<@adamwill:fedora.im>
16:01:08
!startmeeting F42-blocker-review
<@meetbot:fedora.im>
16:01:08
Meeting started at 2025-03-31 16:01:08 UTC
<@meetbot:fedora.im>
16:01:09
The Meeting name is 'F42-blocker-review'
<@adamwill:fedora.im>
16:01:12
!topic Roll Call
<@adamwill:fedora.im>
16:01:16
who's around for blocker fun?
<@boniboyblue:fedora.im>
16:02:05
!hi
<@zodbot:fedora.im>
16:02:06
Christopher Boni (boniboyblue)
<@geraldosimiao:matrix.org>
16:02:11
!hi
<@zodbot:fedora.im>
16:02:14
Geraldo S. Simião Kutz (geraldosimiao) - he / him / his
<@derekenz:fedora.im>
16:02:17
!hi
<@zodbot:fedora.im>
16:02:18
Derek Enz (derekenz)
<@decathorpe:fedora.im>
16:03:04
!hi
<@zodbot:fedora.im>
16:03:05
Fabio Valentini (decathorpe) - he / him / his
<@lruzicka:matrix.org>
16:03:40
Aloha
<@adamwill:fedora.im>
16:04:57
alrighty, let's get rolling with some shiny shiny boilerplate
<@adamwill:fedora.im>
16:05:39
!topic Introduction
<@adamwill:fedora.im>
16:05:39
!info The criteria for release blocking bugs can be found at:
<@adamwill:fedora.im>
16:05:39
!link https://fedoraproject.org/wiki/Fedora_42_Final_Release_Criteria
<@adamwill:fedora.im>
16:05:39
!link https://fedoraproject.org/wiki/Fedora_42_Beta_Release_Criteria
<@adamwill:fedora.im>
16:05:39
!link http://qa.fedoraproject.org/blockerbugs/current
<@adamwill:fedora.im>
16:05:39
!info The bugs up for review today are available at:
<@adamwill:fedora.im>
16:05:39
!link https://fedoraproject.org/wiki/QA:SOP_Blocker_Bug_Meeting
<@adamwill:fedora.im>
16:05:39
!info We'll be following the process outlined at:
<@adamwill:fedora.im>
16:05:39
!link https://fedoraproject.org/wiki/Basic_Release_Criteria
<@adamwill:fedora.im>
16:05:39
!info Our purpose in this meeting is to review proposed blocker and nice-to-have bugs and decide whether to accept them, and to monitor the progress of fixing existing accepted blocker and nice-to-have bugs.
<@adamwill:fedora.im>
16:05:39
Why are we here?
<@adamwill:fedora.im>
16:06:06
!info for F42 Final, we have:
<@adamwill:fedora.im>
16:06:07
!info 8 Accepted Blockers
<@adamwill:fedora.im>
16:06:07
!info 3 Proposed Blockers
<@adamwill:fedora.im>
16:06:11
!info 2 Accepted Freeze Exceptions
<@adamwill:fedora.im>
16:06:11
!info 1 Proposed Freeze Exceptions
<@adamwill:fedora.im>
16:06:21
who wants to secretarialize?
<@adamwill:fedora.im>
16:09:33
...or me
<@adamwill:fedora.im>
16:09:38
!info adamw will secretarialize
<@adamwill:fedora.im>
16:09:48
let's get started with:
<@adamwill:fedora.im>
16:09:54
!topic Proposed Final blockers
<@adamwill:fedora.im>
16:10:02
!link https://pagure.io/fedora-qa/blocker-review/issue/1817
<@adamwill:fedora.im>
16:10:02
!info Proposed Blocker, abrt, NEW
<@adamwill:fedora.im>
16:10:02
!topic (2356237) "Package isn't signed with proper key" for all packages
<@adamwill:fedora.im>
16:10:02
!link https://bugzilla.redhat.com/show_bug.cgi?id=2356237
<@adamwill:fedora.im>
16:10:05
sounds like this one is brand new
<@lbrabec:matrix.org>
16:10:19
!hi
<@zodbot:fedora.im>
16:10:20
Lukas Brabec (lbrabec)
<@lbrabec:matrix.org>
16:10:33
Sorry, I'm late again...
