fedora_atomic_wg
LOGS
16:30:03 <jbrooks> #startmeeting fedora_atomic_wg
16:30:03 <zodbot> Meeting started Wed Dec 13 16:30:03 2017 UTC.  The chair is jbrooks. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:30:03 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
16:30:03 <zodbot> The meeting name has been set to 'fedora_atomic_wg'
16:30:08 <jbrooks> #topic roll call
16:30:36 <maxamillion> .hello2
16:30:37 <zodbot> maxamillion: maxamillion 'Adam Miller' <maxamillion@gmail.com>
16:30:42 <sanja> .hello2
16:30:43 <zodbot> sanja: sanja 'Sanja Bonic' <sanja@redhat.com>
16:30:44 <ashcrow> .hello smilner
16:30:46 <zodbot> ashcrow: smilner 'None' <smilner@redhat.com>
16:30:46 <dustymabe> .hello2
16:30:49 <zodbot> dustymabe: dustymabe 'Dusty Mabe' <dustymabe@redhat.com>
16:30:50 <jberkus> .hello jberkus
16:30:52 <zodbot> jberkus: jberkus 'Josh Berkus' <josh@agliodbs.com>
16:31:01 <rubao> .hello rubao
16:31:01 <jlebon> .hello jlebon
16:31:02 <zodbot> rubao: rubao 'rubao' <rubao.net@hotmail.com>
16:31:05 <zodbot> jlebon: jlebon 'None' <jlebon@redhat.com>
16:31:10 <miabbott> .hello2
16:31:11 <zodbot> miabbott: miabbott 'Micah Abbott' <miabbott@redhat.com>
16:31:13 <ksinny> .hello sinnykumari
16:31:14 <zodbot> ksinny: sinnykumari 'Sinny Kumari' <ksinny@gmail.com>
16:31:16 <walters> .hello
16:31:17 <zodbot> walters: (hello <an alias, 1 argument>) -- Alias for "hellomynameis $1".
16:31:28 <walters> .hello walters
16:31:31 <zodbot> walters: walters 'Colin Walters' <walters@redhat.com>
16:31:43 <davdunc> .hello davdunc
16:31:46 <zodbot> davdunc: davdunc 'David Duncan' <davdunc@amazon.com>
16:32:27 <jlebon> i can't figure out why my name is "None"
16:32:36 <jlebon> things look ok in fedora accounts at least
16:33:16 <jbrooks> #chair  maxamillion davdunc walters ksinny miabbott jlebon rubao jberkus dustymabe ashcrow sanja
16:33:16 <zodbot> Current chairs: ashcrow davdunc dustymabe jberkus jbrooks jlebon ksinny maxamillion miabbott rubao sanja walters
16:33:21 <jbrooks> Did I miss anyone?
16:33:21 <dustymabe> jlebon: must be a canada thing
16:33:27 <ashcrow> lol
16:33:31 <rubao> dustymabe++
16:33:32 <zodbot> rubao: Karma for dustymabe changed to 7 (for the f27 release cycle):  https://badges.fedoraproject.org/tags/cookie/any
16:33:36 <ashcrow> jlebon: I have the same issue to :-)
16:33:46 <dustymabe> canada/florida
16:33:47 <kushal> .hellomynameis kushal
16:33:48 <zodbot> kushal: kushal 'Kushal Das' <mail@kushaldas.in>
16:33:56 <ashcrow> dustymabe: haha
16:34:05 <jbrooks> #chair kushal
16:34:05 <zodbot> Current chairs: ashcrow davdunc dustymabe jberkus jbrooks jlebon ksinny kushal maxamillion miabbott rubao sanja walters
16:34:07 <jlebon> heh
16:34:40 <jbrooks> #topic action items from last meeting
16:35:14 <jbrooks> jbrooks to retest asciibinder rpm
16:35:24 <jbrooks> maxamillion to try to implement the (hopefully trivial) work for #376
16:35:48 <jbrooks> Hmmm, I did re-rest asciibinder
16:36:08 <jbrooks> It worked, I'm pretty sure I told the packager about that
16:36:18 <jbrooks> maxamillion, Did you get to your item?
