16:31:19 <jlebon> #startmeeting fedora_coreos_meeting
16:31:19 <zodbot> Meeting started Wed Oct 19 16:31:19 2022 UTC.
16:31:19 <zodbot> This meeting is logged and archived in a public location.
16:31:19 <zodbot> The chair is jlebon. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions.
16:31:19 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
16:31:19 <zodbot> The meeting name has been set to 'fedora_coreos_meeting'
16:31:24 <jlebon> #topic roll call
16:31:25 <bgilbert> .hi
16:31:26 <zodbot> bgilbert: bgilbert 'Benjamin Gilbert' <bgilbert@backtick.net>
16:31:32 <marmijo> .hi
16:31:33 <zodbot> marmijo: marmijo 'Michael Armijo' <marmijo@redhat.com>
16:31:34 * bgilbert waves at zodbot
16:31:54 <jlebon> #chair bgilbert marmijo
16:31:54 <zodbot> Current chairs: bgilbert jlebon marmijo
16:33:01 <travier> .hello siosm
16:33:02 <zodbot> travier: siosm 'Timothée Ravier' <travier@redhat.com>
16:33:28 <jlebon> #chair travier
16:33:28 <zodbot> Current chairs: bgilbert jlebon marmijo travier
16:33:55 <dustymabe> .hi
16:33:56 <zodbot> dustymabe: dustymabe 'Dusty Mabe' <dusty@dustymabe.com>
16:34:31 <jlebon> #chair dustymabe
16:34:31 <zodbot> Current chairs: bgilbert dustymabe jlebon marmijo travier
16:34:51 <jlebon> small group today :)  let's wait another minute
16:35:25 <mnguyen> .hello mnguyen
16:35:26 <zodbot> mnguyen: mnguyen 'Michael Nguyen' <mnguyen@redhat.com>
16:35:53 <jlebon> #chair mnguyen
16:35:53 <zodbot> Current chairs: bgilbert dustymabe jlebon marmijo mnguyen travier
16:36:02 <jlebon> alrighty, let's get going!
16:36:04 <jlebon> #topic Action items from last meeting
16:36:06 <spresti[m]> .hi
16:36:06 <copperi[m]> .hi
16:36:06 <zodbot> spresti[m]: Sorry, but user 'spresti [m]' does not exist
16:36:09 <zodbot> copperi[m]: Sorry, but user 'copperi [m]' does not exist
16:36:18 <jlebon> looks like there were none! :tada:
16:36:25 <jlebon> #chair spresti[m] copperi[m]
16:36:25 <zodbot> Current chairs: bgilbert copperi[m] dustymabe jlebon marmijo mnguyen spresti[m] travier
16:36:56 <jlebon> #topic Boot partition can easily run out of space on upgrade
16:36:59 <jlebon> #link https://github.com/coreos/fedora-coreos-tracker/issues/1247
16:37:21 <jlebon> no update here, this is just a call for help to implement what was decided in https://github.com/coreos/fedora-coreos-tracker/issues/1247#issuecomment-1261286872
16:37:22 <walters> .hi
16:37:22 <spresti[m]> .hello spresti
16:37:22 <zodbot> walters: walters 'Colin Walters' <walters@redhat.com>
16:37:25 <zodbot> spresti[m]: spresti 'Steven Presti' <spresti@redhat.com>
16:37:33 <jlebon> it's currently preventing us from shipping ppc64le
16:37:38 <jlebon> #chair walters
16:37:38 <zodbot> Current chairs: bgilbert copperi[m] dustymabe jlebon marmijo mnguyen spresti[m] travier walters
16:38:18 <jlebon> so if anyone is interested in learning about the initrd, and possibly ostree, this could be a good opportunity
16:38:27 <jmarrero> .hi
16:38:29 <zodbot> jmarrero: jmarrero 'Joseph Marrero' <jmarrero@redhat.com>
16:38:37 <jlebon> #chair jmarrero
16:38:37 <zodbot> Current chairs: bgilbert copperi[m] dustymabe jlebon jmarrero marmijo mnguyen spresti[m] travier walters
16:38:44 <jlebon> dustymabe: anything else you want to say on this before we move on?
