - she / her / hers

!hi

!hi

dustymabe: Dusty Mabe (dustymabe) - he / him / his

rapneset: Rolv Apneseth (rapneset)

Hey everyone, I guess there is no action items from the last meeting so we can talk about our current fedora release schedule

F45 is already in the next stream! ✨️🎉

Thanks apiaseck for working on the relase 🌟

Thanks marmijo for the nice announcement email

Any topics related to F45 release?

Yasmin Valim de Souza: I added a topic to the meeting checklist. The meeting label was added after the issue was created.

Nice, we can talk about it in the tracker topics.

I just have a quick reminder related to it and also a invite to everyone

Thank you marmijo!

dustymabe has already given cookies to jbtrystram during the F44 timeframe

dustymabe has already given cookies to ydesouza during the F44 timeframe

Next week we have our FCOS 45 test week and I hope everyone can join us!

dustymabe has already given cookies to ydesouza during the F44 timeframe

dustymabe has already given cookies to marmijo during the F44 timeframe

acervera has already given cookies to ydesouza during the F44 timeframe

rapneset has already given cookies to ydesouza during the F44 timeframe

marmijo gave a cookie to ydesouza. They now have 16 cookies, 8 of which were obtained in the Fedora 44 release cycle

We have a live meeting Monday, September 21 from 10:00 AM – 11:00 AM EDT (2:00 PM – 3:00 PM UTC)

tlbueno gave a cookie to ydesouza. They now have 17 cookies, 9 of which were obtained in the Fedora 44 release cycle

If you cannot join the meeting, feel free to test it anyway. Here is some helpful links.

That's it! Let's go to our topics from the tracker!

First one is from Rolv Apneseth

That one's from me, added it only shortly before this meeting. We're considering dropping all DTBs in aarch64 based on what was said in that issue. Do people have strong opinions about this?

It's to avoid the problems we're having with aarch64's space usage in /boot

rapneset: I think maybe all DTBs is the ultimate destination.. it would have been better to drop them prior to F45 beta (i.e. test week)

So we would only want to drop some for now or what do you mean?

it would be good for the image that people do testing on to have the change if we drop them all

so we could make the change and spit out a new release for `next` early next week

OR we be more conservative and just target say `apple` for now and make the change next cycle

it would be good for the image that people do testing on to have the change if we drop them all

so we could make the change and spit out a new release for `next` early next week

OR we be more conservative and just target say `apple` for now and make the bigger change next cycle

WDYT?

I would go for the big change now and produce an image before monday

I would be happy with just apple - we need to find other ways to reduce the size in the boot partition anyway

Apple would only give us 2MiB. I think if we want a meaningful improvement I'd go with the big change now

2 MiB per deployment isn't bad either though

ok. we'd likely want to send out a followup email to the announcement marmijo just sent out with the new information

I am hoping that we can find something else to remove from all architectures (I am not sure I see anything ppc64le specific that we can remove to fix that architecture)

But removing all would be ideal ofc, since then we dont have to worry about aarch64 anymore

if IoT don't ship any, i feel the likeliness of this change breaking anyone is low ?

jbtrystram: we'll find out :)

I think aarch64 is much more important than ppc64le (at the very least for FCOS). For ppc we're looking into compressing the kernel right?

yes, but from what I have read it seems like it may not be possible without changes to petitboot. Waiting to hear back from people with more knowledge on this tho.

It seems like a reasonable thing to ask of petitboot

regardless, I do think that aarch64 is a lot more important, so I'm very happy with removing all dtbs too

So, how do we proceed here then? Should we vote on if we want to get that into this coming test week?

We can make a proposed "test to agree upon" and then vote if you like

We can make a proposed "text to agree upon" and then vote if you like

Do we want to propose this now here?

Yes we can do one now

Would we have to cut a new next release to land this before the test week?

