17:01:17 #startmeeting fedora_atomic_wg 17:01:17 Meeting started Wed Jun 28 17:01:17 2017 UTC. The chair is dustymabe. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:01:17 Useful Commands: #action #agreed #halp #info #idea #link #topic. 17:01:17 The meeting name has been set to 'fedora_atomic_wg' 17:01:23 #topic roll call 17:01:27 .hello yzhang 17:01:28 yzhang: yzhang 'Yu Qi Zhang' 17:01:40 .hellomynameis kushal 17:01:41 kushal: kushal 'Kushal Das' 17:01:51 .hello maxamillion 17:01:52 maxamillion: maxamillion 'Adam Miller' 17:02:18 .hello jberkus 17:02:19 jberkus: jberkus 'Josh Berkus' 17:03:25 .hello dustymabe 17:03:26 dustymabe: dustymabe 'Dusty Mabe' 17:03:52 .hello rubao 17:03:53 rubao: rubao 'rubao' 17:04:07 welcome everyone 17:04:14 .hello mmicene 17:04:15 nzwulfin: mmicene 'Matt Micene' 17:04:33 .hello sayanchowdhury 17:04:36 sayan: sayanchowdhury 'Sayan Chowdhury' 17:04:41 #chair sayan nzwulfin rubao jberkus maxamillion kushal yzhang 17:04:41 Current chairs: dustymabe jberkus kushal maxamillion nzwulfin rubao sayan yzhang 17:05:02 .hello suiwenfeng 17:05:03 suiwenfeng: suiwenfeng 'suiwenfeng' 17:05:10 hi suiwenfeng, welcome 17:05:59 It`s my first meeting at atomic wg.|-_-\ 17:06:22 .hellomynameis jzb 17:06:22 jzb: jzb 'Joe Brockmeier' 17:06:27 .hello roshi 17:06:28 roshi: roshi 'Mike Ruckman' 17:07:04 #topic list of people to ping for meetings 17:07:13 #link https://pagure.io/atomic-wg/pull-request/288# 17:07:33 I created a list of people that I'll ping before the meetings start - this is more of a friendly reminder 17:07:46 if you want your name added to the list just add a comment in that PR and I'll get you added 17:08:05 make sure to let me know what your IRC nick is (in case it is different from your FAS account username) 17:08:56 do we change meeting time often? 17:09:13 nope 17:09:17 at least we haven't 17:09:41 This is a good time for most of us I think. 17:09:43 yzhang: no we haven't change the meeting time in a while 17:09:53 do note that we don't change for daylight savings, though 17:09:57 ok, was just curious, thanks 17:10:03 so we stay at the same UTC time year round 17:10:18 roshi, I will contact you after finish reading some documents. 17:10:19 if you live in a country/tz that changes with daylight savings then the meeting will "move" 17:10:21 for ou 17:10:30 sounds good suiwenfeng 17:10:30 .hello miabbott 17:10:31 miabbott: miabbott 'Micah Abbott' 17:10:45 ok i'll move on to previous meeting action items 17:10:51 #topic previous meeting action items 17:10:53 * maxamillion to add summary to ticket 17:10:55 https://pagure.io/atomic-wg/issue/284 17:10:57 * dustymabe to propose a VFAD for wiki and docs discussion 17:10:59 * jberkus to update wg on final approval of PRD 17:11:01 * jberkus to create new issues, specs for docs.projectatomic.io 17:11:03 publication 17:11:39 #action dustymabe to propose a VFAD for wiki and docs discussion 17:11:44 ^^ i promise to do that today 17:11:56 * sayan is having flaky internet, goes to find the issue 17:12:27 sayan: issue? related to action items? 17:13:04 welcom gbraad_ gholms :) 17:13:13 Heya 17:13:20 #chair suiwenfeng jzb roshi miabbott gbraad_ gholms 17:13:20 Current chairs: dustymabe gbraad_ gholms jberkus jzb kushal maxamillion miabbott nzwulfin roshi rubao sayan suiwenfeng yzhang 17:13:33 dustymabe: I'm looking at this ticket and I don't remember what needs a summary update :( 17:13:41 dustymabe: I'll ping some folks and try to get that sorted 17:13:45 Sorry; I'm normally double-booked now. 17:13:47 * maxamillion adds it to his TODO list for the day 17:13:59 gholms: same :/ 17:14:02 maxamillion: