17:00:37 #startmeeting fedora_atomic_wg 17:00:37 Meeting started Wed Nov 16 17:00:37 2016 UTC. The chair is trishnag. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:37 Useful Commands: #action #agreed #halp #info #idea #link #topic. 17:00:37 The meeting name has been set to 'fedora_atomic_wg' 17:00:40 .hello bowlofeggs 17:00:41 bowlofeggs: bowlofeggs 'Randy Barlow' 17:00:41 #topic Roll Call 17:00:49 .hello trishnag 17:00:51 trishnag: trishnag 'Trishna Guha' 17:00:53 .hello sayanchowdhury 17:00:57 sayan: sayanchowdhury 'Sayan Chowdhury' 17:00:59 .hello dustymabe 17:01:00 dustymabe: dustymabe 'Dusty Mabe' 17:01:02 .hellomynameis kushal 17:01:03 kushal: kushal 'Kushal Das' 17:01:38 .fas jasonbrooks 17:01:39 jbrooks: jasonbrooks 'Jason Brooks' 17:02:04 trishnag, before we get into meeting, I want to say something :) 17:02:08 #chair bowlofeggs sayan dustymabe kushal jbrooks 17:02:08 Current chairs: bowlofeggs dustymabe jbrooks kushal sayan trishnag 17:02:11 * meeting tickets etc 17:02:14 kushal: sure go ahead 17:02:53 What about we dedicate next meeting about the features/topics relevant to atomic working group during F26 release? 17:03:17 That should be after the release date. 17:03:41 kushal: i'd like to see an agenda proposed for that 17:03:46 kushal: +1 17:03:54 if there hasn't been one already put forth 17:03:56 dustymabe, yup, we can work on the agenda from tonight :) 17:04:06 i.e. let's query the list and get input 17:04:15 and formulate an agenda and invite interested parties 17:04:30 kushal: next week might not be good because of holidays in the US 17:04:35 following week may be best 17:04:44 dustymabe, Oka 17:04:45 Okay 17:04:51 Holidays are good. 17:04:53 always good :) 17:04:59 hehe :) 17:05:00 but certainly let's query the list and start to prepare an agenda 17:05:03 Yes. 17:05:07 dustymabe, I will drop a mail 17:05:08 Right 17:05:26 #action kushal will write to list for the agenda of the features meeting for Atomic group 17:06:34 kushal: Can I move to action items? or you have anything else? :) 17:06:42 .hello maxamillion 17:06:43 maxamillion: maxamillion 'Adam Miller' 17:06:43 trishnag, sorry, go ahead :) 17:06:56 cool 17:07:02 #chair maxamillion 17:07:02 Current chairs: bowlofeggs dustymabe jbrooks kushal maxamillion sayan trishnag 17:07:07 #topic Action items from last meeting 17:07:20 * jbrooks to file issue on kube situation w/ f25 atomic 17:07:37 https://pagure.io/atomic-wg/issue/176 17:07:40 trishnag: jbrooks filed a very detailed issue 17:07:40 that's it 17:07:44 jbrooks: thank you 17:08:01 jbrooks: Thanks. 17:08:30 We will have to make a lot noise in social media + blogs + docs 17:08:49 kushal: right, but we still don't have an actual answer yet 17:08:56 system containers are "coming" 17:09:06 dustymabe, we don't need those 17:09:13 kushal: Also we need people to review the docs. 17:09:14 in the mean time I am working on getting the openshift-ansible installer to work against F25 atomic host 17:09:15 but they'd be cool to have :D 17:09:17 we can use package layering for etcd and flannel 17:09:23 dustymabe: +1 17:09:42 dustymabe, how does that look? that's probably going to be the best fully fedora solution 17:09:51 dustymabe: openshift-ansible is an impressive pile of ansible but omg it's *giant* 17:10:09 We can doc how to use kube w/ upstream containers, but we don't have them from fedora itself 17:10:11 dustymabe: the rabbit hole is impressive in depth ;) 17:10:13 jbrooks: yeah package layering could work, but I hate the solution of boot your machine, run a command, reboot your machine, then you can run 17:10:24 Yeah 17:10:32 dustymabe: +1 17:10:33 It is what it is 17:10:36 :/ 17:10:53 The system containers do work fine 17:10:58 kube is a bigger issue, though 17:11:27 We just need to add those etcd and flannel system containers to Fedora-Dockerfiles, the rpms they need are in the repos 17:11:31 jbrooks: did you get an update on newer kube in F25? 