16:31:01 #startmeeting fedora_atomic_wg 16:31:01 Meeting started Wed Feb 14 16:31:01 2018 UTC. The chair is miabbott. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:31:01 Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:31:01 The meeting name has been set to 'fedora_atomic_wg' 16:31:18 #topic roll call 16:31:27 .hello2 16:31:28 strigazi: strigazi 'Spyros Trigazis' 16:31:30 .hello jlebon 16:31:31 jlebon: jlebon 'None' 16:31:34 .hello gscrivano 16:31:35 .hello rubao 16:31:35 giuseppe: gscrivano 'Giuseppe Scrivano' 16:31:37 .hello sinnykumari 16:31:38 .hello2 16:31:39 rubao: rubao 'rubao' 16:31:42 ksinny: sinnykumari 'Sinny Kumari' 16:31:45 miabbott: miabbott 'Micah Abbott' 16:31:45 .hello2 16:31:51 dustymabe: dustymabe 'Dusty Mabe' 16:32:32 .hello smilner 16:32:33 ashcrow: smilner 'None' 16:32:48 #chair miabbott strigazi jlebon gscrivano ksinny 16:32:48 Current chairs: gscrivano jlebon ksinny miabbott strigazi 16:32:50 .hello2 16:32:51 sanja: sanja 'Sanja Bonic' 16:33:18 #chair rubao dustymabe sanja smilner 16:33:18 Current chairs: dustymabe gscrivano jlebon ksinny miabbott rubao sanja smilner strigazi 16:33:32 .hello jasonbrooks 16:33:33 jbrooks: jasonbrooks 'Jason Brooks' 16:34:00 #chair jbrooks 16:34:00 Current chairs: dustymabe gscrivano jbrooks jlebon ksinny miabbott rubao sanja smilner strigazi 16:34:13 .hello walters 16:34:14 walters: walters 'Colin Walters' 16:34:30 #chair walters 16:34:30 Current chairs: dustymabe gscrivano jbrooks jlebon ksinny miabbott rubao sanja smilner strigazi walters 16:35:01 did i miss anyone? 16:35:23 i'll just assume i nailed it 16:35:29 #chair ashcrow 16:35:29 Current chairs: ashcrow dustymabe gscrivano jbrooks jlebon ksinny miabbott rubao sanja smilner strigazi walters 16:35:34 #topic previous meeting action items 16:35:48 - dustymabe to create new ticket for evaluating podman in a few months 16:35:55 - sanja to create ticket in atomic-wg pagure for coordinating devconf.us talks 16:36:03 dustymabe: you first 16:36:08 #info dusty created ticket for evaluating podman in a few months 16:36:11 #link https://pagure.io/atomic-wg/issue/422 16:36:24 fantastic 16:36:32 sanja: how about your action? 16:37:02 The creating ticket thing? 16:37:06 Done during last meeting. 16:37:08 yep 16:37:11 I think we linked it...? 16:37:23 found it 16:37:24 #link https://pagure.io/atomic-wg/issue/421 16:37:27 #link https://pagure.io/atomic-wg/issue/421 16:37:29 :) 16:37:39 lots of links 16:37:47 :-) 16:37:53 ok, i think that is all the action items from last meeting 16:38:03 I'll follow up on that next week when people add what they wanna talk about in case there's some missing. @mclasen maybe? 16:38:15 .hello2 16:38:16 lorbus: lorbus 'Christian Glombek' 16:38:18 * mclasen looks 16:38:22 * lorbus is sorry for being late 16:38:23 sounds good to me 16:38:30 #chair lorbus 16:38:30 Current chairs: ashcrow dustymabe gscrivano jbrooks jlebon ksinny lorbus miabbott rubao sanja smilner strigazi walters 16:38:44 onward to topics 16:38:55 #topic Design Meetings over Video 16:39:16 #link https://pagure.io/atomic-wg/issue/418 16:39:19 sanja: this was re: devconf.us ? 16:39:28 dustymabe: it's assigned to you...any update? 16:39:38 i don't have any update on this right now 16:40:07 i need to get it on the calendar and take topics 16:40:14 .hello2 davdunc 16:40:15 davdunc: davdunc 'David Duncan' 16:40:25 I have a feeling when we have the first item that needs discussing in that way then updates will happen pretty quickly. 16:40:31 #action dustymabe to schedule february VFAD a CFP for proposed topics 16:40:32 welcome davdunc! 16:40:36 mclasen yes 16:40:37 .hello2 jdoss 16:40:38 jdoss: jdoss 'Joe Doss' 16:40:45 i'll keep the meeting tag on the issue and we can discuss at the next meeting perhaps? 