#meeting:fedoraproject.org log

<@davide:cavalca.name>

17:01:48

!startmeeting CentOS Hyperscale SIG

<@meetbot:fedora.im>

17:01:50

Meeting started at 2024-11-06 17:01:48 UTC

<@meetbot:fedora.im>

17:01:51

The Meeting name is 'CentOS Hyperscale SIG'

<@davide:cavalca.name>

17:02:04

morning folks

<@davide:cavalca.name>

17:02:16

as usual the DST transition is not smooth

<@davide:cavalca.name>

17:02:17

!topic Roll call

<@salimma:fedora.im>

17:03:36

!hi

<@zodbot:fedora.im>

17:03:37

Michel Lind (salimma) - he / him / his

<@rcolebaugh:matrix.org>

17:03:53

!hi

<@davide:cavalca.name>

17:04:58

let's get started

<@zodbot:fedora.im>

17:05:01

Raymond Colebaugh (rcolebaugh) - he / him / his

<@davide:cavalca.name>

17:05:01

!topic Roll call

<@davide:cavalca.name>

17:05:14

!topic Followups

<@davide:cavalca.name>

17:05:41

Conan Kudo: and I presented at CentOS Showcase earlier this week

<@davide:cavalca.name>

17:05:49

I assume the video will be up on youtube at some point

<@davide:cavalca.name>

17:06:29

anybody else has followups to share?

<@conan_kudo:matrix.org>

17:07:44

!hi

<@zodbot:fedora.im>

17:07:46

Neal Gompa (ngompa) - he / him / his

<@conan_kudo:matrix.org>

17:08:09

I don't have anything else as a followup

<@salimma:fedora.im>

17:08:11

I had a related presentation too :P

<@conan_kudo:matrix.org>

17:08:35

well, I guess we talked about making hyperscale composes at the showcase

<@conan_kudo:matrix.org>

17:09:04

so that cat's out of the bag: it's something I've been working on to build because producing our artifacts in a timely and reliable fashion is annoying

<@davide:cavalca.name>

17:09:18

ah yes, the infra for that should be more or less ready, and I have a ticket to get the domain hooked up: https://pagure.io/centos-infra/issue/1530

<@conan_kudo:matrix.org>

17:11:11

we'll have more details on that to share later, but it's some exciting stuff for CentOS SIG folks :)

<@davide:cavalca.name>

17:11:34

yeah we should plan a blog post when it's all in production

<@conan_kudo:matrix.org>

17:11:45

and a CentOS Connect talk

<@davide:cavalca.name>

17:12:10

on that note, let's move into

<@davide:cavalca.name>

17:12:11

!topic Announcements

<@davide:cavalca.name>

17:12:18

CentOS Connect cfp should be opening soon

<@conan_kudo:matrix.org>

17:12:46

there is also the round of spring conferences to consider

<@davide:cavalca.name>

17:13:06

the cfp for SCALE is closed now

<@conan_kudo:matrix.org>

17:13:06

offhand: while SCaLE's CFP is closed, LinuxFest Northwest is still open as is Red Hat Summit

<@conan_kudo:matrix.org>

17:13:29

not sure what's up with Texas LinuxFest yet, I don't see anything there

<@davide:cavalca.name>

17:13:40

I'm not sure I'll do LFNW this round, we'll see

<@davide:cavalca.name>

17:13:50

we'll definitely do something for Summit though

<@conan_kudo:matrix.org>

17:14:07

mostly just pointing it out for everyone here

<@davide:cavalca.name>

17:15:40

in other news, I just tagged for release a numactl backport

<@davide:cavalca.name>

17:15:53

thanks to Jonathan Wright for getting the latest version updated in Fedora

<@davide:cavalca.name>

17:16:07

I've also filed tickets on jira to get it rebased in c9s and c10s

<@conan_kudo:matrix.org>

17:19:52

I think I'm going to have a solution to unblock kernel builds in CBS implemented this week

<@salimma:fedora.im>

17:20:03

nice

<@conan_kudo:matrix.org>

17:20:06

so hopefully I'll be able to finally make a new kernel release that won't fail because of NVRA conflicts

<@salimma:fedora.im>

17:20:09

ngompa++

<@zodbot:fedora.im>

17:20:11

salimma gave a cookie to ngompa. They now have 149 cookies, 2 of which were obtained in the Fedora 41 release cycle

<@davide:cavalca.name>

17:21:58

moving on

<@davide:cavalca.name>

17:21:59

!topic Tickets

<@davide:cavalca.name>

17:22:15

we still need to tidy up the tickets, I don't think there's anything actionable here at the moment

<@conan_kudo:matrix.org>

17:22:36

we should probably spend a part of our next hangout just crushing them

<@conan_kudo:matrix.org>

17:23:06

otherwise I don't think we'll ever do it

<@davide:cavalca.name>

17:23:40

yeah good idea

<@davide:cavalca.name>

17:23:44

!topic Membership

<@davide:cavalca.name>

17:23:51

don't think we have anything here either this week

<@davide:cavalca.name>

17:24:35

!topic Misc

<@conan_kudo:matrix.org>

17:24:36

nope

<@davide:cavalca.name>

17:24:43

anything else folks want to discuss?

<@conan_kudo:matrix.org>

17:24:49

yeah, one thing from me

<@davide:cavalca.name>

17:24:56

go for it

<@conan_kudo:matrix.org>

17:25:01

so I just approved this PR for c10s systemd v255 thing

<@conan_kudo:matrix.org>

17:25:17

but... why are we shipping systemd v255 to c10s, when c10s started with v256?

<@davide:cavalca.name>

17:26:03

this is in a separate branch; Ryan Wilson is working on it because we need an interim bugfix to be able to do update from 255 to 256 cleanly

<@davide:cavalca.name>

17:26:28

this came up as we were doing the internal rollout at meta

<@ryantimwilson:matrix.org>

17:26:40

Meta internal still uses 255 for C10 :/

<@ryantimwilson:matrix.org>

17:26:56

so yea i put it on a separate branch cuz we're behind

<@conan_kudo:matrix.org>

17:27:03

oof

<@conan_kudo:matrix.org>

17:27:14

we need to be careful with such things though, because it's rather risky

<@conan_kudo:matrix.org>

17:27:36

under normal circumstances, they are basically untestable because the system doesn't really _do_ downgrades naturally

<@davide:cavalca.name>

17:27:43

hopefully it's just a transitional oneoff

<@conan_kudo:matrix.org>

17:28:09

we also need to do some serious cleanup of old builds of things

<@conan_kudo:matrix.org>

17:28:26

there's a lot of old kernels and such that I know we're not using, but they add a lot of weight to the repos

<@davide:cavalca.name>

17:28:49

yeah I've been untagging things as they get superseded by CentOS proper

<@conan_kudo:matrix.org>

17:29:01

like kernels from back when we were based on the rhel kernel kind of old

<@conan_kudo:matrix.org>

17:29:16

I noticed it when I was doing some compose prod testing last week

<@conan_kudo:matrix.org>

17:29:43

Michel Lind 🎩 UTC-5 asked me to not just go and untag things, so instead I'm bringing it up here

<@davide:cavalca.name>

17:31:17

anything else to discuss, or should we end early today?

<@conan_kudo:matrix.org>

17:32:16

nothing from me

<@davide:cavalca.name>

17:32:30

alright, let's get some time back

<@davide:cavalca.name>

17:32:33

thanks everyone!

<@davide:cavalca.name>

17:32:35

!endmeeting