<@sgallagh:fedora.im>
17:00:57
!startmeeting FESCO (2024-08-27)
<@sgallagh:fedora.im>
17:00:58
!meetingname fesco
<@meetbot:fedora.im>
17:00:59
Meeting started at 2024-08-27 17:00:57 UTC
<@meetbot:fedora.im>
17:00:59
The Meeting name is 'FESCO (2024-08-27)'
<@meetbot:fedora.im>
17:01:00
The Meeting Name is now fesco
<@sgallagh:fedora.im>
17:01:06
Chairs: @conan_kudo:matrix.org, @ngompa:fedora.im, @nirik:matrix.scrye.com, @humaton:fedora.im, @zbyszek:fedora.im, @sgallagh:fedora.im, @jistone:fedora.im, @dcantrell:fedora.im, @decathorpe:fedora.im, @salimma:fedora.im
<@zbyszek:fedora.im>
17:01:08
!hi
<@zodbot:fedora.im>
17:01:09
Zbigniew Jędrzejewski-Szmek (zbyszek)
<@sgallagh:fedora.im>
17:01:11
!topic Init Process
<@sgallagh:fedora.im>
17:01:23
!hi
<@zodbot:fedora.im>
17:01:24
Stephen Gallagher (sgallagh) - he / him / his
<@dcantrell:fedora.im>
17:01:32
!hi
<@zodbot:fedora.im>
17:01:34
David Cantrell (dcantrell) - he / him / his
<@nirik:matrix.scrye.com>
17:01:35
morning
<@conan_kudo:matrix.org>
17:01:43
!hi
<@zodbot:fedora.im>
17:01:46
Neal Gompa (ngompa) - he / him / his
<@zbyszek:fedora.im>
17:01:51
I need to drop in ~20 minutes...
<@salimma:fedora.im>
17:02:22
!hi
<@zodbot:fedora.im>
17:02:23
Michel Lind (salimma) - he / him / his
<@sgallagh:fedora.im>
17:03:14
OK, we've got quorum so let's get started and others can join as they arrive.
<@sgallagh:fedora.im>
17:03:41
!topic #3262 Permanent Updates Policy exception for uv
<@sgallagh:fedora.im>
17:03:46
!fesco 3264
<@zodbot:fedora.im>
17:03:46
● **Last Updated:** 2 hours ago
<@zodbot:fedora.im>
17:03:46
● **Assignee:** Not Assigned
<@zodbot:fedora.im>
17:03:46
● **Opened:** 21 hours ago by amoloney
<@zodbot:fedora.im>
17:03:46
<@zodbot:fedora.im>
17:03:46
**fesco #3264** (https://pagure.io/fesco/issue/3264):**Incomplete Changes Report for Fedora Linux 41**
<@jistone:fedora.im>
17:03:50
!hi
<@zodbot:fedora.im>
17:03:50
Josh Stone (jistone) - he / him / his
<@salimma:fedora.im>
17:04:15
wait, are we doing 3262 or 3264?
<@sgallagh:fedora.im>
17:04:24
Oops, mistyped
<@sgallagh:fedora.im>
17:04:37
!fesco 3262
<@zodbot:fedora.im>
17:04:38
**fesco #3262** (https://pagure.io/fesco/issue/3262):**Permanent Updates Policy exception for uv**
<@zodbot:fedora.im>
17:04:38
● **Assignee:** Not Assigned
<@zodbot:fedora.im>
17:04:38
● **Last Updated:** 3 days ago
<@zodbot:fedora.im>
17:04:38
● **Opened:** 5 days ago by music
<@zodbot:fedora.im>
17:04:38
<@sgallagh:fedora.im>
17:04:53
Michel Lind 🎩: Thanks for catching that
<@jistone:fedora.im>
17:05:17
is `uv` ever a `BuildRequires`? (now or in the future)
<@jistone:fedora.im>
17:05:22
or more of an end user thing?
<@sgallagh:fedora.im>
17:05:22
So, the discussion in the ticket is generally in favor of allowing the exception, but I wanted to put a constraint that the exception ends once upstream stabilizes.
<@sgallagh:fedora.im>
17:06:03
Josh Stone: Seems like it's more of an end-user thing
<@zbyszek:fedora.im>
17:06:22
I don't quite see the point, but I'm fine with the constraint too. +1 either way.
<@conan_kudo:matrix.org>
17:06:54
I don't see much of a point in the constraint either, but meh
<@conan_kudo:matrix.org>
17:07:01
I'm +1 with or without
<@salimma:fedora.im>
17:07:28
Once it hits 1.0 hopefully it does not need a permanent exception anymore, right? so it's nice having it expire I guess. I have a slight preference for having the constraint, but I'm ok either way
<@dcantrell:fedora.im>
17:08:06
my gut feeling is that uv and things like it will evolve rapidly and we'll never need the exception, but either way +1 from me
<@jistone:fedora.im>
17:08:31
+1 with or without the constraint, but I think that could also be handled by revoking approval if it gets mismanaged
<@sgallagh:fedora.im>
17:08:31
dcantrell: Never need the exception or the expiration of the exception?
<@dcantrell:fedora.im>
17:09:32
either, honestly. I feel like uv is the sort of thing that will undergo some significant implementation changes that we'll just be back here talking about, hypothetically, moving to uv2 because that's what everyone did but how do we deal with the uv stuff in Fedora now
<@sgallagh:fedora.im>
17:09:37
Josh Stone: You make a fair point. We can always revoke it later. So I guess I'll just drop my pitch for the constraint so we can move on.
<@sgallagh:fedora.im>
17:10:38
dcantrell: I don't understand. We already have requests for a stable updates exception.
<@sgallagh:fedora.im>
17:10:56
This ticket is about whether we just give them a "permanent" exception to ignore the stable updates policy.
<@dcantrell:fedora.im>
17:12:01
right, and I think that's ultimately what will have to be allowed because of the rate of change in the space of Python module management. this stuff changes so rapidly. I don't think there will ever be a point that we would consider stable, so we're really just deciding to allow it or not in Fedora. and I think we need to allow it
<@music:fedora.im>
17:12:21
Hi, dropping in here in case I can answer any questions.
