16:30:14 <dustymabe> #startmeeting fedora_coreos_meeting
16:30:14 <zodbot> Meeting started Wed Oct  3 16:30:14 2018 UTC.
16:30:14 <zodbot> This meeting is logged and archived in a public location.
16:30:14 <zodbot> The chair is dustymabe. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:30:14 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
16:30:14 <zodbot> The meeting name has been set to 'fedora_coreos_meeting'
16:30:19 <bgilbert> .hello2
16:30:19 <dustymabe> #topic roll call
16:30:19 <zodbot> bgilbert: bgilbert 'Benjamin Gilbert' <bgilbert@backtick.net>
16:30:24 <dustymabe> .hello2
16:30:24 <jbrooks> .fas jasonbrooks
16:30:25 <lorbus> .hello2
16:30:25 <zodbot> dustymabe: dustymabe 'Dusty Mabe' <dusty@dustymabe.com>
16:30:27 <zodbot> jbrooks: jasonbrooks 'Jason Brooks' <jbrooks@redhat.com>
16:30:29 <mskarbek> .hello2
16:30:30 <zodbot> lorbus: lorbus 'Christian Glombek' <c@petersen-glombek.de>
16:30:31 <rfairley> .hello rfairleyredhat
16:30:33 <zodbot> mskarbek: mskarbek 'None' <redhat@skarbek.name>
16:30:36 <zodbot> rfairley: rfairleyredhat 'Robert Fairley' <rfairley@redhat.com>
16:30:36 <ksinny> .hello sinnykumari
16:30:39 <zodbot> ksinny: sinnykumari 'Sinny Kumari' <ksinny@gmail.com>
16:30:40 <mnguyen_> .hello mnguyen
16:30:45 <zodbot> mnguyen_: mnguyen 'Michael Nguyen' <mnguyen@redhat.com>
16:31:01 <ajeddeloh> .hello2
16:31:02 <zodbot> ajeddeloh: ajeddeloh 'Andrew Jeddeloh' <andrew.jeddeloh@redhat.com>
16:31:03 <yzhang> .hello2
16:31:05 <zodbot> yzhang: yzhang 'Yu Qi Zhang' <jzehrarnyg@gmail.com>
16:32:18 <jlebon> .hello2
16:32:19 <zodbot> jlebon: jlebon 'None' <jonathan@jlebon.com>
16:32:30 <dustymabe> #chair yzhang ajeddeloh mnguyen_ ksinny rfairley mskarbek lorbus jbrooks bgilbert jlebon
16:32:30 <zodbot> Current chairs: ajeddeloh bgilbert dustymabe jbrooks jlebon ksinny lorbus mnguyen_ mskarbek rfairley yzhang
16:32:34 <dustymabe> woot
16:32:39 <miabbott> .hello miabbott
16:32:40 <zodbot> miabbott: miabbott 'Micah Abbott' <miabbott@redhat.com>
16:32:47 * miabbott really just lurking
16:32:53 <dustymabe> #chair miabbott
16:32:53 <zodbot> Current chairs: ajeddeloh bgilbert dustymabe jbrooks jlebon ksinny lorbus miabbott mnguyen_ mskarbek rfairley yzhang
16:33:13 <dustymabe> #topic Action items from last meeting
16:33:23 <dustymabe> * dustymabe to update python ticket w/ acked proposals
16:33:50 <geoff-> geoff-: Geoff Levand <geoff@infradead.org>
16:34:19 <dustymabe> #info dusty updated python ticket with acked proposals (https://github.com/coreos/fedora-coreos-tracker/issues/32#issuecomment-425145681) and opened a PR to update the design doc (https://github.com/coreos/fedora-coreos-tracker/pull/56)
16:34:26 <dustymabe> please review the above linked PR ^^
16:35:06 <dustymabe> any comments about that before we move to meeting tickets?
