13:01:55 <mattdm> #startmeeting Council (2018-03-14) 13:01:55 <zodbot> Meeting started Wed Mar 14 13:01:55 2018 UTC. The chair is mattdm. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:01:55 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic. 13:01:55 <zodbot> The meeting name has been set to 'council_(2018-03-14)' 13:01:56 <bexelbie> mattdm, good to hear 13:01:57 <mattdm> #meetingname council 13:01:57 <zodbot> The meeting name has been set to 'council' 13:02:01 <bexelbie> .hello bex 13:02:02 <zodbot> bexelbie: bex 'Brian (bex) Exelbierd' <bex@pobox.com> 13:02:04 <dperpeet> .hello dperpeet 13:02:05 <mattdm> #chair mattdm jkurik jwb langdon robyduck bexelbie dperpeet Amita nb dgilmore pbrobinson 13:02:05 <zodbot> Current chairs: Amita bexelbie dgilmore dperpeet jkurik jwb langdon mattdm nb pbrobinson robyduck 13:02:05 <zodbot> dperpeet: dperpeet 'None' <dperpeet@redhat.com> 13:02:18 <alciregi> .hello 13:02:18 <zodbot> alciregi: (hello <an alias, 1 argument>) -- Alias for "hellomynameis $1". 13:02:23 <Amita> .hello amsharma 13:02:24 <zodbot> Amita: amsharma 'Amita Sharma' <amsharma@redhat.com> 13:02:31 <alciregi> .hello alciregi 13:02:32 <zodbot> alciregi: alciregi 'Alessio Ciregia' <alciregi@gmail.com> 13:02:45 <dgilmore> hey mattdm 13:02:51 <Amita> hello everyone 13:02:52 <jwb> mattdm, 1/3 here. double duty on a different meeting 13:02:54 <mattdm> #topic Introductions, Welcomes 13:03:29 <mattdm> So, this is our open floor meeting 13:03:39 <mattdm> We totally failed to find a time that works for everyone 13:03:49 <mattdm> so I suggest that as one topic of conversation 13:03:54 <langdon> .hello2 13:03:55 <zodbot> langdon: langdon 'Langdon White' <langdon@redhat.com> 13:04:24 <mattdm> topic #Today's Open Floor Agenda 13:04:44 <mattdm> Other than the meeting time, we'd talked about having a standing open floor topic of checkins from the objective leads 13:04:58 <mattdm> and if robyduck is here an update on mindshare would be nice 13:05:06 <mattdm> or bexelbie can do that otherwise 13:05:38 <mattdm> anyone have anything else? 13:05:56 <bexelbie> or mattdm can do that :P 13:06:21 <bexelbie> we need to move forward with buegeting - but no action during this meeting 13:06:24 <mattdm> yeah but you'll do a better job :) 13:06:32 <bexelbie> I'll be sending email soon I hope - working on last bit of the paperwork catchup 13:06:43 <bexelbie> but you're views are always critical mattdm :D 13:06:58 <mattdm> I do like to criticize! 13:07:48 <mattdm> okay, so, let's talk about the meeting time 13:08:12 <mattdm> #topic Meeting Time 13:08:22 <mattdm> here's the results code for the latest whenisgood 13:08:29 <mattdm> http://whenisgood.net/fedora-council-2018-take2/results/7peqrra 13:08:45 <mattdm> note that that's the admin link so everyone can edit responses so don't do that :) 13:09:25 <mattdm> The point is that there are *mostly* times when 7 or more of us can't make it 13:10:10 <mattdm> and the best times are ones when still three people can't make it 13:10:21 <bexelbie> mattdm, is that on GMT? 13:10:30 <bexelbie> there is a two person blocked time ... 13:10:32 <mattdm> bexelbie: it is on eastern time 13:10:34 <bexelbie> 3pm Wed 13:10:47 <bexelbie> ick .. /me hates that time but apparently I voted for it :P 13:10:49 <mattdm> yeah, so that's dperpeet and Amita who can't make that one 13:11:07 <Amita> 3PM GMT? 13:11:09 <mattdm> dperpeet and Amita -- is it *possible* to open that up? 13:11:15 <mattdm> Amita: no, sadly, 3PM US/Eastern 13:11:36 <dperpeet> mattdm, I could make it sometimes 13:11:45 <mattdm> We decided to follow US/Eastern because it makes the daylight saving time switch mildy less confusing 13:12:03 <mattdm> dperpeet: sometimes on a predictable basis? 