<@gotmax23:fedora.im>
17:02:31
!startmeeting FESCO (2026-09-15)
<@meetbot:fedora.im>
17:02:31
Meeting started at 2026-09-15 17:02:31 UTC
<@meetbot:fedora.im>
17:02:31
The Meeting name is 'FESCO (2026-09-15)'
<@gotmax23:fedora.im>
17:02:34
!meetingname fesco
<@meetbot:fedora.im>
17:02:34
The Meeting Name is now fesco
<@gotmax23:fedora.im>
17:02:37
!info Agenda: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/thread/Q45LJTBANQWBCPKEYKQUJP3J4DAFGEMU/#Q45LJTBANQWBCPKEYKQUJP3J4DAFGEMU
<@gotmax23:fedora.im>
17:02:41
!topic Init Process
<@gotmax23:fedora.im>
17:02:46
!group members fesco
<@gotmax23:fedora.im>
17:02:48
!hi
<@zodbot:fedora.im>
17:02:53
gotmax23: Maxwell G (gotmax23) - he / him / his or they / them / theirs
<@zbyszek:fedora.im>
17:02:54
!hi
<@zodbot:fedora.im>
17:02:55
zbyszek: Zbigniew Jędrzejewski-Szmek (zbyszek)
<@decathorpe:fedora.im>
17:02:58
!hi
<@zodbot:fedora.im>
17:02:59
Fabio Valentini 🌈: Fabio Valentini (decathorpe) - he / him / his
<@nirik:matrix.scrye.com>
17:03:24
morning
<@salimma:fedora.im>
17:03:47
!hi
<@zodbot:fedora.im>
17:03:48
Michel Lind ☘ UTC+1 😷: Michel Lind (salimma) - he / him / his
<@gotmax23:fedora.im>
17:05:53
oh, did zodbot never do the ping?
<@gotmax23:fedora.im>
17:05:54
!group members fesco
<@zodbot:fedora.im>
17:05:56
Members of fesco: Fabio Valentini, Máirín Duffy, gotmax23 (@gotmax:matrix.org, @gotmax23:fedora.im), Jef Spaleta, Kevin Fenzi, ngompa (@conan_kudo:matrix.org, @ngompa:fedora.im, @pharaoh_atem:opensuse.org, @ngompa:kde.org, @ngompa:almalinux.im), Michel Lind, Timothée Ravier, Simon de Vlieger, Zbigniew Jędrzejewski-Szmek
<@gotmax23:fedora.im>
17:06:28
I copy-pasted the lines really fast, but I did enter them separately this time...
<@nirik:matrix.scrye.com>
17:06:37
Yeah, the hi apparently was the only thing expanded there.
<@conan_kudo:matrix.org>
17:07:03
!hi
<@zodbot:fedora.im>
17:07:05
Conan Kudo 😌: Neal Gompa (ngompa) - he / him / his
<@gotmax23:fedora.im>
17:07:26
okay, I guess 6 will do
<@gotmax23:fedora.im>
17:07:37
!topic Next week's chair
<@salimma:fedora.im>
17:08:35
well, good news, next week my canteen is finally doing the decent thing and serving dinner early
<@salimma:fedora.im>
17:08:39
so I can chair without starving
<@gotmax23:fedora.im>
17:08:51
!action Michel Lind ☘ UTC+1 😷 to chair next week
<@gotmax23:fedora.im>
17:08:57
thank you
<@gotmax23:fedora.im>
17:09:48
it would be nice to try to rotate the chair role around to other members
<@gotmax23:fedora.im>
17:09:53
it would be nice to try to rotate the chair role around more to other members
<@gotmax23:fedora.im>
17:09:55
anyway
<@nirik:matrix.scrye.com>
17:09:56
Its good for them! (spicy take)
<@gotmax23:fedora.im>
17:10:00
!topic #3682 F45 Incomplete Changes Report
<@gotmax23:fedora.im>
17:10:11
!fesco 3682
<@zodbot:fedora.im>
17:10:12
**fesco/tickets #3682** (https://forge.fedoraproject.org/fesco/tickets/issues/3682):**F45 Incomplete Changes Report**
<@zodbot:fedora.im>
17:10:12
<@zodbot:fedora.im>
17:10:12
● **Opened:** a month ago by amoloney
<@zodbot:fedora.im>
17:10:12
● **Last Updated:** 49 minutes ago
<@zodbot:fedora.im>
17:10:12
● **Assignee:** Not Assigned
<@gotmax23:fedora.im>
17:10:15
!info Working list: https://forge.fedoraproject.org/fesco/tickets/issues/3682#issuecomment-1355057
<@gotmax23:fedora.im>
17:10:18
!topic Change: GRUB EFI for Confidential Computing
<@gotmax23:fedora.im>
17:10:22
!bug 2499989
<@zodbot:fedora.im>
17:10:24
RHBZ#2499989 (https://bugzilla.redhat.com/2499989): [Changes Tracking]: GRUB EFI for Confidential Computing
<@gotmax23:fedora.im>
17:10:26
!info Current status: Package review completed in https://bugzilla.redhat.com/show_bug.cgi?id=2513599 but not yet imported
<@gotmax23:fedora.im>
17:10:51
I just got a response from the Change owner on the bug, so it looks like they are working on it now
<@nirik:matrix.scrye.com>
17:11:16
yeah, so I think we can let this land, as long as it's not too much longer
<@conan_kudo:matrix.org>
17:11:23
I don't really care one way or another
<@nirik:matrix.scrye.com>
17:11:24
say final freeze.