<@geraldosimiao:matrix.org>
16:11:02
just in time for the fun, dont worry
<@adamwill:fedora.im>
16:11:23
feel like secretarying, Lukas Brabec ?
<@lbrabec:matrix.org>
16:11:37
sure
<@adamwill:fedora.im>
16:12:55
!info in a late substitution, Lukas Brabec will secretarialize
<@adamwill:fedora.im>
16:13:08
so, i'd be +1 on this bug in theory, but seems like reproducing isn't easy so far?
<@kparal:matrix.org>
16:13:37
it's very easy for me
<@kparal:matrix.org>
16:13:42
just install the latest compose
<@geraldosimiao:matrix.org>
16:14:35
it is "testable" on a live session?
<@geraldosimiao:matrix.org>
16:14:42
I'll give it a try
<@adamwill:fedora.im>
16:15:01
should be, i think
<@adamwill:fedora.im>
16:15:08
if you have enough ram and crash something small
<@geraldosimiao:matrix.org>
16:15:19
we already have Fedora-Workstation-Live-42-20250331.n.0.x86_64.iso
<@kparal:matrix.org>
16:15:21
adamw: FYI, I just proposed a second abrt blocker
<@adamwill:fedora.im>
16:15:40
yay
<@boniboyblue:fedora.im>
16:15:42
I tried on 20250329n and couldn't replicate.
<@kparal:matrix.org>
16:16:36
Signature : (none)
<@kparal:matrix.org>
16:16:36
kparal@fedora:~$ rpm -qi gnome-calculator | grep Signature
<@kparal:matrix.org>
16:16:41
that's the broken session
<@kparal:matrix.org>
16:16:54
kparal@localhost-live:~$ rpm -qi gnome-calculator | grep Signature
<@kparal:matrix.org>
16:16:54
this is the working session:
<@kparal:matrix.org>
16:16:54
```
<@kparal:matrix.org>
16:16:54
Signature : RSA/SHA256, Mon 24 Mar 2025 02:05:45 PM CET, Key ID c8ac4916105ef944
<@kparal:matrix.org>
16:16:54
```
<@kparal:matrix.org>
16:17:13
so yay, we have unsigned packages when I install them from the live image, for some reason
<@adamwill:fedora.im>
16:17:17
so...the bug is that the package is not signed?
<@lruzicka:matrix.org>
16:17:44
Seems not abrty to me.
<@kparal:matrix.org>
16:18:10
hmm, when I downloaded the iso before the compose was finalized, could it affect it?
<@adamwill:fedora.im>
16:18:23
shouldn't do, no
<@adamwill:fedora.im>
16:18:36
the package at https://kojipkgs.fedoraproject.org/compose/branched/Fedora-42-20250331.n.0/compose/Everything/x86_64/os/Packages/g/gnome-calculator-48.0.2-1.fc42.x86_64.rpm is signed for sure
<@kparal:matrix.org>
16:19:20
sha256sum matches on my ISO
<@kparal:matrix.org>
16:19:48
I just booted the Live image, the package is not signed
<@kparal:matrix.org>
16:20:08
so, not ABRT's fault
<@kparal:matrix.org>
16:20:19
possibly still a problem, though 🙂
<@kparal:matrix.org>
16:21:09
nothing is signed there
<@kparal:matrix.org>
16:21:20
it seems, I just tested a few random packages
<@farribeiro:matrix.org>
16:21:34
!ho
<@nirik:matrix.scrye.com>
16:21:39
So, I suspect this is an issue we have had where kiwi gets the buildroot repo enabled, so it installs unsigned stuff from it. ;(
<@nirik:matrix.scrye.com>
16:21:52
I hate this bug, but it's probibly back.