16:38:56 <dustymabe> nope.. mostly just a call for help
16:39:14 <jlebon> ack, let's move on then
16:39:18 <jlebon> #topic remove console=ttyS0 on metal
16:39:21 <jlebon> #link https://github.com/coreos/fedora-coreos-tracker/issues/567
16:39:45 <jlebon> dustymabe: you tagged this one, so I'll tag you :)
16:40:30 <dustymabe> right. mostly wanted to bring this up because of the question posted by jcpowermac at the bottom
16:40:51 <dustymabe> i.e. should we make any changes for vSphere?
16:41:08 <dustymabe> I don't have a strong opinion
16:41:19 <bgilbert> yeah, that's on my list.  I think the answer is probably yes
16:41:33 <dustymabe> bgilbert: should we have the user open a new issue?
16:41:41 <bgilbert> nah
16:41:45 <bgilbert> FCOS docs currently have "how to get to the serial console" instructions for both vmware and virtualbox
16:42:11 <bgilbert> so we either have to delete those, or reinstate the serial console there
16:42:18 <bgilbert> removing it in the first place was a bit of a judgment call
16:42:49 <bgilbert> I'm inclined to make it secondary rather than primary, which means the instructions wouldn't be 100% useful
16:42:53 <bgilbert> but would still catch most things
16:42:56 <bgilbert> interested in thoughts though
16:42:59 <dustymabe> so how do we track this bit? i'm pretty close to marking this particular issue as completed
16:43:20 <bgilbert> it's on my plate
16:43:26 <dustymabe> i.e. when the changes rollout to testing/stable
16:43:34 <dustymabe> ok so keep it in this issue then?
16:43:40 <bgilbert> yeah, I'll get a PR up
16:43:56 <dustymabe> WFM - anyone else have input?
16:44:04 <jlebon> bgilbert: what happens if no serial port is connected to the VM?
16:44:19 <bgilbert> that's been the situation historically and we haven't had any problem reports that I know of
16:44:29 <bgilbert> we don't enable a serial port by default in either platform
16:44:46 <bgilbert> (well, vmware might but I don't think we ask for one)
16:44:52 <jlebon> is that an OVF setting?
16:45:13 <bgilbert> https://docs.fedoraproject.org/en-US/fedora-coreos/provisioning-vmware/#_troubleshooting_first_boot_problems
16:45:20 <bgilbert> presumably
16:45:35 <jlebon> would it make sense to specify it there?
16:45:46 <bgilbert> why?
16:46:08 <bgilbert> the user still needs to configure it to route the output somewhere
16:46:43 <jlebon> ahh ok, I thought the instructions there was the same as what the OVF change would do, but on the fly
16:47:09 <jlebon> but right, we don't want to be hardcoding a console log file path in there
16:47:48 <bgilbert> +1
16:47:50 <jlebon> if adding a console arg can't cause any issues when it's unused, then WFM too
16:48:03 <jlebon> bgilbert: can i action you to follow up on the issue?
16:48:25 <bgilbert> sure
16:48:33 <jlebon> #action bgilbert will follow up on https://github.com/coreos/fedora-coreos-tracker/issues/567 re. VMware
16:48:37 <jlebon> ok let's move on
16:48:54 <jlebon> #topic tracker: Fedora 37 changes considerations
16:48:57 <jlebon> #link https://github.com/coreos/fedora-coreos-tracker/issues/1222
16:49:05 <walters> Do we still dump e.g. ignition failures to the console, or it's 100% silent now?