<@conan_kudo:matrix.org>
17:11:34
it seemed they were debating making it a subpackage of the regular grub2 package as well
<@conan_kudo:matrix.org>
17:11:38
I saw commits in grub2 for it
<@gotmax23:fedora.im>
17:11:55
proposed info The package review is complete, and the Change Owner can work on importing/building/submitting the package for f45 before the final freeze
<@decathorpe:fedora.im>
17:11:59
it's not included on any images is it? so it would be fine to land as a post GA update in theory ...
<@decathorpe:fedora.im>
17:12:05
but ack
<@gotmax23:fedora.im>
17:12:05
yeah
<@conan_kudo:matrix.org>
17:12:07
it's not on any images
<@nirik:matrix.scrye.com>
17:12:07
humf. Thats a change in plans and is not great.
<@salimma:fedora.im>
17:12:08
ack
<@conan_kudo:matrix.org>
17:12:09
ack
<@zbyszek:fedora.im>
17:12:14
It's an independent package, so do we really care? Even if it's soon after release it'd be OK…
<@conan_kudo:matrix.org>
17:12:22
we don't really care
<@salimma:fedora.im>
17:12:30
as long as they don't suddenly ask for it to be a blocker
<@conan_kudo:matrix.org>
17:12:32
it's not a big deal how they do it as long as the outputs are what the change expects
<@nirik:matrix.scrye.com>
17:12:37
I think we only care in that if we advertise it and it's not there it's bad
<@gotmax23:fedora.im>
17:12:50
was it not always intended to be a separate source package?
<@conan_kudo:matrix.org>
17:12:56
I don't think so
<@conan_kudo:matrix.org>
17:13:05
I think it was originally supposed to just be another alternate build in grub2
<@zbyszek:fedora.im>
17:13:56
"An independent separate incarnation (package) of the GRUB bootloader"
<@zbyszek:fedora.im>
17:14:02
That's what the Change page says
<@nirik:matrix.scrye.com>
17:14:07
well, if they make it a grub2 subpackage they need to revisit the change...
<@conan_kudo:matrix.org>
17:14:27
again I don't really care
<@nirik:matrix.scrye.com>
17:14:28
but anyhow, we can just ack and move on for now...
<@conan_kudo:matrix.org>
17:14:32
these details don't matter
<@salimma:fedora.im>
17:14:39
yes, please move on
<@gotmax23:fedora.im>
17:14:48
I guess we can set the final freeze deadline for now and check again next week? The package review was approved, so I don't think there is a delay that'd need more time than that
<@gotmax23:fedora.im>
17:14:51
okay
<@gotmax23:fedora.im>
17:15:15
!info The package review is complete, and the Change Owner can work on importing/building/submitting the package for f45 before the final freeze
<@gotmax23:fedora.im>
17:15:21
!topic Change: Relocate RPM repository configs to /usr
<@gotmax23:fedora.im>
17:15:24
!bug 2481848
<@zodbot:fedora.im>
17:15:26
RHBZ#2481848 (https://bugzilla.redhat.com/2481848): [Changes Tracking]: Relocate RPM repository configs to /usr
<@gotmax23:fedora.im>
17:15:28
!info Current status: The important parts of the Change are all implemented. We will watch progress of the changes necessary to fedora-third-party.
<@conan_kudo:matrix.org>
17:15:43
I'm working on fedora-third-party code now, I'm hoping to have a PR later this week
<@conan_kudo:matrix.org>
17:16:04
I'm actually trying to get a Kinoite install going so I can test it, but the installer is choking on installing :(
<@conan_kudo:matrix.org>
17:16:11
so yay :(
<@zbyszek:fedora.im>
17:16:21
More things to debug ;)
<@conan_kudo:matrix.org>
17:16:28
yeah
<@conan_kudo:matrix.org>
17:16:35
I think it probably doesn't take 2 hours to install an ostree image
<@conan_kudo:matrix.org>
17:16:39
so something definitely broke
<@gotmax23:fedora.im>
17:17:06
shall we reuse the info from last week and revisit next week?
<@conan_kudo:matrix.org>
17:17:09
yes
<@gotmax23:fedora.im>
17:17:15
okay
<@gotmax23:fedora.im>
17:17:28
proposed info The important parts of the Change are all implemented. We will watch progress of the changes necessary to fedora-third-party.
<@conan_kudo:matrix.org>
17:17:35
ack
<@nirik:matrix.scrye.com>
17:17:48
ack
<@decathorpe:fedora.im>
17:17:50
ack
<@gotmax23:fedora.im>
17:18:17
!info The important parts of the Change are all implemented. We will watch progress of the changes necessary to fedora-third-party.