<@farribeiro:matrix.org>
16:21:53
!hi
<@zodbot:fedora.im>
16:21:55
Fábio Ribeiro (farribeiro) - he / him / his
<@adamwill:fedora.im>
16:21:56
fun
<@adamwill:fedora.im>
16:22:16
so, hmm, this is awkward
<@adamwill:fedora.im>
16:22:33
unsigned packages on the release ISOs would definitely be blocker-y for me
<@adamwill:fedora.im>
16:22:41
but we don't have the release ISOs yet...
<@nirik:matrix.scrye.com>
16:23:02
hum, rawhide seems ok in container tho. so I could be wrong.
<@adamwill:fedora.im>
16:23:13
i think i'd still want to take this as +1 blocker to indicate 'we want to be fairly sure this isn't gonna happen to the final compose'...
<@nirik:matrix.scrye.com>
16:23:16
or is it only lives?
<@adamwill:fedora.im>
16:23:28
i don't think we know for sure yet
<@adamwill:fedora.im>
16:23:38
the ws live from today's 42 compose is the image we know is affected
<@kparal:matrix.org>
16:23:38
I updated the bug
<@geraldosimiao:matrix.org>
16:23:57
kamil said its on a fresh installed system
<@lruzicka:matrix.org>
16:23:57
Punt and see how it will be on newer composes?
<@kparal:matrix.org>
16:23:59
yesterday's compose
<@kparal:matrix.org>
16:24:12
0330, to be exact
<@lruzicka:matrix.org>
16:24:30
I will check the 0331 in a couple of minutes
<@kparal:matrix.org>
16:24:36
upgrading the packages from bodhi makes then signed, so this is a compose bug
<@adamwill:fedora.im>
16:24:38
ah, sorry
<@geraldosimiao:matrix.org>
16:25:22
strange, I tested here on live iso and ABRT don't even get the crashed gnome-calculator
<@nirik:matrix.scrye.com>
16:25:27
yeah, not sure whats going on from a quick glance, but we can dig more
<@adamwill:fedora.im>
16:25:46
geraldosimiao you should jsut be able to do `rpm -qi gnome-calculator` and see whether it shows a signature
<@geraldosimiao:matrix.org>
16:26:27
Version : 48.0.2
<@geraldosimiao:matrix.org>
16:26:27
Name : gnome-calculator
<@geraldosimiao:matrix.org>
16:26:27
Release : 1.fc42
<@geraldosimiao:matrix.org>
16:26:27
Architecture: x86_64
<@geraldosimiao:matrix.org>
16:26:27
Install Date: Mon 31 Mar 2025 09:45:09 AM UTC
<@geraldosimiao:matrix.org>
16:26:27
Group : Unspecified
<@geraldosimiao:matrix.org>
16:26:27
Size : 7092212
<@geraldosimiao:matrix.org>
16:26:27
License : GPL-3.0-or-later AND CC-BY-SA-3.0 AND CC0-1.0
<@geraldosimiao:matrix.org>
16:26:27
Signature : RSA/SHA256, Mon 24 Mar 2025 01:05:45 PM UTC, Key ID c8ac4916105ef944
<@geraldosimiao:matrix.org>
16:26:27
Summary : A desktop calculator
<@geraldosimiao:matrix.org>
16:26:27
Bug URL : https://bugz.fedoraproject.org/gnome-calculator
<@geraldosimiao:matrix.org>
16:26:27
URL : https://wiki.gnome.org/Apps/Calculator
<@geraldosimiao:matrix.org>
16:26:27
Vendor : Fedora Project
<@geraldosimiao:matrix.org>
16:26:27
Packager : Fedora Project
<@geraldosimiao:matrix.org>
16:26:27
Build Host : buildvm-x86-09.iad2.fedoraproject.org
<@geraldosimiao:matrix.org>
16:26:27
Build Date : Mon 24 Mar 2025 10:48:18 AM UTC
<@geraldosimiao:matrix.org>
16:26:27
Source RPM : gnome-calculator-48.0.2-1.fc42.src.rpm
<@boniboyblue:fedora.im>
16:26:30
I don't have this issue on 20250329.