<@sgallagh:fedora.im>
17:12:50
dcantrell: In short: you're voting +1 for the permanent exception. Yes?
<@dcantrell:fedora.im>
17:12:59
yes
<@conan_kudo:matrix.org>
17:13:14
for user-facing tools like black; ruff; and uv, yeah this makes a lot of sense
<@conan_kudo:matrix.org>
17:13:41
(I believe we already have exceptions for the former two, so this is just a new one to the pile)
<@sgallagh:fedora.im>
17:13:54
I count +7 in favor in-ticket and in meeting, combined.
<@sgallagh:fedora.im>
17:14:51
!agreed The "uv" package is granted a permanent stable updates exception (+7, 0, -0)
<@sgallagh:fedora.im>
17:15:13
!topic #3264 Incomplete Changes Report for Fedora Linux 41
<@sgallagh:fedora.im>
17:15:19
!fesco 3264
<@zodbot:fedora.im>
17:15:20
● **Opened:** 21 hours ago by amoloney
<@zodbot:fedora.im>
17:15:20
● **Last Updated:** 2 hours ago
<@zodbot:fedora.im>
17:15:20
● **Assignee:** Not Assigned
<@zodbot:fedora.im>
17:15:20
<@zodbot:fedora.im>
17:15:20
**fesco #3264** (https://pagure.io/fesco/issue/3264):**Incomplete Changes Report for Fedora Linux 41**
<@sgallagh:fedora.im>
17:16:00
Looks like there are 18 Changes marked as Incomplete right now.
<@nirik:matrix.scrye.com>
17:16:19
there were some updates from the orig ticket comment
<@sgallagh:fedora.im>
17:16:19
Oh, wait. That may not be accurate.
<@sgallagh:fedora.im>
17:16:45
How do we want to go through these? Just one by one from the top of the ticket?
<@zbyszek:fedora.im>
17:16:51
Let's just go one-by-one.
<@nirik:matrix.scrye.com>
17:16:57
yeah.
<@conan_kudo:matrix.org>
17:16:59
as topics, yeah, that's pretty much the way to go
<@conan_kudo:matrix.org>
17:17:06
and have info statements as needed
<@conan_kudo:matrix.org>
17:17:15
that way there will be summarized info
<@zbyszek:fedora.im>
17:17:18
Yeah, topics so that the minutes are legible.
<@sgallagh:fedora.im>
17:17:19
!topic Incomplete Change: Python 3.13
<@sgallagh:fedora.im>
17:17:28
<@zbyszek:fedora.im>
17:17:42
!info There are still 59 packages in Fedora 41 that have not been rebuilt with Python 3.13 yet.
<@sgallagh:fedora.im>
17:17:43
I think this one is in a Testable state, no?
<@conan_kudo:matrix.org>
17:17:58
I think we usually call this "done"
<@zbyszek:fedora.im>
17:17:59
Yeah, that's like 99% done.
<@conan_kudo:matrix.org>
17:18:18
so ON_QA makes sense for this
<@nirik:matrix.scrye.com>
17:18:24
+1
<@zbyszek:fedora.im>
17:18:45
Are we voting on something?
<@conan_kudo:matrix.org>
17:18:56
I don't think so
<@sgallagh:fedora.im>
17:19:21
Given the number of tickets, I suggest we note the current state and ask whether anyone wants to pull the emergency cord on the change.
<@sgallagh:fedora.im>
17:19:28
If no one does, we proceed to the next.
<@nirik:matrix.scrye.com>
17:19:37
no, I just was saying "I agree with Conan Kudo"
<@zbyszek:fedora.im>
17:19:44
kk
<@sgallagh:fedora.im>
17:20:15
!info Change is ON_QA and sufficiently complete at this stage
<@sgallagh:fedora.im>
17:20:31
!topic IPU6 Camera Support
<@sgallagh:fedora.im>
17:20:38
<@zbyszek:fedora.im>
17:20:52
!info Ticket is in MODIFIED state.
<@zbyszek:fedora.im>
17:21:17
I'd just leave it as it. It seems to be on trakc.
<@nirik:matrix.scrye.com>
17:21:28
yeah.
<@sgallagh:fedora.im>
17:21:29
The description from Hans seems to suggest it should be ON_QA, really
<@conan_kudo:matrix.org>
17:21:38
it's a fiddly kernel thing 🤷‍♂️
<@conan_kudo:matrix.org>
17:21:45
but yeah, it does sound like it should be ON_QA
<@sgallagh:fedora.im>
17:22:13
!info Change is feature complete, but with some open bugs.
<@sgallagh:fedora.im>
17:22:28
!topic RPM 4.20
<@sgallagh:fedora.im>
17:22:36
<@zbyszek:fedora.im>
17:22:40
!info Ticket was updated to ON_QA today.
<@sgallagh:fedora.im>
17:22:52
Good enough for me.
<@zbyszek:fedora.im>
17:22:59
Everything seems to be on track, bugs are resolved as they are discovered.
<@sgallagh:fedora.im>
17:23:35
!topic Golang 1.23
<@sgallagh:fedora.im>
17:23:41
<@sgallagh:fedora.im>
17:24:08
(Stop me if I go too quickly, but we have a lot to cover so I'm not going to spend a lot of time on cases that look fine)
<@jistone:fedora.im>
17:24:11
there's a build in f41: https://koji.fedoraproject.org/koji/buildinfo?buildID=2531786
<@zbyszek:fedora.im>
17:24:18
golang-1.23.0-3.fc41
<@conan_kudo:matrix.org>
17:24:31
This looks on track
<@zbyszek:fedora.im>
17:24:38
Update to Go 1.23.0
<@zbyszek:fedora.im>
17:24:38
Alejandro Sáez • 13 days ago
<@sgallagh:fedora.im>
17:24:48
The Change says that a mass-rebuild of all dependent packages is required.
<@sgallagh:fedora.im>
17:24:57
This landed before or after the main Mass Rebuild?
<@conan_kudo:matrix.org>
17:25:06
after
<@sgallagh:fedora.im>
17:26:05
I'm not real thrilled with the current state. Looking at the "Dependencies" section it lists ~2000 packages that need to be rebuilt.