16:40:48 :-) thnaks @ashcrow 16:40:50 #chair davdunc 16:40:50 Current chairs: ashcrow davdunc dustymabe gscrivano jbrooks jlebon ksinny lorbus miabbott rubao sanja smilner strigazi walters 16:40:51 miabbott: +1 16:41:09 #chiar imcleod 16:41:12 foo 16:41:12 sanja: will think about it 16:41:14 #chair imcleod 16:41:14 Current chairs: ashcrow davdunc dustymabe gscrivano imcleod jbrooks jlebon ksinny lorbus miabbott rubao sanja smilner strigazi walters 16:41:21 good enough for me 16:41:33 #topic Identify users for docker hub auto builds for system containers 16:41:34 Not the first time I've been called a chiar.... 16:41:40 #link https://pagure.io/atomic-wg/issue/379 16:41:45 #chair jdoss 16:41:45 Current chairs: ashcrow davdunc dustymabe gscrivano imcleod jbrooks jdoss jlebon ksinny lorbus miabbott rubao sanja smilner strigazi walters 16:41:49 I can talk to this one ... 16:42:08 A bit ago we agreed to start getting automatic builds happening in dockerhub ... 16:42:11 ashcrow: you opened this issue...not assigned to anyone, but it looks like there is still discussion/work to be done 16:42:41 Lol imcleod 16:42:59 but before we are able to do this a few things need to happen. The first is what user should be used to handle the access. Since this was set up before I joined the team I don't really know if one was already created, or if a new one is needed, etc... 16:43:14 Thanks for confirming I'm a weirdo by making me laugh out loud in a cafe 16:43:25 that issue is asking @jberkus @dustymabe @jasonbrooks @giuseppep @walters (and anyone really) if they know of one. 16:43:51 ashcrow, this is a user on dockerhub, right? 16:43:58 I don't know, but the atomicapp ppl must 16:44:05 They were using the namespace 16:44:22 yeah I have some vague memory of an account we used 16:44:24 jbrooks: it's a user on dockerhub and on github as they have to have a 1:1 for autobuilding 16:44:46 will we use the docker hub autobuild? 16:44:49 we don't have to have the answer in the room, I just wanted to get this in front of people as the ticket was starting to fade 16:45:05 or do we build the images somewhere else and just push them to docker.io? 16:45:17 it would be nice if FLIBS did this 16:45:27 giuseppe: we could do either, but the original work to look into it for the wg was about using dockerhub build. But we don't have to do that. 16:45:50 ashcrow, my desire for these images has not faded, though ;) 16:46:02 jbrooks: neither has mine 16:46:16 and since we had a blog post about the change go out late last year I would like to make some progress :-) 16:46:26 agreed 16:46:36 dockerhub will probably be easier, we will need only to setup the webhooks in github 16:47:07 there is a projectatomic org on docker hub 16:47:45 sanja: do you have the keys to the projectatomic org on docker hub? 16:47:49 or know who does? 16:47:55 I can see the list of people who are in it 16:47:57 dustymabe: true, but the org and the user would be different 16:48:01 looks like I should be able to add people 16:48:09 Nope so guessing jberkus or someone else 16:48:13 I'd assume we would have a bot account, but if we don't then we can create one 16:48:18 *cough* or me 16:48:26 * dustymabe staring at web interface right now 16:48:29 Exactly. 16:48:49 hook us up, dustymabe 16:49:02 :-) 16:49:33 I'd like to weed out this list a bit. I'm going to send an email to current members and see who doesn't need it. 16:49:43 who from here would like to request access? 16:50:12 for this work 16:50:22 dustymabe: I would like access stevemilnerrh 16:50:22 I'd like 16:50:33 gscrivano 16:50:37 if we're doing "arbitrary containers" as distinct from the "must be RPMs" FLIBS, has any thought been given to CentOS Container Pipeline? 16:50:44 dustymabe, I can help jasonbrooks 16:51:15 walters, that could work, I don't know if they build immediately or if it's nightly 16:51:48 #action dusty to send email to existing members of PA docker hub org and proposed new members to solidify membership 16:52:28 thanks dusty 16:52:47 i think that is a good first step. do we want to sign up for anything else on this ticket? 