<@geraldosimiao:matrix.org>
16:26:48
thats on Fedora-Workstation-Live-42-20250331.n.0.x86_64.iso
<@adamwill:fedora.im>
16:26:53
huh
<@adamwill:fedora.im>
16:26:55
so only 0330 is bad>
<@adamwill:fedora.im>
16:27:00
so only 0330 is bad?
<@kparal:matrix.org>
16:27:05
it seems so
<@geraldosimiao:matrix.org>
16:27:07
so it seems
<@kparal:matrix.org>
16:27:19
I can test 0330 KDE in a moment
<@kparal:matrix.org>
16:27:28
too bad we don't have rpm on netinst-style images
<@adamwill:fedora.im>
16:28:02
https://kojipkgs.fedoraproject.org//work/tasks/1910/130901910/root.log shows all packages as coming from the repo 'c76fb725f9c04d1eb18c01135c708e15'
<@adamwill:fedora.im>
16:28:09
we wipe the rpm db on netinsts, iirc
<@adamwill:fedora.im>
16:28:11
to save space
<@kparal:matrix.org>
16:28:53
I know, maybe we should just stop. But that's for another discussion 🙂
<@kparal:matrix.org>
16:29:39
0330 KDE seems signed
<@conan_kudo:matrix.org>
16:30:01
!hi
<@kparal:matrix.org>
16:30:09
am I just lucky or what, to select the single broken image in a long time 🙂
<@zodbot:fedora.im>
16:30:16
Neal Gompa (ngompa) - he / him / his
<@geraldosimiao:matrix.org>
16:30:24
confirmed here, only 0330 workstation
<@geraldosimiao:matrix.org>
16:30:32
rpm -qi gnome-calculator
<@adamwill:fedora.im>
16:30:35
0330 and 0331 both seem to have the same repo config - the compose repo, plus a f42-kiwi-build repo
<@geraldosimiao:matrix.org>
16:30:40
Release : 1.fc42
<@geraldosimiao:matrix.org>
16:30:40
Bug URL : https://bugz.fedoraproject.org/gnome-calculator
<@geraldosimiao:matrix.org>
16:30:40
URL : https://wiki.gnome.org/Apps/Calculator
<@geraldosimiao:matrix.org>
16:30:40
Vendor : Fedora Project
<@geraldosimiao:matrix.org>
16:30:40
Packager : Fedora Project
<@geraldosimiao:matrix.org>
16:30:40
Build Host : buildvm-x86-09.iad2.fedoraproject.org
<@geraldosimiao:matrix.org>
16:30:40
Build Date : Mon 24 Mar 2025 10:48:18 AM UTC
<@geraldosimiao:matrix.org>
16:30:40
Source RPM : gnome-calculator-48.0.2-1.fc42.src.rpm
<@geraldosimiao:matrix.org>
16:30:40
Summary : A desktop calculator
<@geraldosimiao:matrix.org>
16:30:40
Signature : (none)
<@geraldosimiao:matrix.org>
16:30:40
License : GPL-3.0-or-later AND CC-BY-SA-3.0 AND CC0-1.0
<@geraldosimiao:matrix.org>
16:30:40
Size : 7092212
<@geraldosimiao:matrix.org>
16:30:40
Name : gnome-calculator
<@geraldosimiao:matrix.org>
16:30:40
Version : 48.0.2
<@geraldosimiao:matrix.org>
16:30:40
Architecture: x86_64
<@geraldosimiao:matrix.org>
16:30:40
Install Date: Sun 30 Mar 2025 08:39:32 AM UTC
<@geraldosimiao:matrix.org>
16:30:40
Group : Unspecified
<@lruzicka:matrix.org>
16:30:43
That's your attitude.
<@geraldosimiao:matrix.org>
16:30:47
from the 0330 iso
<@kparal:matrix.org>
16:30:50
firefox is signed on KDE, not signed on Workstation, both 0330
<@kparal:matrix.org>
16:31:41
something failed during the compose process. If 0331 is ok, I guess we can just skip this bug (and make sure the final images are OK)
<@kparal:matrix.org>
16:31:55
that's a good check for automation, I guess
<@adamwill:fedora.im>
16:32:02
yeah...
