20:00:14 <sgallagh> #startmeeting Server SIG Weekly Meeting (2017-08-15)
20:00:14 <sgallagh> #meetingname serversig
20:00:14 <sgallagh> #chair nirik sgallagh mhayden dperpeet smooge jds2001 vvaldez adamw mjwolf
20:00:14 <sgallagh> #topic Roll Call
20:00:15 <zodbot> Meeting started Tue Aug 15 20:00:14 2017 UTC.  The chair is sgallagh. Information about MeetBot at http://wiki.debian.org/MeetBot.
20:00:15 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
20:00:15 <zodbot> The meeting name has been set to 'server_sig_weekly_meeting_(2017-08-15)'
20:00:15 <zodbot> The meeting name has been set to 'serversig'
20:00:15 <zodbot> Current chairs: adamw dperpeet jds2001 mhayden mjwolf nirik sgallagh smooge vvaldez
20:00:22 <sgallagh> .hello sgallagh
20:00:24 <zodbot> sgallagh: sgallagh 'Stephen Gallagher' <sgallagh@redhat.com>
20:00:24 <mhayden> .hello mhayden
20:00:25 <adamw> .hello adamwill
20:00:26 <zodbot> mhayden: mhayden 'Major Hayden' <major@mhtx.net>
20:00:26 <nirik> .hello kevin
20:00:29 <zodbot> adamw: adamwill 'Adam Williamson' <awilliam@redhat.com>
20:00:32 <adamw> but unless someone pings me specifically i'm not likely to be watching, sorry
20:00:32 <zodbot> nirik: kevin 'Kevin Fenzi' <kevin@scrye.com>
20:00:35 <adamw> too much stuff exploding
20:00:42 <sgallagh> adamw: Uh oh
20:02:13 <mjwolf> .hello mjwolf
20:02:13 <zodbot> mjwolf: mjwolf 'Michael Wolf' <mjwolf@us.ibm.com>
20:02:28 <langdon> .hello2
20:02:29 <zodbot> langdon: langdon 'Langdon White' <langdon@redhat.com>
20:02:38 * langdon was looking for the mtg in #f-m
20:03:16 <sgallagh> Wait, when did .hello2 become a thing?
20:03:43 <sgallagh> anyway...
20:03:49 <sgallagh> #topic Agenda
20:04:09 <sgallagh> We have one proposed item for the agenda today: langdon wants to discuss Modularity plans for Fedora 27
20:04:10 <langdon> sgallagh, don't know.. but it works..
20:04:17 * linuxmodder here
20:04:19 <langdon> yeah.. such as it is..
20:04:30 <sgallagh> I'm sure this will be a very short and precise meeting /s
20:05:07 <sgallagh> Do we have other pressing topics to discuss today?
20:05:33 <sgallagh> Alright, let's get started.
20:05:43 <sgallagh> #topic Magical Modularity Tour
20:05:47 <langdon> ha
20:05:48 <sgallagh> langdon: ?
20:06:01 <langdon> ok.. so basically, we just wanted to confirm: https://github.com/fedora-modularity/f27-content-tracking
20:06:04 <langdon> with y'all
20:06:31 <langdon> i also realized it would be good to track all the other work going on as issues on that repo
20:06:57 <langdon> but.. i only just thought of that
20:07:53 <langdon> as sgallagh knows.. there is also significantly more in the platform module than there used to be.. so some of your items may be covered there
20:07:56 <langdon> e.g. less
20:08:04 <langdon> comments, thoughts?
20:08:15 * nirik looks at link
20:08:19 <sgallagh> You're saying that less is more?
20:08:21 <sgallagh> ;-)
20:08:48 <langdon> exactly
20:08:52 <langdon> and pine is not elm
20:08:53 <smooge> no none is more
20:09:31 <sgallagh> Kidding aside, I think the section on Server WG requirements looks accurate from a "let's not regress" perspective
20:09:43 <nirik> yeah
20:09:48 <adamw> list looks OK to me, at least as regards critical content.
