17:00:36 #startmeeting fedora_cloud_wg 17:00:36 Meeting started Wed Sep 21 17:00:36 2016 UTC. The chair is sayan. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:36 Useful Commands: #action #agreed #halp #info #idea #link #topic. 17:00:36 The meeting name has been set to 'fedora_cloud_wg' 17:00:48 #topic Roll Call 17:00:54 .hello trishnag 17:00:55 trishnag: trishnag 'Trishna Guha' 17:00:56 .hello sayanchowdhury 17:00:58 sayan: sayanchowdhury 'Sayan Chowdhury' 17:01:03 .hello jberkus 17:01:04 jberkus: jberkus 'Josh Berkus' 17:01:07 .hello bowlofeggs 17:01:08 bowlofeggs: bowlofeggs 'Randy Barlow' 17:01:10 .hello mmicene 17:01:10 nzwulfin: mmicene 'Matt Micene' 17:02:16 #chair trishnag bowlofeggs nzwulfin jberkus 17:02:16 Current chairs: bowlofeggs jberkus nzwulfin sayan trishnag 17:02:27 .hello scollier 17:02:28 scollier: scollier 'Scott Collier' 17:02:33 #chair scollier 17:02:33 Current chairs: bowlofeggs jberkus nzwulfin sayan scollier trishnag 17:02:45 * dustymabe sneaks in 17:03:00 .hellomynameis dustymabe 17:03:02 dustymabe: dustymabe 'Dusty Mabe' 17:03:03 #chair dustymabe 17:03:03 Current chairs: bowlofeggs dustymabe jberkus nzwulfin sayan scollier trishnag 17:03:12 bowlofeggs: \o/ 17:03:16 .hellomynameis kushal 17:03:17 kushal: kushal 'Kushal Das' 17:03:18 .hello subho 17:03:21 subho: subho 'None' 17:03:31 #chair kushal 17:03:31 Current chairs: bowlofeggs dustymabe jberkus kushal nzwulfin sayan scollier trishnag 17:03:35 #chair subho 17:03:35 Current chairs: bowlofeggs dustymabe jberkus kushal nzwulfin sayan scollier subho trishnag 17:03:47 .fas jasonbrooks 17:03:48 jbrooks: jasonbrooks 'Jason Brooks' 17:04:03 #chair jbrooks 17:04:03 Current chairs: bowlofeggs dustymabe jberkus jbrooks kushal nzwulfin sayan scollier subho trishnag 17:04:12 #topic Action items from last meeting 17:04:21 * Kushal will work on moving to Pagure from Trac 17:04:23 * kushal will create the vote ticket, and update the list :) 17:05:03 kushal: any updates? 17:05:17 I know you deployed the non-gating test one 17:05:44 .fas rtnpro 17:05:45 rtnpro: rtnpro 'Ratnadeep Debnath' 17:05:49 sayan, the first item I will do next week. 17:05:52 could not finish 17:05:52 #chair rtnpro 17:05:52 Current chairs: bowlofeggs dustymabe jberkus jbrooks kushal nzwulfin rtnpro sayan scollier subho trishnag 17:06:01 sayan, readd that action 17:06:06 ok, re-actioning it 17:06:19 #action Kushal will work on moving to Pagure from Trac 17:06:34 Moving onto tickets 17:06:51 #topic vagrant boxes fixups https://fedorahosted.org/cloud/ticket/136 17:07:40 .hello maxamillion 17:07:40 maxamillion: maxamillion 'Adam Miller' 17:07:55 imcleod isn't here, so moving to next ticket 17:08:02 #chair maxamillion 17:08:02 Current chairs: bowlofeggs dustymabe jberkus jbrooks kushal maxamillion nzwulfin rtnpro sayan scollier subho trishnag 17:08:24 #topic design, deploy and document Fedora OpenShift Playground (FOSP) https://fedorahosted.org/cloud/ticket/153 17:08:50 Next work session is Sept 23 so skipping this one too 17:09:32 jberkus: anything you would like to add? 17:09:52 sayan: no, except that the work session has been postponed 17:09:58 scollier: ^^^ ? 17:10:17 jberkus, looking 17:10:34 sayan, i pushed the meeting out one week. 17:10:43 i'll update the ticket. 17:10:55 scollier: sure 17:11:28 #topic make Fedora Atomic download page clearer https://fedorahosted.org/cloud/ticket/154 17:12:01 jberkus: update on this one? ^^ 17:12:25 sayan: see the whole long discussion about the new getfedora pages 17:12:29 and the wiki page 17:13:15 are we ready to discuss getfedora? 17:13:38 so, here: https://fedoraproject.org/wiki/Cloud/getfedora_update 17:14:04 jberkus: what about the aws picker for cloud base? 17:14:15 here's a draft from the designer: https://raw.githubusercontent.com/fedoradesign/nextweb-assets/master/Mockups/Brochure%20Site/brochure-mock-atomic_WIP.png 17:14:23 sayan: I can't answer that. someone else has to 17:15:37 the mockup looks good to me 17:16:06 jberkus, looks nice 17:16:12 just so everyone's up to date, the problem is that this redesign orphans the AWS picker for Cloud base images (as opposed to Atomic) 17:16:21 I like it 17:16:22 +1 17:16:29 presumably that now needs to live somewhere on Server 17:16:42 but someone more central to Fedora than me needs to take it to them 17:16:48 So, is cloud base going to server for 25? Or is that 26? 