<@lruzicka:matrix.org>
16:32:04
0331 is signed
<@lruzicka:matrix.org>
16:32:14
at least for gnome-calculator
<@adamwill:fedora.im>
16:32:23
i can stuff a check in openqa, i guess
<@lruzicka:matrix.org>
16:32:28
and firefox
<@lruzicka:matrix.org>
16:32:36
I can do it.
<@adamwill:fedora.im>
16:32:44
no! i can!
<@adamwill:fedora.im>
16:32:44
:D
<@lruzicka:matrix.org>
16:33:00
as you wish
<@adamwill:fedora.im>
16:33:02
(if you do it i will nitpick your approach for two weeks and then forget about the pr for two months)
<@lruzicka:matrix.org>
16:33:27
yeah, that's possible :D
<@geraldosimiao:matrix.org>
16:34:23
gnome-software have same behavior (not signed at 0330 and signed at 0331)
<@adamwill:fedora.im>
16:35:12
ok, so...shall we declare this not exactly a blocker, but try and remember to keep an eye on it for final?
<@adamwill:fedora.im>
16:35:21
and ask neal and kevin to try and figure out what happened?
<@geraldosimiao:matrix.org>
16:35:27
so punt?
<@adamwill:fedora.im>
16:35:33
i'd say rejected
<@geraldosimiao:matrix.org>
16:35:38
ok
<@adamwill:fedora.im>
16:35:42
but...punt might work
<@geraldosimiao:matrix.org>
16:35:43
FB -1
<@kparal:matrix.org>
16:36:03
blocker -1, at least until we see it repeat
<@lruzicka:matrix.org>
16:37:24
-1 and watch out for the next one
<@boniboyblue:fedora.im>
16:37:55
FinalBlocker -1
<@conan_kudo:matrix.org>
16:38:04
-1
<@adamwill:fedora.im>
16:38:49
proposed !agreed 2356237 - RejectedBlocker (Final) - this appears to have been a blip with unsigned packages appearing in a single ISO in a single compose. If we see it repeat, or figure out the cause and decide it endangers the official compose, we will re-consider this. We will check this in all RC releases before signing off
<@geraldosimiao:matrix.org>
16:39:04
ack
<@derekenz:fedora.im>
16:39:08
ack
<@boniboyblue:fedora.im>
16:39:13
ack
<@farribeiro:matrix.org>
16:39:33
Ack
<@adamwill:fedora.im>
16:39:43
!agreed 2356237 - RejectedBlocker (Final) - this appears to have been a blip with unsigned packages appearing in a single ISO in a single compose. If we see it repeat, or figure out the cause and decide it endangers the official compose, we will re-consider this. We will check this in all RC releases before signing off
<@adamwill:fedora.im>
16:40:08
!info Ticket vote: FinalBlocker (+4,0,-0) (+geraldosimiao, +boniboyblue, +derekenz, +nielsenb)
<@adamwill:fedora.im>
16:40:08
!link https://bugzilla.redhat.com/show_bug.cgi?id=2354798
<@adamwill:fedora.im>
16:40:08
!topic (2354798) "no usable disks" after re-creating MDRAID
<@adamwill:fedora.im>
16:40:08
!info Proposed Blocker, anaconda, NEW
<@adamwill:fedora.im>
16:40:08
!link https://pagure.io/fedora-qa/blocker-review/issue/1809
<@adamwill:fedora.im>
16:40:19
this has +4, but i left it undecided as i kinda wanted to see if kparal can answer my question
<@kparal:matrix.org>
16:40:39
I'm trying to test it now
<@kparal:matrix.org>
16:40:57
I'll need a few minutes
<@adamwill:fedora.im>
16:40:58
and i wanted to cite what I vaguely recall as a precedent...when gtkui was new, i think we adopted a convention that we wouldn't take bugs as blockers if, to trigger them, you had to do something wrong and attempt to fix it, all in one install attempt
<@adamwill:fedora.im>
16:41:21
this was kind of a practical thing because otherwise we would never have shipped the damn thing, iirc
<@geraldosimiao:matrix.org>
16:42:20
humm thats a wise procedure
<@conan_kudo:matrix.org>
16:43:17
though back then our install DVD still existed and failure cases were doom-worthy
<@adamwill:fedora.im>
16:43:18
deciding advanced storage blockers gets pretty squishy because you're almost always going to be able to break it if you poke enough buttons
<@adamwill:fedora.im>
16:43:36
we still have an install dvd, though it doesn't have webui on it yet
<@kparal:matrix.org>
16:43:53
gah, found more bugs...