16:36:47 <dustymabe> jbrooks: so no more action items for asciibinder? I know it was possibly going to get into fedora as an official rpm at one point
16:36:59 <markllama> .hello markllama
16:37:00 <dustymabe> depending on dependencies getting into fedora as well
16:37:00 <zodbot> markllama: markllama 'Mark Lamourine' <markllama@gmail.com>
16:37:10 <dustymabe> markllama: \o/
16:37:13 <jbrooks> dustymabe, No action items for us, I'd say
16:37:30 <dustymabe> jbrooks: so what was the final state there? use it from copr?
16:38:03 <jbrooks> #info https://lists.fedoraproject.org/archives/list/ruby-sig@lists.fedoraproject.org/message/6HYR5BTBHQNDN46DSZGUX54O4GZ6WUPV/
16:38:03 <maxamillion> jbrooks: oh, I think that might be done already ... puiterwijk was kind enough to just put a config in the proxies to make it work for now and then it will be taken over and manage by bodhi once the bodhi+containers work is done
16:38:52 <jbrooks> dustymabe, I don't know -- it doesn't really matter, we can use the current container that just uses the gem -- that's the official upstream method, or we can wait till it's a real pkg, or we can use copr
16:39:03 <jbrooks> maxamillion, awesome
16:39:34 <dustymabe> does someone want to volunteer to confirm the fix is in place and close ticket #376 ?
16:40:10 <jbrooks> maxamillion, can you see about closing #376?
16:40:11 <dustymabe> jbrooks: the important thing is we don't block on it. but ultimately prefer real pkg in 'long run'
16:40:13 * cverna waves
16:40:21 <dustymabe> that way the rest of fedora can benefit from it
16:40:27 * dustymabe waves at cverna
16:40:56 <maxamillion> jbrooks: can do
16:40:59 <jbrooks> thanks
16:41:02 <jbrooks> OK
16:41:06 <dustymabe> should we #action that?
16:41:22 <maxamillion> sure
16:41:34 <jbrooks> #action maxamillion to confirm and close #376
16:42:03 <jbrooks> #topic meeting tickets
16:42:46 <jbrooks> #topic 377 - Voting Process
16:42:55 <jbrooks> #info https://pagure.io/atomic-wg/issue/377
16:43:05 <jbrooks> jberkus, This is yours?
16:43:26 <ashcrow> I assigned it to him
16:43:42 <jberkus> actually, it's more the quorum which is in question
16:43:44 <ashcrow> it's something we had talked about quite awhile ago in the meeting
16:44:00 <jberkus> the voting process is pretty straightforwards
16:44:06 <jberkus> let me find that ticket
16:44:19 <jbrooks> #topic 363 Quorum Rules
16:44:23 <ashcrow> https://pagure.io/atomic-wg/issue/363
16:44:28 <ashcrow> jbrooks: that one?
16:44:30 <ashcrow> jberkus: ^
16:44:31 <jbrooks> Yep
16:45:07 <jberkus> yah, let me post a revised proposal based on ashcrow's suggestions
16:45:23 <ashcrow> sounds good
16:45:33 <jbrooks> OK, cool
16:45:44 <jbrooks> #action jberkus to post a revised quorum proposal
16:46:11 <jbrooks> #topic 360 -  Decide strategy for including container runtimes in Atomic Host
16:46:20 <jbrooks> #info https://pagure.io/atomic-wg/issue/360
16:46:41 <jbrooks> I think we're looking for survey data to inform this discussion
16:47:41 <ashcrow> I believe jberkus noted Monday he will be working on it shortly and should have something late this week, early next
16:47:48 <ashcrow> jberkus: ^^ did I remember that correctly?
16:47:49 <jbrooks> OK, we
16:47:57 <jberkus> yah
16:47:58 <jbrooks> OK, we'll discuss next week
16:48:05 <sanja> yes
16:48:16 <jbrooks> #topic Open Floor
16:48:26 <jbrooks> Who has items to discuss?
16:48:34 * ksinny has
16:48:42 <jbrooks> shoot, ksinny
16:48:49 <markllama> re container runtimes in atomic, please include RKT in the discussion.