16:36:09 <slowrie> .hello2
16:36:10 <zodbot> slowrie: slowrie 'Stephen Lowrie' <slowrie@redhat.com>
16:36:15 <dustymabe> #chair slowrie
16:36:15 <zodbot> Current chairs: ajeddeloh bgilbert dustymabe jbrooks jlebon ksinny lorbus miabbott mnguyen_ mskarbek rfairley slowrie yzhang
16:36:19 <dustymabe> #topic set up logging for #fedora-coreos IRC channel
16:36:23 <dustymabe> #link https://github.com/coreos/fedora-coreos-tracker/issues/11
16:36:41 <bgilbert> whoops, I reopened that but didn't remove the meeting label
16:36:50 <dustymabe> ok so we set up logging for #fedora-coreos using botbot.me in the past
16:36:58 <dustymabe> bgilbert: nope, I just added the meeting label to it
16:37:02 <bgilbert> cool
16:37:36 <dustymabe> botbot.me is shutting down I think it was because ENOTIME and also GPDR compliance issues (worried about getting sued)
16:38:12 <lorbus> are there any similar services we could use?
16:38:27 <dustymabe> we'd like to have #fedora-coreos logged. I queried our legal council about setting up a botbot.me instance for Fedora and they said they'd have the same concerns about GPDR compliance
16:38:44 <bgilbert> what's the difference between that and zodbot?
16:38:48 <dustymabe> any ideas on how we move forward ?
16:39:08 <bgilbert> re GDPR
16:39:24 <dustymabe> bgilbert: good question.. I don't think zodbot logs by default (only during meetings like this), but that is a question we could take back and ask them
16:39:39 <jbrooks> https://github.com/BotBotMe
16:39:48 <jbrooks> Our own instance could be an option
16:39:50 <geoff-> GDPR
16:40:12 <dustymabe> jbrooks: right. that was the question i asked to our legal council. i.e. if we set up our own would it be OK
16:40:22 <jbrooks> oh right
16:41:06 <dustymabe> so this was mostly to raise awareness of the issue and to have anyone propose ideas (i.e. if other services exist that we could use, etc..)
16:41:24 <dustymabe> also any ideas we want to take back and ask legal we can
16:41:31 <dustymabe> so populate those in the ticket please
16:42:11 <dustymabe> any comments before next ticket ?
16:42:15 <slowrie> is there some sort of workaround you can do to essentially have the publicly listed channel point to something that isn't logged but says that most developer discussion happens inanother channel which is logged?
16:43:03 <dustymabe> i'm really not sure :)
16:43:21 <ajeddeloh> Does anyone know offhand the timeline for a gdpr request?
16:43:41 <ajeddeloh> i.e. if it's "you have a week to take things down" could we just log the last week
16:43:46 <slowrie> I think it's 40 calendar days
16:44:12 <ajeddeloh> Idk about others, but I personally don't really care about logs older than 40 days
16:44:16 <dustymabe> i.e. if we only keep things temporarily then it might not be a concern
16:44:18 <bgilbert> +1
16:44:49 <ksinny> sounds good
16:44:57 <dustymabe> ok thats a good question to ask legal
16:45:00 <dustymabe> will add it to the ticket
16:45:28 <dustymabe> moving to next ticket...
16:45:54 <dustymabe> #topic Major release and update cycle for Fedora CoreOS
16:45:58 <dustymabe> #link https://github.com/coreos/fedora-coreos-tracker/issues/22
16:46:17 <dustymabe> ok we discussed this last week in the meeting
16:46:30 <dustymabe> we had quite the discussion. thanks @bgilbert and everyone involved
16:46:46 <bgilbert> since last week I've added some clarification on what I meant by "backport"
16:47:05 <bgilbert> and a new point about testing upcoming kernel releases in the "next" stream
16:47:33 <dustymabe> +1
16:47:58 <ajeddeloh> +1
16:48:25 <dustymabe> I think the short story there is that we may actually pull some kernels from rawhide to test in our next stream after they've hit an 'rc level' we're comfortable with
16:48:39 <bgilbert> right
16:48:51 <bgilbert> because otherwise new kernels would bake in testing for two weeks and then immediately promote to stable
16:49:00 <bgilbert> which isn't enough time
16:49:03 <dustymabe> this should help us testing more stable and stable more stable :)
16:49:14 <dustymabe> s/us/us make/
16:49:28 <bgilbert> so last week we left off discussing the "testing" and "stable" streams.  should we continue that now?