<@geraldosimiao:matrix.org>
16:44:12
someone do somthing about Kamil...
<@geraldosimiao:matrix.org>
16:44:20
😁
<@adamwill:fedora.im>
16:44:29
we've found the *real* blocker
<@kparal:matrix.org>
16:45:04
is it me?
<@kparal:matrix.org>
16:45:25
I think I reproduced the error even when the RAID was already existing
<@derekenz:fedora.im>
16:45:47
Yeah its Fedora thats bugged lol
<@geraldosimiao:matrix.org>
16:45:57
blocker bug meetings: the place where the real QA fun takes place ;)
<@kparal:matrix.org>
16:48:34
yes, the bug is there even for pre-existing RAID
<@adamwill:fedora.im>
16:48:44
ok
<@kparal:matrix.org>
16:48:46
boot with a RAID present, delete it, create a new one -> no usable disks found
<@adamwill:fedora.im>
16:48:47
with that i'm fine to be +1
<@adamwill:fedora.im>
16:49:17
at least until the devs argue :D
<@adamwill:fedora.im>
16:49:27
so, anyone want to vote -1? we have sufficient +1s to accept atm
<@boniboyblue:fedora.im>
16:49:49
Nah, I'll stick with my +1.
<@kparal:matrix.org>
16:49:58
+1 blocker
<@derekenz:fedora.im>
16:50:00
Im good
<@farribeiro:matrix.org>
16:50:05
+1
<@geraldosimiao:matrix.org>
16:50:53
+1
<@adamwill:fedora.im>
16:51:29
proposed !agreed 2354798 - AcceptedBlocker (Final) - this is accepted as a violation of the Beta and Final custom partitioning criteria
<@adamwill:fedora.im>
16:51:45
ehhh
<@conan_kudo:matrix.org>
16:52:06
ack
<@geraldosimiao:matrix.org>
16:52:10
ack
<@derekenz:fedora.im>
16:52:11
ack
<@farribeiro:matrix.org>
16:52:13
Ack
<@boniboyblue:fedora.im>
16:52:34
ack
<@geraldosimiao:matrix.org>
16:53:16
and maybe a waive candidate for latter? based on what adamw said about the former experience on the gktui?
<@adamwill:fedora.im>
16:53:30
proposed !agreed 2354798 - AcceptedBlocker (Final) - this is accepted as a violation of Beta custom partition criterion "...the installer must be able to: Correctly interpret, and modify as described below ... software RAID arrays at RAID levels 0, 1 and 5 containing ext4 partitions ... Remove existing storage volumes ... Create mount points backed by ext4 partitions, LVM volumes or btrfs volumes, or software RAID arrays at RAID levels 0, 1 and 5 containing ext4 partitions"
<@adamwill:fedora.im>
16:53:41
there, i'll be less lazy
<@kparal:matrix.org>
16:54:03
ack
<@boniboyblue:fedora.im>
16:54:12
Ack.