<@sgallagh:fedora.im>
17:26:17
If that happened after the regular mass-rebuild, I'm inclined to punt this.
<@jistone:fedora.im>
17:26:40
I'm not sure why a rebuild is required though
<@nirik:matrix.scrye.com>
17:26:48
yeah... thats a lot of churn after freeze has started. I don't know that bodhi will like an update with 2000 builds.
<@conan_kudo:matrix.org>
17:27:25
strictly speaking, it's not required, but since Go programs (like Rust programs) embed the standard runtime into their binaries, it's preferred
<@jistone:fedora.im>
17:27:37
that's just for leaf packages then
<@conan_kudo:matrix.org>
17:27:43
yup
<@conan_kudo:matrix.org>
17:28:13
more specifically only for ones that produce binaries
<@sgallagh:fedora.im>
17:28:20
So all the built packages will effectively NOT be golang 1.23, meaning this Change will have had no effect.
<@salimma:fedora.im>
17:28:24
so it's tricky right. a mass rebuild is not really required, but it was part of the change
<@sgallagh:fedora.im>
17:28:29
Other than to land a new compiler package
<@conan_kudo:matrix.org>
17:28:52
I mean, we don't do it for rust either and nobody bats an eye except for when firefox fails to build with it
<@salimma:fedora.im>
17:29:06
given that there is a request to upgrade golang in f40, I guess it will be silly to revert this
<@jistone:fedora.im>
17:29:06
we also approved a stable rebase without further rebuilds: https://pagure.io/fesco/issue/3261
<@sgallagh:fedora.im>
17:29:07
Proposal: Reject the Change as incomplete due to the lack of a rebuild, but don't block inclusion of the compiler
<@zbyszek:fedora.im>
17:29:13
-1
<@salimma:fedora.im>
17:29:17
ah right, that's the issue I was thinking of
<@conan_kudo:matrix.org>
17:29:50
I dunno, I think it'd be worth punting and asking them first what they want to do
<@zbyszek:fedora.im>
17:29:59
To explain: I think it's important that we have the updated compiler, so that users get the latest. It might also be needed for some packages and programs.
<@jistone:fedora.im>
17:30:05
I think the Change still makes sense as a marketing point, even if it's just the compiler
<@conan_kudo:matrix.org>
17:30:15
I don't think it's unreasonable for them to still do the mass build of the go stack even now
<@zbyszek:fedora.im>
17:30:37
Having the mass rebuild would be nice, so I think we should talk to the Change Owner and coordiante a mass rebuild.
<@conan_kudo:matrix.org>
17:30:41
if it was after beta and we're in final freeze, that would be a different story
<@salimma:fedora.im>
17:30:50
we can punt then? ask them to either commit to finishing a mass rebuild, or editing the change to only say the new compiler is shipped
<@zbyszek:fedora.im>
17:31:08
But no matter if the mass rebuild happens or not, the Change should be left in, since the part that's visible to users is implemented.
<@conan_kudo:matrix.org>
17:31:15
yes
<@nirik:matrix.scrye.com>
17:31:15
and do you all expect to land that before beta? ie make it a blocker?
<@sgallagh:fedora.im>
17:31:24
NO
<@salimma:fedora.im>
17:31:31
I would not delay the beta, no
<@sgallagh:fedora.im>
17:31:36
I'm firmly against blocking on a rebuild
<@conan_kudo:matrix.org>
17:31:40
no delaying the beta
<@salimma:fedora.im>
17:31:43
mass rebuild will happen in a side tag anyway right? so it's not really affected by the beta freeze
<@zbyszek:fedora.im>
17:31:47
No, the individual packages can be updated separately.
<@sgallagh:fedora.im>
17:31:59
Michel Lind 🎩: But whether that build is tagged into Beta matters
<@conan_kudo:matrix.org>
17:32:07
I'm just saying timeline wise it's there's room to complete a mass build
<@zbyszek:fedora.im>
17:32:12
A side tag is not needed, because the updated compiler version is already in stable.
<@salimma:fedora.im>
17:32:18
I'll be ok allowing a FE for this
<@salimma:fedora.im>
17:32:23
but not blocking
<@sgallagh:fedora.im>
17:32:31
Michel Lind 🎩: I'm honestly against that too.
<@salimma:fedora.im>
17:32:38
zbyszek: true, but there might be dependency chains
<@nirik:matrix.scrye.com>
17:32:47
I think its too late for a mass rebuild....
<@zbyszek:fedora.im>
17:32:57
It'd be better to do the compilations and create updates and let them land after Beta.
<@conan_kudo:matrix.org>
17:32:59
heck, it's also important to note we can skip any packages built after go 1.23 landed too
<@conan_kudo:matrix.org>
17:33:04
so the subset of packages may be even smaller
<@conan_kudo:matrix.org>
17:33:27
we don't know that because we're not computing what the set is right now, but we can ask them to identify it
<@salimma:fedora.im>
17:33:40
at this point we should really ask the Golang change owners to see what they can commit to doing, yeah
<@jistone:fedora.im>
17:33:44
I don't think dependency chains matter either -- again, we're only talking about leaf binaries
<@sgallagh:fedora.im>
17:34:11
Proposal: We allow them to revise the Change to be just about shipping the compiler. We don't advertise anything implying that packages were rebuilt USING that compiler. We allow packages to build as their maintainers see fit, within the normal confines of the Freezes and Stable Updates Policy.
<@nirik:matrix.scrye.com>
17:34:24
+1
<@jistone:fedora.im>
17:34:34
+1
<@dcantrell:fedora.im>
17:34:39
+1
<@salimma:fedora.im>
17:34:49
assuming all the 'library' source only packages builds fine, yeah, mass rebuilding everything once works. if you need to patch some of them that's another story
<@salimma:fedora.im>
17:34:51
=1
<@conan_kudo:matrix.org>
17:34:53
+1
<@salimma:fedora.im>
17:34:56
+1
<@conan_kudo:matrix.org>
17:35:30
it's too bad the Go and Fedora schedules are poorly aligned
<@sgallagh:fedora.im>
17:35:38
!agreed We allow them to revise the Change to be just about shipping the compiler. We don't advertise anything implying that packages were rebuilt USING that compiler. We allow packages to build as their maintainers see fit, within the normal confines of the Freezes and Stable Updates Policy. (+6, 0, -0)
<@sgallagh:fedora.im>
17:36:11
!topic Systemd Security Hardening
<@sgallagh:fedora.im>
17:36:18
<@conan_kudo:matrix.org>
17:37:00
This is one of those ongoing changes, no? I'm not sure there's anything we can specifically do one way or another here?