17:12:23 dustymabe, I'm still working on testing it -- I'm trying w/ etcd, flannel and kubelet via pkg layering, and the rest in containers, based on what rhel does 17:12:27 But it's not working yet 17:12:39 That has to do w/ the dockerfiles 17:12:51 But we don't have those dockerfiles in fedora anyway 17:13:10 That's why I think origin on fedora might be the best story for now, assuming that works 17:13:27 jbrooks: so giuseppe is working on adding some system containers to projectatomic/ on docker hub 17:13:34 Or, using upstream containers and like kubeadm does work, it's just not all home-built 17:13:54 so we could possibly use those 17:13:59 Fine with me 17:14:06 Is that legit fedora enough for us? 17:14:15 jbrooks: yeah, that is a good question 17:14:34 I think longer term (once system containers are more legit) we can make them more of a real thing produced by fedora releng 17:14:48 maxamillion: do you have an opinion on this? 17:14:50 Does releng produce any containers at this point? 17:15:17 I've kind of struggled with this myself about how in a containerized world with things coming from all sorts of external places, is it still "Fedora" (similar questions for things like Flatpak apps and such) 17:15:59 jbrooks: technically? yes ... just not in production yet for roughly another couple weeks 17:16:28 I mean, we can in production now, but only in limited capacity 17:16:28 maxamillion: yes. things are certainly starting to get a lot more blurry 17:16:43 part of me is scared. part of me is not 17:16:50 :) 17:17:06 dustymabe, Maybe I'll shift to origin and we can try and make that work 17:17:10 First 17:17:21 dustymabe: I think it rounds back to the question from Alexander Larssen of "where does the operating system end and the application begin?" and if Fedora only caters to the operating system side of that question, do we particularly care where the application comes from? 17:17:23 jbrooks: well I got it working using the installer 17:17:37 dustymabe, Cool, just w/ those workarounds? 17:17:45 the issues you linked 17:17:50 jbrooks: pretty much 17:18:42 OK, that's good, I'll keep going on kube stuff. I figure we want to document: here's plain docker, here's origin, here's kube 17:18:53 jbrooks: yep 17:19:19 maxamillion, do you know what other layer imaged builds are coming in the next release? 17:19:30 Once I get the 1.4.5 pkgs actually working, I'll step up my poking about getting them available for f25 17:21:06 kushal: I'm going to be doing capacity planning shortly so we'll know what we can actually handle but I'd like to target everything currently in Fedora-Dockerfiles github repo 17:21:29 maxamillion, that is a great news. 17:21:40 maxamillion, and still within Fedora packages, correct? 17:21:40 .hello scollier 17:21:41 scollier: scollier 'Scott Collier' 17:21:44 and scollier is here :) 17:21:48 dang, sorry about that 17:21:51 kushal: I don't understand the question 17:21:52 it's been soooo long. 17:21:52 #chair scollier 17:21:52 Current chairs: bowlofeggs dustymabe jbrooks kushal maxamillion sayan scollier trishnag 17:22:09 scollier: hiya, you haven't missed much - we were just chatting about https://pagure.io/atomic-wg/issue/176 17:22:14 maxamillion, I mean only the containers with official packages inside 17:23:49 kushal: correct 17:24:17 Okay, thanks :) 17:24:48 kushal: curated content from external sources is unfortunately way down the priority list and it's not a small amount of work to accomplish 17:25:03 maxamillion, True, I was just asking. 17:25:11 Anyway our build systems do not allow that iirc. 17:26:12 kushal: right, no part of koji builds allow that today so we targeted that initially 17:27:35 * walters bets no one would notice if someone started backing up their vacation pictures in the lookaside cache 17:28:03 :) 17:28:12 icloud - go 17:28:13 anyways, just a quick note i'm looking at the ISO thing now 17:29:13 walters: that's probably fair 17:29:43 walters: however, we could at least audit it better than 'curl https://foo.io/install.sh | bash' in a Dockerfile 17:31:31 Ok Let's move to meeting items. 17:31:42 #topic Fedora Cloud Base Vagrant Boxes Don't Allow Vagrant User To su To Other Users https://pagure.io/atomic-wg/issue/173 17:32:14 We have no assignee for this ticket. 