13:12:28 <dperpeet> mattdm, >=50% probably 13:12:32 <bexelbie> mattdm, but you asked us to vote on GMT ... 13:12:43 <mattdm> bexelbie: the voting thing let you set your timezone 13:12:47 <mattdm> the results our in my timezone 13:13:03 <mattdm> that's how when is good works 13:13:05 <mattdm> see http://whenisgood.net/fedora-council-2018-take2 13:13:08 <Amita> so it will be 11:30 PM for me 13:13:08 <mattdm> there's a dropdown 13:13:16 <bexelbie> yes, but I thought we didn't move the meeting for timezone changes is my point 13:13:19 <mattdm> Amita: yeah that is pretty miserable 13:13:31 <mattdm> bexelbie: we move it along with the US timezone change 13:13:35 <dperpeet> mattdm, I could almost always make time when my input is planned via agenda 13:13:37 <mattdm> at least, we had been. 13:13:58 <mattdm> dperpeet: okay, yeah, and I want to plan that for you *anyway*, so okay, cool. 13:14:43 <mattdm> dperpeet: as objective lead, I think we'll *need* you for objective checkin meetings once a month, and want you around for input on other things at least one other time 13:14:57 <dperpeet> that sounds very doable 13:15:02 * bexelbie is going to have to find a better mobile IRC app as I am sure I'll be on a tram during this meeting sometimes :D 13:16:02 <mattdm> Amita: It's inevitable because of this whole "world is round" thing, but your times are all exceedingly early in the morning in the US :( 13:16:30 <Amita> mattdm, ok, np, I can also adjust 13:16:35 <Amita> and will join sometimes 13:16:50 <mattdm> One possibility is to have one meeting a month at an early time 13:16:54 <Amita> and if there is something specific for me, I can join in for sure 13:17:07 * dgilmore notes that he cannot join any earlier than this Monday to Wednesday 13:17:20 <dperpeet> mattdm, no need to do that on my account, shifting times are worse on my schedule than a known bad time :) 13:17:30 <dperpeet> but of course the time is worse for Amita 13:17:47 * dgilmore has to get a kid up, fed, and ready for school those days and is distracted by that 13:18:03 <mattdm> Amita: can you make late at night work once a month? 13:18:20 <Amita> mattdm, I can try for sure more than once 13:18:52 <dgilmore> mattdm: my schedule is currently in a state of flux, I will know what it will look like in about 3-4 weeks 13:19:03 <Amita> and please feel free to update me before if my presence is important for any particular day 13:19:03 <mattdm> okay. cool. let's try that then and see if it's going ok 13:19:23 <mattdm> Amita: yeah. I'll try to be better about advance agendas especially for the tickets meetings 13:19:34 <Amita> mattdm, cool, thanks 13:19:59 <mattdm> #info We're going to aim for 3pm US/Eastern (currently 19:00 GMT) on Wednesdays as a time that most people can make it 13:20:20 <mattdm> I did the GMT math right, right? 13:20:36 <Amita> oh wait 13:20:40 <Amita> that is 12:30 AM IST 13:20:44 <Amita> night 13:20:46 <Amita> WOW 13:20:48 <Amita> :( 13:21:26 <mattdm> Hmmmmmm. Ok, that's even worse. 