I am on the deck to do the release next week

Should we skip this because the test week or not?

you're up for the week of the 29th, which is in two weeks :)

Ah, right! Thats it! So no releases next week?

marmijo: yes. we'd have to do that I think

basically we could do it after.. but the test week would be a really good indicator of if anything is going to break with this change or not

PROPOSE: Do we want to remove the DBTs before the test week so we can test it properly?

+1

+1

+1

I agree that would be good to test it.

+1

+1

+1

+1

+1

I think we would have to use "!propose" keyword?

I don't think it works

!propose .

!idea works now IIRC

agreed! We agreed in this meeting to remove the DBTS before the FCOS 45 test week so we can test and be sure that anything is going to break with this change or not.

!idea: Do we want to remove the DBTs before the test week so we can test it properly?

oh nice, the bot picks up the edits !

I think it's new :)

Is not in the right order, but I guess we can understand it in the logs after the meeting :)

Ok, so after this we have one more topic to discuss in the tracker

marmijo: Is this one you said that you added?

No, I added the previous topic to the checklist

For this one, we've just kept the meeting label on it for a while now

Since we're well past the deadlines, there are no new change considerations to review. We also just sent out beta with some changes highlighted in the announcement.

Ah, right! So I guess we don't have any other things to discuss about the F45 release? Or there is any topic you want to bring?

Looks like we decided to defer change 228 to F46: https://fedoraproject.org/wiki/Changes/CoreOS_default_to_enable_systemd-oomd_and_swap_on_Zram

https://fedoramagazine.org/announcing-fedora-linux-45-beta/ mentions the support systemd-oomd and zram swap support to Fedora CoreOS.

We should probably get that edited out.

woops :)

Yeah I've seen other places online saying CoreOS supports that now in F45

The PR is ready otherwise, we could just get it in now

yeah, I realize that from another announcement as well.

I will take care of this for the Fedora Docs Release page if we are going to defer it.

I will take care of this for the Fedora Docs Release Notes page if we are going to defer it.

Thanks, Hristo Marinov :)

Can we go to our open floor?

Or do you have any other insights about this topic?

There's no way we can get it in to F45 no?

i'm not sure of the process, but the change request already shows deferred to F46

if we are rebuilding a new image next week we can just get it. But heh since publicly said we deferred it let's not flip-flop

Alright

Nice. Let's use our last minutes to do a quick open floor!

I released CLHM to rawhide today, since the selinux-policy PR that was blocking us got merged and released

We should get systemd-confext support now :)

jbtrystram: nice

is that selinux -policy coming to f45/f44 too? should we fast-track?

Nice work jbtrystram 🎉

rapneset gave a cookie to jbtrystram. They now have 9 cookies, 6 of which were obtained in the Fedora 44 release cycle

We have a brainstorming doc for the unification discussion: https://pad.riseup.net/p/RGQAC_uRxVRCxaHw_OIQ

it's released in f45

Note that https://github.com/coreos/fedora-coreos-tracker/issues/2217 means we have selinux-policy pinned

I only released CLHM to rawhide for now

thanks for the reminder rapneset

Nice reminder, Hristo Marinov! Let's keep the unification discuss present in our meetings and I hope we can unify more the two community soon.

rapneset: we need to push upstream to get fixes on those issues. Can we reach out to the devs?

rapneset: ahh just saw it's fixed https://github.com/fedora-selinux/selinux-policy/issues/3392

I did yeah. They commented that the latest version of `container-selinux` fixed this but I couldn't get it to work (noted in https://github.com/coreos/fedora-coreos-config/pull/4383)

I guess let's comment there then

Sure yeah

Well, we are already out of time. If there is any other subjects or discussions we can keep then in our Matrix channel or even in the issues comments.

Thank you all for being here and participate!

Thanks for hosting Yasmin Valim de Souza

And please, feel free to contribute in our Test Week! 🩵

thanks for hosting!

thanks Yasmin Valim de Souza