16:52:55 walters: the intial work is to remove all unused / unupdated container images from our namespace in dockerhub, and then populate development builds to that namespace 16:52:59 jbrooks, I think they check the git repository quite often, I added docker and crio there and when something is changed in git I get a notification of a new build almost immediately 16:53:33 ashcrow: right, definitely agree with the goal 16:53:38 giuseppe, I have no complaints with using them, I've got kube containers there already, it works reliably 16:54:22 miabbott: I think that's it 16:54:39 thanks everyone for jumping in, looking, and discussing it! 16:54:57 ok, those are the only two issues with the 'meeting' label 16:55:28 #topic open floor 16:57:00 anyone got something to discuss? 16:57:04 update on AWS Marketplace: FAH 27.72 is making the way through preliminary scans. 16:57:12 I pinged dustymabe about this but Azure has no official or unofficial Fedora Cloud or Fedora Atomic images. 16:57:31 Uuuh neat 16:57:32 davdunc: nice 16:57:53 #info update on AWS Marketplace: FAH 27.72 is making the way through preliminary scans. 16:57:54 jdoss: a problem there is WALinuxAgent... 16:57:55 I he said that that is because Azure requires support for images. Is there any way we could get a community image up that comes from a trusted source? 16:57:57 i.e. https://github.com/Azure/WALinuxAgent 16:58:19 walters: it looks like cloud-init is starting to be a thing. 16:58:21 it's a beast 16:58:38 https://docs.microsoft.com/en-us/azure/virtual-machines/linux/using-cloud-init 16:58:38 walters: yeah. WALinuxAgent is a hard requirement any longer for Azure 16:58:55 we actually did work to get the new cloud-init into f27 before release 16:58:58 yeah there's some stub cloud-init for Azure support 16:58:58 specifically for that 16:59:16 which, don't get me wrong, there's a lot of things to dislike about cloud-init but at least it's a known standard-ish quantity 16:59:29 yeah. that's my general sentiment towards it 16:59:39 I can build images on my end, but I was bummed that there was no Fedora Cloud/Atomic from an end user standpoint. 16:59:45 davdunc: what kind of 'scans' do images have to go through? 17:00:19 dustymabe: they look for ssh keys, logs, specific CVE's 17:00:35 k 17:00:46 any chance that scan pipeline could be public? 17:00:52 or at least the test results from it? 17:01:00 would be nice to run every compose through it 17:01:21 It sounds like there is a lot of redtape for any kind of official image. I am not sure if there is much that can be done. 17:01:23 there is a seller's guide for the process. The results are not public b/c they only disclose the guidelines. 17:01:51 kk, yeah I was looking more for a pass/fail type of thing 17:02:01 we don't want to do a release and then find out later it failed the scans 17:02:07 but the AWS MP expectations are public and "pass" is a hard requirement for publication. 17:02:08 davdunc: is that scan pipeline actually aware of ostree? I wonder if it e.g. just looks for `rpm -q bash` without realizing the OS actually lives in `/ostree/deploy/...` 17:02:10 but we'll probably have to go with that for now 17:02:25 davdunc: related https://bugzilla.redhat.com/show_bug.cgi?id=1102241 17:02:26 walters: won't rpm -q bash work though? 17:02:43 ahh, you mean in a non-booted image 17:02:51 dustymabe: worst case, people would start with a older version until we fix it, and upgrade it 17:02:51 dustymabe: it depends on whether the pipeline *boots* the system or just mounts the filesystem using something like libguestfs 17:03:07 misc: right 17:03:16 walters: there will be an opportunity for me to fine tune their scanning process. 