13:21:51 <Amita> mattdm, is it not possible to do it earlier 13:21:57 <Amita> as it is afternoon for US 13:22:26 <mattdm> Amita: there's a slot 3 hours earlier whcih works for many people too 13:22:41 <mattdm> but robyduck is missing too and he's not here 13:22:45 <Amita> that will be great, if we can choose that one 13:23:06 <Amita> let's ask him over mail, if he agrees 13:23:13 <Amita> where is from? 13:23:15 <mattdm> What if we make one of the open floor meetings at that time, and the rest at the later time? 13:23:20 <mattdm> Amita: he is from Italy 13:23:38 <mattdm> So it is possibly during his commute home from work 13:24:00 <Amita> you want me to drop him a quick mail CC you 13:24:03 <Amita> about it? 13:24:19 <mattdm> Amita: I'll post to the council list. let's try the staggered time and see how it goes 13:24:33 <mattdm> The tickets and ongoing stuff, you can keep up with in the tickets 13:24:53 <mattdm> #info We need to do more work *not just at meeting time* in order for this to work 13:24:54 <Amita> Sure 13:25:20 <mattdm> #info Let's try also having one of the open floor meetings an earlier time to be more inclusive 13:25:32 <mattdm> I am not missign the irony that it's hard to include our inclusion person :-/ 13:25:48 <mattdm> okay, so anyway, that took half the meeting. let's move on. :) 13:25:56 <mattdm> #topic CI Objective Update 13:26:05 <mattdm> dperpeet: you still around? :) 13:26:09 <dperpeet> yup 13:26:20 <mattdm> Can you give a brief rundown of how this is going? 13:26:42 <dperpeet> overall the mindset change towards including tests is looking good 13:27:07 <dperpeet> packagers are adding tests and gating in Fedora appears to be accepted 13:27:17 <Amita> great 13:27:21 <dperpeet> testing non-Atomic packages is almost ready 13:27:36 <dperpeet> the pipeline for that is up in a testing environment, but not production ready yet 13:27:54 <dperpeet> this should work in the next two weeks 13:27:58 <mattdm> dperpeet: that's amazing. do we have numbers on this? 13:28:15 <mattdm> I'm looking at https://fedoraproject.org/wiki/Objectives/Continuous_Integration_and_Delivery_of_Fedora_Atomic_Host#Key_Results 13:28:25 <mattdm> It says "100 packages in Fedora Atomic host have at least one test stored in dist-git according to the spec." 13:28:33 <mattdm> I'd be curious to know what percent we are at :) 13:28:46 <mattdm> #info packagers are adding tests and gating in Fedora appears to be accepted 13:28:59 <mattdm> #info testing non-Atomic packages is almost ready -- should work in next two weeks 13:29:04 <dperpeet> well, our statistics page seems to be broken right now :) 13:29:09 <dperpeet> #link https://fedoraproject.org/wiki/CI/Tests/stat 13:29:18 <mattdm> oops :) 13:29:52 * bexelbie wonders if we need a test to gate the statistics 13:29:52 <mattdm> if it *were* working, what would it say? 13:29:55 * bexelbie ducks 13:30:07 <dperpeet> bexelbie, working on it :) 13:30:16 <dperpeet> I think the numbers were between 20 and 30% 13:30:22 <dperpeet> for Atomic Host packages 13:30:39 <mattdm> ok, so not *quite* a hundred yet :) 13:30:43 <dperpeet> yeah 13:30:51 * langdon got very distracted.. apologies.. reading scrollback 13:30:57 <dperpeet> and we're working on the compose-level tests also 13:30:59 <dperpeet> right now 13:31:11 <mattdm> Do you think that is likely to ramp up quickly? Do we need to do a publicity campaign or something? 13:31:12 <dperpeet> so the ones that come ones the packages have been put together 13:31:36 <dperpeet> #link https://fedoramagazine.org/continuous-integration-fedora/ 13:31:45 <dperpeet> we've published an article on the tests very recently 13:32:04 <dperpeet> and updated the wiki 13:32:08 <dperpeet> #link https://fedoraproject.org/wiki/CI 13:32:11 <dgilmore> dperpeet: in your up in two weeks are you accounting for the fact that infrastructure is frozen until Beta is released? 