17:17:03 25 17:17:42 i guess a question about that 17:18:05 so does the server WG now own it? are we no longer going to talk about "cloud" issues in this channel and in this meeting? 17:18:27 dustymabe: +1 17:18:54 would seem kind of odd 17:18:54 that's presumably the idea, yeah. the problem is that while the council members talked about cloud --> atomic, there wasn't any serious talk about how that changes server 17:19:11 does the server WG know about the issue and the plan they would be taking over? 17:19:20 I don't know that they do 17:19:32 Well, that's an important bit 17:19:45 yes, which is why I keep trying to raise it 17:20:39 yeah I don't know what the best situation is 17:20:51 seems almost like server should be broken out into two targets 17:21:06 server(hardware + cloud) 17:21:14 Whatever happens w/ it, they need to be in on it 17:21:42 I'd expect to use server bits on metal and cloud 17:21:48 Just like I expect that from atomic 17:22:02 or would it make more sense to get a fair # of people from the server WG to join CLOUD meetings and then have a new WG for Atomic 17:22:27 jbrooks: +1 17:22:32 The weird thing is that we chose to switch our focus to atomic a long time back 17:23:20 ok so basically what we need to do is have server WG own the thing, but a fair number of us probably need to put down some roots there as well 17:23:32 dustymabe: I think it would be better for people from server WG to join Cloud meetings 17:23:37 I don't get how a minimal install of fedora that'll run on a cloud even counts as a separate WG 17:23:38 or the vice-versa? 17:23:46 sayan: +1 17:24:01 Whichever, but I'm on team atomic, wherever that is 17:24:16 jbrooks: well I think "Fedora Server" has meant something in the past. i.e. not a minimal install 17:24:31 jbrooks: yeah, the choice was made but then nothing ever happened ... I think the fear of leaving people behind lead to maintaining things as they were 17:24:38 Yeah 17:25:05 Well, maybe we need to do a new atomic wg 17:25:09 so that is why I said that 17:25:13 And we need to do the prd and all that 17:25:25 i think a new wg would make sense 17:25:27 And we haven't done it yet, so we're still the old cloud wg 17:25:38 right that would be a more official switch 17:25:43 jbrooks: we started during Flock, but it didn't get finished 17:26:01 and the people who are just interested in the 'old' cloud can still come to the same places they have in the past 17:26:04 I think the problem with creating a new group is getting people to show up and actually do work 17:26:13 jberkus, are there any notes from that? 17:26:15 we have enough trouble with that now 17:26:23 I will show up for atomic 17:26:29 That's why I've been here 17:26:33 me too 17:26:35 jbrooks: thank you for being here 17:26:39 I will as well, but I'll stop showing up for cloud 17:26:42 Heh, thank you 17:26:51 Me too, probably 17:26:56 I think what would happen is there'd be an exodus of people doing work for cloud 17:26:59 so the idea from the PRD meeting was not to create a 4th group 17:27:07 I mean, I want there to be a vagrant image of fedora base 17:27:09 so we would effectively abandon the users just as we would have otherwise 17:27:12 but to replace Cloud with Atomic and merge Cloud Base into Server 17:27:19 I think it's crazy that fedora wouldn't just have that 17:27:48 right. so we'd at least have to have some minimal team of people who make sure that happen. part of the server WG or not 17:27:51 which I'm ok with 17:27:57 just want to make sure it has a place 17:28:05 dustymabe, correct. 17:28:12 But sadly we will not have time for it. 17:28:12 dustymabe: rightly said 17:28:14 :( 17:28:22 the Server WG needs to accept the responsibility for the Cloud Base 17:28:32 OK, who's the server WG point person? 17:28:36 we can't just be like "here, you deal with it" and then they will magically take on the extra work :) 17:28:54 Or maybe we ping the FPL 17:28:56 :) 17:29:14 Fix everything, FPL! 17:29:21 well. does it really make sense though? a lot of the "workflows" for building/testing cloud base are so similar to Atomic 17:29:39 In centos, the base stuff is core 17:29:40 * dustymabe head explodes 17:30:08 hello I am from the server WG 17:30:14 smooge: hi 17:30:24 hello I am from the bowlofeggs WG 17:30:29 (hehe) 17:30:30 heh 17:30:33 bowlofeggs: hehe 17:30:41 hehe 17:30:42 bowlofeggs, where's your dang prd 17:30:51 bowlofeggs hey, your WG release is late. I'm hungry. 