20:10:19 <langdon> just to be clear.. you also don't have to be 100% right now.. just this is what we are working towards..
20:10:33 <langdon> and we would love help.. but, we can't really tell you how to help til next week :/
20:12:14 <sgallagh> langdon: For context, you are referring to the document you just submitted to FESCo?
20:12:48 <langdon> sgallagh, correct.. and needing adam's (poorly named IMO) dependency report dealio to be done
20:13:34 <sgallagh> For the minutes:
20:13:35 <langdon> sgallagh, unless you know how people could help now? or have ideas how they could?
20:13:47 <sgallagh> #link https://github.com/fedora-modularity/f27-content-tracking
20:13:48 <sgallagh> #link https://fedoraproject.org/wiki/Module:Review_Process
20:14:30 <sgallagh> langdon: The set of people currently serving on the Server SIG would be a good set of people to review those docs
20:14:45 <sgallagh> #link https://github.com/fedora-modularity/dependency-report
20:15:09 <nirik> s/SIG/WG/ but thats nitpicking. ;)
20:15:15 <langdon> ahh yes.. please do..
20:16:01 <langdon> file issues that you can't just "fix" here: https://pagure.io/modularity/issues
20:16:22 <langdon> wiki isn't very conducive to having issues/comments (imo)
20:16:25 <sgallagh> With my FESCo cap on, anyone who could provide feedback on the Module Review Process this week before FESCo discusses it would be greatly appreciated
20:17:41 <langdon> it is basically a clone of the container one (thanks maxamillion :) ) .. which, i think, was a clone of the rpm one..
20:17:50 <sgallagh> Also (at the risk of self-promotion), please consult https://sgallagh.wordpress.com/2017/05/30/sausage-factory-an-introduction-to-building-modules-in-fedora/ for basic information on creating modules.
20:17:56 <sgallagh> Followed by https://sgallagh.wordpress.com/2017/06/30/sausage-factory-advanced-module-building-in-fedora/
20:18:21 <langdon> unfortunately, reading neither of those ACTUALLY results in getting any sausage :/
20:18:44 <nirik> the sausage is a lie!
20:18:46 <sgallagh> Anyone who has a vested interest in seeing the Server modules get delivered is advised to join #fedora-modularity and assist us
20:19:41 <sgallagh> Also, if you happen to have access to atypical hardware architectures, it wouldn't hurt to get some testing of both builds and actual binaries on them
20:19:48 <langdon> +1 .. we are way behind
20:19:48 * sgallagh looks directly at mjwolf
20:19:54 <langdon> ha
20:20:31 <mjwolf> mjwolf: I will take a look
20:20:53 <adamw> sgallagh: we will be able to test on ppc in openQA fairly soon (it'd probably all be deployed already if i didn't have five other goddamn things going wrong with openqa to worry about)
20:21:14 <sgallagh> adamw: ack and thank you
20:21:17 <langdon> adamw, otherwise, you would get bored..
20:21:43 <sgallagh> I'm actually terrified of the day adamw is bored
20:21:51 <langdon> sgallagh, right?!!??
20:22:16 <sgallagh> langdon: Do you want to call out any specific actions people could volunteer to take?
20:22:21 <sgallagh> (Or be volunteered FOR)
20:22:25 * adamw dearly wishes to be bored right now
20:22:42 <langdon> ha.. /me thinks
20:23:45 <langdon> anything anyone can contribute to the infra work would definitely help.. bodhi changes, arbitrary branching, etc.. nirik, I think you would know if there is anything anyone could "do" that is low hanging fruit on the factory 2 work.
20:24:14 <nirik> not entirely sure... this last week has been a blur. Lots of little stuff to fix from pagure over dist git...
20:24:30 <nirik> but I think we are finally starting to catch up. Hopefully before the next truck hits us. ;)
20:25:15 <langdon> here's wishing...