<@jistone:fedora.im>
17:37:23
it was already postponed from F40
<@conan_kudo:matrix.org>
17:38:03
probably this is going to have to be deferred
<@conan_kudo:matrix.org>
17:38:16
changing systemd units is a fairly invasive thing
<@salimma:fedora.im>
17:38:20
yeah
<@salimma:fedora.im>
17:38:31
I have seen some of these for packages I comaintain and don't feel comfortable merging them
<@nirik:matrix.scrye.com>
17:39:00
This seems like something that should just advertise 'chunks' of what gets done in that cycle...
<@jistone:fedora.im>
17:39:24
yeah, it doesn't really have a good finish line
<@sgallagh:fedora.im>
17:39:49
Proposal: Ask them to revise the title to something indicating "F41 progress" and edit Release Notes section with highlights.
<@salimma:fedora.im>
17:40:45
+1
<@jistone:fedora.im>
17:40:46
+1
<@zbyszek:fedora.im>
17:40:49
+1
<@dcantrell:fedora.im>
17:41:11
+1
<@conan_kudo:matrix.org>
17:41:12
+1
<@sgallagh:fedora.im>
17:42:01
!agreed Ask them to revise the title to something indicating "F41 progress" and edit Release Notes section with highlights. (+6, 0, -0)
<@sgallagh:fedora.im>
17:42:18
!topic Netavark Nftables Default
<@sgallagh:fedora.im>
17:42:24
<@zbyszek:fedora.im>
17:42:32
I need to drop now. Sorry!
<@sgallagh:fedora.im>
17:42:38
Looks like that's ON_QA now.
<@sgallagh:fedora.im>
17:43:19
Any concerns here or shall we move ahead?
<@conan_kudo:matrix.org>
17:43:33
nah
<@nirik:matrix.scrye.com>
17:43:34
move on
<@sgallagh:fedora.im>
17:43:47
!topic Haskell GHC 9.6 and Stackage LTS 22
<@sgallagh:fedora.im>
17:43:52
<@sgallagh:fedora.im>
17:44:24
Looks like this was moved to ON_QA today.
<@nirik:matrix.scrye.com>
17:44:36
cool.
<@sgallagh:fedora.im>
17:44:49
!info Ticket was moved to ON_QA today.
<@conan_kudo:matrix.org>
17:45:03
oof we forgot to info the last one too
<@sgallagh:fedora.im>
17:45:21
!info PREVIOUS TOPIC: Ticket was moved to ON_QA
<@sgallagh:fedora.im>
17:45:33
I'll adjust it when I send the minutes out.
<@conan_kudo:matrix.org>
17:45:40
sounds good
<@sgallagh:fedora.im>
17:45:56
!topic Make Tuned the Default Power Profile Management Daemon
<@sgallagh:fedora.im>
17:46:03
<@sgallagh:fedora.im>
17:47:11
Conan Kudo: You probably know best: what's the plan WRT powerDevil?
<@sgallagh:fedora.im>
17:47:23
Conan Kudo: You probably know best: what's the plan WRT PowerDevil?
<@conan_kudo:matrix.org>
17:48:26
I will update powerdevil to `Recommends: ppd-service` and `Suggests: tuned-ppd`
<@conan_kudo:matrix.org>
17:48:54
if someone wants to use power-profiles-daemon, by all means
<@conan_kudo:matrix.org>
17:49:02
it should still satisfy the dependency
<@sgallagh:fedora.im>
17:50:41
OK, sounds like we can just punt on this and revisit it at Beta Freeze?
<@conan_kudo:matrix.org>
17:50:56
yep
<@sgallagh:fedora.im>
17:51:09
Wait, is Freeze today?
<@sgallagh:fedora.im>
17:52:03
Yes, yes it is.
<@nirik:matrix.scrye.com>
17:52:23
yep
<@sgallagh:fedora.im>
17:52:35
So, asking a different question: should this block Beta?
<@salimma:fedora.im>
17:52:36
so we can punt until the end of the meeting :)
<@sgallagh:fedora.im>
17:53:01
I don't think we want to change power-management between Beta and Final
<@conan_kudo:matrix.org>
17:53:01
I just pushed the commit to rawhide and submitted a build there
<@conan_kudo:matrix.org>
17:53:08
I can submit an update for f41 right now :)
<@sgallagh:fedora.im>
17:53:34
Conan Kudo: Great, but I think we want to decide on whether FESCo wants to consider it blocking.
<@sgallagh:fedora.im>
17:53:52
I'm of the opinion that either it blocks Beta or the Change is postponed to F42.
<@nirik:matrix.scrye.com>
17:54:29
yes.
<@nirik:matrix.scrye.com>
17:55:20
whats all left to do here?
<@nirik:matrix.scrye.com>
17:55:28
besides the powedevil part
<@sgallagh:fedora.im>
17:55:30
Proposal: This change blocks Beta for Fedora 41.
<@conan_kudo:matrix.org>
17:55:35
nothing else is left
<@sgallagh:fedora.im>
17:56:10
If the Change introduces bugs, we can fix them for Final, but I think we have to land it on the Beta media or it won't get real-world testing
<@jistone:fedora.im>
17:56:36
+1
<@salimma:fedora.im>
17:56:56
+1
<@nirik:matrix.scrye.com>
17:57:15
ok, if it's just that... sure, +1
<@dcantrell:fedora.im>
17:57:29
I've got another call, have to drop. +1 from me
<@sgallagh:fedora.im>
17:58:52
Conan Kudo: ?