<@gotmax23:fedora.im>
17:18:30
!topic Change: oo7 Secrets Service Provider
<@gotmax23:fedora.im>
17:18:33
!bug 25036811
<@salimma:fedora.im>
17:18:42
oof
<@salimma:fedora.im>
17:18:51
I need to catch up on what got posted recently in the workstation room
<@conan_kudo:matrix.org>
17:18:51
sigh
<@conan_kudo:matrix.org>
17:19:09
it will be functional in GNOME with 51.0 megaupdate
<@conan_kudo:matrix.org>
17:19:22
Budgie is fixed
<@nirik:matrix.scrye.com>
17:19:23
fedora.exceptions.InfoGatherError: Issue querying Bugzilla: Bug #25036811 does not exist. :)
<@gotmax23:fedora.im>
17:19:26
!bug 2503681
<@zodbot:fedora.im>
17:19:28
RHBZ#2503681 (https://bugzilla.redhat.com/2503681): [Changes Tracking]: oo7 Secrets Service Provider
<@conan_kudo:matrix.org>
17:19:39
and I will be making the appropriate changes to the other desktops
<@conan_kudo:matrix.org>
17:19:49
I'm deferring KDE, it's too much for that one
<@salimma:fedora.im>
17:19:57
we should document this transition for any other desktop that comes along later, I think
<@nirik:matrix.scrye.com>
17:20:09
I got a 'your login didn't unlock the keyring' thing in rawhide recently, but haven't had a chance to look into it much
<@conan_kudo:matrix.org>
17:20:12
getting gnome-keyring out of the hotpath is critical since I expect it to go unmaintained after this cycle
<@salimma:fedora.im>
17:20:17
but yeah this is one of those "we need to do this because GNOME is dropping keyring but omg why this project is awful"
<@conan_kudo:matrix.org>
17:20:25
yes
<@gotmax23:fedora.im>
17:20:25
I'm fine with deferring KDE
<@gotmax23:fedora.im>
17:20:30
Neal do you want to make a proposal
<@gotmax23:fedora.im>
17:20:33
we'd need to vote on that
<@conan_kudo:matrix.org>
17:20:35
yep
<@conan_kudo:matrix.org>
17:21:21
proposed !agreed The oo7 transition is in progress and will be completed for all deliverables except KDE variants. The transition for KDE variants will be deferred to a future Change proposal.
<@nirik:matrix.scrye.com>
17:21:51
ack
<@zbyszek:fedora.im>
17:21:53
wfm +1
<@conan_kudo:matrix.org>
17:21:56
ack
<@salimma:fedora.im>
17:22:09
+1
<@salimma:fedora.im>
17:22:22
so KDE and Kinoite and ... any other variant?
<@decathorpe:fedora.im>
17:22:26
could it still be reverted in F45 if necessary? it sounds painful to get everything lined up ...
<@decathorpe:fedora.im>
17:22:26
but sure, ack ...
<@salimma:fedora.im>
17:22:39
I think Neal doesn't want to screw up RHEL 11
<@conan_kudo:matrix.org>
17:22:44
KDE desktop, mobile and its atomic counterparts, as well as any labs that use KDE desktops
<@salimma:fedora.im>
17:22:47
though technically we should not be concerned with RHEL here...
<@gotmax23:fedora.im>
17:22:55
+1
<@salimma:fedora.im>
17:23:12
if so far for GNOME it's just the dbus activation I think it's not too bad
<@salimma:fedora.im>
17:23:18
pray there are no more bugs
<@conan_kudo:matrix.org>
17:23:44
the tiling wms and cinnamon and mate need some kind of service activation code
<@conan_kudo:matrix.org>
17:23:56
since the pam module can't launch the daemon
<@salimma:fedora.im>
17:24:05
so we identified the commit that broke session activation right
<@conan_kudo:matrix.org>
17:24:09
yes
<@salimma:fedora.im>
17:24:13
can we.... make it runtime configurable?
<@gotmax23:fedora.im>
17:24:20
how hard is it to back this out if necessary? it sounds like some progress was made, so is it even likely that it will be necessary?
<@salimma:fedora.im>
17:24:29
since upstream wrote a lot of things with AI, for this one project I will hold my nose and spend AI token if needed
<@conan_kudo:matrix.org>
17:24:35
it would be quite hard to back it out
<@gotmax23:fedora.im>
17:24:43
vibe-coded kwallet integration that was broken worried me
<@gotmax23:fedora.im>
17:24:48
but if we're skipping KDE this cycle, okay
<@conan_kudo:matrix.org>
17:25:03
the KDE upstream folks weren't happy either when I brought it up to them
<@conan_kudo:matrix.org>
17:25:15
so there's some discussion I'm going to have there to figure out what we're doing about oo7
<@conan_kudo:matrix.org>
17:25:37
the plan is to support it... but it needs to be smooth
<@conan_kudo:matrix.org>
17:25:44
that's why I'm not committing to anything right now
<@gotmax23:fedora.im>
17:25:57
got it
<@conan_kudo:matrix.org>
17:26:00
the gnome-keyring migrations work, so that is all fine and dandy
<@salimma:fedora.im>
17:26:02
with the cheapest Claude model!
<@gotmax23:fedora.im>
17:26:10
at least upstream GNOME seems invested in making this work for GNOME
<@gotmax23:fedora.im>
17:26:21
Shall I post the decision now?
<@conan_kudo:matrix.org>
17:26:24
well, oo7 is developed by a gnome guy, so I sure hope so
<@conan_kudo:matrix.org>
17:26:25
yes
<@gotmax23:fedora.im>
17:26:26
or any other last comments?
<@salimma:fedora.im>
17:26:27
yeah I think backing it off when upstream is going full speed ahead is... just going to make more mess
<@salimma:fedora.im>
17:26:32
yes, I think we are agreed
<@gotmax23:fedora.im>
17:26:53
proposed !agreed The oo7 transition is in progress and will be completed for all deliverables except KDE variants. The transition for KDE variants will be deferred to a future Change proposal. (+6, 0, -0)
<@nirik:matrix.scrye.com>
17:27:07
ack
<@decathorpe:fedora.im>
17:27:09
ack
<@gotmax23:fedora.im>
17:27:13
!agreed The oo7 transition is in progress and will be completed for all deliverables except KDE variants. The transition for KDE variants will be deferred to a future Change proposal. (+6, 0, -0)
<@gotmax23:fedora.im>
17:27:26
!action ngompa to update Change Proposal accordingly
<@gotmax23:fedora.im>
17:27:32
!topic Change: Encapsule devel containers
<@gotmax23:fedora.im>
17:27:35
!bug 2517958
<@zodbot:fedora.im>
17:27:37
RHBZ#2517958 (https://bugzilla.redhat.com/2517958): [Changes Tracking]: Encapsule isolated devel containers
<@gotmax23:fedora.im>
17:27:37
!info Current status: This is only blocked by package review and can land at any time before final freeze.