16:49:21 <jlebon> dustymabe: i'll let you take that one too
16:50:11 <dustymabe> jlebon: I think we're mostly done here. Just want to make sure there isn't anything we should follow up on
16:50:23 <dustymabe> look for the ⚠️ in the issue description
16:51:48 <bgilbert> walters: we should dump them to both primary and secondary consoles
16:52:27 <dustymabe> jlebon: for the "Preset All Systemd Units on First Boot" one we have https://github.com/coreos/ignition/issues/1440 broken out
16:53:22 <dustymabe> i guess in f37 that will be enabled and we can start to experiment with dropping the ignition bits
16:53:38 <jlebon> yup, indeed
16:53:45 <jlebon> i think the only other one that stands out is https://github.com/coreos/fedora-coreos-tracker/issues/1230
16:54:12 <jlebon> we'll want to make sure we follow-up on that after GA to be ready for f38
16:54:40 <dustymabe> jlebon: I wonder if they've already switched rawhide over
16:55:24 <jlebon> oh yeah, good point. we'll need to dig in there
16:55:48 <jlebon> dustymabe: for the group do you want to do a TL;DR of the next few releases
16:56:06 <dustymabe> going all the way to the change page it looks like the changes won't actually reach stable users until F39
16:57:12 <dustymabe> still on the changes topic.. It would be nice to dig into this too: https://github.com/coreos/fedora-coreos-tracker/issues/1231
16:57:28 <dustymabe> i.e. dropping our reliance on syslinux
16:57:53 <dustymabe> jlebon: yeah I can do a summary of the releases coming up
16:58:02 <jlebon> ahh right, so there is a possible time-sensitive aspect to that one too
16:58:18 <bgilbert> do we know whether syslinux will be dropped in the same release?
16:58:21 <bgilbert> i.e. can we lag a bit
16:58:23 <jlebon> i feel like we should tag those differently
16:58:55 <dustymabe> bgilbert: we'll definitely know when syslinux gets dropped because things will start failing in rawhide early early
16:59:03 <bgilbert> well yes :-)
16:59:08 <dustymabe> but I'm not sure what the plan is on what release they are going to try to drop it
16:59:25 <dustymabe> and... other community members might pick it up, so it might not go away anytime soon
16:59:45 <dustymabe> but I think getting away from it if possible without too much pain, would be ideal.
16:59:53 <bgilbert> sure
17:01:16 <dustymabe> new topic for releases?
17:01:29 <dustymabe> there was another topic right about hetzner
17:01:45 <jlebon> wasn't planning on it, but we can if you'd like
17:02:03 <jlebon> we can do it at the end if you prefer
17:02:32 <dustymabe> either way. doesn't matter
17:02:34 <jlebon> #topic Platform Request: Hetzner
17:02:37 <jlebon> #link https://github.com/coreos/fedora-coreos-tracker/issues/1324
17:02:58 * dustymabe vaguely recalls other requests for hetzner in the past
17:03:00 <jlebon> we've finally had a request for adding support for Hetzner
17:03:07 <bgilbert> :-)
17:03:13 <jlebon> i feel like we've been circling it for a while
17:03:21 <bgilbert> yeah
17:03:42 <jlebon> my understanding is that there's a coreos-installer flow possible currently
17:04:29 <jlebon> bgilbert: hah, love how there's not one, but two Ignition PRs for it
17:05:08 <jlebon> it feels like the first step is to get someone to find the answers to the questions in the template
17:05:29 <bgilbert> I think we can extract most of the info we need from the old PR discussions
17:05:37 <bgilbert> e.g. hostname and metadata service
17:05:52 <bgilbert> I think everything was originally blocked on the metadata endpoints not being stable, but they were later stabilized
17:06:17 <jlebon> for the rest, do we have contact from anyone who works there?
17:06:17 <dustymabe> the biggest obstacle is probably just having somebody with enough perserverance to work on it until all the work is done
17:06:28 <bgilbert> in sum, I don't know if we have 100% of the info we need, but we can probably get close
17:06:30 <bgilbert> dustymabe: +1
17:06:42 <jlebon> indeed, yup
17:07:08 <dustymabe> i'm pretty sure travier has used them in the past - so he'd be familiar, but probably too busy
17:07:33 <jlebon> ok, let me try a proposed
17:08:18 <jlebon> #proposed we would like to add support for Hetzner, but currently don't have the cycles for it. we need someone willing to pick it up and push it forward.
17:09:32 <jlebon> or is anyone opposed to adding the platform at all even if we had someone offer?