<@derekenz:fedora.im>
16:54:14
ack
<@lruzicka:matrix.org>
16:54:21
ack
<@geraldosimiao:matrix.org>
16:54:27
ack
<@adamwill:fedora.im>
16:54:35
!agreed 2354798 - AcceptedBlocker (Final) - this is accepted as a violation of Beta custom partition criterion "...the installer must be able to: Correctly interpret, and modify as described below ... software RAID arrays at RAID levels 0, 1 and 5 containing ext4 partitions ... Remove existing storage volumes ... Create mount points backed by ext4 partitions, LVM volumes or btrfs volumes, or software RAID arrays at RAID levels 0, 1 and 5 containing ext4 partitions"
<@adamwill:fedora.im>
16:54:52
!link https://pagure.io/fedora-qa/blocker-review/issue/1818
<@adamwill:fedora.im>
16:54:52
!topic (2356257) Can't open settings to adjust Bugzilla apikey
<@adamwill:fedora.im>
16:54:52
!info Proposed Blocker, gnome-abrt, NEW
<@adamwill:fedora.im>
16:54:52
!link https://bugzilla.redhat.com/show_bug.cgi?id=2356257
<@adamwill:fedora.im>
16:54:55
another kparal special
<@kparal:matrix.org>
16:55:39
freshly from the oven
<@geraldosimiao:matrix.org>
16:55:39
strange that I didn't noticed this before, abrt now really don't have a "config" button
<@derekenz:fedora.im>
16:56:22
Yes can confirm
<@adamwill:fedora.im>
16:57:07
what's under the burger menu there?
<@adamwill:fedora.im>
16:57:11
just the list of crashes?
<@boniboyblue:fedora.im>
16:57:18
Only About.
<@adamwill:fedora.im>
16:57:22
heh
<@geraldosimiao:matrix.org>
16:57:23
so how can a user set this the graphical way?
<@geraldosimiao:matrix.org>
16:57:37
only in CLI
<@lbrabec:matrix.org>
16:58:36
https://github.com/abrt/gnome-abrt/issues/370
<@kparal:matrix.org>
16:58:44
this not only affects expired apikeys, but also the cases where the user typed in some invalid string sequence during their first bug report, I guess
<@kparal:matrix.org>
16:59:44
I can test it, but I think it would make much difference, it's just an additional use case
<@kparal:matrix.org>
16:59:55
I can test it, but I don't think it would make much difference, it's just an additional use case
<@adamwill:fedora.im>
17:00:33
mmmm. i would say this seems like it can be fixed pretty well with a post-release update...
<@adamwill:fedora.im>
17:00:41
not *sure* it's bad enough to block release, though it's obviously sucky
<@lruzicka:matrix.org>
17:01:10
It depends whether your API key still works.
<@lruzicka:matrix.org>
17:01:35
If not, the app is really useless and it would violate the criterion, won't it?
<@adamwill:fedora.im>
17:02:21
i mean, i guess that's the decision we're making
<@adamwill:fedora.im>
17:02:30
how commonly will someone enter a no-longer-valid API key
<@adamwill:fedora.im>
17:02:31
or typo it
<@kparal:matrix.org>
17:02:33
we should also consider upgrades
<@conan_kudo:matrix.org>
17:02:38
I would probably consider it violated if it's not possible to set an API key
<@conan_kudo:matrix.org>
17:02:58
since that makes the whole problem reporting flow broken
<@derekenz:fedora.im>
17:03:06
Agreed
<@geraldosimiao:matrix.org>
17:03:13
as far as I see, we canot entre the first one for the start
<@kparal:matrix.org>
17:03:16
the apikeys now have one year expiration time. So if somebody wants to report a bug after a long time, and they upgraded to F42 before that, they can no longer report the bug
<@geraldosimiao:matrix.org>
17:03:19
only using the CLI
<@geraldosimiao:matrix.org>
17:03:49
using the graphical interface it isnt possible by now
<@geraldosimiao:matrix.org>
17:04:12
or if it is a new user
<@kparal:matrix.org>
17:04:41
a new user can enter his/her first apikey just fine
<@adamwill:fedora.im>
17:04:47
for a new user you get one shot to enter your API key and get it right
<@boniboyblue:fedora.im>
17:05:08
If the key does expiry does it prompt for a new one?