<@gotmax23:fedora.im>
17:27:50
Shall we reuse this info from last week?
<@gotmax23:fedora.im>
17:27:57
I don't think anything has changed here
<@gotmax23:fedora.im>
17:28:10
or we can make it less strict? this is a leaf package and can be a zero day if needed
<@salimma:fedora.im>
17:28:33
this is Jens', right?
<@gotmax23:fedora.im>
17:28:36
yes
<@gotmax23:fedora.im>
17:28:45
I think it's haskell which can be hard to find reviewers for I think
<@salimma:fedora.im>
17:28:46
I can review it tomorrow if it's blocking (but I was wondering why he does not proactively seek reviewers)
<@gotmax23:fedora.im>
17:28:58
cool, thanks
<@salimma:fedora.im>
17:29:07
!action Michel will review this package
<@gotmax23:fedora.im>
17:29:29
proposed info This is only blocked by package review and can land at any time before final freeze **or as a zero-day update if needed**.
<@gotmax23:fedora.im>
17:29:32
the bolded part is new
<@gotmax23:fedora.im>
17:29:36
do we want to add that?
<@nirik:matrix.scrye.com>
17:29:52
I guess... but we would need to make sure that happened.
<@conan_kudo:matrix.org>
17:29:56
let's have it for final freeze
<@conan_kudo:matrix.org>
17:29:59
leave the bolded part out
<@decathorpe:fedora.im>
17:30:13
yeah let's keep it for until final freeze for now
<@gotmax23:fedora.im>
17:30:17
okay
<@gotmax23:fedora.im>
17:30:24
proposed info This is only blocked by package review and can land at any time before final freeze
<@gotmax23:fedora.im>
17:30:25
ack
<@conan_kudo:matrix.org>
17:30:28
ack
<@decathorpe:fedora.im>
17:30:31
ack
<@salimma:fedora.im>
17:30:38
ack
<@zbyszek:fedora.im>
17:30:47
ack
<@nirik:matrix.scrye.com>
17:31:18
ack
<@gotmax23:fedora.im>
17:31:24
!info This is only blocked by package review and can land at any time before final freeze
<@gotmax23:fedora.im>
17:31:27
!topic Change: LLVM 23
<@gotmax23:fedora.im>
17:31:30
!bug 2499980
<@zodbot:fedora.im>
17:31:32
RHBZ#2499980 (https://bugzilla.redhat.com/2499980): [Changes Tracking]: LLVM 23
<@gotmax23:fedora.im>
17:31:50
this still hasn't landed?
<@nirik:matrix.scrye.com>
17:32:02
it's in testing
<@decathorpe:fedora.im>
17:32:04
it's in rawhide at least
<@nirik:matrix.scrye.com>
17:32:33
https://bodhi.fedoraproject.org/updates/FEDORA-2026-06a5023802
<@gotmax23:fedora.im>
17:32:35
!link https://bodhi.fedoraproject.org/updates/FEDORA-2026-06a5023802
<@nirik:matrix.scrye.com>
17:32:39
having some gating issues?
<@gotmax23:fedora.im>
17:32:40
See adamw's comment
<@nirik:matrix.scrye.com>
17:32:43
but yeah
<@gotmax23:fedora.im>
17:32:53
I guess I was tagged and didn't see it 🙈
<@decathorpe:fedora.im>
17:32:58
missing tests for llvm itself? ...
<@decathorpe:fedora.im>
17:33:11
ABSENT required results for llvm
<@decathorpe:fedora.im>
17:33:15
that's not good :D
<@conan_kudo:matrix.org>
17:33:23
🙈
<@salimma:fedora.im>
17:33:42
oy
<@nirik:matrix.scrye.com>
17:33:50
Anyhow, lets let that get sorted out of meeting... and we can just check this next week?
<@conan_kudo:matrix.org>
17:34:02
they have tested defined as required for gating that don't exist anymore probably
<@decathorpe:fedora.im>
17:34:09
why is it every time that I look at a package that has custom gating tests they're borked? ...
<@gotmax23:fedora.im>
17:34:11
!fesco 3629
<@zodbot:fedora.im>
17:34:13
**fesco/tickets #3629** (https://forge.fedoraproject.org/fesco/tickets/issues/3629):**Change: LLVM 23**
<@zodbot:fedora.im>
17:34:13
<@zodbot:fedora.im>
17:34:13
● **Closed** 2 months ago
<@zodbot:fedora.im>
17:34:13
● **Opened:** 3 months ago by amoloney
<@zodbot:fedora.im>
17:34:13
● **Last Updated:** a day ago
<@zodbot:fedora.im>
17:34:13
● **Assignee:** Not Assigned
<@conan_kudo:matrix.org>
17:34:26
I am starting to wonder if we should ban them
<@conan_kudo:matrix.org>
17:34:30
nobody is maintaining them
<@conan_kudo:matrix.org>
17:34:45
it's not like debian with autopkgtests where there's a culture of discipline around them
<@gotmax23:fedora.im>
17:34:49
I maintain mine
<@salimma:fedora.im>
17:34:50
do these packagers... maintain their own gating tests?
<@gotmax23:fedora.im>
17:35:03
but for some packages they often fail and block things and nobody looks at them
<@salimma:fedora.im>
17:35:07
well yeah, in Debian you get carrot and stick
<@nirik:matrix.scrye.com>
17:35:16
There are any number of reasons test results might not be there...