17:31:08 hahaha 17:31:31 So, smooge, does the server WG want to / plan to adopt fedora cloud base? 17:31:39 my prd is deployed all across the world in various diners and restaurants 17:31:40 smooge: so the plan is that the Server WG take over cloud base 17:31:49 yes I have heard of this plan 17:32:18 I think we are looking for a more formal request and statement of what it means 17:32:47 smooge, OK, and that'd be the cloud WG that would be doing the requesting? 17:33:14 well it is your base 17:33:24 smooge: how do you think the integration could go? 17:33:27 and I don't believe all your base belong to us 17:33:33 heh 17:34:45 so I am not sure what the "cloud" base means. How does it overlap with the Server base? Are we talking about images that run in cloud or images that create a cloud infrastructure 17:35:06 smooge: it's just a "minimal" fedora image 17:35:14 for running in the cloud 17:35:14 yep 17:35:17 or as local VM 17:35:47 And, btw, the sort of thing I'd start w/ on metal, too, if that was an option 17:36:17 well the on-metal is what the Server WG already does 17:36:23 dustymabe, ok and it needs to be built into various "forms": vbox, libvirt, ?? 17:36:23 jbrooks: can't you just select what groups you want installed in anaconda 17:36:24 they'd basically just add cloud images 17:36:33 smooge: right 17:37:20 dustymabe, yeah, I guess I'd use the netinstall 17:37:38 dustymabe, I just can't get over the 1.7GB iso 17:37:49 I guess I'm being old and cranky or something 17:38:40 so should someone from the Cloud WG file a request with the Server WG more info? 17:39:24 "with the Server WG more info" ? 17:39:48 this might sound crazy but it almost seems like more work to "move" cloud to the server WG than it is to just keep it in this WG and have a few people maintain it 17:39:58 let's make a wiki page or gobby doc or something that we can collectively as a group edit and review it before shipping the request to the server WG 17:39:58 sayan, maxamillion could someone file a ticket or a request on the mailing list for what is wanted 17:40:10 maxamillion: I meant ticket with more information 17:40:11 i get that the "focus" would change 17:40:12 a detailed one 17:40:18 dustymabe: ehhhhh 17:40:32 dustymabe: I thing OK with helping to maintain it 17:40:46 s/thing/am 17:40:52 dustymabe: there are some people who spend a respectable amount of time on it during various periods of the release cycle 17:41:31 if the cloud base image moves to the server WG, what happens to the current suite of automated tests? 17:41:36 maxamillion: do you think if it moved to server WG the people that work onit would be different people> 17:41:36 by the way, don't take my questions as "we don't want it." it is more of "what do you want us to take over" so that we don't end up saying "here is server.img" and you wanted something else 17:43:08 dustymabe: I don't know, I just don't want to shift focus and then people who are doing work to work on the new thing and the old things just fall on the floor .... or lets do that, make a hard line in the sand and just say "we're not doing $thing anymore" and be done with it, I don't really have a preference but we need to have a well defined objective of what we're trying to accomplish and 17:43:14 how we would like to accomplish it 17:44:03 maxamillion, One thing I keep thinking, is how can the cloud wg decide when most of the official membership isn't here / invovled 17:44:34 Most of us couldn't even vote, officially, to drop this 17:44:44 do we have people who are doing work on cloud images who aren't in the meetigns? 17:44:45 jbrooks: that is when we vote to update the WG members 17:45:05 dustymabe: can we do that though? I think that'd need to go to the Council 17:45:20 ehh. I think we've done it in the past 17:45:29 adimania hasn't been here in like a year 17:45:40 roshi will keep his spot, as he will be back 17:46:26 TBH delivering the cloud image could be done by like 1 person if that was their focus 17:46:54 maybe we'll break off a small group from this one - still part of the same WG 17:46:58 but with a focus on cloud base 17:47:08 so we can have two separate meetings during the week 17:48:02 dustymabe, but most of us will go to the container/atomic meeting. 