16:31:14 #startmeeting fedora_atomic_wg 16:31:14 Meeting started Wed Dec 20 16:31:14 2017 UTC. The chair is dustymabe. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:31:14 Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:31:14 The meeting name has been set to 'fedora_atomic_wg' 16:31:17 #topic roll call 16:31:25 .hello dustymabe 16:31:25 .hello jberkus 16:31:26 dustymabe: dustymabe 'Dusty Mabe' 16:31:26 .hello jlebon 16:31:28 jberkus: jberkus 'Josh Berkus' 16:31:32 jlebon: jlebon 'None' 16:31:38 sort of hi. 16:31:46 .hello sinnykumari 16:31:47 ksinny: sinnykumari 'Sinny Kumari' 16:31:52 .hello jasonbrooks 16:31:53 jbrooks: jasonbrooks 'Jason Brooks' 16:31:59 puiterwijk: glad to have you here! 16:32:03 hi ksinny! 16:32:11 .hello Esgee 16:32:12 dustymabe: Hey! 16:32:15 dustymabe: well, only partly. Most of my concentration is still in the thousands of lines of curl code 16:32:16 EsGee: Sorry, but you don't exist 16:32:39 puiterwijk: fun 16:32:45 .hello2 16:32:46 strigazi: strigazi 'Spyros Trigazis' 16:33:01 strigazi: woor! 16:33:03 .hello2 16:33:04 woot! 16:33:04 miabbott: miabbott 'Micah Abbott' 16:33:23 #chair jberkus jlebon puiterwijk ksinny jbrooks EsGee strigazi miabbott 16:33:23 Current chairs: EsGee dustymabe jberkus jbrooks jlebon ksinny miabbott puiterwijk strigazi 16:33:39 anybody seen sayan lately? 16:34:23 ok let's get started 16:34:24 Not me 16:34:28 EsGee: that means you're not registered with fedoraproject.org, that's all 16:34:41 #topic previous meeting action items 16:34:57 * maxamillion to confirm and close #376 16:34:59 * jberkus to post a revised quorum proposal 16:35:01 * cverna to talk to releng about doing a container release since we just 16:35:03 released FAH 16:35:05 * jberkus to review/merge the PR to update official docs from f25 to f27 16:35:09 .hello markllama 16:35:10 markllama: markllama 'Mark Lamourine' 16:35:38 #info maxamillion confirmed and closed #376 16:36:33 looks like cverna is not here but I can update that 16:36:35 you wanna take these in order? 16:36:39 .hello sayanchowdhury 16:36:40 sayan: sayanchowdhury 'Sayan Chowdhury' 16:36:42 sayan: yay! 16:36:53 jberkus: sure 16:37:00 you can update on 2nd item 16:37:10 lemme find the issue 16:37:16 there's a revised proposal ... 16:38:33 dustymabe: not keeping well lately :( this week has been better 16:38:49 https://pagure.io/atomic-wg/issue/363 16:38:58 sayan: that's unfortunate. I'm glad you're here though. wanted to discuss AMIs/fedimg later in meeting 16:39:06 dustymabe: yes, sure 16:39:07 looks like we still need to set rules for how many people have to vote in pagure 16:39:25 jberkus: ok. 16:39:28 I'll put a proposal on there 16:39:35 we can discuss that in more depth when we are in tickets 16:39:40 since this one has the meeting tag? 16:40:29 * dustymabe thinks jberkus is writing the proposal right now 16:40:32 so moving to next AI 16:40:40 * cverna to talk to releng about doing a container release since we just released FAH 16:40:58 #info dustymabe opened ticket to request new container release https://pagure.io/releng/issue/7217 16:41:06 I talked to mboddu maxamillion and puiterwijk 16:41:38 mainly this is waiting on automation to be moved to prod, which according to puiterwijk should be pretty easy to do and EOD yesterday was the target 16:41:42 Oh, yeah, I had intended to finish setting everything up yesterday, I need to fix one more thing and then hand it over to Mohan 16:41:45 I imagine that got held up 16:41:58 puiterwijk: do we think we can do that today? 16:41:59 Yeah, sorry. Curl really got me annoyed 16:42:03 dustymabe: yes. 16:42:13 puiterwijk: yeah curl is a rat hole for sure 16:42:16 :( 16:42:27 #info hoping to have a new container release later today! 16:42:40 ok last action item from last meeting for jberkus 16:42:41 HTTP/2 is a really nice protocol. But if you ignore the server saying "Please reconnect" and instead error out... weird things happen 16:42:49 * jberkus to review/merge the PR to update official docs from f25 to f27 16:43:02 merged 16:43:16 #info jberkus merged PR to update official docs from f25 to f27 16:43:22 so if there's still f25 stuff in the site, it needs updating 16:43:35 jberkus: does that mean our docs are updated and shouldn't have old content? 