17:03:28 now, we also need to know that the pipeline failed 17:03:37 hmm, thinking about it i'd guess they use cloud-init to inject a scan agent 17:03:40 anything else would be a huge maintenance pain 17:03:48 I'll let everyone know the results. 17:04:09 Thanks davdunc 17:04:09 and we can make a decision on how to respond or update as needed. 17:04:14 davdunc: +1. the only other thing to note is that we did release 27.81 yesterday 17:04:30 so :) - might have another one to run through when they get done with that one 17:04:37 once I have the steps in the process documented, I'll work with Sayan to make it a part of the automation. 17:04:48 #action davdunc to inform WG about AWS scan results of Fedora Atomic Host 27.72 17:05:05 I have another item for open floor 17:05:33 dustymabe: before you start, are we finished discussing Azure? 17:05:45 we've been getting a lot of interest lately in the channel. probably related to the workstation buzz and also the aquisition red hat made recently 17:05:49 oops 17:05:57 anything we can do to inch closer to getting Fedora images there? 17:06:05 miabbott: it's mostly legal 17:06:19 i pretty much have to wait for any input from mattdm on it 17:06:35 we do have a proposed blog post showing people how to create their own images 17:06:39 that we never got published 17:06:45 but is pretty much completely written 17:07:03 #info waiting on input from mattdm about publishing Fedora images to Azure 17:07:07 see https://pagure.io/atomic-wg/issue/364 17:07:19 Where is the blog post Dusty? 17:07:22 Kk 17:07:23 is it this? - https://gist.github.com/baude/03ddc7198a43fe1a007b10a1fecd268f 17:07:48 yeah 17:08:17 ok, since we are in a holding pattern there, let's proceed with other items...go ahead dusty 17:08:22 :) 17:08:27 we've been getting a lot of interest lately in the channel. probably related to the workstation buzz and also the aquisition red hat made recently 17:08:38 which also means we get a lot of users hitting common problems 17:09:03 hopefully we'll get them cleaned up soon, but until then I wrote up an email detailing some of the common ones we see 17:09:08 https://lists.projectatomic.io/projectatomic-archives/atomic-devel/2018-February/msg00029.html 17:09:10 Dusty - what you wrote in the email recently is seeping into an FAQ on the website. 17:09:15 we might make this part of an FAQ 17:09:18 or something 17:09:24 sanja: ahh. perfect 17:09:52 See above. 17:10:00 #info a list of common problems affecting users: https://lists.projectatomic.io/projectatomic-archives/atomic-devel/2018-February/msg00029.html 17:10:10 #info common problems users are hitting with Atomic Host/Workstation 17:10:11 #link https://lists.projectatomic.io/projectatomic-archives/atomic-devel/2018-February/msg00029.html 17:10:12 nice list, dusty! 17:10:14 sanja: +1 17:10:33 thanks jlebon 17:10:41 Tomorrow around 2pm UTC or earlier I expect the website switch - assuming everything works smoothly. Depends on misc and my code and OpenShift. And cosmic rays. 17:10:56 cosmisc ray ? 17:11:10 sanja: maybe we can put the docs site payload on a falcon heavy and send it into the cloud that way 17:11:18 sanja: \o/ 17:11:21 ^^ 17:11:27 Then you'll see what you wrote as part of an FAQ and I'd encourage everyone who has answers to these common questions to put that in the ideas folder on the new repo or send an email to atomic-devel mailing list like dustymabe did 17:11:42 Lol dustymabe 17:11:56 dustymabe: you know those launches are not THAT far from me ... :-D 17:11:59 (not gonna do a /nick elon_misc) 17:12:03 misc: HA! 17:12:05 dustymabe +1 for the composition of the q&a mail 17:12:25 that's it from me 17:12:44 #info submit ideas for docs/guides/faqs on projectatomic site 17:12:46 #link https://lists.projectatomic.io/projectatomic-archives/atomic-devel/2018-February/msg00029.html 17:12:48 #unlink 17:12:48 sanja: when you say "website switch", you mean change the DNS ? (cause that might take a bit more time) 17:12:54 #link https://github.com/projectatomic/atomic-site-new/tree/master/ideas 17:12:57 i don't even know if unlink is a thing... 