16:30:08 <bgilbert> #startmeeting fedora_coreos_meeting
16:30:08 <zodbot> Meeting started Wed May 22 16:30:08 2019 UTC.
16:30:08 <zodbot> This meeting is logged and archived in a public location.
16:30:08 <zodbot> The chair is bgilbert. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:30:08 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
16:30:08 <zodbot> The meeting name has been set to 'fedora_coreos_meeting'
16:30:11 <bgilbert> #topic roll call
16:30:13 <bgilbert> .hello2
16:30:14 <zodbot> bgilbert: bgilbert 'Benjamin Gilbert' <bgilbert@backtick.net>
16:30:36 <geoff-> geoff-: Geoff Levand <geoff@infradead.org>
16:30:39 <dustymabe> .hello2
16:30:39 <zodbot> dustymabe: dustymabe 'Dusty Mabe' <dusty@dustymabe.com>
16:30:44 <sayan> .hello sayanchowdhury
16:30:45 <zodbot> sayan: sayanchowdhury 'Sayan Chowdhury' <sayan.chowdhury2012@gmail.com>
16:30:53 <slowrie> .hello2
16:30:54 <zodbot> slowrie: slowrie 'Stephen Lowrie' <slowrie@redhat.com>
16:31:30 <kaeso> .hello lucab
16:31:31 <zodbot> kaeso: lucab 'Luca Bruno' <lucab@redhat.com>
16:31:53 <ajeddeloh> .hello2
16:31:54 <zodbot> ajeddeloh: ajeddeloh 'Andrew Jeddeloh' <andrew.jeddeloh@redhat.com>
16:32:03 <jbrooks> .fas jasonbrooks
16:32:04 <zodbot> jbrooks: jasonbrooks 'Jason Brooks' <jbrooks@redhat.com>
16:32:48 <red_beard> .hello redbeard
16:32:49 <zodbot> red_beard: redbeard 'Brian 'redbeard' Harrington' <bharring@redhat.com>
16:33:08 <ksinny> .hello sinnykumari
16:33:09 <zodbot> ksinny: sinnykumari 'Sinny Kumari' <ksinny@gmail.com>
16:33:36 <bgilbert> #chair geoff- dustymabe sayan slowrie kaeso ajeddeloh jbrooks red_beard ksinny
16:33:36 <zodbot> Current chairs: ajeddeloh bgilbert dustymabe geoff- jbrooks kaeso ksinny red_beard sayan slowrie
16:33:52 <bgilbert> #info Please add items to the meeting agenda at https://public.etherpad-mozilla.org/p/20190522-FCOS-Meeting
16:33:57 <bgilbert> #topic Action items from last meeting
16:34:21 <bgilbert> ksinny to add relevant 'exclude's in the treefile to make sure we don't accidently pull in Python dependent packages in future
16:34:27 <bgilbert> ajeddeloh to create a tracker issue for the *-wrapper topic
16:34:34 <bgilbert> bgilbert will file a ticket on conditionally disabling SMT, with links to PRs
16:34:51 <bgilbert> #info bgilbert filed https://github.com/coreos/fedora-coreos-tracker/issues/181
16:35:10 <ajeddeloh> #info ajeddeloh filed https://github.com/coreos/fedora-coreos-tracker/issues/178
16:35:16 <ksinny> #info we discussed later in the previous meeting that this feature is not yet available and need to get implemneted. Ticket already exist https://github.com/projectatomic/rpm-ostree/issues/1768
16:35:50 <bgilbert> cool, moving on to...
16:35:52 <bgilbert> #topic SMT conditionally disabled by default
16:36:16 <jlebon> .hello2
16:36:17 <zodbot> jlebon: jlebon 'None' <jonathan@jlebon.com>
16:36:18 <rfairley> .hello2
16:36:21 <zodbot> rfairley: rfairley 'None' <rfairley@redhat.com>
16:36:23 <bgilbert> #chair jlebon rfairley
16:36:23 <zodbot> Current chairs: ajeddeloh bgilbert dustymabe geoff- jbrooks jlebon kaeso ksinny red_beard rfairley sayan slowrie
16:36:43 <bgilbert> pursuant to #181, SMT is now automatically disabled in FCOS on CPUs that are vulnerable
16:36:55 <bgilbert> according to the kernel's heuristics
16:37:12 <bgilbert> which means that a future kernel update might disable SMT if new vulnerabilities come out.
16:37:18 <bgilbert> we'll document that default and provide ways to change it.
16:37:42 <bgilbert> further to that, we're discussing ways to change kernel args from an Ignition config
16:37:56 <bgilbert> #link https://github.com/ostreedev/ostree/issues/479
16:38:01 <bgilbert> #link https://github.com/coreos/ignition-dracut/issues/81
16:38:11 <bgilbert> the first one is the kargs part, and the second one is for rebooting after changing kargs
16:38:18 <bgilbert> so the first boot has the same kargs as the others
16:38:28 <bgilbert> that's all I have on that topic.  questions/comments?