<@salimma:fedora.im>
17:35:18
if you have autopkgtests it removes the amount of time your package is held back from testing
<@decathorpe:fedora.im>
17:35:44
I'd be OK with "let's poke people to look at the failing tests and look again next week if the update is stable"
<@zbyszek:fedora.im>
17:36:07
I recently learned that we get "absent" if there's more than 20 packages in the update. So maybe that's just that.
<@nirik:matrix.scrye.com>
17:36:13
but it won't be... read the comments from adam
<@gotmax23:fedora.im>
17:36:15
should we set a one week deadline?
<@gotmax23:fedora.im>
17:36:23
re-target Change if not stable by then?
<@conan_kudo:matrix.org>
17:36:35
it already has to land before final freeze
<@conan_kudo:matrix.org>
17:36:47
but yeah, I'm fine with tightening the screws on this one
<@nirik:matrix.scrye.com>
17:36:54
The timing of llvm releases has always been horrible. I'm ok with letting them land it now, but it's unfortunate...
<@gotmax23:fedora.im>
17:37:09
> FESCo will be discussing what to do, see https://forge.fedoraproject.org/fesco/tickets/issues/3682#issuecomment-1355057 . I don't think this should be pushed stable until FESCo has considered it.
<@conan_kudo:matrix.org>
17:37:12
well, unfortunately our own maintainers didn't back me up when I brought it upstream
<@conan_kudo:matrix.org>
17:37:26
so LLVM didn't take our concerns seriously
<@gotmax23:fedora.im>
17:37:43
It sounds like there was an understanding that this was supposed to be pushed during the Beta freeze, not after
<@gotmax23:fedora.im>
17:37:51
https://bodhi.fedoraproject.org/updates/FEDORA-2026-06a5023802#comment-4769172
<@decathorpe:fedora.im>
17:37:59
tighten-screws+1
<@decathorpe:fedora.im>
17:37:59
backing out from the llvm 23 update *now* would be quite annoying :(
<@salimma:fedora.im>
17:38:01
I think we should *make* our maintainers care about timelines otherwise upstream will never care
<@salimma:fedora.im>
17:38:06
so yes +1 too
<@nirik:matrix.scrye.com>
17:38:24
thats what we approved but the change says "Our plan is to push 23.1.0 into Fedora 45 during the Beta Freeze with a Beta Freeze exception." but I guess that plan was delayed...
<@conan_kudo:matrix.org>
17:38:27
I would like to say it needs to land by this Friday
<@conan_kudo:matrix.org>
17:38:37
and no later
<@zbyszek:fedora.im>
17:38:52
Proposal: The update can be pushed to stable until this Friday. Maintainers must look at the test results and fix things or waive them.
<@salimma:fedora.im>
17:38:52
right. so basically no further delays, right? if they miss the beta freeze window then that's it
<@nirik:matrix.scrye.com>
17:38:59
they may have depended on the beta being delayed?
<@conan_kudo:matrix.org>
17:39:00
yes
<@conan_kudo:matrix.org>
17:39:10
nobody should depend on the beta being delayed
<@nirik:matrix.scrye.com>
17:39:13
+1 to the proposal
<@conan_kudo:matrix.org>
17:39:19
+1 to the proposal
<@zbyszek:fedora.im>
17:39:22
IIUC, at this point it's just a matter of waving the two tests.
<@gotmax23:fedora.im>
17:39:38
I was in the middle of writing one, but okay
<@gotmax23:fedora.im>
17:39:39
+1
<@conan_kudo:matrix.org>
17:39:41
if we have people depending on delays, then we need to tighten the screws even more to make it clear that's not acceptable
<@nirik:matrix.scrye.com>
17:39:41
s/waiving/fixing/ possibly... but yeah
<@salimma:fedora.im>
17:39:46
+1
<@salimma:fedora.im>
17:39:58
yes, otherwise they have an incentive to *cause* a delay
<@decathorpe:fedora.im>
17:40:01
I don't see us hard-coding anywhere that this *had* to land *during* the beta freeze. it seems to have been the rough plan, but I don't see references to it in our decision.
<@decathorpe:fedora.im>
17:40:13
+1
<@nirik:matrix.scrye.com>
17:40:20
well, it's in the change page...
<@conan_kudo:matrix.org>
17:40:33
it was part of us accepting it, it's written in the Change itself
<@salimma:fedora.im>
17:40:42
I guess we can make it explicit that we as FESCo are deciding to hold the Change Owners to their plan
<@conan_kudo:matrix.org>
17:40:45
every cycle I grumble about it, so I definitely remember us accepting it :P
<@conan_kudo:matrix.org>
17:41:29
I also make it a point to tell the gcc people how much I appreciate them making an effort to do things on time these days :)
<@salimma:fedora.im>
17:42:00
the more we deal with LLVM the more I am glad we have people like Mamoru who valiantly fix packages that prefer Clang so they still build with GCC
<@gotmax23:fedora.im>
17:42:03
proposed !agreed The update can be pushed to stable until this Friday. Maintainers must look at the test results and address failures. (+6, 0, -0)
<@salimma:fedora.im>
17:42:08
ack
<@nirik:matrix.scrye.com>
17:42:09
ack
<@conan_kudo:matrix.org>
17:42:13
ack
<@zbyszek:fedora.im>
17:42:17
ack
<@decathorpe:fedora.im>
17:42:26
ack
<@gotmax23:fedora.im>
17:42:32
!agreed The update can be pushed to stable until this Friday. Maintainers must look at the test results and address failures. (+6, 0, -0)
<@gotmax23:fedora.im>
17:42:40
!topic #3685 Upgrade to libxml2-2.15.4 in F45
<@gotmax23:fedora.im>
17:42:43
!fesco 3685
<@zodbot:fedora.im>
17:42:46
**fesco/tickets #3685** (https://forge.fedoraproject.org/fesco/tickets/issues/3685):**Upgrade to libxml2-2.15.4 in F45**
<@zodbot:fedora.im>
17:42:46
<@zodbot:fedora.im>
17:42:46
● **Opened:** 3 hours ago by zbyszek
<@zodbot:fedora.im>
17:42:46
● **Last Updated:** 36 minutes ago
<@zodbot:fedora.im>
17:42:46
● **Assignee:** Not Assigned
<@nirik:matrix.scrye.com>
17:42:52
oh the fun one.