basically you were going to add a summary of the discussion we had last meeting 17:14:05 to the ticket 17:14:08 dustymabe: ah +1 17:14:21 maxamillion: logs from that meeting are here 17:14:22 https://meetbot.fedoraproject.org/teams/fedora_atomic_wg/fedora_atomic_wg.2017-06-21-17.03.log.html 17:14:47 ooops, lemme check on PRD 17:14:52 #action maxamillion to look at logs from last meeting and add summary of our discussion to ticket https://pagure.io/atomic-wg/issue/284 17:14:58 dustymabe: you're the best 17:15:04 dustymabe++ 17:15:05 * dustymabe blushes 17:15:18 maxamillion: save that cookie until after f26 release :) 17:15:18 :D 17:15:23 dustymabe: lol 17:15:25 Heh 17:15:31 Our PRD has been approved 17:15:32 dustymabe: I'm sure you'll get another one 17:15:35 jberkus: \o/ 17:15:40 jberkus: woot! 17:15:49 jberkus: roshi thanks for running that to term 17:15:55 and jbrooks as well 17:16:04 #info PRD has been approved 17:16:10 we were thanked for making the structure of our PRD match Server so that it's easy to comapre them 17:16:19 what'd I do? 17:16:21 so thanks for that roshi 17:16:23 roshi++ 17:16:24 roshi: wiki magic 17:16:31 dustymabe, what will be discussed in this meeting? 17:16:32 I can explain! 17:16:34 I tripped 17:16:37 there were wires 17:16:39 #info Matching Fedora Server's PRD format was helpful 17:16:39 I didn't mean to! 17:16:45 have mercy! 17:16:55 suiwenfeng: we typically go over some announcements, then previous meeting action items, then we go over tickets 17:17:09 tickets with the meeting keyword in our issue tracker 17:17:12 https://pagure.io/atomic-wg/issues?status=Open&tags=meeting 17:17:22 and then open floor, for anyone with anything new 17:17:44 suiwenfeng: any questions about that structure? 17:18:05 jberkus: any update on jberkus to create new issues, specs for docs.projectatomic.io 17:18:20 dustymabe, no, please assign me one of them if possible after the meeting... 17:18:52 dustymabe: not at this point, please re-action 17:18:55 #action jberkus to create new issues, specs for docs.projectatomic.io publication 17:19:01 suiwenfeng: the QA is much the same 17:19:03 suiwenfeng: ok we'll chat 17:19:41 dustymabe, Thanks. 17:20:07 #topic clarify policy on atomic host support for older Fedora "number" releases 17:20:11 #info https://pagure.io/atomic-wg/issue/228 17:20:27 * roshi saw a blog post about that the other day 17:20:29 Looks like the fedoramagazine post is up. 17:20:31 #info the fedora magazine post has landed https://fedoramagazine.org/upcoming-fedora-atomic-lifecycle-changes/ 17:20:47 the ticket can now be closed. anyone opposed? 17:21:06 dustymabe, nope. 17:21:08 Keep in mind that without support for N-1 releases working N+1 builds are more important than ever. 17:21:18 nope, close it 17:21:27 gholms: indeed and we are hoping to test more and more 17:21:34 in more automated ways 17:22:21 ok moving on to next ticket 17:22:43 Builds are just as important as tests. ;) 17:22:54 #topic fedimg: don't use 'builder' instance for uploading AMIs 17:23:00 #link https://pagure.io/atomic-wg/issue/269 17:23:22 this will most likely wait until after fedora 26 release because we are now in freeze 17:23:29 #info this will most likely wait until after fedora 26 release because we are now in freeze 17:23:34 sayan: any other update? 