#meeting:fedoraproject.org log

<@salimma:fedora.im>

15:07:47

!startmeeting CentOS Hyperscale SIG

<@meetbot:fedora.im>

15:07:49

Meeting started at 2025-04-23 15:07:47 UTC

<@meetbot:fedora.im>

15:07:49

The Meeting name is 'CentOS Hyperscale SIG'

<@salimma:fedora.im>

15:07:59

!topic Roll call

<@salimma:fedora.im>

15:08:02

!hi

<@zodbot:fedora.im>

15:08:03

Michel Lind (salimma) - he / him / his

<@nhanlon:beeper.com>

15:08:11

!hi

<@zodbot:fedora.im>

15:08:13

Neil Hanlon (neil) - he / him / his

<@salimma:fedora.im>

15:08:14

our chairs are running a bit late, so just starting the check in

<@nhanlon:beeper.com>

15:08:29

g'day Michel, hope all is well

<@salimma:fedora.im>

15:09:14

hi Neil! reasonably well, just have to watch the kid at home for a few more mins

<@rcolebaugh:matrix.org>

15:09:14

!hi

<@zodbot:fedora.im>

15:09:16

Raymond Colebaugh (rcolebaugh) - he / him / his

<@conan_kudo:matrix.org>

15:10:01

!hi

<@zodbot:fedora.im>

15:10:03

Neal Gompa (ngompa) - he / him / his

<@davide:cavalca.name>

15:10:19

!hi

<@zodbot:fedora.im>

15:10:20

Davide Cavalca (dcavalca) - he / him / his

<@davide:cavalca.name>

15:13:35

morning folks

<@davide:cavalca.name>

15:13:40

let's get started

<@davide:cavalca.name>

15:13:42

!topic Followups

<@davide:cavalca.name>

15:13:52

anybody has followups from the last meeting?

<@conan_kudo:matrix.org>

15:15:06

we have our first composes!

<@salimma:fedora.im>

15:15:07

I remembered too late we want to start cleaning up inactive members

<@conan_kudo:matrix.org>

15:15:08

they exist

<@conan_kudo:matrix.org>

15:15:13

and things and stuff

<@conan_kudo:matrix.org>

15:15:25

made with kurchu

<@salimma:fedora.im>

15:15:35

so I'll try to get a PR for membership process before the next meeting

<@salimma:fedora.im>

15:15:47

now we need to release kurchu

<@conan_kudo:matrix.org>

15:15:57

I'm working on a blog post to introduce kurchu to the world

<@conan_kudo:matrix.org>

15:16:00

yes

<@conan_kudo:matrix.org>

15:16:59

we already have an existing process to drop people, we just haven't been doing it

<@conan_kudo:matrix.org>

15:17:22

we need to just go ahead and prune everyone with no activity for a year or something like that

<@salimma:fedora.im>

15:18:19

we say 6 months here, it turns out https://sig.centos.org/hyperscale/policies/governance/

<@salimma:fedora.im>

15:18:51

I think we just need to flesh out the actual process - e.g. what counts as contributions, how often we run the purge, etc.?

<@conan_kudo:matrix.org>

15:18:56

so let's just do it

<@conan_kudo:matrix.org>

15:19:14

it's going to be kind of painful, but membership bankruptcy is important esp since that controls gitlab access

<@salimma:fedora.im>

15:19:26

yeah. so just tiny tweak to the docs, and a tool to automatically grab activity data

<@conan_kudo:matrix.org>

15:19:49

contributions counts as meaningful attendence and feedback or code work

<@salimma:fedora.im>

15:19:56

we can start purging the members that have not been seen for a year, those are definitely inactive, and play it by ear for 6-12 months

<@salimma:fedora.im>

15:20:15

yeah, so the script will need to check the meeting logs too

<@conan_kudo:matrix.org>

15:20:23

qa work will count too

<@conan_kudo:matrix.org>

15:20:45

yup

<@salimma:fedora.im>

15:20:55

that and builds will be easy. Pagure PR comments might need to be added. but we can start simple and do manual checks for those that are false positives

<@conan_kudo:matrix.org>

15:20:55

fortunately MXIDs map nicely to FAS for the most part

<@conan_kudo:matrix.org>

15:21:01

so we can check based on that

<@salimma:fedora.im>

15:21:03

and now gitlab too, yay. ouch

<@salimma:fedora.im>

15:21:20

I think the name shows up in the meeting log? for those that use hi properly anyway

<@salimma:fedora.im>

15:21:44

but it's 20 past, let's move on - anyone else has a follow up?

<@davide:cavalca.name>

15:22:04

I meant to update the release packages and mock configs but haven't gotten around it yet

<@salimma:fedora.im>

15:24:06

on the topic of things that need to be done, I guess I'll pivot soon to prepping Hyperscale GNOME - this will also involve properly launching Proposed Updates' website and announcing it anyway

<@salimma:fedora.im>

15:24:19

and try and _not_ do anything else until Summit, since that's coming up really soon

<@davide:cavalca.name>

15:28:36

next up

<@davide:cavalca.name>

15:28:37

!topic Announcements

<@davide:cavalca.name>

15:28:56

as Michel mentioned, a bunch of us will be at Summit in a few weeks, with several talks planned

<@salimma:fedora.im>

15:31:02

it will be a konflux of talks

<@conan_kudo:matrix.org>

15:31:14

Davide and I need to get heads-down on CentOS Hyperscale Asahi

<@conan_kudo:matrix.org>

15:31:32

🤦‍♂️

<@salimma:fedora.im>

15:31:39

yeah, I suggested just renaming our 7zip review 7zip-standalone and get it unblocked

