Meeting started at 2025-03-19 17:00:12 UTC

The Meeting name is 'fedora-server'

As usual, let's wait a moment for everybody to show up.

I'll post the agenda in 2-3 minutes.

!hello

Emmanuel Seyman (eseyman) - he / him / his

Hey, everybody

Yeah, greetings!

Hm, the period of time change is always difficult.

I left I got home 10 minutes ago

I left $WORK early and I just got home 10 minutes ago...

Yeah, and we had to reorganize our family dinner and all the cooking for it :-)

!hi

John Himpel (jwhimpel)

Welcome John.

hello, John!

It's almost spring here.

You're lucky. There was frost here this morning.

I was sweating in my biking gloves when I arrived at $WORK this morning

Well, we meet our quorum, Nirik is lurking as always, I think, so we should start.

!undo

Well. copy&paste is always dangerous

And !undo seems not to work for !info.

Anything to add!

Anything to add?

No, that seems fine

OK.

I guess we all noticed that Beta is go.

We can start with our beloved testing.

Yes. I will update my laptop and install Server on an SBC this weekend

Thats good!

Another info is that the planned podcast about server is still unclear. No date yet.

But I hope the bottleneck will soon be resolved and that something will happen soon.

That less good...

Yeah.

Let's proceed.

I'm still awaiting feedback

Yes, sorry. Last week was too busy for me to do anything

I will review tonight or tomorrow

Yes, I started today to make use of it. I'm struggeling a bit to set it up properly.

I haven't got around to systematizing my notes in a comprehensible form yet. I'll do that tomorrow.

Peter Boy:Offline, send me an update on what you're having to struggle.

Yes, I will!

And I wrote a draft of an introduction for our documentation (in stg). But for some reason the documentation update seems not to work,

Yes, there are a lot of changes in stg that are not in prod

Not in prod, but also not in stg, unfortunately.

I'll check later the update process. Or maybe it is blocked because of the beta phase?

it's been stuck for a while, it seems

Yes, the last content update is Last content update: 2024-03-27 Something is missing

And Last build: 2025-03-19 15:50:14 UTC, should be 16:50 UTC

We'll check.

So we have no further updates on the NFS topic, but hopefully some news very soon.

Nevertheless, we should also discuss and decide soon how we want to distribute the Ansible roles.

My idea is to offer the playbooks as rpm packages, which then together form a control node area on the workstation. A first draft of this can be found in the stg area

at least as soon as it get's updated

rpm packages is the way to go

role models would be:

John Himpel: ?? do we agree about rpm?

RPMs would be fine. Probably distributed inside the RPM as a module, but I'm inexperienced on how to create those. Currently I just make raw roles. We can talk about this a bit later.

Yes, it is not urgent. But we should decide about a proper directory structure. My draft contains a proposal, basically it is the structure you propose in the documentation you uploaded for the NFS part.

I have a first ansible script for the webservice - together with a friend's son of mine, who has some experience with Ansible. And we have a rough draft of a Mail service Ansilble script, which it rather extensive and complex.

So we have no further updates on the NFS topic, but hopefully some news very soon.

Let's go on.

no tracking issue yet

Unfortunately, we don't yet have a written plan for this. But basically, we have agreed to carry out the installation routines live and to check the documentation in the process.

And as far as I remember, I'm the only one who has machines available for testing. This means that the installation on hardware would be my responsibility.

At least apart from SBCs

Emmanuel Seyman: You wanted to check the accepted changes. And which SBC do you have for testing?

I can test VM (upgrade and install), but it will probably be next week. I am still rather limited in going up/down stairs to get to computer.

I have several here. I'll probably go with the Banana-Pi

John Himpel: that's good. You'll need to check in particular that the LVM problem is solved correctly.

Emmanuel Seyman: I have an Pine64 rockPro and a Radax Rock Pi. not to duplicate our efforts.

Peter Boy: Do you have a BZ for that, so I can be sure I've tested correctly.

John not out of my head. There is a BZ, I'll send it by mail.

My testing will be install via anaconda as there are/were extensive changes to anaconda.

OK, but it would be helpful to test the VM image as well. I think, we need both, the image and Anaconda.

I think you're thinking of https://bugzilla.redhat.com/show_bug.cgi?id=2265422

Yes,The number looks familiar.

Someone else will need to test VM Image install as I'm not sure how to do that (and probably would not spot any errors vs. my error). :-(

But the arm-image-installer is resolved. The other VM images are critical.

John Himpel: OK, I'll do that.

https://bugzilla.redhat.com/show_bug.cgi?id=2247872, maybe ?

Yes! you found it :-)

Emmanuel Seyman: Peter Boy Thanks!

Unfortunately, my IBM s390x zSystem is currently not ready for use. Does anyone else still have one of these in the basement?? :-)

Not me...

/me looks around in various directions

I can test it in KVM on a z system? but it's virt in virt (in virt)

Nice, yeah virt in virt in virt that's IBM with its "partitions"

if that would be helpful, I do have access to some machines in LinuxONE Community Cloud that I can test with

I always feel a bit uncomfortable. We distribute ppc64 and zsystem variants, but nobody seems to test and report back,

I feel the same way (not about fedora necessarily). It's difficult to accept sending things off into the void and "hoping" they work

Yeah, indeed,

But let'smove on

no tracking issue yet

We didn't change the distribution, so this must be a build error, or at least an issue

I didn't had a closer look yet.

Probably a package (or several packages) grew in size.

But that much ??

We had the same issue several times back. It was always an image generation issue.

Any ideas? I'd hate to just change the permissible size. We deliberately limited it to take into account the conditions in critical areas of our world.

I see that we have already exceeded our time.

Let's discuss this problem on the mailing list.

Anything to discuss? Otherwise I would close.

I'm good

Signing off

OK, then bye bye next week!! And I hope a can mail the agenda earlier.