16:44:12 also, a big thank you to contributor Fabaff for doing that! 16:44:31 dustymabe: that means that if you see docs that have f25 or older stuff, you need to add a new PR 16:44:54 that's Fabian Affolter 16:44:58 jberkus: nice 16:45:09 does he hang out in the channel much / 16:45:15 I love me some drive-by contributors 16:45:25 dustymabe: we actually get lots of one-off contribs to the web site 16:45:48 cool good to know.. EsGee if you find anything during your exploration that needs updating please let us know 16:45:53 * cverna waves 16:45:56 ok moving on to tickets 16:45:59 hiya cverna :) 16:46:16 #topic Using buildah in Atomic Host 16:46:23 #link https://pagure.io/atomic-wg/issue/402 16:46:44 dustymabe: o/ 16:47:03 this is a topic that came up recently. Basically trying to find out if there is a use case for having buildah inside the base ostree 16:47:20 please weigh in on the ticket if you have input 16:47:43 we can also have discussion here if anyone is inclined 16:48:49 ok moving to next ticket 16:49:10 #topic Overhaul list of members, Quorum Rules 16:49:15 #link https://pagure.io/atomic-wg/issue/363 16:49:25 see last comment for proposal 16:50:03 proposal in https://pagure.io/atomic-wg/issue/363#comment-485513 16:51:08 jberkus: question 16:51:24 mainly about number of votes 16:51:30 At least 5 people must vote, or 51% of the WG membership, whichever is less. 16:51:55 so if the WG membership is 6 members then the number is 4 16:52:13 but if WG membership is 12 it is 5 ? 16:52:41 correct 16:52:44 ok 16:52:57 sounds good 16:53:07 yah, now we just need membership rules 16:53:10 only other question is Ad-Hoc votes must be open for at least three working days (see below) after the announcement. 16:53:36 my question there is about time sensitive issues (if we ever encountered that situation) 16:54:00 so for example in previous iterations you mentioned ad-hoc IRC votes 16:54:22 which to me typically denotes we're in iRC and something is most likely time sensitve that we're discussing 16:54:51 I thought about that, but I couldn't come up with a likely time-sensitive situation 16:54:55 my suggestion was that we create an issue in pagure for those instances and still have people vote in the ticket (transparency) 16:55:14 but I don't know about blocking for 3 days 16:55:59 What would be an example of a time sensitive item 16:56:52 jbrooks: we are scheduled to release tomorrow, but CVE X came out. CVE X is medium severity. Should we release tomorrow or push release til fix is in place 16:57:07 hmmm, point 16:57:13 I've mostly just been using my judgement on these things 16:57:25 which is fine and I don't know if I want to bog down all of this with 'process' 16:57:38 but someone may demand more transparency on that in the future 16:57:45 which would require tickets/voting etc.. 16:58:04 There should be room for discretion 16:58:05 so there is my contrived time sensitive example 16:58:12 jbrooks: right 16:58:25 dustymabe: ok, sounds like more discussion is required on the ticket 16:58:36 do other WGs have quorum rules? 16:58:45 jberkus: i think we can just add a point that we can override the 3 day rule for issues we deem time sensitive 16:59:01 but it all must still be done in a ticket so it's transparent 16:59:16 jberkus: not sure 16:59:39 anyway, I'd like to close this with "more discussion required in the ticket" 16:59:45 sure 16:59:51 but we are pretty close I think 16:59:58 thanks for working at this jberkus 17:00:04 moving on to next ticket 17:00:14 #topic Decide strategy for including container runtimes in Atomic Host 17:00:21 #link https://pagure.io/atomic-wg/issue/360 17:00:57 * dustymabe is expecting that jberkus will discuss survey results 17:01:22 survey results link: https://docs.google.com/document/d/1aC8uthOMrTLiYK3l5muUVKDnOO7Rich2JxJ0sd4LJ6o/edit?usp=sharing 17:01:46 #info survey results link: https://docs.google.com/document/d/1aC8uthOMrTLiYK3l5muUVKDnOO7Rich2JxJ0sd4LJ6o/edit?usp=sharing 17:02:07 so, questions/comments about the results? 17:02:40 jberkus: it seemed like there were a large number of people who use the builtin docker and who like it baked in 17:02:47 but there were a fair number that were like 17:02:52 please let me use anything other than docker 17:02:56 yah 17:03:04 I was hoping that the survey would be more conclusive 17:03:14 but you could use those results to argue either way 17:03:53 yeah. I have my own opinions but neglected to participate in the survey 17:04:00 as requested 17:04:12 good, because I would have had to scrub your responses (as I did several others) 17:04:29 I'm not confident about our container image delivery right now 17:04:40 jbrooks: for system containers? 