16:48:54 <ksinny> #info Sent PR for updating cloud-utils package in Fedora to 0.30 - https://src.fedoraproject.org/rpms/cloud-utils/pull-request/1
16:49:06 <sanja> +1 for rkt
16:49:12 <jbrooks> markllama, Do we have a pkg for rkt in fedora?>
16:49:17 <markllama> yes,
16:49:31 <markllama> I worked on it.  Can't go in RHEL7 becuase it requires systemd 222
16:49:39 <markllama> RHEL7 is/was stuck at 219
16:49:47 <jbrooks> OK, layering should be an option, and system container, too
16:50:09 <jbrooks> ksinny, Can we help move that along?
16:50:27 <dustymabe> markllama: we definitely haven't had many people ask us about rkt
16:50:37 <markllama> separate issue I have, I'd love a better guide to creating system containers.  "requirements" != "method".  I may be missing resources that already exist.
16:50:52 <ksinny> jbrooks: It will need to get reviewed, I have asked maintainer but having additional review would be great
16:51:00 <ashcrow> markllama: We have a blog post on it. Let me dig it up.
16:51:06 <markllama> FWIW, rkt I think is OCF compliant.
16:51:31 <dustymabe> markllama: OCI ?
16:51:32 <markllama> ashcrow thanks, I have looked and found some but not enough to succeed.
16:51:41 <ashcrow> markllama: https://www.projectatomic.io/blog/2017/06/creating-system-containers/
16:51:42 <markllama> dustymabe: yes....
16:51:49 <jbrooks> dustymabe, can we suggest someone to help review https://src.fedoraproject.org/rpms/cloud-utils/pull-request/1
16:51:52 <sanja> dustymabe: I think rkt is coming though, I know several companies prefer it over Docker and personally  I think it'd be good for Atomic *but* gotta think of added workload naturally, so....
16:52:13 <ashcrow> markllama: I'll create a tracking issue for someone to write up an end to end how to
16:52:20 <jbrooks> As long as we have maintainers for rkt, it's cool
16:52:30 <jberkus> ashcrow: see https://pagure.io/atomic-wg/issue/363 for a quorum proposal for you to mark up
16:52:31 <dustymabe> sanja: yeah, this is good to add to the discussion about container runtimes
16:52:39 <markllama> ashcrow I've read that.  Not really sufficient for real world. I would love to be your guinea pig!
16:53:04 <jbrooks> markllama, I always copy an existing one and then tweak endlessly
16:53:19 <ashcrow> markllama: hrm, we have had success with other people following that. But there are lots of corner cases with containers. We can figure it out together and make a better how to.
16:53:30 <dustymabe> markllama: so there is a rkt rpm in Fedora?
16:53:39 <dustymabe> does package layering the rpm work on atomic host?
16:54:10 <markllama> dustymabe: rkt.x86_64 : CLI for running app containers
16:54:43 <markllama> dustymabe: it did when i got it accepted into Fedora.  My first task after was to build an Atomic image with it
16:54:51 <markllama> but it's been a year or so since I did any work on it.
16:55:24 <dustymabe> markllama: 'build an Atomic image with it' is different than package layering
16:55:28 <jbrooks> Another item I'd like to raise is: can we get a container release?
16:55:43 <markllama> dustymabe: I will need education to answer that question then.
16:55:50 <miabbott> dustymabe: markllama `rkt` appears to layer just fine...not sure if it is functional after that, though
16:55:53 <dustymabe> markllama: grab me after meeting
16:56:02 <jbrooks> maxamillion, do we have a new person to poke about this now?
16:56:04 * markllama nods
16:56:26 <dustymabe> jbrooks: good question. we need some sort of continuity for container releases
16:56:26 <cverna> jbrooks: I ll be the new person :)
16:56:31 <dustymabe> cverna: woot!
16:56:35 <jbrooks> #chair cverna
16:56:35 <zodbot> Current chairs: ashcrow cverna davdunc dustymabe jberkus jbrooks jlebon ksinny kushal maxamillion miabbott rubao sanja walters
16:56:39 <jbrooks> NIiiiice
16:56:43 <dustymabe> cverna++
16:56:43 <zodbot> dustymabe: Karma for cverna changed to 9 (for the f27 release cycle):  https://badges.fedoraproject.org/tags/cookie/any
16:56:51 <cverna> but full time from Jjan 2018
16:56:51 <jbrooks> Heh, I'll be getting to know you, cverna ;)
16:57:03 <dustymabe> cverna: meet jbrooks, you should run when he pings you
16:57:08 <cverna> haha I am sure you will :)
16:57:32 <cverna> added to my survival notes :)
16:57:34 <jbrooks> Sooo, cverna, can we get a container release?