13:32:17 <mattdm> dperpeet: okay, cool. 13:32:21 <dperpeet> moving some of this into fedora docs was moved to a potential follow up 13:32:34 <bexelbie> +1 to docs :) 13:32:37 <dperpeet> dgilmore, getting tests in place doesn't require infrastructure changes 13:32:42 <dgilmore> +100 to docs 13:32:45 <dperpeet> the pipeline puts results into pagure 13:32:51 <dperpeet> and onto the message bus 13:32:54 * mattdm notes that the magazine does not seem to make the date of publication obvious 13:33:04 <dgilmore> dperpeet: okay, so how do the tests get run then? 13:33:11 * dgilmore can look at that offline 13:33:12 <dperpeet> that said, there should be an unfreeze in ~2 weeks 13:33:21 * mattdm afk for two minutes. continue without me please 13:34:04 <dperpeet> not sure how much I should rathole here, but pingou and the others working on this assure me that everything can be ready by the time the freeze is lifted in ~2 weeks :) 13:34:15 <dgilmore> dperpeet: I will take it offline 13:34:40 <dperpeet> thanks, happy to answer more detailed questions, maybe in #fedora-ci 13:34:41 <dgilmore> I assumed things will need to get deployed and I was just wanting to make sure that things were all taken into account 13:34:49 <pingou> dgilmore: they get run in a pipeline that's not frozen basically 13:36:13 <dperpeet> next on the roadmap is then pull request testing 13:36:13 <dgilmore> dperpeet: how have the AtomicHost test been going? 13:36:20 * mattdm is back 13:36:47 <dperpeet> dgilmore, the higher level ones? people started looking into them a few days ago 13:36:55 <dperpeet> I have to admit I haven't checked on the details since then 13:37:20 <dgilmore> dperpeet: no problem, was just curious how effective they were being found 13:37:22 <dperpeet> we should have something working in 2-3 weeks I'd say 13:37:40 <mattdm> dperpeet: What exactly is broken on the status page? 13:37:41 * dgilmore would like to see test set up for the anaconda runtime as a priority 13:37:55 <dperpeet> mattdm, wrong number of packages, and not just Atomic anymore 13:38:03 <mattdm> ah okay. 13:38:05 * dgilmore also does not know why he fails to type tests today 13:38:24 <mattdm> when you get that updated, can you ping me? I'm going to link it from https://docs.fedoraproject.org/fedora-project/project/objectives.html 13:38:29 <dperpeet> dgilmore, noted - due to PTO of the people involved, I expect more progress next week 13:38:36 <dperpeet> mattdm, will do, thank you 13:38:43 <dgilmore> dperpeet: awesome 13:38:46 <mattdm> dperpeet: also, do we have any badges related to CI? 13:38:55 <dperpeet> mattdm, I don't think we do 13:39:01 <pingou> not yet 13:39:03 <dperpeet> but we have testing related ones 13:39:15 <mattdm> dperpeet: let's make some! 13:39:47 <mattdm> You probably know what the specific things we want to emphasize/reward are, and what they might trigger on 13:40:14 <dperpeet> I can help brainstorm a few badge titles / descriptions 13:40:23 <mattdm> here's the link for suggesting new badges: https://pagure.io/Fedora-Badges 13:40:41 <mattdm> dperpeet: awesome. I think that'll help the publicity a little bit, and also will give me a graphic to put in the docs :) 13:40:51 <dperpeet> thank you, sounds good 13:40:55 <mattdm> dperpeet: anything else critical to mention? 13:41:05 <mattdm> if not, let's move on to modularity if langdon is still here.... 13:41:05 <dperpeet> we have some discussions around metadata and where to put tests 13:41:12 <dperpeet> but I think that's ratholing 13:41:29 <dperpeet> #link https://fedoraproject.org/wiki/Flexible_Metadata_Format 13:41:30 <mattdm> yeah *I* don't care :) 13:41:32 <dperpeet> for who's interested 13:41:34 <langdon> im here 13:41:35 <dperpeet> done :) 13:41:41 <mattdm> cool. thanks dperpeet! 