<@conan_kudo:matrix.org>
17:58:58
pushed builds
<@conan_kudo:matrix.org>
17:58:58
+1
<@sgallagh:fedora.im>
17:59:34
!agreed This Change blocks Beta for Fedora 41. (+6, 0, -0)
<@nirik:matrix.scrye.com>
18:00:00
so likely we need a blocker bug, etc...
<@sgallagh:fedora.im>
18:00:00
By my count, we now have exactly five FESCo members left. Enough to continue, but barely.
<@sgallagh:fedora.im>
18:01:01
Conan Kudo: Could you handle opening the blocker BZ for us?
<@conan_kudo:matrix.org>
18:01:08
sure
<@sgallagh:fedora.im>
18:01:50
Thank you
<@sgallagh:fedora.im>
18:02:03
!action Conan Kudo to handle filing a blocker BZ for this Change
<@sgallagh:fedora.im>
18:02:19
!topic Libvirt Virtual Network NFTables
<@sgallagh:fedora.im>
18:02:27
<@nirik:matrix.scrye.com>
18:03:30
so, not sure the state here. No answer from change owner...
<@sgallagh:fedora.im>
18:04:46
Yeah, I don't even have a good way to try to test this right now.
<@nirik:matrix.scrye.com>
18:05:14
looks like it might be already done
<@sgallagh:fedora.im>
18:05:23
Shall we punt for a week to get a reply from the Change owner and delay the Change to F42 if it's not complete?
<@sgallagh:fedora.im>
18:05:38
nirik: Source?
<@nirik:matrix.scrye.com>
18:05:45
- Update to version 10.4.0
<@nirik:matrix.scrye.com>
18:05:45
- Change virtual network backend from iptables to nftables
<@nirik:matrix.scrye.com>
18:05:45
* Wed Jun 05 2024 Daniel P. Berrangé <berrange@redhat.com> - 10.4.0-1
<@nirik:matrix.scrye.com>
18:05:54
https://koji.fedoraproject.org/koji/buildinfo?buildID=2537274
<@nirik:matrix.scrye.com>
18:06:03
yeah, lets ask them to confirm
<@sgallagh:fedora.im>
18:07:13
OK, they've been pinged in the ticket, but I'll drop Dan a line directly as well
<@sgallagh:fedora.im>
18:07:48
!info This is believed to be complete. sgallagh will reach out to the Change Owner for confirmation.
<@sgallagh:fedora.im>
18:08:08
!topic Confidential Virtualization Host with AMD SEV-SNP
<@sgallagh:fedora.im>
18:08:14
<@jistone:fedora.im>
18:08:49
zbyszek suggested postponing: https://pagure.io/fesco/issue/3264#comment-927548
<@salimma:fedora.im>
18:09:38
yeah, < 50% of intended packages in seems too low
<@conan_kudo:matrix.org>
18:09:43
yup
<@sgallagh:fedora.im>
18:09:52
Proposal: Defer Change to Fedora 42
<@conan_kudo:matrix.org>
18:09:56
+1
<@nirik:matrix.scrye.com>
18:09:58
+1
<@jistone:fedora.im>
18:10:37
+1
<@sgallagh:fedora.im>
18:11:08
Michel Lind 🎩: ?
<@sgallagh:fedora.im>
18:11:49
!agreed Defer Change to Fedora 42 (+5, 0, -0)
<@salimma:fedora.im>
18:11:51
+1
<@salimma:fedora.im>
18:11:52
sorry
<@sgallagh:fedora.im>
18:11:58
(I'm interpreting his previous comment as +1)
<@sgallagh:fedora.im>
18:11:59
Or that :)
<@sgallagh:fedora.im>
18:12:26
!topic KDE Plasma Mobile Spin & Fedora Kinoite Spin\
<@sgallagh:fedora.im>
18:12:34
<@sgallagh:fedora.im>
18:13:02
I think I saw that the Live ISO is being built now, so I think this is ready?
<@conan_kudo:matrix.org>
18:13:15
Yes, this is effectively ON_QA
<@conan_kudo:matrix.org>
18:13:31
There is a pending pull request to activate it for F41 but they have successfully built in Rawhide.
<@sgallagh:fedora.im>
18:13:34
Ah, the BZ was updated today to ON_QA
<@conan_kudo:matrix.org>
18:13:46
We are just waiting for a compose to finish things off.
<@sgallagh:fedora.im>
18:13:50
Conan Kudo: I installed the Rawhide build of it today; looking good!
<@conan_kudo:matrix.org>
18:13:55
:D
<@jistone:fedora.im>
18:14:12
@siosm also commented on the incomplete changes ticket
<@sgallagh:fedora.im>
18:14:18
!info Images exist and BZ was moved to ON_QA
<@sgallagh:fedora.im>
18:14:58
!info Kinoite Mobile ISOs will not be available for Fedora 41
<@sgallagh:fedora.im>
18:15:39
!topic Enable bootupd for Fedora Silverblue & Kinoite
<@sgallagh:fedora.im>
18:15:45
<@nirik:matrix.scrye.com>
18:16:08
not available? did you mean now available?
<@sgallagh:fedora.im>
18:16:28
<@decathorpe:fedora.im>
18:16:42
hey o/ sorry, I partially forgot about the meeting, and was stuck in transit
<@sgallagh:fedora.im>
18:16:46
nirik: No, I was clarifying that the Kinoite variant won't be there
<@conan_kudo:matrix.org>
18:16:46
Kinoite Mobile is the ostree one, that isn't going to have ISOs, only bootc images
<@conan_kudo:matrix.org>
18:16:58
Kinoite Mobile is the ostree one, that isn't going to have ISOs, only bootc images this cycle
<@nirik:matrix.scrye.com>
18:17:21
Oh, ok, I see what you are saying then...
<@sgallagh:fedora.im>
18:17:30
My "clarification" may have been less than helpful. Sorry.
<@sgallagh:fedora.im>
18:17:58
!info This is now ON_QA and is testable for Beta
<@sgallagh:fedora.im>
18:18:53
!topic mkosi-initrd
<@sgallagh:fedora.im>
18:18:59
<@sgallagh:fedora.im>
18:19:12
!info Change Owner has deferred this Change to Fedora 42
<@sgallagh:fedora.im>
18:19:34
!topic OstreeNativeContainerStable
<@sgallagh:fedora.im>
18:19:40
<@sgallagh:fedora.im>
18:20:19
This one seems to have been effectively withdrawn in favor of the bootc initiative
<@nirik:matrix.scrye.com>
18:20:25
yeah.