16:58:27 <dustymabe> jbrooks: he already ran away
16:58:28 <maxamillion> cverna: !!!! :D :D :D
16:58:33 <jbrooks> markllama, My first stab at a hello world after layering the pkg https://paste.fedoraproject.org/paste/96wrgedn8bT9JLQJT2heAw
16:58:43 <jbrooks> We'll track him
16:58:45 <cverna> jbrooks: good question :) I am far from being up to speed so my answer for now is I don't know
16:58:49 <jbrooks> He already accepted the cookie
16:58:58 <cverna> :)
16:59:19 <cverna> maxamillion: help help :D
16:59:30 <maxamillion> jbrooks: container releases are handled by RelEng now, ping mboddu and/or Kellin
16:59:40 <jbrooks> maxamillion, Great, thanks!
16:59:51 <jbrooks> Other items for this week?
16:59:52 <markllama> jbrooks.  Yeah.  needs work.
16:59:56 * davdunc - minor topic: Fedora in AWS Marketplace is moving along.
17:00:03 <jbrooks> Nice
17:00:07 <maxamillion> jbrooks: I wrote an ansible-powered release process for them that can be wired up to automation magic once we have authoritative tests for Atomic Host and can actually do a hand-off release of it
17:00:14 <ashcrow> markllama: https://pagure.io/atomic-wg/issue/399
17:00:17 * cverna feels safe for a little bit
17:00:21 <jbrooks> :)
17:00:22 <maxamillion> cverna: :)
17:00:26 <davdunc> I'll have a submission ready by Monday for review.
17:00:46 <jbrooks> davdunc, We'll keep our eyes peeled for that
17:01:07 * ksinny has another info to share
17:01:10 <ksinny> #info Sent PR for adding a post on Composing custom OSTree update for Fedora 27 atomic Host on projectatomic blog - https://github.com/projectatomic/atomic-site/pull/496
17:01:18 <markllama> aschrow just comment to get on the notifications for it? I don't want to be "assigned" until I look at whether I can work on it instead of just throwing darts.
17:01:21 <ksinny> It will be nice if someone can review it :)
17:01:35 <jbrooks> ksinny, I'll have a look at that
17:01:47 <ksinny> jbrooks: thanks
17:01:52 <ksinny> jbrooks++
17:01:52 <zodbot> ksinny: Karma for jasonbrooks changed to 2 (for the f27 release cycle):  https://badges.fedoraproject.org/tags/cookie/any
17:02:04 <jberkus> ksinny: sanja is reviewing your post
17:02:12 <ashcrow> markllama: you are not assigned. You're just referenced so you get updates
17:02:26 <ksinny> jbrooks: Thanks good to know!
17:02:27 <ashcrow> markllama: assignment is on the right of the ticket
17:02:29 <ksinny> sanja++
17:02:29 <zodbot> ksinny: Karma for sanja changed to 2 (for the f27 release cycle):  https://badges.fedoraproject.org/tags/cookie/any
17:02:38 <dustymabe> maxamillion: re: container releases
17:02:44 <markllama> aschrow: turns out I'm already CCd.
17:02:46 <dustymabe> you said to ping Kellin or mohan
17:02:59 <dustymabe> is there still a regularly scheduled container release that will happen?
17:03:03 <dustymabe> including rebuilds
17:04:10 <ashcrow> markllama: yup! I added you to it as a cc.
17:05:37 <jbrooks> Other items for this week?
17:05:50 <dustymabe> jbrooks: was hoping to hear from adam on that re:
17:06:09 <jbrooks> dustymabe, I figure we'll hit them up in fedora-releng
17:06:33 <dustymabe> IOW ping mohan or kellin every time I want container X released is not a good plan. in the past we had releases periodically at least
17:06:50 <maxamillion> dustymabe: yes, they *should* happen following the Atomic Host release (the automation processes I wrote include rebuilds), but if something goes sideways then RelEng is the point of contact now
17:06:50 <dustymabe> the question is. do mohan/kellin plan to release all containers on a periodic schedule or not?