20:25:18 <sgallagh> nirik: Is that truck the branch?
20:25:29 <langdon> ralph is out today who would know best
20:25:35 <sgallagh> ... probably should have made a joke about swinging a branch at your head
20:25:36 <nirik> naw, that truck is here now. ;)
20:26:43 <langdon> but as far as "actual" modules.. dependency help is very useful.. if you know any of the items on the list well, let's #info it and let adam know so he has another channel for questions
20:27:05 <smooge> the branch, the flock, the mass rebuild because we found something broke in gcc, the rebuild of the rebuild because...
20:28:22 <sgallagh> #info sgallagh is a resource for SSSD, FreeIPA and Cockpit dependencies
20:28:26 * langdon is finding flock very inconvenient this year.. if mattdm isn't willing to move f27.. maybe we could just move flock
20:29:12 <sgallagh> langdon: Don't even joke about that
20:29:25 <smooge> I was trying to get it moved to devconf
20:29:29 <langdon> ha
20:29:31 <sgallagh> You would have a riot on your hands
20:29:37 <langdon> nothing like brno in feb
20:30:39 <langdon> so.. any other #info peeps?
20:30:48 <sgallagh> Well, the advantage of Flock this year is that we have time dedicated to hacking on things while everyone is in one place right before the big Freeze
20:31:18 <langdon> #action asamalik to review the meeting minutes for people who can act as experts on the packages he is working with
20:31:19 <sgallagh> Since historically, everything lands 48-to-negative-twelve hours before Freeze, we are in good shape :)
20:31:36 * langdon not #chair.. but the fedbot liked it... am i?
20:32:02 <sgallagh> #action asamalik to review the meeting minutes for people who can act as experts on the packages he is working with
20:32:06 <sgallagh> there
20:32:26 <langdon> cool
20:32:31 <sgallagh> Do we have anyone around with considerable experience with PostgreSQL, NetworkManager or udisks2?
20:32:32 * langdon will forget to tell him
20:32:44 <sgallagh> langdon: Oh, sidebar: storaged merged back to udisks2
20:32:50 <langdon> ha
20:33:06 <nirik> yeah, I expect a new name a year from them now. ;)
20:33:18 <langdon> udisks.next FTW
20:33:32 * nirik has used all those things, but wouldn't say he is any kind of expert on them
20:33:58 <sgallagh> nirik: Personally, I liked the storaged name better, as it was more accurate. Ah well
20:34:06 <nirik> udisks -> storaged -> udisks2 -> storaged2 -> udisks3 :)
20:34:30 <sgallagh> nirik: The actual history of that was more complicated still :)
20:34:35 <nirik> yes, I know
20:35:31 <langdon> nirik, mild sidebar, do you know the fedora-change that is the bodhi changes for modules?
20:36:19 <langdon> i may have found it "bodhi non-rpm artifacts"
20:36:20 <nirik> https://fedoraproject.org/wiki/Changes/BodhiNonRPMArtifacts
20:36:24 <langdon> thanks!
20:36:57 <sgallagh> langdon: Oh, that's another concern
20:37:22 <sgallagh> Last I heard, the work was on the way to make Bodhi capable of non-RPM artifacts, but not specifically prepared for module artifacts.
20:37:23 <sgallagh> Any comment?
20:37:42 <sgallagh> Because I feel like that's a blocker to a supportable release of Server
20:37:51 <sgallagh> err, Boltron
20:38:02 <langdon> sgallagh, yeah.. i think i was in that "thread" .. was supposed to follow up with ralph today .. but couldn't
20:38:05 <smooge> I was thinking we would call the new storage.. StorageTech
20:38:33 <sgallagh> smooge: "Where'dIPutThatAgain"
20:38:45 <smooge> "InfiniteMonkeys"
20:39:05 <sgallagh> smooge: I have a story about that, remind me to tell you at Flock :)
20:39:52 <nirik> "zipdrive100"
20:40:01 <sgallagh> langdon: Well, do we have any idea how likely it is that this will get sorted before Beta Freeze?