<@conan_kudo:matrix.org>
18:20:30
it's basically the same thing anyway
<@conan_kudo:matrix.org>
18:20:43
just a spruced up branding
<@sgallagh:fedora.im>
18:20:50
Shall we just mark this Change as "withdrawn" then?
<@conan_kudo:matrix.org>
18:20:57
yes
<@salimma:fedora.im>
18:21:15
it did mention bootc in the change proposal though?
<@salimma:fedora.im>
18:21:18
but yeah
<@salimma:fedora.im>
18:21:36
I guess bootc took off more than anticiapted
<@sgallagh:fedora.im>
18:21:38
Michel Lind 🎩: The "Bootc Initiative" is a Fedora Council "Initiative"
<@conan_kudo:matrix.org>
18:21:48
this change document was always confusing anyway
<@sgallagh:fedora.im>
18:21:48
Which is a project-wide effort, rather than a Change
<@salimma:fedora.im>
18:21:55
I guess bootc took off more than anticipated
<@conan_kudo:matrix.org>
18:22:00
I remember when we first got it and I kind of panicked over it :)
<@nirik:matrix.scrye.com>
18:22:10
I mean, it's still likely to need changes...
<@sgallagh:fedora.im>
18:22:29
Yes, but this one is too comprehensive (and probably not completable)
<@conan_kudo:matrix.org>
18:22:31
yeah, but the initiative should have these changes broken down into more logical implementable chunks
<@conan_kudo:matrix.org>
18:22:51
even now when I read this I don't know what this change actually does
<@nirik:matrix.scrye.com>
18:22:51
yep
<@sgallagh:fedora.im>
18:23:12
Proposal: Reject this change and request that smaller ones be resubmitted for Fedora 42.
<@decathorpe:fedora.im>
18:23:16
throwing around buzzwords mainly
<@conan_kudo:matrix.org>
18:23:19
+1
<@decathorpe:fedora.im>
18:23:34
+1
<@nirik:matrix.scrye.com>
18:23:46
+1
<@salimma:fedora.im>
18:24:27
+1
<@sgallagh:fedora.im>
18:25:02
!agreed Reject this change and request that smaller ones be resubmitted for Fedora 42. (+5, 0, -0)
<@sgallagh:fedora.im>
18:25:30
!topic PyTorch 2.4
<@sgallagh:fedora.im>
18:25:30
<@sgallagh:fedora.im>
18:26:02
!info This is now ON_QA
<@conan_kudo:matrix.org>
18:26:22
Tom Rix just got back from vacation too. So looks like everything is gravy. :)
<@sgallagh:fedora.im>
18:26:34
!topic ROCm 6.2
<@sgallagh:fedora.im>
18:26:36
<@sgallagh:fedora.im>
18:27:04
Per zbyszek, this doesn't look like it's ready.
<@sgallagh:fedora.im>
18:27:09
Proposal: defer to F42
<@sgallagh:fedora.im>
18:27:30
Actually, let me revise that.
<@sgallagh:fedora.im>
18:28:00
This is pretty self-contained, so it could probably land late without much trouble.
<@conan_kudo:matrix.org>
18:28:23
Yeah, let's ask them to try to get it in before Final Freeze.
<@sgallagh:fedora.im>
18:28:32
I can get behind that.
<@nirik:matrix.scrye.com>
18:29:18
seems reasonable
<@sgallagh:fedora.im>
18:29:24
Proposal: Permit this to land anytime before Final Freeze, revisit decision at that time.
<@conan_kudo:matrix.org>
18:29:29
+1
<@decathorpe:fedora.im>
18:29:32
+1
<@jistone:fedora.im>
18:29:34
+1
<@nirik:matrix.scrye.com>
18:29:37
+1
<@decathorpe:fedora.im>
18:29:48
(is this even on install media? it might make sense to allow it landing even later ...
<@conan_kudo:matrix.org>
18:29:51
Michel Lind 🎩: ping vote?
<@conan_kudo:matrix.org>
18:29:57
it is not as far as I know
<@sgallagh:fedora.im>
18:30:26
Fabio Valentini: I'd not want it to be part of the release announcement if it's not ready at Freeze.
<@jsteffan:fedora.im>
18:30:55
it shouldn't be no
<@decathorpe:fedora.im>
18:30:57
well, release announcement can be tweaked until GA too ... but yes
<@sgallagh:fedora.im>
18:31:26
!agreed Permit this to land anytime before Final Freeze, revisit decision at that time. (+5, 0, -0)
<@sgallagh:fedora.im>
18:31:40
!topic Reduce the amount of "dontaudit" rules pertaining to unlabeled_t
<@sgallagh:fedora.im>
18:31:46
<@salimma:fedora.im>
18:32:39
I like this change. no status update though...
<@conan_kudo:matrix.org>
18:32:41
this looks like nothing happened?
<@conan_kudo:matrix.org>
18:32:48
but yeah I like this change too
<@sgallagh:fedora.im>
18:33:28
I see no clear indication in the changelog for selinux-policy that this happened.
<@conan_kudo:matrix.org>
18:33:43
let's defer this to F42
<@sgallagh:fedora.im>
18:33:59
Want to make that a "Proposal:"?
<@decathorpe:fedora.im>
18:34:09
(assuming it didn't already happen, that is?)
<@conan_kudo:matrix.org>
18:34:46
Proposal: Defer this Change to F42
<@conan_kudo:matrix.org>
18:35:00
+1 from me of course
<@sgallagh:fedora.im>
18:35:19
+1
<@sgallagh:fedora.im>
18:35:33
We can retcon this decision if it turns out it's already in place.