17:33:25 trishnag: this will have to wait until after F25 17:33:53 so either someone can volunteer, or we can wait 17:34:05 trishnag: this would be a good thing to add to the tests 17:34:13 dustymabe: ack. I am going to add on_hold tag to it. 17:34:41 dustymabe: cool. I will create an issue to tunirtests for the same. 17:35:29 #action trishnag will create an issue on tunirtests for https://pagure.io/atomic-wg/issue/173 17:36:05 moving to next ticket. 17:36:07 #topic Finish new Fedora Cloud PRD https://pagure.io/atomic-wg/issue/170 17:36:48 jberkus doesn't seem to be around. 17:37:55 #topic Spec for IRC bot to notify about blockers https://pagure.io/atomic-wg/issue/169 17:39:16 i don't think anyone has anything for this ticket 17:39:27 Looks like this is on-hold as well. 17:39:51 #topic Ship fedora-motd in F24 atomic image https://pagure.io/atomic-wg/issue/160 17:40:14 rtnpro: kushal Update on this? 17:40:16 i'd say rather than irc, we have a mechanism to assign a person to failures or something like that 17:41:35 trishnag, I did not manage to work on it based on walters feedback. I will update on it once I have the fixes 17:41:47 rtnpro: ok thanks. 17:42:27 #topic Proposed website changes for Cloud Base → Atomic Host switch https://pagure.io/atomic-wg/issue/158 17:43:21 dustymabe: Thanks for the comment. 17:43:21 trishnag: mattdm and I are going to meet with robyduc today on this 17:43:40 i also gave feedback on the atomic download page 17:43:58 some of my suggestions won't make it in until after F25 release. but we are getting things fixed that were "wrong" 17:44:15 I'm going to create a few new tickets for the websites team based on our discussion today 17:44:23 i'll update this issue with that information when it happens 17:45:06 dustymabe: great! 17:45:30 #topic make Fedora Atomic download page clearer https://pagure.io/atomic-wg/issue/154 17:46:17 trishnag: i think these were comments on the F24 download page 17:46:41 which is about to be obsolete - I'll update this ticket to say that and that josh will have to give new feedback on the new download page 17:46:41 yup https://fedoraproject.org/wiki/Cloud/getfedora_update 17:47:05 dustymabe++ 17:47:07 trishnag: Karma for dustymabe changed to 5 (for the f24 release cycle): https://badges.fedoraproject.org/tags/cookie/any 17:47:44 #topic design, deploy and document Fedora OpenShift Playground (FOSP) https://pagure.io/atomic-wg/issue/153 17:48:37 scollier: Do you have anything to say for this ticket? 17:48:43 trishnag, looking 17:49:33 trishnag, no, i haven't had a chance to sync with misc on this, looks like he's working with puiterwijk on some auth issues. 17:49:40 the recent update was from misc and walters. 17:49:44 trishnag, I will follow up with them and update the ticket. 17:50:00 scollier: Cool Thanks :). 17:50:38 #topic vagrant boxes fixups https://pagure.io/atomic-wg/issue/136 17:51:48 I suggested a fix for this, Ian commented in irc, he was ambivalent about my solution, not sure what the next step might be 17:53:08 jbrooks: did you open a PR? 17:54:05 dustymabe, No, I can though 17:54:22 that would be the logical next step. and the discussion can be public 17:54:37 dustymabe: +1 17:55:06 https://github.com/redhat-imaging/imagefactory/pull/392 17:55:49 jbrooks: might be good to add a description to the commit that describes why this change is useful 17:55:59 for those browsing history in the future :) 17:56:24 also a link to the issue https://pagure.io/atomic-wg/issue/136 :) 17:57:10 done 17:57:22 Awesome! 17:57:38 We can probably move to Open floor now. :-) 17:57:46 #topic Open Floor 17:59:03 scollier: ^^ 17:59:14 trishnag, how long do we have, couple of min? 17:59:21 60 seconds ... ish 17:59:22 scollier: yes we have 17:59:36 not long enough, but for next meeting, i'd like to come to decision on fedora-dockerfiles 17:59:53 right now it's suffering from bit-rot, and it's not fair to the folks opening PRs 18:00:21 or we can move the discussion to our channel scollier 18:00:35 trishnag, happy to do so. 18:00:41 scollier: cool 18:00:51 ending the meeting as we are running out of time :) 18:00:55 5 18:00:56 4 18:00:58 3 18:01:00 2 18:01:02 1 18:01:04 #endmeeting