17:23:44 * kushal is wondering if sayan's network is working 17:23:53 dustymabe, network problem this side. 17:24:13 kushal: k. I think i had enough information 17:24:14 Yeah, so, import-volume involves writing to S3, FWIW. 17:24:33 dustymabe, He was working on it, I think he will ask for a FBR 17:24:54 kushal: hmm - i would actually prefer no FBR for this issue in particular 17:25:11 there was another thing that needs fixing with fedimg that we can FBR 17:25:21 for those interested FBR == Freeze Break Request 17:25:23 We also can't use the ec2-* tools thanks to their license, but I can stuff whatever functionality we need into euca2ools. 17:25:43 it's basically what you have to do to get things merged for changes to fedora infrastructure during "Freeze" 17:25:55 dustymabe, Okay. 17:26:09 gholms: cool. maybe we can get with sayan when he gets back to talk about it 17:26:13 moving on 17:26:16 dustymabe, I thought both of them will get into the same release. 17:26:17 Sounds good 17:26:28 #topic Set architecture label in the base container image 17:26:35 #link https://pagure.io/atomic-wg/issue/276 17:27:02 we discussed this last meeting, probably should have removed the meeting tag 17:27:06 removing it now 17:27:25 for anyone interested here was the comment from last meeting: 17:27:31 Discussed in the atomic working group meeting today. The proposal in the description sounds reasonable. Assuming we need to set an architecture label, then making layered images inherit from the base image makes sense. This assumes that there is not some technical reason why this can't be achieved. 17:27:31 hi 17:27:32 +1 17:27:34 DimGR: welcome 17:27:37 #chair DimGR 17:27:37 Current chairs: DimGR dustymabe gbraad_ gholms jberkus jzb kushal maxamillion miabbott nzwulfin roshi rubao sayan suiwenfeng yzhang 17:27:48 pagure seems to be having some troubles here 17:27:57 yzhang: yeah, that happens :( 17:28:07 it is improving, though 17:28:08 * sayan is back 17:28:11 sayan: yay 17:28:29 sayan: we moved past your ticket, but gholms wants to talk to you after the meeting 17:28:32 dustymabe: I will discuss the progress in open floor 17:28:39 ok 17:28:45 #topic Figure out comprehensive strategy for atomic host container storage 17:29:07 #link https://pagure.io/atomic-wg/issue/281 17:29:33 still dead for me 17:29:45 RIP pagure 17:29:53 so this ticket is mostly about moving atomic host to a larger root partition in the future 17:30:03 pagure was giving me 404 errors last night 17:30:09 and having the container-storage-setup match other platforms in fedora 17:30:21 using overlay2 by default on the root partition 17:30:33 you can easily configure it to put the storage somewhere else, but that will be the default 17:30:41 Other platforms? 17:30:55 other platforms == fedora sever fedora workstation etc.. 17:31:02 sorry, should have clarified 17:31:02 Ah 17:31:46 anyone have any input on this change? 17:31:47 dustymabe: that seems like a fine solution to me, and its what we talked about for F27 17:31:52 are we looking to move the schedule up? 17:32:20 It's awfully late to move it up. 17:32:57 no - just for f27 17:33:03 dustymabe, That sounds good. 17:33:18 so this hinges on overlayfs proving stable in f26 17:33:28 Mhm 17:33:28 i've been using it for quite a while and all has gone well 17:33:34 in f25 and f26 17:34:04 ok is anyone opposed to this change? 17:34:17 haven't followed this close 17:34:30 so the proposal is to have all of Fedora on overlayfs? 17:34:50 roshi: well - all of fedora will default to overlayfs for docker in fedora 26 already 17:34:59 this change is more for how storage is set up 17:35:05 ok, so for docker 17:35:12 docker on any Fedora host would be overlay 17:35:25 +1 to that then :) 17:35:30 right now in f26atomic host /var/lib/docker will be mounted on another filesystem on another logicial volume 17:35:40 just wanted to make sure I grokked what was going on :) 17:35:49 we would change the default to not do that 17:36:02 we already have done for f27 17:36:11 which basically means our storage setup is more simple for the default 17:36:12 (and i'd vote to port this to f26 at some point too) 17:36:51 yeah not sure about changing f26 mid release - but that is a separate discussion 17:36:59 We will have to add test for no regression on that part. 17:37:20 ok - anyone opposed? 