17:04:49 For any fedora container images 17:05:07 jbrooks: here's another question: if we have alternate container runtimes in a system container, how difficult is it to install them "over" the built-in one? 17:05:10 jbrooks: yeah I hear your concerns on that 17:05:28 puiterwijk: and mboddu should be making that better today 17:05:28 jberkus, It's easy 17:05:34 based on maxamillions automation work 17:06:14 and then there is work to have bodhi be able to promote container content as well 17:06:45 We can use override to allow for layering 17:07:01 It's more complicated, but still doable 17:07:27 yes. i'm hoping we can make it easier in the future too 17:07:37 i need to talk to jlebon about possibilities there 17:07:38 I don't think the benefits of changing from the status quo outweigh the negatives 17:07:53 jbrooks: I agree 17:08:04 my thoughts are 17:08:14 1 - leave docker in (for now anyways) 17:08:28 2 - make cri-o easily installable/usable 17:08:39 3 - make newer versions of docker easily installable/usable 17:08:46 I'm OK with that 17:09:08 4 - possibly include cri-o in base in the future if we want (and hopefully we get the size down a bit) 17:09:49 I would love us to start on creating an alternate OStree "minimal" image for the IoT userbase, and strip docker out of it, and also a lot of other stuff 17:09:58 but that's a bigger project 17:10:02 It'd be cool if we could have another ref, yeah 17:10:09 jberkus: you should set up some time to talk to pbrobinson 17:10:27 he's been talking about that for a while, but haven't seen him chat in #atomic about it much 17:10:30 dustymabe: yah, the problem is that everyone thinks this is a good idea, but nobody has time to work on it 17:10:30 would love to hear his plans 17:11:15 jberkus: do you mind making a ticket in our pagure for this? 17:11:18 it will start the discussion 17:11:21 sure 17:11:36 so, if we're not removing docker, do we need to take a vote? 17:11:37 another thing. we should start announcing some tickets we create on the atomic-devel list 17:12:00 i.e. I just created this ticket which we'd like input on. please discuss there 17:12:08 since those notifs don't go to that list 17:12:31 jberkus: i think we should put forth a proposal in the ticket 17:12:37 ok 17:12:53 .hello2 17:12:54 sanja: sanja 'Sanja Bonic' 17:13:02 hey, made it! 17:13:05 i don't know if we need a vote persay, but we should say what we're thinking and allow others to input 17:13:26 Sorry guys, only now managed to collect the remaining water from my flat. 17:13:42 sanja: we're just concluding that we don't want to remove docker from the base image in f27-28 cycle 17:13:43 jberkus: it might be good to follow up reply to your email about the survey as well with the results link and a link to this ticket and the proposal we came up with 17:13:54 water pipe broke while I had to rush for my social security number appointment so sorry for that, back to business 17:14:03 fun 17:14:20 kk I'll have to read up how we came to that conclusion 17:14:22 * dustymabe imagines sanja with a bucket 17:14:28 lol 17:14:45 jberkus: what do you think of what I wrote? 17:15:06 * sanja definitely didn't have that on her bucket list today. 17:15:25 #action jberkus to follow up reply to atomic-devel email about survey with results link and a link to pagure ticket with the proposal we came up with 17:15:28 ^^ 17:15:50 dustymabe: wrote where? 17:16:07 jberkus: what's the q? 17:16:20 * jberkus is confused 17:16:26 you sent an email to atomic-devel asking people to fill out the survey, correct? 17:16:32 sanja: sounds like you needed a "bucket" list 17:16:42 dustymabe: right. and with the results 17:16:47 ahh, you did? 17:16:54 yeah, check the list 17:16:59 indeed. I missed that 17:17:02 #undo 17:17:02 Removing item from minutes: ACTION by dustymabe at 17:15:25 : jberkus to follow up reply to atomic-devel email about survey with results link and a link to pagure ticket with the proposal we came up with 17:17:09 ok new action 17:17:11 Yeah, it was a good post, jberkus 17:17:14 well, we still need to make a proposal 17:17:25 I liked the cah love ;) 17:17:36 and apropos of nothing, I love Jupyter Notebook 17:17:41 #action jberkus to add proposal to pagure ticket and link atomic-devel mail thread to proposal for potential discussion 17:17:52 jberkus: better? 