13:41:55 <mattdm> #topic Modularity objective update 13:42:10 <langdon> ohh good gracious 13:42:21 <mattdm> well that's an interesting start :) 13:42:37 <langdon> well.. normally i type a lot off to the side and c/p :) 13:42:57 <dgilmore> langdon: you re on the spot? 13:43:24 <mattdm> So I see that modular content *is* landing at 13:43:25 <langdon> so.. modularity is having an awesome time.. first and foremost.. we don't have support for modules in any package manager "not dnf" .. which is problematic if the module is carrying nvrs greater than the everything repo 13:43:26 <mattdm> https://dl.fedoraproject.org/pub/fedora/linux/development/28/Modular/x86_64/os/Packages/ 13:43:36 <langdon> howeveer, we are getting modular content 13:43:52 <langdon> and if you install fedora-repos-modular.. you get modules 13:44:09 <mattdm> dgilmore: is that subpackage subpackaged in the latest nightly? 13:44:17 <langdon> if you want to try it out easily docker run --rm -it langdon/addon-modular-boltron /bin/bash being built daily 13:44:30 <langdon> mattdm: last i knew it was in updates-testing for 28 13:44:37 <nb> hello 13:44:43 <mattdm> langdon: is that different from if I have a system running *regular* nightly? 13:44:48 <mattdm> oh hi nb! 13:44:50 <langdon> the docker container is actually built from a playbook you can look at 13:44:59 <mattdm> you may want to scroll back to the schedule discussion 13:45:09 <mattdm> langdon: that does not answer my question :) 13:45:22 <dgilmore> mattdm: should be yes 13:45:27 <langdon> mattdm: regular nightly of 28? no, i wouldn't think so.. or look at my (really sgallagh's) playbook 13:45:48 <langdon> for me.. i don't have enough detail on what "regular nightly" means to know for sure 13:46:17 <mattdm> langdon: a build of F28 Server that is basically what the release would look like if we released today 13:46:24 <bexelbie> to make sure I am separating humor from concern, are we supposed to have support for modules in dnf ... I thought that was the plan 13:46:44 <langdon> bexelbie: the problem is all the OTHER package managers.. e.g package kit 13:46:49 <sgallagh> mattdm: Yes, it was pushed to the stable repo yesterday and should be in the next successful compose 13:46:54 <mattdm> bexelbie: The problem is that we *only* have support in DNF. The plan was to have support in *libdnf* so packagekit and microdnf would also pick it up 13:46:57 <mattdm> but that did not happen 13:47:01 <sgallagh> Last night's broke for other reasons 13:47:18 * bexelbie isn't going to ask about why we have so many package managers :D 13:47:23 <bexelbie> do we have a plan for correcting that? 13:47:26 <mattdm> sgallagh: last night's compose or last night's modularity something? 13:47:31 <bexelbie> or are we moving forward with a plan B? 13:47:31 <sgallagh> compose 13:47:53 <mattdm> I think the current plan is to only enable modularity for the Server release, where DNF is the standard package manager anyway 13:47:54 <sgallagh> bexelbie: Workaround right now is that we won't ship the modular repos anywhere but Server Edition 13:48:01 <mattdm> that is, not Workstation or Atomic 13:48:09 <langdon> right 13:48:10 <sgallagh> But longer-term, we're going to fix that in libdnf. We don't have a committed timetable on that yet 13:48:11 <bexelbie> sgallagh, ahh 13:48:24 <mattdm> However, this is problematic for container images, where microdnf is supposed to