<@conan_kudo:matrix.org>
17:42:55
ugh
<@salimma:fedora.im>
17:43:29
a) this should not have been sprung on to the meeting last week and b) IDK how the miscommunication that this was approved for f45 happened
<@salimma:fedora.im>
17:43:45
everyone was clear last week that we approved attempting to fix f46 only and f45 is treated separately
<@zbyszek:fedora.im>
17:44:00
This is exactly what is happening.
<@zbyszek:fedora.im>
17:44:04
I don't understand the issue.
<@nirik:matrix.scrye.com>
17:44:05
yes, and this is the request for f45 now?
<@salimma:fedora.im>
17:44:21
yes but the tracking bug got changed to target f45 before this meeting
<@nirik:matrix.scrye.com>
17:44:28
rebuilding and updating ~500 packages post beta is... ugh.
<@salimma:fedora.im>
17:44:32
gotmax23changed it back, but that did happen and we should not pretend it did not
<@zbyszek:fedora.im>
17:44:59
Aoife changed the bug to f45 and I immediately reassigned it back.
<@decathorpe:fedora.im>
17:45:27
can we summon adamw or is he off today?
<@zbyszek:fedora.im>
17:45:39
I was indending to bring this up here during the meeting and didn't want to have things flipping back and forths.
<@decathorpe:fedora.im>
17:46:09
you brought it up last week after the 120 time limit was already over, that wasn't ... ideal timing
<@gotmax23:fedora.im>
17:46:30
we only agreed to re-target it to f46. that was done by the FOA since that was the meeting vote; it probably shouldn't have been changed back
<@gotmax23:fedora.im>
17:46:44
anyway, let's discuss the current state of this, since we have a new ticket now
<@gotmax23:fedora.im>
17:47:03
I am concerning about doing this post-branching this late in the cycle
<@zbyszek:fedora.im>
17:47:08
I brought it up during open floor. I know long meetings are tiring, but that's not really a reason to hold it against topics that are later in the agenda.
<@gotmax23:fedora.im>
17:47:41
I am concerned about doing this post-branching this late in the cycle
<@zbyszek:fedora.im>
17:47:54
Right, so I think that it's bad style to release F45 with an important package with known vulnerabilities unfixed. But if y'all think that it's too much hassle to update, then just vote against the proposal and let's move on.
<@nirik:matrix.scrye.com>
17:48:00
yeah, thats a ton of change after beta...
<@conan_kudo:matrix.org>
17:48:13
I personally don't actually care whether it's a lot of change or not
<@salimma:fedora.im>
17:48:19
the problem is this could have been addressed earlier
<@conan_kudo:matrix.org>
17:48:25
I have a decent amount of confidence in our infrastructure for that
<@decathorpe:fedora.im>
17:48:25
I don't understand. I see Aoife switching the tracking bug from blocking F45 to F46.
<@decathorpe:fedora.im>
17:48:25
<@decathorpe:fedora.im>
17:48:25
> Blocks: 2520385
<@decathorpe:fedora.im>
17:48:25
> No longer blocks: 2402320
<@zbyszek:fedora.im>
17:48:36
I don't think it's that much change. The packages rebuild cleanly and seem to work fine with the new version of the library.
<@salimma:fedora.im>
17:48:43
now it's too late. as I said last week, if we allow this we definitely should allow Neal's Change that got delayed because releng had a mishap
<@zbyszek:fedora.im>
17:49:46
Also, I don't like not doing the update, because it most likely means that F45 will never get updates for most libxml2 bugs, and since they keep coming at a regular clip, I think we can expect more in the next 13 months.
<@gotmax23:fedora.im>
17:49:54
the lack of communication and tardiness surrounding this Change is frustrating, but if we can get it done in time without creating a lot of disruption, meh
<@nirik:matrix.scrye.com>
17:49:57
well, by change I mean... a lot of things could have unintended changes or be broken and people could test beta and say "oh, all good"...
<@salimma:fedora.im>
17:49:59
they're not connected, but the point is we should not cut some change a lot of slack after the change owner basically torpedoed his own change by disappearing
<@gotmax23:fedora.im>
17:50:24
what are other "stable" distros doing? there's really no backports that we can borrow?
<@decathorpe:fedora.im>
17:50:37
did we get any communication from the actual Change owner other than "retarget to f46 or drop"? https://bugzilla.redhat.com/show_bug.cgi?id=2523668#c2
<@gotmax23:fedora.im>
17:50:52
it sounds like he was in favor of re-targeting it?
<@nirik:matrix.scrye.com>
17:50:56
former change owner now I guess?
<@salimma:fedora.im>
17:50:59
btw the wiki still says ChangeAcceptedF45 so depending on what we decide today we need to retarget that
<@salimma:fedora.im>
17:51:32
both Dave and zbyszek are listed as owners now so still current I guess
<@nirik:matrix.scrye.com>
17:52:28
ok.