16:49:40 <dustymabe> sure
16:50:21 <bgilbert> there was some concern about packages in Fedora `updates` taking a while to propagate to `stable`
16:50:49 <dustymabe> i.e. the possible 4 week delay ?
16:51:01 <bgilbert> max 4 weeks, right
16:51:08 <bgilbert> except in cases where we backport
16:51:13 <dustymabe> +1
16:51:38 <jlebon> i guess that's part of making stable more stable?
16:51:44 <bgilbert> that's the idea, yeah
16:52:23 <jlebon> if we want to be more stable than Fedora stable repos, then delaying it is part of the design
16:52:59 <dustymabe> bgilbert: done? ready for comments ?
16:54:02 <bgilbert> you mean, did I have more words?  not at this time :-)
16:54:07 <dustymabe> kk :)
16:54:23 <dustymabe> so with the more clarification on backports that you did in the issue I think I'm ok with this point
16:55:02 <dustymabe> the tricky part is when there is some minor regression in userland behavior that makes it into testing (and the fix lands right after in bodhi stable)
16:55:29 <dustymabe> the conflict is whether we spin a new testing (and thus delay the next stable) or do we just tell people to pick up the next testing when it lands
16:56:33 <ksinny> what if a recent package update has a fix which we need in stable? Are we going to wait for 2 weeks to stay package in testing or pull in into stable as soon as it is avaialble in Fedora-updates?
16:57:04 <bgilbert> dustymabe: oh, that's not how I'd thought of it.  I'm in CL headspace :-)
16:57:31 <bgilbert> dustymabe: my assumption was that we'd spin a new testing _with just the fix_, and that that wouldn't reset the stable promotion timer
16:58:01 <bgilbert> ksinny: we'd pull the fix into stable, either by backporting just the fix or by accepting the whole package update depending on circumstances
16:58:23 <ksinny> +1
16:58:25 <dustymabe> bgilbert: oh, cool +1
16:59:06 <dustymabe> bgilbert: when we write up a summary that detail would be good to include
16:59:10 <jlebon> so to summarize, are we looking at 3 official refs + 1 unofficial ref (updates-testing) ?
16:59:14 <bgilbert> dustymabe: +1
16:59:24 <jbrooks> It'll be nice to be somewhat decoupled from bodhi, we'll need some tooling or process for how that works
16:59:42 <jbrooks> I'm +1 to what you're describing
17:00:13 <dustymabe> jlebon: 3 official refs, 3 unofficial nightly refs (updates, updates-testing, rawhide)
17:00:24 <dustymabe> bgilbert: ^^ that would be my vote
17:00:54 <bgilbert> dustymabe: in CL we sometimes accept bugfixes/security fixes directly at the point of a stable promotion, rather than putting them in beta first.  not sure which approach is preferred here
17:00:55 <jlebon> is updates basically a nightly testing?
17:01:08 <dustymabe> we could even name the updates/updates-testing ones something specific to bodhi
17:01:27 <bgilbert> can we put the unofficial refs in a different repo?