17:13:05 miabbott: I know #undo is 17:13:09 don't know about #unlink 17:13:15 oh well, i tried 17:13:22 A for effort 17:13:30 :-) 17:13:37 ooh ooh 17:13:41 I have another Item 17:13:48 I mean change the website from new to original domain and RIP old one. Unless we can't make that work for tomorrow then I'll link the new one and we do a redirect later 17:14:05 #info the first Atomic Workstation SIG meeting will be on Monday 02/19 17:14:12 #link https://fedoraproject.org/wiki/Workstation/AtomicWorkstation/SIG/Agenda 17:14:14 I have another topic after that 17:14:29 * misc will discuss on website change outside of the meeting 17:14:29 additions to the agenda more than welcome 17:14:58 mclasen: yep. unfortunately I'm going to be AFK that day, but I'll plan to make the next one 17:15:09 and I'll read the notes if any are taken 17:15:22 I'll try to make it for some of the meeting. It conflicts with another meeting which I can't drop. 17:15:25 long weekends are the best. I'll make sure we take notes 17:15:49 So in essence, we started the SIG and you see that the momentum is really good right now. If you're interested in metrics it doesn't only feel like we gained more people, we've gained significantly more people (on IRC, Twitter, etc - we just have to make use of it and push further) 17:16:03 mclasen++ sanja++ thanks for helping bootstrap that WG 17:16:25 Huge thanks mclasen for your blog postd 17:16:25 *s 17:16:25 hat tip to otaylor too 17:16:37 mclasen++ 17:16:37 sanja: Karma for mclasen changed to 2 (for the f27 release cycle): https://badges.fedoraproject.org/tags/cookie/any 17:16:39 dustymabe++ 17:16:50 we also have a whole new interesting problem set in the crossover/relationship between FA{H,W} and Fedora {Server,Workstation} and Modularity 17:16:52 sanja: :) you already gave me a cookie 17:17:08 walters: indeed 17:17:13 walters: *nods* 17:17:14 You deserve ALL THE COOKIES 17:17:29 walters: do you think there is enough content their to discuss during a VFAD? 17:17:35 sigh 17:17:37 there* 17:17:38 it's a good problem to have though! 😄 17:17:49 Ok we should probably talk about that problem ...not today though 17:17:59 once I start mixing up their/there/they're then it's time for me to call it a day 17:18:16 didn't even make it to lunch 17:18:31 it's only noon 17:18:36 (eastern) 17:18:38 Ok so my topic fits right in 17:18:53 Can we please delete the redundant mailing lists and rename fedora-atomic channel 17:19:17 sounds like this is going to require some debating 17:19:28 dustymabe: content in the docs ideas? Probably? 17:19:29 we can create a list to debate on the lists 17:19:29 sanja: want to make a ticket? 17:19:43 yes, more lists 17:19:44 I'm working on really streamlining our contribution process and making it all clear and reduce confusion 17:19:44 yeah I'd like a ticket. I think it will make the outcome smoother. 17:19:57 sanja: +1 17:20:07 +1 for ticket, so we document which is going, to avoid issue :) 17:20:21 #action sanja to create ticket on eliminating redundant mailing lists + irc channels 17:20:48 yeah. although we just recently did this: https://pagure.io/atomic-wg/issue/295 17:21:30 Good. Another thing is that misc and I discussed at length whether to move away from IRC. 17:21:31 Behold, this is the decision: we will stick with IRC unless there are objections. 17:21:32 I will create a tutorial on IRC and using it with Riot or having a bouncer so that newcomers to IRC have a smooth transition from Slack & Co which both misc and I decidedly didn't want 17:21:58 sanja: sounds good 17:22:15 sanja: that'd be awesome! 