<@conan_kudo:matrix.org>
17:53:00
indeed yes
<@conan_kudo:matrix.org>
17:53:25
like I actually don't have a problem with letting this through, but I'm really annoyed because my change was so simple and got kicked because of a releng error
<@salimma:fedora.im>
17:53:45
I just double checked, and yes, zbyszek changed it *back* to f45
<@salimma:fedora.im>
17:54:05
but let's discuss this on the merits
<@gotmax23:fedora.im>
17:54:20
yes, please
<@conan_kudo:matrix.org>
17:54:21
and really what is likely to happen is that if we don't do this now, there might be an argument to do it post-GA as a massive zero day update, which I would argue is worse
<@zbyszek:fedora.im>
17:54:23
Yes, as I wrote above, I undid the reassignment based on the understanding that it'll be discussed further this week.
<@salimma:fedora.im>
17:54:30
regardless of what happened before this, a) I trust zbyszek to do the change properly for f46 but b) I think it's too late for a rebuild for f45
<@conan_kudo:matrix.org>
17:54:52
I can't find any examples of someone dealing with these CVEs without the abi break
<@conan_kudo:matrix.org>
17:55:05
I am trying to check debian atm to see if they did something special though
<@gotmax23:fedora.im>
17:55:08
I see some commits in https://gitlab.com/redhat/centos-stream/rpms/libxml2/-/commits/c10s
<@zbyszek:fedora.im>
17:55:24
I can't find the Debian source tracker for this package :(
<@salimma:fedora.im>
17:55:50
https://tracker.debian.org/pkg/libxml2
<@conan_kudo:matrix.org>
17:56:23
it looks like debian's libxml2 packaging is kind of a weird mess?
<@conan_kudo:matrix.org>
17:56:36
`2.12.7+dfsg+really2.9.14` is... oofy
<@nirik:matrix.scrye.com>
17:56:52
+omgwtfbbq
<@conan_kudo:matrix.org>
17:56:53
looks like it's been that way for a few release cycles
<@decathorpe:fedora.im>
17:56:58
either way, I stand by my -1 for F45 if there's going to be a vote - but I need to stop looking at screens for a bit
<@salimma:fedora.im>
17:57:04
it's fixed properly in 2.15.4 at least
<@salimma:fedora.im>
17:57:09
but yeah I"m also -1
<@zbyszek:fedora.im>
17:57:17
" Accepted libxml2 2.12.7+dfsg+really2.9.14-2.1+deb13u3 (source) into proposed-updates " hmm
<@nirik:matrix.scrye.com>
17:57:41
so, with 2 -1's there's not enough of us left to pass it even if we all voted +1 right?
<@conan_kudo:matrix.org>
17:57:51
I really feel for this, we really need to do something about it
<@conan_kudo:matrix.org>
17:57:54
I just don't know what
<@decathorpe:fedora.im>
17:57:59
yup
<@nirik:matrix.scrye.com>
17:58:08
yeah, the security side of it is... troubling too.
<@decathorpe:fedora.im>
17:58:21
let's just skip Fedora 45 and pretend it never happened
<@salimma:fedora.im>
17:58:27
we can see what's in the 2.12.7/2.9.14 one https://salsa.debian.org/xml-sgml-team/libxml2/-/tree/debian/2.12.7+dfsg+really2.9.14-2.1+deb13u3?pow_referer=&ref_type=tags
<@conan_kudo:matrix.org>
17:58:30
I'm pissed because this whole situation is due to a maintainer who proposed the change and yeeted into the sunset
<@gotmax23:fedora.im>
17:58:31
maintainers of foundational libraries being non-responsive is really becoming a problem
<@gotmax23:fedora.im>
17:58:52
many of the nonresponsive maintainer tickets recently have involved important packages
<@conan_kudo:matrix.org>
17:59:04
should we trigger a mass check?
<@conan_kudo:matrix.org>
17:59:09
we haven't done those in a while
<@salimma:fedora.im>
17:59:10
and FESCo being down to 5-6 people during this critical period :(
<@nirik:matrix.scrye.com>
17:59:30
check of what?
<@salimma:fedora.im>
17:59:31
so I suspect what happened on the Debian side is it's a weird epoch bump
<@salimma:fedora.im>
17:59:41
"I want to downgrade this package while letting people have an upgrade path"
<@gotmax23:fedora.im>
17:59:44
at this point, I'm kind of inclined to approve an f45 rebuild if we can create a reasonable plan to do it and handle branch divergences and merge back changes/communicate with maintainers
<@gotmax23:fedora.im>
17:59:48
but this is really not great
<@conan_kudo:matrix.org>
17:59:51
maintainer aliveness
<@conan_kudo:matrix.org>
18:00:13
we can run a check against all maintainers in the critpath dependency web
<@conan_kudo:matrix.org>
18:00:27
and flush out all the nonresponsive ones
<@nirik:matrix.scrye.com>
18:00:30
we have never done that (aside packagers/provenpackagers every cycle we do)
<@conan_kudo:matrix.org>
18:00:35
it will be a bad day, but we might come out good on the other side
<@nirik:matrix.scrye.com>
18:00:44
we do this every cycle.
<@conan_kudo:matrix.org>
18:01:00
we don't do it for packagers every cycle
<@conan_kudo:matrix.org>
18:01:07
we only do it for proven packagers every other cycle
<@salimma:fedora.im>
18:01:09
for critpath it does make sense, if a package is undermaintained we can focus some attention on those / encourage them to take on comaintainers
<@nirik:matrix.scrye.com>
18:01:20
https://docs.fedoraproject.org/en-US/fesco/Policy_for_inactive_packagers/
<@salimma:fedora.im>
18:01:23
and group maintainers don't count if it's a large packaging SIG
<@nirik:matrix.scrye.com>
18:01:35
it's every cycle. I process them. ;)
<@salimma:fedora.im>
18:01:38
anyway, should we discuss that later and wrap up deciding on libxml215 now?