17:37:36 nope 17:38:11 dustymabe, Nope. 17:38:14 #proposed changing fedora 27 and beyond to default to overlayfs on the root partition will help us simplify our storage setup and align better with other fedora variants. This seems like a reasonable change to make assuming that overlayfs proves stable for the f26 cycle. 17:38:20 acks? 17:39:06 dustymabe, Aye to that. 17:39:20 +1 17:39:53 +1 17:40:02 anyone else? 17:40:10 sorry, multi-tasking badly 17:40:16 maxamillion: happens 17:40:18 I always have this hour double-booked :/ 17:40:23 +1 17:40:33 ok will update the ticket 17:41:27 #agreed changing fedora 27 and beyond to default to overlayfs on the root partition will help us simplify our storage setup and align better with other fedora variants. This seems like a reasonable change to make assuming that overlayfs proves stable for the f26 cycle. 17:41:38 #topic Atomic Host images omit many common locales that all other flavors include 17:41:45 #link https://pagure.io/atomic-wg/issue/282 17:41:46 at some point we might want to establish a "minimum number of attendees for a meeting to continue with the meeting" so that we can define what's quorum for voting on things 17:41:51 just a thought 17:42:15 maxamillion: yeah - so far we've been pretty unstructured 17:42:27 which has its pros and cons 17:42:43 yeah 17:42:51 for QA, or blocker, we need 3 17:42:59 seems like we have some technical discussion going on in this ticket - anyone have anything to add for now ? 17:42:59 dustymabe: agreed 17:43:08 we generally have more than that for this meeting though 17:43:27 roshi: right, I'd just like to come up with some baseline and add it to the meetings wiki page 17:43:36 +1 to that 17:43:39 im catching up , hence being silent now 17:43:43 roshi: maxamillion want to take that action? 17:44:14 #action maxamillion roshi to come up with guidelines for meeting quorum for the atomic working group 17:44:17 too late 17:44:39 ok we can move past this ticket 17:44:40 dustymabe: +1 17:44:56 #topic Identify group of people interested in maintaining kubernetes 17:45:03 #link https://pagure.io/atomic-wg/issue/282 17:45:08 #undo 17:45:08 Removing item from minutes: 17:45:14 #link https://pagure.io/atomic-wg/issue/287 17:45:43 not I 17:45:46 so I created this ticket to try to find a group of people who use and want to help maintain kubernetes in Fedora 17:45:54 I don't know enough about it to help maintain it 17:45:57 seems like there's more than a few 17:46:10 yeah we've had a few people step up already (woot!) 17:46:29 dustymabe, I am in? 17:46:34 glad to reach out to the community and get new contributors 17:46:40 I've recently started to tinker around with kubernetes 17:46:44 suiwenfeng: :) i don't know if it is exclusive 17:46:54 but not much of a knowledge yet. 17:46:56 basically anyone is in - if they want to help 17:47:21 so what we are doing is finding a list of names and then we'll decide what to do with it 17:47:25 * gholms guarantees one will have more interest from outside of the red hat universe, provided someone manages to maintain it 17:47:33 we may make a new specialized group where kuberetes issues can be discussed 17:47:41 or that group may just piggy back off of this group 17:48:06 gholms: one == ? 