17:09:45 <dustymabe> i'm not, sounds like a nice platform
17:10:02 <travier> I know about Hetzner but I've never used them. I've seen some guides for manual setup with cosi and linked them somewhere in an issue in the docs repo
17:10:07 <dustymabe> though IIUC they have a lot of offerings, so it would help if we decide on what we want to focus on there
17:10:15 <bgilbert> +1 to adding the platform
17:10:19 <dustymabe> travier: sorry. I was mistaken
17:10:24 <bgilbert> I'm tempted to weaken "don't have the cycles" a bit
17:10:28 <travier> +1 to adding the platform. We should reach out to them
17:10:28 <bgilbert> and just say we'd like volunteers
17:10:54 <jlebon> bgilbert: WFM
17:11:04 <dustymabe> +1
17:11:13 <copperi[m]> +1
17:11:22 <jlebon> #proposed we would like to add support for Hetzner. we are looking for volunteers to pick it up and push it forward.
17:11:27 <bgilbert> +1
17:11:47 <spresti[m]> +1
17:12:12 <jlebon> #agreed we would like to add support for Hetzner. we are looking for volunteers to pick it up and push it forward.
17:12:38 <travier> +1
17:12:38 <jlebon> ok, moving on unless anyone wants to add anything
17:13:06 <jlebon> #topic Fedora 37 rollout schedule
17:13:10 <travier> 👍
17:13:21 <jlebon> dustymabe: take it away
17:13:54 <dustymabe> 👋
17:14:15 <dustymabe> This is mostly a reminder of our strategy around "Major Fedora Version Rebases"
17:14:28 <dustymabe> #link https://github.com/coreos/fedora-coreos-tracker/blob/main/Design.md#major-fedora-version-rebases
17:15:05 <dustymabe> assuming Fedora 37 is approved as "Go" on thursday of this week we'll immediately cut/ship a `next` release on that day
17:15:22 <dustymabe> that content will then be promoted to `testing` next week
17:15:29 <dustymabe> and `stable` two weeks later
17:16:38 <jlebon> to clarify, if we're Go this thursday, GA should be on tuesday next week, matching our testing release
17:16:39 <dustymabe> of course if the decision is delayed until next week then the schedule moves
17:17:03 <dustymabe> jlebon: right - `testing` would be on F37 next tuesday
17:17:53 <dustymabe> any questions?
17:18:05 <jlebon> cool, thanks dustymabe for keeping track of this
17:18:19 <dustymabe> 🙂
17:18:42 <jlebon> will move on to open floor if there's nothing else
17:18:57 <jlebon> #topic Open Floor
17:19:20 <jlebon> nice, we actually have time for an open floor :)
17:19:51 <dustymabe> #info thanks copperi[m] for helping out with docker ecosystem packaging! link: https://src.fedoraproject.org/rpms/runc/pull-request/4
17:19:51 <travier> Looks good
17:20:30 <bgilbert> copperi[m]++
17:20:42 <travier> copperi[m]++
17:20:44 <travier> thanks!
17:20:45 <dustymabe> i'll get that package update into the `next` release next week
17:21:01 <jlebon> just a random thought i had: a common pain point with FCOS development is a lot of things get stuck at decided/pending-action. i wonder if we should do a larger effort to reach out to the Fedora community to see if anyone is interested.
17:21:19 <jlebon> e.g. cross-posting on the community forum and the devel list
17:21:38 <bgilbert> jlebon: yeah, wouldn't hurt.  I don't think that's 100% a bug though
17:22:04 <dustymabe> yeah - outreach would be nice. unfortunately outreach (and the mentoring that comes along with it) is also work :(
17:22:40 <bgilbert> we're maintaining a nice pipeline of work for anyone looking for some :-)
17:22:52 <jlebon> dustymabe: indeed, the mentoring part is the expensive bit
17:24:02 <dustymabe> #info we started producing azure aarch64 artifacts
17:24:22 <dustymabe> got an instance to run it yesterday
17:24:26 <jlebon> bgilbert: right, i think there's a lot of tickets there with varying degrees of priority. i could imagine some being worth the overhead and others not.
17:25:15 <bgilbert> yeah, and often we have to discuss it here to figure out how much overhead there is
17:25:53 <jlebon> the overhead overhead
17:26:13 <jlebon> heh i feel like i just explained your joke
17:26:34 <dustymabe> 🤣
17:26:45 <jlebon> ok, closing in 52s unless anyone has anything else
17:26:48 <bgilbert> :-P
17:27:26 <walters> I think posting, but having people offer to mentor is good
17:27:40 <jlebon> #endmeeting