<@salimma:fedora.im>
18:01:44
we're one hour in
<@conan_kudo:matrix.org>
18:01:59
I'm okay with doing the rebuild, but... with a lot of finger wagging
<@conan_kudo:matrix.org>
18:02:04
this is a really screwed up situation
<@zbyszek:fedora.im>
18:02:13
It is.
<@nirik:matrix.scrye.com>
18:02:29
it's not able to pass now, hand back to voting in ticket?
<@zbyszek:fedora.im>
18:02:33
It'd have been better to notice this issue brewing a month ago and take action then. Alas.
<@conan_kudo:matrix.org>
18:02:44
yeah let's push the voting back to the ticket
<@salimma:fedora.im>
18:03:07
maybe we can email the absent members directly and strongly encourage them to vote
<@decathorpe:fedora.im>
18:03:15
and ping QA
<@conan_kudo:matrix.org>
18:03:15
yes, that would be nice
<@zbyszek:fedora.im>
18:03:16
gotmax23: please check if my answer in the ticket is satisfactory. I'm not sure if I got the point of the question.
<@salimma:fedora.im>
18:03:19
otherwise we have not been able to make hard decisions for the whole month
<@gotmax23:fedora.im>
18:03:25
Maybe we can frame this as an Updates Policy exception
<@conan_kudo:matrix.org>
18:03:34
it probably would be
<@gotmax23:fedora.im>
18:03:34
and yes, I think we need to move voting back to the ticket, since we're deadlocked
<@gotmax23:fedora.im>
18:03:41
also +1 to involving QA
<@conan_kudo:matrix.org>
18:03:43
I suspect if we don't do it now, it will wind up being that later
<@gotmax23:fedora.im>
18:03:48
I already tagged them in the ticket
<@salimma:fedora.im>
18:03:56
ok I'm willing to vote +1 if QA is OK supporting this
<@salimma:fedora.im>
18:04:03
so I'll wait for Adam to reply before casting a vote
<@gotmax23:fedora.im>
18:04:15
proposed !info we will move voting back to the ticket due to lack of a majority
<@nirik:matrix.scrye.com>
18:04:21
ack
<@decathorpe:fedora.im>
18:04:35
ack
<@conan_kudo:matrix.org>
18:04:42
ack
<@salimma:fedora.im>
18:04:42
ack
<@zbyszek:fedora.im>
18:04:48
ack
<@gotmax23:fedora.im>
18:05:01
!info we will move voting back to the ticket due to lack of a majority
<@gotmax23:fedora.im>
18:05:20
!action let's figure out a way to prevent cases like this in the future in a separate ticket
<@gotmax23:fedora.im>
18:05:32
!topic Council happenings
<@gotmax23:fedora.im>
18:05:40
I posted a short summary in the meeting agenda
<@salimma:fedora.im>
18:05:41
"fun"
<@gotmax23:fedora.im>
18:05:52
Unless there's any commentary here, I suppose we can !endmeeting
<@gotmax23:fedora.im>
18:06:09
I did not plan for Open Floor this meeting due to time constraints
<@gotmax23:fedora.im>
18:06:29
and the only other agenda addition was the one we just discussed
<@salimma:fedora.im>
18:06:34
the meeting is supposed to be... 90 mins? I don't mind ending early though
<@nirik:matrix.scrye.com>
18:06:51
it's blocked out for 2 hours I think... but yeah, early is good. ;)
<@conan_kudo:matrix.org>
18:07:08
it's a 120 minute meeting
<@conan_kudo:matrix.org>
18:07:14
but ending early is always good
<@salimma:fedora.im>
18:07:17
yeah - double checked, I am sure it was at least 90 mins but it is indeed 120 mins
<@gotmax23:fedora.im>
18:07:30
!info - Final draft of Fedora Forge usage policy:
<@gotmax23:fedora.im>
18:07:30
https://forge.fedoraproject.org/council/tickets/issues/558#issuecomment-1355304
<@gotmax23:fedora.im>
18:07:39
!info - Ongoing discussions on Conflicts of Interest:
<@gotmax23:fedora.im>
18:07:39
https://discussion.fedoraproject.org/t/conflict-of-interest-guidelines-policy-draft-feedback-please/197508
<@gotmax23:fedora.im>
18:07:53
!info - Ongoing discussions on Innovation Lifecycle Proposal. There's another meeting on Thursday
<@gotmax23:fedora.im>
18:07:59
!info - Fedora's & Digital Public Goods Alliance:
<@gotmax23:fedora.im>
18:07:59
https://forge.fedoraproject.org/council/tickets/issues/576
<@gotmax23:fedora.im>
18:08:04
!info - Draft Flock to Fedora 2027 CfP Themes will be brought to FESCo:
<@gotmax23:fedora.im>
18:08:04
https://forge.fedoraproject.org/council/tickets/issues/578
<@gotmax23:fedora.im>
18:08:35
If someone has a pressing open floor topic, I suppose we do technically have time for it
<@salimma:fedora.im>
18:09:08
if nobody raises something for a couple of minutes I guess it's fair to end
<@conan_kudo:matrix.org>
18:09:34
let's end it all :)
<@conan_kudo:matrix.org>
18:09:38
I want to go for a walk
<@gotmax23:fedora.im>
18:09:42
okay, bye everyone
<@gotmax23:fedora.im>
18:09:46
!endmeeting