<@decathorpe:fedora.im>
18:36:45
+1
<@nirik:matrix.scrye.com>
18:36:53
+1
<@salimma:fedora.im>
18:37:16
+1
<@salimma:fedora.im>
18:37:39
idempotent deferral :)
<@sgallagh:fedora.im>
18:37:42
!agreed Defer this Change to F42 (+5, 0, -0)
<@sgallagh:fedora.im>
18:38:01
!topic Taskwarrior 3
<@sgallagh:fedora.im>
18:38:11
<@sgallagh:fedora.im>
18:38:29
(Every time I see this, my brain translates it to Mechwarrior 3)
<@salimma:fedora.im>
18:39:19
I'm confused
<@salimma:fedora.im>
18:39:26
the tracking bug was only created... 15 days ago?
<@sgallagh:fedora.im>
18:40:12
This doesn't look complete yet.
<@sgallagh:fedora.im>
18:40:30
The `task2` package exists, but `task` hasn't been updated to Taskwarrior 3
<@nirik:matrix.scrye.com>
18:40:38
task add work on task change
<@conan_kudo:matrix.org>
18:40:40
some of the tracking bugs didn't exist because Aoife Moloney - On Vacation, Back 2nd Sept's scripts were crashing
<@conan_kudo:matrix.org>
18:40:48
I debugged it with her and got it working
<@conan_kudo:matrix.org>
18:40:56
that's why some of the bugs are so new
<@salimma:fedora.im>
18:40:56
ah
<@salimma:fedora.im>
18:41:07
I have something to propose about that in open floor later :P
<@nirik:matrix.scrye.com>
18:41:12
Anyhow, since this is somewhat of a leaf here, lets ping them and ask and revisit next week?
<@sgallagh:fedora.im>
18:41:20
task2 was only created today, which is concerning
<@salimma:fedora.im>
18:41:20
this one is self contained so I guess yeah we can defer
<@conan_kudo:matrix.org>
18:41:21
worksforme
<@sgallagh:fedora.im>
18:41:27
I disagree
<@salimma:fedora.im>
18:41:51
I suspect some owners only get reminded when a task is opened, and this one was a bit late?
<@sgallagh:fedora.im>
18:41:54
Actually, hang on. I think I misread the upgrade section.
<@sgallagh:fedora.im>
18:42:09
Looks like existing users will be upgraded to `task2`, not v3
<@sgallagh:fedora.im>
18:42:17
So I guess it's fine to punt
<@salimma:fedora.im>
18:42:18
existing users get task2 swapped in place of task v2, yeah
<@salimma:fedora.im>
18:42:39
so you have to explicitly opt in to get to task v3, and... landing task2 first is the right order
<@sgallagh:fedora.im>
18:42:45
Yes
<@sgallagh:fedora.im>
18:43:08
So I think this is proceeding and it should be safe for existing users.
<@sgallagh:fedora.im>
18:43:15
I vote we punt and reconsider it next week.
<@conan_kudo:matrix.org>
18:43:36
worksforme
<@decathorpe:fedora.im>
18:43:45
+1
<@conan_kudo:matrix.org>
18:43:52
+1
<@jistone:fedora.im>
18:44:02
+1
<@sgallagh:fedora.im>
18:44:10
That's the last on the list that Aoife created, but there's a few more tickets on Miro's comment.
<@sgallagh:fedora.im>
18:44:31
!info Punting the decision on this Change until next week. (+5, 0, -0)
<@sgallagh:fedora.im>
18:45:04
!topic GNU Toolchain Update (gcc 14+, binutils 2.42+, glibc 2.40, gdb 14+)
<@sgallagh:fedora.im>
18:45:14
<@sgallagh:fedora.im>
18:45:22
This has been in place for months, so I think we're good here.
<@conan_kudo:matrix.org>
18:45:29
yup
<@sgallagh:fedora.im>
18:46:02
!info This has been in place since March.
<@sgallagh:fedora.im>
18:46:38
I'm skipping the ones on the list that are struck out.
<@sgallagh:fedora.im>
18:46:45
!topic Fedora Miracle
<@sgallagh:fedora.im>
18:46:52
<@sgallagh:fedora.im>
18:47:34
This being non-blocking and a leaf, we can probably just revisit at Final Freeze, I think
<@conan_kudo:matrix.org>
18:47:44
This is also pretty much ready to go
<@sgallagh:fedora.im>
18:47:59
<@conan_kudo:matrix.org>
18:48:00
All the definitions are in place and we're just waiting for composes to include it
<@conan_kudo:matrix.org>
18:48:15
well, and nirik needs to merge my PR to turn it on for 41 :P
<@conan_kudo:matrix.org>
18:48:23
but otherwise everything is ready
<@sgallagh:fedora.im>
18:48:26
Ah ok
<@sgallagh:fedora.im>
18:48:38
!info This is pretty much ready, just awaiting a compose.
<@nirik:matrix.scrye.com>
18:49:05
once we get a rawhide, yep.
<@sgallagh:fedora.im>
18:49:09
!topic LLVM 19
<@sgallagh:fedora.im>
18:49:31
<@jistone:fedora.im>
18:49:53
I asked Tom about this one -- it's building in copr now, and planning to head to koji when that's done
<@jistone:fedora.im>
18:49:56
<@conan_kudo:matrix.org>
18:50:27
I don't love that it's so late, but hey, they're trying :)
<@sgallagh:fedora.im>
18:50:35
We originally approved this with the expectation that it would be landing during Beta Freeze, so I don't see any reason to relitigate it now
<@decathorpe:fedora.im>
18:50:37
isn't LLVM always late? ...
<@conan_kudo:matrix.org>
18:50:37
another one of those misaligned schedules
<@conan_kudo:matrix.org>
18:50:44
yes
<@jistone:fedora.im>
18:50:47
right
<@conan_kudo:matrix.org>
18:50:57
for all intents and purposes it's basically where we expect it to be right now
<@jistone:fedora.im>
18:51:07
there was also a plan to have all compat packages in place early, but they ran into trouble with soname conflicts in rpm provides
<@sgallagh:fedora.im>
18:51:17
!info The late landing of LLVM was planned for in the original proposal. It will land during Beta Freeze.