16:39:39 <red_beard> seems reasonable
16:40:01 <bgilbert> #topic created new repo for releng automation for coreos-pool koji tag
16:40:19 <dustymabe> this is mostly just an #info
16:40:31 <dustymabe> #info created new repo for releng automation for coreos-pool koji tag: https://pagure.io/dusty/coreos-koji-tagger
16:40:37 <dustymabe> will move it to new location (possibly GH if infra openshift can pull from there) when appropriate
16:41:13 <dustymabe> the input to this tool will be the lockfile json - the output will be interacting with koji to tag in builds into the coreos-pool tag
16:41:34 <bgilbert> +1
16:42:56 <bgilbert> #topic releng approved GitHub as source for FCOS configs
16:43:48 <dustymabe> #info releng approved GitHub as source for FCOS configs
16:44:06 <ksinny> dustymabe++
16:44:07 <dustymabe> jlebon had a concern about this when closing out the previous releng related ticket
16:44:23 <dustymabe> they discussed during the meeting today and approved: https://pagure.io/releng/issue/8373#comment-571672
16:45:04 <bgilbert> nice
16:46:49 <jlebon> dustymabe: nice, thanks for following up on that!
16:46:57 <dustymabe> np
16:47:02 <bgilbert> #topic Open Floor
16:47:07 <bgilbert> anyone have anything else?
16:47:30 <red_beard> so to follow up on one of my hare brained ideas.... thoughts on
16:47:55 <red_beard> "spins" of FCOS.  once we get to the point of releases, what tolerance does the overall group have for that?
16:48:13 <red_beard> aka more opinionated variants of the build for laser specific purposes.
16:48:18 <cleber> hi, is there an agenda for the Fedora Cloud Workgroup Meeting?
16:48:53 <bgilbert> cleber: this meeting is for Fedora CoreOS.  not sure if anyone here can answer your question
16:49:11 <bgilbert> I'll give my opinion on red_beard's question
16:49:25 <jlebon> red_beard: as in, spins we produce ourselves, or whether we're open for our tooling to support others doing this?
16:49:27 <bgilbert> can't speak for the group, and I've actually already talked to red_beard offline :-)
16:49:28 <cleber> bgilbert: thanks, AFAICT it's supposed to start in 11 minutes - sorry to disturb
16:49:29 <dustymabe> hi cleber we haven't been having that meeting for a while - i should remove it from the calendar
16:49:33 <ajeddeloh> red_beard: my thought is that with COSA people who want to make a spin can. Not sure about how hosting updates and such would work
16:49:58 <dustymabe> cleber: ping me in #fedora-cloud if you want to chat
16:50:08 <cleber> dustymabe: cool, thanks
16:50:09 <bgilbert> the tooling is open and people can build whatever they want
16:50:24 <ajeddeloh> but our goal should not be to make spins. i.e. don't prevent others from doing that, but don't support them directly
16:50:30 <bgilbert> FCOS is already supposed to be a pretty focused distro
16:50:50 <bgilbert> if people need to do their own spins for something inside our declared use cases, we're doing something wrong
16:51:04 <red_beard> :thumbsup: i just wanted to make sure that some of the offline discussions got captured (with the sentiment) so that it's part of the public record :D
16:51:14 <kaeso> I think we aren't moving far away from ContainerLinux position on that
16:51:21 <bgilbert> and if they need to do spins for something outside our use cases, that's great but we probably won't have any reason to publicize those :-)
16:51:37 <bgilbert> yup, agree with ajeddeloh and kaeso
16:51:41 <jlebon> ajeddeloh: i'd reword it as, allow our tools to support that, as long as it doesn't conflict with how we use the tools ourselves
16:51:46 <dustymabe> yeah - that being said it seems like a spin outside our use case could possibly use some of the same tooling
16:51:55 <ajeddeloh> jlebon++
16:51:56 <jlebon> (where "conflict" includes increased maintenance burden)
16:51:57 <red_beard> kaeso:agreed this is around things like the kubelet, etc
16:51:58 <zodbot> ajeddeloh: Karma for jlebon changed to 1 (for the current release cycle):  https://badges.fedoraproject.org/tags/cookie/any
16:52:07 <dustymabe> for example: Fedora IoT ?
16:52:26 <red_beard> also a very good point
16:52:52 <red_beard> the number of requsts for CoreOS/Container Linux on RPis was immense over the years
16:53:36 <dustymabe> i think the *spins* thing is something we definitely shouldn't concern ourselves with right now while we focus on getting something shipped
16:53:50 <bgilbert> re RPi, I wonder if that'd actually be more doable for FCOS than CL
16:53:51 <ajeddeloh> dustymabe++
16:53:57 <bgilbert> dustymabe: +1
16:54:40 <dustymabe> that being said - I think ostree itself lends us to the ability to have common base with other use cases and then a layer on top for each specific use case
16:54:41 <bgilbert> conversely, spins/forks/downstreams can be useful for proving out a concept that then gets integrated back into the origin
16:54:58 <bgilbert> so I'd encourage people to experiment if they want to experiment
16:55:01 <dustymabe> but again - that would be at least 6 months away
16:55:20 <bgilbert> and see where things go
16:55:22 <red_beard> jlebon: for what it's worth this is more just having the *possibility* to do something congruent with Fedora (standard) not in the foreseeable future for it being a ratified thing.  as dustymabe pointed out, we need to ship it before we get too crazy with it.
16:55:24 <jlebon> for folks who haven't seen it, they might be interested in the Fedora Silverblue repo: https://pagure.io/workstation-ostree-config/tree/master
16:55:38 <jlebon> there's various "overlays" for different spins
16:56:11 <jlebon> though AFAIK only the vanilla GNOME one is built in releng
16:56:53 <jlebon> red_beard: +1
16:58:15 <bgilbert> any other topics?
16:58:36 <kaeso> nack
17:00:08 <bgilbert> thanks for coming, all!
17:00:09 <bgilbert> #endmeeting