17:48:23 maxamillion, We can update our membership as required iirc. 17:48:23 kushal that is fine 17:48:51 that is no different than having a server WG with like 2 people that care about cloud base image 17:49:11 same difference in my opinion 17:49:27 To me, a server product that doesn't run in clouds... it's much less compelling 17:49:41 I expect to be able to run a server distro in the cloud or on metal 17:49:56 jbrooks: I guess I need to understand server a bit mor 17:49:59 more 17:50:19 Like, understand that WG or product? 17:50:26 WG 17:50:29 It seems like it was trying to be more like Windows Server 17:50:35 with the roles and such 17:50:37 I'll say this, I have no real interest in the base cloud image so I don't really pay it much attention, and as someone who doesn't really have a "dog in the fight" I'm going to duck out of the discussion around the base image's fate 17:50:38 yeah 17:50:44 And then it seems like that was sort of dropped 17:51:03 for my cloud image I don't want anything extra 17:51:17 just a minimal base and then i can use cloud-init to get to where I want to be 17:51:29 ^ that 17:51:36 scollier++ 17:51:37 dustymabe: Karma for scollier changed to 2 (for the f24 release cycle): https://badges.fedoraproject.org/tags/cookie/any 17:52:04 so jbrooks I agree with you. having a server product that runs in the cloud is beneficial too 17:52:07 dustymabe, but on metal you want more? 17:52:09 * tflink is very interested in the base image (or at least the tooling that makes it) as that is a critical bit of taskotron 17:52:22 but I would suggest starting with cloud base image and then easily configured that using the roles to get to a desired state 17:52:35 jbrooks: not necessarily 17:52:38 Yeah, so, just call it server and be done 17:52:48 I personally always want to start minimal and then add the things I want 17:52:52 Me too 17:53:08 but I think the server WG had a different idea on that 17:53:10 smooge: ^^ 17:53:41 * dustymabe just looked at the time - sorry for derailment 17:53:52 let's take this to #fedora-cloud after the meeting 17:53:55 sayan please continue 17:54:00 sure 17:54:19 should I move to open floor? 17:54:29 probably unless there was something else big 17:54:30 or quickly go through other tickets? 17:55:09 #topic Decide on post-GA update cadence for various deliverables https://fedorahosted.org/cloud/ticket/155 17:55:14 maxamillion: jbrooks ^^ 17:55:18 dustymabe, I don't know what the server WG has in mind exactly.. beyond we want to be able to deploy stuff on an image somehow the way people want it 17:55:42 smooge: i'll be in fedora-cloud 17:55:44 I'll update that ticket and close it 17:55:47 sayan: ^ 17:55:47 dustymabe, which why I want to make sure a formal idea is put forth so no one feels like they got bum-shafted 17:55:51 maxamillion: thanks 17:55:58 sayan: we have that sorted out, I'm hoping to start work on it by the end of the week 17:56:20 #topic Need complete Kickstart docs for Atomic Host https://fedorahosted.org/cloud/ticket/156 17:56:39 jberkus: any update on this one? ^^ 17:57:24 been busy with other stuff, new kickstarts are awaiting testing 17:57:57 Ok 17:58:06 moving onto next then 17:58:18 #topic Put OpenShift Origin in Container with Fedora Base https://fedorahosted.org/cloud/ticket/168 17:58:31 scollier: ^^ 17:58:51 yup, far as i know, we are not going to block the FOSP work on the FAO work. 17:58:59 sayan, i'm happy to close that ticket. 17:59:14 sounds like maxamillion has that part covered. 17:59:14 scollier: thanks 17:59:19 yep 17:59:22 scollier: I do? 17:59:33 maxamillion, yup, origin in system containers as part of FAO 17:59:49 scollier: oh ... ehhhh, if I ever find time I'll get back to it :) 17:59:54 since we are less on time moving to open floor :) 18:00:00 #topic Open Floor 18:00:17 Anybody have anything to share in Open Floor? 18:00:49 I want to schedule a test day 18:00:57 jbrooks++ 18:00:58 I can send a msg to the list about that 18:01:04 jbrooks: +! 18:01:06 +1 18:01:15 jbrooks: +1 18:01:40 we need regular mails to explain all of these sort forms :( 18:01:43 But I only want to test the server product 18:01:46 heh j/k 18:01:56 #action jbrooks to send out an email to the list regarding Test Day 18:02:25 :) 18:02:35 * maxamillion has to run to get food before next meeting 18:02:40 sayan: thanks for hosting! 18:02:47 Anything else? 18:02:59 maxamillion: thanks for coming 18:03:07 ending in 18:03:08 3. 18:03:10 2. 18:03:13 1. 18:03:17 #endmeeting