<@sgallagh:fedora.im>
18:52:03
!topic fedora-repoquery tool
<@sgallagh:fedora.im>
18:52:13
<@salimma:fedora.im>
18:52:20
this is the one that's just a new package :)
<@sgallagh:fedora.im>
18:52:25
The ticket hasn't been updated, but this is in F41 stable
<@sgallagh:fedora.im>
18:52:33
So it should be ON_QA
<@salimma:fedora.im>
18:52:33
yeah, fedora-packager pulls it in now
<@nirik:matrix.scrye.com>
18:52:33
I think it's 'done'
<@sgallagh:fedora.im>
18:53:05
Oh, it was moved to ON_QA today.
<@sgallagh:fedora.im>
18:53:07
All good.
<@sgallagh:fedora.im>
18:53:17
!topic This Change is ON_QA now
<@sgallagh:fedora.im>
18:53:25
... Last one:
<@sgallagh:fedora.im>
18:53:28
!topic Switch to dnf5
<@sgallagh:fedora.im>
18:53:43
<@nirik:matrix.scrye.com>
18:54:03
I think this is done? whats left?
<@conan_kudo:matrix.org>
18:54:11
nothing really
<@conan_kudo:matrix.org>
18:54:25
even the images built with kiwi now use dnf5
<@conan_kudo:matrix.org>
18:54:38
and all the artifacts produced on every tool now include dnf5
<@nirik:matrix.scrye.com>
18:54:44
well, they were working on history and dist upgrades...
<@conan_kudo:matrix.org>
18:54:59
I hope we have that...
<@nirik:matrix.scrye.com>
18:55:23
proposal: ping maintainers for a update and revisit next week?
<@nirik:matrix.scrye.com>
18:55:33
but I don't think we want to go back at this point...
<@salimma:fedora.im>
18:56:31
I'm missing some nice-to-have third party plugins like etckeeper, but I suspect those won't be ported *until* a distro ships dnf5
<@jistone:fedora.im>
18:56:38
I guess the update will just shape how much is marketable?
<@nirik:matrix.scrye.com>
18:57:03
yeah...
<@jistone:fedora.im>
18:57:25
+1
<@salimma:fedora.im>
18:57:45
+1
<@conan_kudo:matrix.org>
18:57:48
+1
<@sgallagh:fedora.im>
18:58:14
Sure, +1
<@sgallagh:fedora.im>
18:58:41
!agreed ping maintainers for a update and revisit next week (+5, 0, -0)
<@sgallagh:fedora.im>
18:58:50
!topic Next Week's Chair
<@sgallagh:fedora.im>
18:59:10
I think I've covered my chairing responsibility for the next while :)
<@zodbot:fedora.im>
18:59:43
jistone gave a cookie to sgallagh. They now have 237 cookies, 12 of which were obtained in the Fedora 40 release cycle
<@sgallagh:fedora.im>
19:00:01
Who can take next week?
<@zodbot:fedora.im>
19:00:27
ngompa has already given cookies to sgallagh during the F40 timeframe
<@conan_kudo:matrix.org>
19:00:42
I can do it.
<@sgallagh:fedora.im>
19:01:01
Thank you
<@conan_kudo:matrix.org>
19:01:03
fortunately my flight to Germany is the day after :)
<@sgallagh:fedora.im>
19:01:29
!action Conan Kudo to chair the 2024-09-03 meeting
<@sgallagh:fedora.im>
19:01:36
!topic Open Floor
<@sgallagh:fedora.im>
19:02:44
Going once.
<@jistone:fedora.im>
19:03:11
didn't Michel Lind 🎩 have something earlier?
<@zodbot:fedora.im>
19:03:26
sgallagh gave a cookie to ngompa. They now have 136 cookies, 23 of which were obtained in the Fedora 40 release cycle
<@salimma:fedora.im>
19:03:49
oh yeah
<@salimma:fedora.im>
19:04:06
this release cycle seems a bit... tight, partly because Flock happened just before a lot of deadlines
<@salimma:fedora.im>
19:04:17
uh... how can we make sure this does not happen for f42 and later?
<@nirik:matrix.scrye.com>
19:04:57
well, flock might move. ;) (see survey!)
<@decathorpe:fedora.im>
19:05:11
survey? 👀
<@sgallagh:fedora.im>
19:05:41
Michel Lind 🎩: Well, we won't have a Flock during F42, so that helps :)
<@salimma:fedora.im>
19:05:47
yeah, still need to do the Flock survey
<@salimma:fedora.im>
19:05:51
oh true. more a F43 issue
<@jflory7:fedora.im>
19:06:03
Announcements!
<@sgallagh:fedora.im>
19:06:06
Oh dear, we summoned Justin!
<@jflory7:fedora.im>
19:06:10
The survey is super, super important 😉
<@salimma:fedora.im>
19:06:12
otoh if F42 is delayed it's a disaster in the US as we have Thanksgiving :)
<@nirik:matrix.scrye.com>
19:06:13
https://discussion.fedoraproject.org/t/your-feedback-matters-flock-to-fedora-2024-post-event-survey/129743
<@decathorpe:fedora.im>
19:06:34
I thought it was a retrospective survey, not something about the future ...
<@decathorpe:fedora.im>
19:06:50
not really clear from the title or announcement
<@jflory7:fedora.im>
19:06:51
We are testing some new ideas for 2025 in the survey. Even if you did not attend in-person, please complete it as a virtual attendee.
<@jflory7:fedora.im>
19:06:57
All feedback welcome 🙏🏻
<@sgallagh:fedora.im>
19:07:10
<@sgallagh:fedora.im>
19:07:18
!info Please fill out the survey!
<@zodbot:fedora.im>
19:07:32
jflory7 has already given cookies to sgallagh during the F40 timeframe
<@sgallagh:fedora.im>
19:08:02
Michel Lind 🎩: Is that a sufficient answer for today? We're over the two hour mark at this point.
<@salimma:fedora.im>
19:09:12
yeah
<@sgallagh:fedora.im>
19:09:23
Thanks
<@salimma:fedora.im>
19:09:29
I'm putting the scheduling as feedback for next year :)
<@sgallagh:fedora.im>
19:09:40
Alright, thank you for joining us on this epic journey today!
<@sgallagh:fedora.im>
19:09:49
!endmeeting