17:22:23 yeah. those of us who donate money to open source projects should consider matrix/riot as a candidate if we don't want to see slack take over the world 17:22:23 :o, nice! 17:22:37 Mattermost is yet another app and thing to use etc so we are doing it this way 17:23:00 #action sanja to create irc tutorial to onboard newcomers from slack, etc 17:23:12 sanja++ 17:23:25 sanja: you are signing up for a lot of work ;) 17:23:39 She's probably in 3 meetings right now too 17:23:42 :-D 17:23:44 So we're sticking with IRC and making it easier on newcomers, BUT I'd really appreciate a vote on the channel and ML now. Because it's the same as the ticket dustymabe just linked before 17:23:48 Hahaha 17:23:55 No I'm super dedicated and just in the bus right now 17:23:56 😂 17:24:05 +1 to IRC 17:24:07 the fedmsg bus ? 17:24:10 the *crazy bus* 17:24:17 Rofl 17:24:18 we're all in it, look around 17:24:24 we've got 7 minutes left in this meeting and i feel like the vote on channel/ML is going to take longer than that 17:24:37 The bus that brings me out of my cave to the tennis club 17:24:39 vote + debate 17:24:51 yeah. +1 for ticket 17:25:01 Ok next time then but no new ticket link Dusty's old one 17:25:08 but please consider the discussion that already happened before we add new judgements 17:25:11 -1 for MOAR tickets and lists 17:25:23 ok ok 17:25:52 Aren't you glad you got a deletist amongst you now? 17:25:55 i'm not going to try to re-action for that...i'll just let sanja sort it out 17:26:02 Lol 17:26:07 Can we clarify what we are voting on then in the old ticket before we start voting there? 17:26:09 so, we're going to reopen that ticket? 17:26:10 Good 17:26:18 prefer not to re-open it 17:26:21 Yes 17:26:21 Ok 17:26:25 Then fine I'll make a new one 17:26:28 :) 17:26:30 Buuut read old one 17:26:34 sorry. this happens when I get hangry 17:26:35 ok 17:26:43 So we don't re-enact the same discussions 17:26:47 cannot be avoided 17:26:47 !feed dustymabe 17:26:54 Hahaha 17:26:58 * misc is gonna open a ticket on the irc bot 17:27:22 #action misc to open ticket on the irc bot 17:27:24 now it is official 17:27:28 ha 17:27:54 are we good on those items? 17:27:54 Good meeting 17:27:55 ah ah 17:28:01 any more for open floor? 17:28:05 None here 17:28:06 * jbrooks is curious about the web site 17:28:15 I have a talk accepted for FAW next month 17:28:20 misc: nice 17:28:28 but I will wait for the date to tell it again in meeting 17:28:41 (also in french, so I could link anything and say "that's it") 17:28:48 Yep cool 17:28:49 real quick. there was a fedmag article that went out with a few videos from devconf 17:28:53 Hehe 17:28:54 jlebon's FAW talk made it in there 17:29:08 jbrooks http://new.projectatomic.io 17:29:09 #link https://fedoramagazine.org/3-fedora-videos-devconf-cz-2018/ 17:29:13 But it's only interim state 17:29:19 dustymabe: oh yeah, noticed that yesterday! 17:29:21 sanja, are we ditching the blog? 17:29:22 #info new Project Atomic site 17:29:24 was pleasantly surprised :) 17:29:27 #link http://new.projectatomic.io/ 17:29:32 Do not come now with all the change suggestions, wait for tomorrow I'm migrating and changing content 17:29:52 jlebon++ 17:29:52 rubao: Karma for jlebon changed to 4 (for the f27 release cycle): https://badges.fedoraproject.org/tags/cookie/any 17:30:05 Oh how cool @ jlebon talk 17:30:24 jbrooks I didn't intend to ditch the blog, should we? 17:30:51 I think it's pretty useful but actually most content goes to docs 17:31:01 sanja, the existing site code generates the blog posts and such -- is this just about new content, or a new everything 17:31:34 blog is good for hype....can link to the posts on social media, etc 17:32:04 Rehaul entirely. 17:32:16 we can talk about it in #atomic 17:32:45 Yep 17:32:47 yup 17:32:48 Thanks everyone for the meeting 17:32:54 alrighty...ending meeting in 60s 17:33:00 we set the fuse 17:33:06 and mission impossible music 17:33:28 thanks everyone! 17:33:30 This suspense. 17:33:47 #endmeeting