17:48:23 kubernetes 17:48:43 gholms: rightr 17:48:45 Just how it appears from over here, at least 17:48:58 yep, that is very understood 17:48:58 I think that's a fair statement 17:48:59 gholms, Same everywhere. 17:49:39 so summary: add your name if you're interested :) 17:49:46 next topic 17:49:50 #topic open floor 17:49:54 who all has something for open floor? 17:50:14 * gholms pokes sayan 17:50:52 Yes, so over the last few days, I have been fixing the AMI public/private issues 17:51:02 eh, I would appreciate help with realizing: https://github.com/gbraad/minishift-atomic-iso 17:51:11 that's all 17:51:18 :-D 17:51:25 I need to step afk to grab some food before my next meeting, apologies all 17:51:28 dustymabe: thanks for hosting! 17:51:35 gbraad_: can you open an issue under https://pagure.io/atomic-wg ? 17:51:50 I deployed the fix on Monday. But, found an issue where the snapshot copied to other regions are still private 17:51:54 gbraad_: what is it? 17:51:54 gbraad_: might be nice to have the rest of the group updated on progress for that 17:52:08 jberkus: basically we want minishift to run on atomic host 17:52:13 ah, ok 17:52:20 i.e. the little minishift OS is atomic host 17:52:29 ok 17:52:30 which openshift running as a container 17:52:34 s/which/with/ 17:52:57 it would also be nice to have a minishift which runs on atomic host without a VM ... 17:53:06 #info Work continues on fixing EC2 image visibility, as copied images need their permissions changed 17:53:21 targeting a non-VM environment is on the roadmap, but not for now 17:53:31 sayan: any luck with the snapshot permissions ? 17:53:51 #info if you haven't yet tried out minishift, you should check it out 17:54:08 dustymabe: sadly, libcloud does not have ways to query the snapshot. But, I have a written a work-around 17:54:11 #undo 17:54:11 Removing item from minutes: INFO by dustymabe at 17:53:51 : if you haven't yet tried out minishift, you should check it out 17:54:15 #info if you haven't yet tried out minishift, you should check it out https://www.openshift.org/minishift/ 17:54:19 any docker/container host is our goal, but for predictability we run a VM at the moment 17:54:46 dustymabe: right now, I am testing that in the dev instance I have 17:55:01 gbraad_: glad to have you at this meeting. looking forward to seeing you more in the future 17:55:11 sayan++ 17:55:15 you will 17:55:21 After the fix is deployed, I'll make some patches to the libcloud to have method to query the snapshots 17:55:45 ok - anyone have anything else for open floor ? 17:55:57 have mostly been doing things for atomic in the bacground 17:56:18 is there i could test on openstack and fed atomic ? 17:56:23 *something 17:57:05 gbraad_: btw, thanks so much for helping with Atomic at LinuxCon China. did you get your shirt? 17:57:24 DimGR: basically running fedora or centos atomic host on openstack and reporting issues 17:57:35 or any difference in behavior you see vs non openstack 17:57:36 eh... no? 17:57:38 jberkus: was that shirt for gbraad? 17:57:39 oops 17:57:57 DimGR: have you tried out fedora 26 atomic host on openstack yet? 17:58:00 LOL 17:58:02 yzhang: there should have been multiple shirts, two for you and one for gbraad_ 17:58:10 jberkus: I got the 2 for me 17:58:14 dammit 17:58:21 language 17:58:24 I was under the impression the other one was for Gordon 17:58:27 Oh, right. Libcloud. 17:58:32 Well, there was only 2 but 17:58:33 dustymabe only downloaded the image today but didnt have time to fire up instances , will do tomorrow 17:58:33 ok i think we are done with meeting DimGR hit me up in #fedora-cloud 17:58:36 or #atomic 17:58:42 ok 17:58:46 #endmeeting