17:17:53 jbrooks: yeah, that survey really shows that just because CAH users are invisible doesn't mean they don't exist 17:18:05 typical centos users 17:18:17 jberkus: i'm sure there are a lot of people who participate in this community who primarily use CAH 17:18:35 they know alot of what we do flows down 17:18:54 and I'm happy to have them around 17:19:30 ok more topics 17:19:47 #topic open floor 17:19:55 hey sayan - i'm hoping to get an update on fedimg 17:20:17 would like to be able to upload to new AWS regions sooon! 17:20:30 dustymabe: hey 17:21:01 dustymabe: I am working on the comments by pingou on the PR 17:21:15 before I can merge the PR 17:21:38 * sayan digs the link to the PR 17:22:05 #link https://github.com/fedora-infra/fedimg/pull/62#discussion_r156349603 17:22:11 sayan: thanks 17:22:16 do you know an ETA for this work? 17:23:09 dustymabe: this would be done by tomorrow, and then ping pingou for the final review of the PR 17:23:20 ok thaks 17:23:21 thanks 17:23:26 anyone else with anything for open floor 17:23:57 sanja: jlebon: am i correct in my understanding that you *both* have talks accepted at devconf to talk about atomic workstation? 17:24:52 dustymabe: that is correct 17:25:02 * jlebon checks schedule 17:25:33 that's cool that we got two talks accepted on that topic. 17:25:37 Yes, Devconf and FOSDEM for me. 17:25:42 About Atomic Workstation. 17:25:46 I wonder if we shouldn't have you two collaborate on the talk 17:25:49 so the Atomic BOF and Retrospective was accepted 17:26:08 I proposed this, but by devconf sanja will be MC'ing it 17:26:27 I think he meant jlebon and me? 17:26:54 right. jlebon got a talk accepted on FAW and sanja did too 17:27:07 to add to the confusion, there's also "Atomic Workstation" by klember 17:27:13 sanja: FAW == Fedora Atomic Workstation 17:27:28 yep, figured that out :D thanks 17:27:30 whoa 3 atomic workstation talks? 17:27:36 for the BOF, we need various folks to do 2-minute "state of" presentations for various atomic projects 17:27:41 i'm surprised they were all accepted! 17:27:41 sanja: :) 17:27:46 well, looks like that Atomic Workstation is big in 2018 :P 17:27:48 sanja: what's your talk called? 17:28:00 jlebon: voting for devconf was a bit total chaos this year 17:28:05 "You want a clean desktop OS? Containerize it." 17:28:21 jlebon: the voting system broke down and the committee had to finish with spreadsheets 17:28:31 sanja: as it should be! 17:28:32 sounds great...almost as good as my water pipe 17:28:42 oh wow, and there's also "Desktop OS of the future: ostree + flatpak" 17:28:43 sanja: strikingly similar 17:28:45 should we reach out to klember about his talk? 17:28:50 does anyone know who that is ? 17:28:55 i think that's from the endless folks 17:29:14 dustymabe: i reached out during the voting process, but hadn't heard back 17:29:44 hmm. he must be from the workstation side 17:29:46 puiterwijk: may know 17:29:50 ok we are mostly at time 17:29:54 Hmm? 17:30:03 puiterwijk: klember submitted a talk on FAW 17:30:04 it's Kalev Lember, he works on Gnome stuff 17:30:05 anyway, for the BOF I need folks to participate, look for an email from me & sanja 17:30:09 ahh ok 17:30:15 jberkus: you know I'll be there 17:30:22 i had one more item for open floor 17:30:59 ksinny submitted a patch to cloud-utils rpm in fedora and got it accepted 17:31:09 it was a huge patch to move to newer version of cloud-utils 17:31:20 the maintainer reviewed it and they got some fixups in and now it's merged 17:31:25 great job ksinny !!! 17:31:27 ksinny++ 17:31:32 cloud-utils-0.30 build is alaso available in F27 and rawhide 17:31:38 dustymabe: Thanks! 17:32:10 ksinny: sweet. did we get an update submitted for it? 17:32:31 ksinny: do you mind updating the atomic wg ticket we have ? 17:32:43 dustymabe: Last I checked it wasn't. Maybe I will ping mainatiner tomorrow if it is still not done 17:32:51 dustymabe: sure, will do 17:33:04 ok thanks everyone for coming today! 17:33:10 great meeting right before the holiday 17:33:17 I assume we won't be meeting next week 17:33:36 I'll send an email letting people know 17:33:46 #info no meeting next week due to holidays 17:33:53 #endmeeting