shine 13:48:31 <bexelbie> I can see that 13:48:49 <bexelbie> I was also thinking it is a shame to not have our new hotness in the other newish thing too :D 13:49:00 <langdon> microdnf install dnf 13:49:28 * mattdm still thinks dnf should *be* microdnf and it should automatically install full dnf if you make any calls it doesn't understand 13:49:29 <langdon> snark aside .. FROM fedora-minimal \n RUN microdnf install dnf is still a TON smaller than FROM fedora 13:49:57 <mattdm> okay, so, I guess we can live with that 13:50:13 <mattdm> #info For F28, Modular repo will only be enabled by default for Server 13:50:32 <mattdm> #info You can opt-in by installing the repo package on other releases, but beware issues with packagekit and microdnf 13:50:45 <mattdm> Other than that, are we basically on track? 13:51:16 <mattdm> dgilmore: do you know what's up with the isolinux/iso directories in the Modular tree? 13:51:20 <langdon> modules themselves are kinda behind.. we have "some" but not a lot.. 13:51:30 <langdon> the content can land post f28 ga though 13:51:54 <langdon> and we need to fixx issues with the ~40 we had before 27 to make them use the new arch 13:52:11 <mattdm> langdon: is there a burndown-style tracker for that? 13:52:21 <langdon> mattdm: not that i am aware of 13:52:26 <mattdm> langdon: can you make one? :) 13:52:38 <langdon> yeah.. 13:52:42 <mattdm> thanks :) 13:52:56 <dgilmore> mattdm: yes, we got apatch in pungi for it yesterday 13:53:10 <dgilmore> mattdm: we need to get a new pungi and make a config change to turn it off 13:53:24 <mattdm> dgilmore: okay, cool. because while it techically doesn't hurt beyond wasting space, I can see it really confusing some people who go browsing around the tree 13:53:34 <mattdm> "hey, this modular install doesn't work" 13:53:36 <dgilmore> mattdm: indeed 13:53:39 <bexelbie> you and your url "hacking" mattdm ... 13:53:40 <bexelbie> :P 13:54:06 <mattdm> yes, clicking that "Parent directory" link. Such a hacker! 13:54:12 <mattdm> okay, five minutes left.... 13:54:16 <mattdm> #topic Mindshare FAD 13:54:39 <mattdm> robyduck isn't here... but nb and bexelbie and langdon and I were all also at that FAD 13:54:46 * x3mboy is lurking 13:55:01 <mattdm> rather than summarizing things here, bexelbie, there's an commblog article coming soon? is that right? 13:55:07 <bexelbie> yes 13:55:17 <bexelbie> the draft is due by Friday for review by the group (x3mboy is writing it) 13:55:22 <bexelbie> and we plan to publish next week 13:55:30 <mattdm> #info Stay tuned for Mindshare update next week 13:55:34 <bexelbie> It, like the Docs FAD, was super productive 13:55:37 <bexelbie> that's my teaser 13:55:41 <mattdm> cool :) 13:56:00 <pingou> spoiler! 13:56:00 <mattdm> okay, peter doesn't seem to have shown up, so we'll talk about IoT next time 13:56:01 <x3mboy> The draft will be ready tomorrow night in my TZ 13:56:07 <mattdm> x3mboy++ 13:56:07 <zodbot> mattdm: Karma for x3mboy changed to 21 (for the f27 release cycle): https://badges.fedoraproject.org/tags/cookie/any 13:56:11 <bexelbie> x3mboy++ 13:56:31 <mattdm> okay, so, I'm going to go update the meeting times 13:56:43 <langdon> mattdm: whats the finall answer? 13:56:47 <langdon> im confused... 13:56:55 <mattdm> langdon: for meeting times? 13:56:57 <langdon> right 13:57:03 <langdon> 3pm eastern weds? 13:57:10 <langdon> 12? 13:57:17 <mattdm> 3pm est wed three weeks, with one week at 12 13:57:36 <mattdm> we'll see how insane that is :) 13:57:43 <langdon> ok 13:58:07 <mattdm> thanks everyone 13:58:09 <mattdm> #endmeeting