09:29 <travier[m]> I need some reasonably easy review for this one: https://github.com/coreos/fedora-coreos-docs/pull/459

11:49 <c4rt0> @travier done.

12:37 <travier[m]> thanks!

15:45 <jlebon> dustymabe: refreshing my memory, the reason i had put `broadcast-fedmsg.py` is in the releng-auto repo is that it's very specific to FCOS. `versionary.py` falls in the same boat. i'm thinking we could just have them be in the pipeline repo instead. then we'll already have it since we check it out anyway. it's not a great fit since we're trying to de-FCOS-ify it, but it feels better than in cosa.

15:56 <dustymabe> jlebon: I think we have a lot of FCOS specific things already in COSA, so let's just pile it in. For example: https://github.com/coreos/coreos-assembler/blob/main/src/cosalib/fedora_messaging_request.py

15:56 <dustymabe> If we wanted to extricate all FCOS things from COSA that should probably be part of a larger effort IMO

15:57 <dustymabe> but yeah, pipeline repo works too

15:58 <jlebon> maybe broadcast-fedmsg.py in cosa since it uses cosa APIs and versionary.py in the pipeline repo

16:31 <dustymabe> aaradhak davdunc dustymabe gursewak jaimelm jbrooks jcajka jdoss jlebon jmarrero lorbus miabbott nasirhm ravanelli saqali skunkerk walters

16:31 <dustymabe> FCOS community meeting in #fedora-meeting-1

16:31 <dustymabe> If you don't want to be pinged remove your name from this file: https://github.com/coreos/fedora-coreos-tracker/blob/main/meeting-people.txt

18:02 <jlebon> dustymabe: ended up going with cosa since we don't actually have the pipeline repo checked out in the cosa container

18:02 <jlebon> https://github.com/coreos/coreos-assembler/pull/3113

18:02 <jlebon> https://github.com/coreos/fedora-coreos-pipeline/pull/643

18:15 <dustymabe> jlebon: i'll take a look after I get back from lunch

19:39 <walters> oh wow I'd missed that podman grew `podman container runlabel` which was like a key feature of the old `atomic` command, and it's now used by e.g. https://en.opensuse.org/openSUSE:ALP/Workgroups/SysMngmnt/Cockpit#Cockpit_at_ALP

19:40 <walters> this is like the inverse of using things like butane to generate systemd units to run containers

19:42 <MatthieuSaulnier> yes, but are the containers launched at startup like a background service (a systemd service)?

19:44 <walters> the effect is entirely under the control of the container content: ```skopeo inspect -n docker://registry.opensuse.org/suse/alp/workloads/tumbleweed_containerfiles/suse/alp/workloads/cockpit-ws:latest | grep INSTALL

19:44 <walters> ```

19:44 <walters> "INSTALL": "docker run --rm --privileged -v /:/host -v /etc/machine-id:/etc/machine-id:ro -e IMAGE=${IMAGE} ${IMAGE} /container/label-install ${IMAGE}",

19:44 <walters> * the effect is entirely under the control of the container content:... (full message at <https://libera.ems.host/_matrix/media/r0/download/libera.chat/222c7790ac218a89d29910e4e6ad3e03aedd2753>)

20:05 <dustymabe> jlebon: should we wait for https://github.com/coreos/coreos-assembler/pull/3113#issuecomment-1261400215 or merge away?

20:06 <jlebon> dustymabe: merged and marked dependent PRs as ready :)

20:09 <dustymabe> jlebon: you're going to make my local changes for secrets updates fun to rebase when this merges.. but hey, you can't say I didn't ask for it

20:11 <jlebon> dustymabe: heh. i'm cool holding the pipeline PR if you prefer

20:12 <dustymabe> ehh, I've thought about it but I think I decided against it. actually let me do a test run to see how hard it is to rebase on what you've got

20:39 <dustymabe> jlebon: ok I rebased on top of it.. we should be good to merge your stuff once it passes review

20:41 <jlebon> dustymabe: ahh hadn't seen your comment

20:42 <dustymabe> jlebon: which one :)

20:46 <jlebon> dustymabe: to split into two commits. did that now!

20:46 <jdoss> walters: fine fine fine! you win! (thank you for explaining things to me in #1197)

21:02 <dustymabe> jlebon: https://github.com/coreos/fedora-coreos-pipeline/pull/643#issuecomment-1261460877

21:03 <jlebon> dustymabe: hmm, not sure I follow

21:03 <jlebon> even if we keep running it on the builder, it'll be in the cosa image now, right?

21:03 <jlebon> we only do this for FCOS where we don't need to worry about old cosa

21:03 <dustymabe> jlebon: just isolating on the first commit (thinking about it standing on its own)

21:04 <dustymabe> if you no longer run the `send-ostree-import-request.py` on the remote then there is no need to `cosa remote-session sync {,:}/var/tmp/fcos-releng/coreos-ostree-importer/send-ostree-import-request.py`

21:04 <dustymabe> so let's remove that in the first commit too

21:04 <jlebon> the first commit is about the fact that cosa now ships the scripts, so there's no need to copy it in

21:04 <dustymabe> ahh I see

21:05 <dustymabe> ok that's fine

21:05 <dustymabe> I had the order of the commits wrong

21:05 <dustymabe> I think maybe let's merge this in the morning

21:06 <dustymabe> I need to do open heart surgery on the pipeline tomorrow anyway to fully test out my changes so we'll just do them all at the same time

21:06 <jlebon> ttyl!