17:06:59 <dustymabe> maxamillion: perfect
17:07:06 <dustymabe> so we did a FAH release yesterday
17:07:15 <dustymabe> is it reasonable to expect a container release today/tomorrow?
17:08:19 <cverna> if we have an ansible playbook that does the job it should not be to hard to make a regular release
17:08:41 <jberkus> would be nice, I haven't updated the list of containers since August
17:08:45 <dustymabe> cverna: can you talk to releng and see if they plan to do a container release?
17:08:58 <dustymabe> and if not what is blocking them?
17:09:26 <cverna> sounds good to me
17:09:40 <dustymabe> #action cverna to talk to releng about doing a container release since we just released FAH
17:09:44 <dustymabe> thanks cverna
17:09:50 <dustymabe> ok one final item for today jbrooks
17:09:50 <jbrooks> cool
17:09:58 <jbrooks> Go ahead
17:09:58 <cverna> that will be a good opportunity to get more familliar with that
17:10:17 <dustymabe> Is there anyone here that hasn't got 'hands-on' with atomic host before?
17:10:27 <dustymabe> don't be afraid to raise hands :)
17:10:42 <dustymabe> markllama: I think it's been a while for you
17:10:45 <sanja> Me actually!
17:10:46 <davdunc> I am certainly not hand-on enough.
17:10:58 <dustymabe> :)
17:11:03 <jbrooks> dustymabe, markllama just did a thing at lisa on atomic
17:11:11 * cverna raises his hand too
17:11:11 <markllama> dustymabe: yeah... I have worked with atomic host recently
17:11:20 * markllama raises hand
17:11:37 <markllama> I need to re-learn how to do builds manually
17:11:37 <dustymabe> well this is pretty much for anyone here
17:12:07 <dustymabe> If you haven't done it before you should definitely take an hour and run through the lab that I put on for flock last year
17:12:13 <dustymabe> err.. this year
17:12:17 <dustymabe> not 2018 yet
17:12:21 <dustymabe> https://dustymabe.com/2017/08/29/atomic-host-101-lab-part-0-preparation/
17:12:26 <dustymabe> it starts with that post
17:12:40 <dustymabe> you can do it on your laptop (on a plane if you want to)
17:12:57 <dustymabe> so if you haven't played with package layering
17:13:21 <dustymabe> check it out, there is a section on it
17:13:28 <jbrooks> It's a good lab
17:13:41 <deuscapturus> It would be nice of all of the official docs for Atomic on the website created for Fedora 25 were updated to Fedora 27
17:13:50 <cverna> dustymabe++ for the great lab
17:13:51 <zodbot> cverna: Karma for dustymabe changed to 8 (for the f27 release cycle):  https://badges.fedoraproject.org/tags/cookie/any
17:13:56 <jberkus> deuscapturus: there's a PR, I need to merge it
17:14:06 <dustymabe> markllama: if you want a guide on building your own atomic host ksinny just open a PR on atomic-site
17:14:13 <dustymabe> markllama: you could guinea pig it for review
17:14:35 <dustymabe> https://github.com/projectatomic/atomic-site/pull/496
17:14:54 <jberkus> deuscapturus: sorry, I've been tied up with the changes to move the site to Openshift v3
17:14:58 <ksinny> dustymabe: thanks!
17:14:58 <dustymabe> #action jberkus to review/merge the PR to update official docs from f25 to f27
17:15:16 <markllama> dustymabe: still dipping my toes and learning social conventions, but yeah.
17:15:31 <dustymabe> cool
17:15:37 <dustymabe> markllama: should we expect to see you around more often?
17:15:50 * dustymabe is done with topics
17:16:06 <markllama> Personally, as often as possible.  work dedication... probably very sporadically, when I can dedicate personal time
17:16:24 <dustymabe> well glad to have you, either way
17:16:26 <ashcrow> markllama: look forward to seeing you more often :-)
17:16:45 <jbrooks> +1
17:16:46 <davdunc> indeed markllama
17:16:52 <jbrooks> OK, I think we're all set here
17:16:54 <dustymabe> markllama: the biggest way you can help in that case is to convert any servers that you have to use atomic host today
17:16:57 <miabbott> markllama: come visit in #atomic
17:17:10 <jbrooks> #endmeeting