17:01:30 <jlebon> yeah, naming is going to get confusing fast :)
17:01:44 <bgilbert> e.g. fcos and fcos-devel
17:01:53 <dustymabe> bgilbert: we could
17:02:16 <dustymabe> right now we have a unified repo with everything in it
17:02:18 <jlebon> we could, at the expense of losing some storage efficiencies on the server side
17:02:18 <bgilbert> I'd be much more comfortable having arbitrary things in a separate -devel repo
17:02:28 <bgilbert> as a signal to people using it
17:02:49 <dustymabe> we also have a "compose" repo that is operated on by the build system
17:03:08 <dustymabe> so we could just put content into the compose repo and not sync it to the prod repo (for the unofficial refs)
17:03:32 <jlebon> an alternative is to keep it in the same repo, but just not in the summary file
17:03:42 <jlebon> so you don't really know it's there unless you know about it
17:03:59 <dustymabe> these are mostly implementation details we can discuss when we get closer to implementing them :)
17:04:03 <bgilbert> dustymabe: +1
17:04:10 <jlebon> +1
17:04:17 <ajeddeloh> +1\
17:05:10 <dustymabe> bgilbert: since this topic is a bit involved can I get you to write up a proposal in the ticket (i.e. your description + modifications based on our conversations) and we can all read it and vote on it in the ticket
17:05:19 <dustymabe> then we can take that to the Design doc with a PR
17:05:24 <bgilbert> sounds good
17:06:01 <dustymabe> #action bgilbert to write up a proposal in the ticket (i.e. your description + modifications based on our conversations) and we can all read it and vote on it in the ticket #22
17:06:13 <dustymabe> anyone else with anything for this ticket
17:06:45 <dustymabe> #topic Determine how to handle automatic rollback
17:06:49 <dustymabe> #link https://github.com/coreos/fedora-coreos-tracker/issues/47
17:07:12 <dustymabe> oh where oh where did lorbus go ? lorbus[m]  :)
17:07:32 <dustymabe> so we had stalled this conversation a bit to wait on lorbus[m] to bring us his wisdom
17:07:50 <dustymabe> he's now with us again but I don't know if he's had a chance to digest this ticket
17:08:03 <dustymabe> was going to ask him but I think his net connection just dropped
17:08:20 <dustymabe> shall we move to open floor or is there anything anyone else would like to bring up on this topic ?
17:08:57 <ajeddeloh> I can bring up some stuff, but its not critical and it'd be better for lorbus[m] to be here
17:09:13 <dustymabe> k.
17:09:25 <dustymabe> punt til next time, he'll be better prepared by then anyways
17:09:30 <dustymabe> #topic open floor
17:09:31 <ajeddeloh> +!
17:09:41 <dustymabe> we actually get an open floor today!
17:09:56 <dustymabe> reminder to anyone who may be interested
17:10:22 <dustymabe> we have devconf.cz coming up in january (https://devconf.info/cz)
17:10:50 <walters> .hello2
17:10:51 <zodbot> walters: walters 'Colin Walters' <walters@redhat.com>
17:11:04 <dustymabe> please submit a talk if you are interested in the immutable operating system space (there is a whole track this year
17:11:40 <dustymabe> does anyone have anything for open floor ?
17:11:49 <dustymabe> welcome back lorbus :)
17:12:21 <lorbus> dustymabe: thx^^ sucky internet connectivity for me today :/
17:12:47 <dustymabe> no problem we skipped over https://github.com/coreos/fedora-coreos-tracker/issues/47 and will re-visit next week
17:13:27 <dustymabe> ok will close out the meeting in 2 minutes if no takers on open floor announcements/topics
17:14:09 <geoff-> I see the logo looks different now...
17:14:09 <dustymabe> #chair walters
17:14:09 <zodbot> Current chairs: ajeddeloh bgilbert dustymabe jbrooks jlebon ksinny lorbus miabbott mnguyen_ mskarbek rfairley slowrie walters yzhang
17:14:20 <geoff-> https://github.com/coreos/coreos.fedoraproject.org/blob/master/public/images/fedoracoreos-logo.svg
17:14:47 <dustymabe> geoff-: yep. sanja has been working on it
17:15:06 <dustymabe> you can send feedback to her (although I don't see her on IRC right now)
17:16:01 <lorbus> ok thanks!
17:16:14 <dustymabe> looks like https://coreos.fedoraproject.org/ is under maintenance right now
17:16:20 <dustymabe> i'm guessing misc is working on it :)
17:16:28 <dustymabe> ok closing the meeting out!
17:16:30 <dustymabe> #endmeeting