20:40:24 <sgallagh> Because as of right now, I'd call our status "orange" (yellow tinged with red) over this
20:40:24 <langdon> sgallagh, which? the bodhi stuff?
20:40:27 <sgallagh> yes
20:40:53 <langdon> we don't actually need it by beta freeze, right? we don't need it til ga+1-second?
20:41:47 <langdon> but.. i would say we are kinda orange already
20:41:54 <langdon> rather.. for lots of reasons
20:42:15 <sgallagh> langdon: Well, we need it by bodhi activation point
20:42:57 <langdon> sgallagh, sorry.. i am not sure what you mean by "activation point"
20:42:59 <sgallagh> Which is actually *before* Beta Freeze :-|
20:43:14 <sgallagh> #link https://fedoraproject.org/wiki/Releases/27/Schedule
20:43:36 <sgallagh> langdon: There's a point in the dev cycle where we stop inheriting from Koji automatically and all builds need to go through bodhi to end up in a repo
20:44:05 <sgallagh> Now, it's possible that this is not true for the MBS pipeline, but I don't have that information
20:44:09 <langdon> sgallagh, ohh.. right.. i see.. i don't think i knew what that "step" was called
20:44:11 <sgallagh> And we need to
20:44:26 <sgallagh> That date is currently two weeks from today
20:44:30 <sgallagh> (just prior to Flock)
20:44:41 <langdon> right
20:46:00 <bowlofeggs> sgallagh, langdon: bodhi is ready for modules with one exception - it needs a way to "mash" them (for lack of a better term)
20:46:26 <bowlofeggs> sgallagh, langdon: there's an open pull request that needs more work to do that from mcurlej - afaik he's working on it and considers it high priority
20:46:42 <langdon> bowlofeggs, ahh right
20:46:45 <bowlofeggs> https://github.com/fedora-infra/bodhi/pull/1697
20:47:03 <langdon> i knew there was a reason it wasn't keeping me up at night
20:47:24 <bowlofeggs> it should be possible to create module updates in bodhi already
20:47:32 <bowlofeggs> you just can't "ship" them when they get stable yet
20:47:52 <sgallagh> ok, that's a little less frightening
20:47:59 <sgallagh> bowlofeggs: Do we have an ETA on that?
20:48:04 <bowlofeggs> very little integration testing has been done though, because koji in fedora stg doesn't have modules in it yet
20:48:17 <dgilmore> bowlofeggs: yes it does
20:48:22 <bowlofeggs> sgallagh: i've not been told an ETA - i'd recommend reaching out to the factory 2.0 team to find out
20:48:26 <dgilmore> bowlofeggs: just may not have any built
20:48:27 <nirik> I thought they tested it in stg a bunch
20:48:32 * sgallagh reaches
20:48:44 <dgilmore> afaik they tested extensively in stg
20:48:45 <bowlofeggs> dgilmore: oh neat, in that case, we should try to create a module update in bodhi stg
20:50:20 <langdon> bowlofeggs, any reason you can think of that i can't tag you in pagure either by your nic or last name?
20:50:44 <bowlofeggs> langdon: no, i should be @bowlofeggs
20:50:52 <bowlofeggs> (that's my FAS name)
20:50:56 <langdon> weird.. ok.. ill try again
20:51:38 <langdon> yeah.. works fine.. just no popup :/
20:52:36 <sgallagh> OK, do we have anything further to discuss in this group today?
20:52:44 <sgallagh> We're getting close to the top of the hour
20:55:10 <sgallagh> I'll take that as a no
20:55:23 <sgallagh> #topic Open Floor
20:55:33 <sgallagh> I'll leave the floor open for a couple minutes, then close the meeting.
20:57:52 <sgallagh> Thank you for coming.
20:57:52 <sgallagh> #endmeeting