<@davide:cavalca.name>

15:33:07

on that note, I also came up with this abomination to workaround a missing subpackage for pipewire in el10: https://bugzilla.redhat.com/show_bug.cgi?id=2361814

<@salimma:fedora.im>

15:33:42

I'm starting to really, really hate -epel packages

<@salimma:fedora.im>

15:34:01

oh related to that, we'll need to bring iptables to Hyperscale, *but* will only be making a hs+fb build

<@conan_kudo:matrix.org>

15:35:18

seems fine to me

<@conan_kudo:matrix.org>

15:35:35

that should be easy, just make sure you branch it appropriately so people know

<@salimma:fedora.im>

15:36:39

ah yes, that's interesting - we have not discussed that for the new gitlab branching structure

<@salimma:fedora.im>

15:36:56

I guess cXs-hs-fb ? or can we use + symbols

<@salimma:fedora.im>

15:38:03

I don't see a rule against + so it might be fine, we'll see

<@conan_kudo:matrix.org>

15:38:58

you can use +

<@conan_kudo:matrix.org>

15:39:04

you just can't use ~ and @ I think

<@salimma:fedora.im>

15:40:42

there's a bunch more rules but I don't think we violate them anyway

<@davide:cavalca.name>

15:40:58

did we write up the branching policy for gitlab somewhere?

<@salimma:fedora.im>

15:41:04

^, :, ?, [, .., etc.

<@salimma:fedora.im>

15:41:09

and gitlab has additional rules too

<@salimma:fedora.im>

15:41:17

I think we're still waiting for the gitlab contribution guide

<@davide:cavalca.name>

15:41:34

on the Meta side, we have a few folks that want to start using it for other packages, so it'd be good to have everything formalized

<@salimma:fedora.im>

15:41:39

https://docs.gitlab.com/user/project/repository/branches/#name-your-branch

<@conan_kudo:matrix.org>

15:41:48

I've got a branch to push as a PR for this

<@conan_kudo:matrix.org>

15:41:59

I've just been in firefighting mode up until F42 released last week

<@davide:cavalca.name>

15:44:58

anything else for announcements?

<@salimma:fedora.im>

15:47:01

crickets

<@davide:cavalca.name>

15:47:21

!topic Tickets

<@davide:cavalca.name>

15:48:47

!topic Membership

<@davide:cavalca.name>

15:50:26

Tommy was trying to join but he's having matrix trouble

<@salimma:fedora.im>

15:50:43

we can wait a bit

<@salimma:fedora.im>

15:51:15

"having matrix trouble" sums up my last month quite well

<@conan_kudo:matrix.org>

15:52:11

at least they seem to be calming down for me

<@conan_kudo:matrix.org>

15:52:27

I think most of the trouble came from the transition to MAS

<@salimma:fedora.im>

15:52:30

yeah but Matrix's paid offering being shittier than matrix.org takes the cake

<@salimma:fedora.im>

15:52:42

it's like they're trying so hard to fail

<@salimma:fedora.im>

15:53:05

Tommy is here! yay

<@tommy-u:matrix.org>

15:53:12

Wow, I never thought I’d make it here 😂

<@davide:cavalca.name>

15:53:20

https://pagure.io/centos-sig-hyperscale/sig/issue/189

<@davide:cavalca.name>

15:53:28

tommy-u: wanna introduce yourself to the group?

<@tommy-u:matrix.org>

15:55:35

<@tommy-u:matrix.org>

15:55:35

Hello friends, I’m Tommy, I recently joined meta after spending way too much time building custom OSs in academia.

<@tommy-u:matrix.org>

15:55:35

I used to intern with Red Hat a bunch and worked on a wacky project to turn Linux into a unikernel

<@tommy-u:matrix.org>

15:55:35

<@tommy-u:matrix.org>

15:55:35

It’s a pleasure to meet everyone :)

<@salimma:fedora.im>

15:56:07

welcome Tommy

<@salimma:fedora.im>

15:56:14

want to say a bit about what you plan to work on in Hyperscale?

<@rcolebaugh:matrix.org>

15:56:20

Hi Tommy, welcome! :)

<@tommy-u:matrix.org>

15:59:21

Sorry I’m very much a hyper scale noob. I’ve been doing some work on integration testing systemd. Does that project fall under the hyper scale unbrella 🤦‍♂️

<@davide:cavalca.name>

15:59:41

yep

<@davide:cavalca.name>

16:01:41

we're out of time; any objections to Tommy joining the Hyperscale SIG ?

<@rcolebaugh:matrix.org>

16:01:54

None from me!

<@tommy-u:matrix.org>

16:02:24

Ok cool, I’m involved with the project of using systemd to launch containers within our data centers. Hopefully some of this integration testing will help make sure we don’t break upstream as we move forward adding this functionality :)

<@salimma:fedora.im>

16:03:08

is that using nspawn?

<@salimma:fedora.im>

16:03:15

would be neat to see it used at scale

<@salimma:fedora.im>

16:03:53

but yeah we can continue this in #centos-hyperscale:fedoraproject.org and free up the room

<@davide:cavalca.name>

16:04:03

!agreed welcome tommy-u to the Hyperscale SIG

<@davide:cavalca.name>

16:04:12

I will take care of the paperwork after this

<@tommy-u:matrix.org>

16:04:12

Currently we’re using systemd-run, but nspawn keeps coming up and I’d like to look into it as an alternative

<@davide:cavalca.name>

16:04:23

thanks everyone!

<@davide:cavalca.name>

16:04:25

!endmeeting