<@kparal:matrix.org>
17:05:29
no. At least for revoked keys. I can't wait a year to test an expired one 🙂
<@geraldosimiao:matrix.org>
17:07:03
but where? if the GUI doesn't have a button os place to set it?
<@adamwill:fedora.im>
17:07:19
a prompt comes up
<@kparal:matrix.org>
17:07:22
The first apikey request pops up a dialog for you
<@adamwill:fedora.im>
17:07:25
if there's no API set at all
<@geraldosimiao:matrix.org>
17:07:30
oh, ok then
<@adamwill:fedora.im>
17:07:37
if there's no API key set at all
<@geraldosimiao:matrix.org>
17:07:38
its a one shot thing
<@adamwill:fedora.im>
17:08:03
but if there is an API key that's set but isn't valid for whatever reason (typoed, expired, revoked) it tells you the report failed but doesn't prompt you to change it (which is arguably also a bug, but hey)
<@kparal:matrix.org>
17:08:35
a typo is a showstopper
<@kparal:matrix.org>
17:08:44
but no more request for a new one
<@kparal:matrix.org>
17:08:44
```
<@kparal:matrix.org>
17:08:44
{"error":true,"message":"The API key you specified is invalid. Please check that you typed it correctly.","code":306,"documentation":"https://bugzilla.redhat.com/docs/en/html/api/index.html"}
<@kparal:matrix.org>
17:08:44
```
<@geraldosimiao:matrix.org>
17:09:18
so, its just a bad user experience then...
<@geraldosimiao:matrix.org>
17:09:27
oh whel
<@adamwill:fedora.im>
17:10:29
geraldosimiao the big problem is that rhbz API keys expire after a year
<@kparal:matrix.org>
17:10:52
for me it's +1 blocker
<@geraldosimiao:matrix.org>
17:10:54
what the criterion says about this? what is the purpose on this?
<@adamwill:fedora.im>
17:11:07
so whenever you report your first bug through abrt, that clock starts ticking. a year after that, that API key will be invalid and you will no longer be able to report via abrt (if this isn't fixed)
<@geraldosimiao:matrix.org>
17:11:12
I mean, the ABRT usefullnes?
<@adamwill:fedora.im>
17:11:23
well, kparal cited the generic cirterion
<@adamwill:fedora.im>
17:11:39
https://fedoraproject.org/wiki/Fedora_42_Final_Release_Criteria#Default_application_functionality
<@adamwill:fedora.im>
17:11:48
which says the app "must start successfully and withstand a basic functionality test"
<@geraldosimiao:matrix.org>
17:12:09
so, think on a newcommer use case
<@adamwill:fedora.im>
17:12:10
i don't think we have any specific criteria for crash reporting outside of the installer (which doesn't use gnome-abrt)
<@lruzicka:matrix.org>
17:12:19
I would also add that it hinders testability
<@kparal:matrix.org>
17:12:25
if you want to include the time component, add the criterion about upgraded systems...
<@adamwill:fedora.im>
17:12:33
right, yeah
<@geraldosimiao:matrix.org>
17:12:39
they wanna use fedora and help, and get this abrt popup
<@adamwill:fedora.im>
17:12:53
beta critera say upgrades must work and 'The upgraded system must meet all release criteria'
<@geraldosimiao:matrix.org>
17:13:45
and learn they must create a bugzilla account for it to finish, and dont get the api popup never again because they close the window to create there account on bugzilla
<@lruzicka:matrix.org>
17:14:25
Would help if the API was set in `/etc/libreport/plugins/bugzilla.conf`
<@geraldosimiao:matrix.org>
17:14:41
oh, thats a ggod information
<@geraldosimiao:matrix.org>
17:14:49
oh, thats a good information
<@adamwill:fedora.im>
17:15:14
geraldosimiao i think the popup would keep appearing so long as you hadn't actually saved an API key
<@adamwill:fedora.im>
17:15:20
as soon as you've saved one, though, you won't see it again
<@lruzicka:matrix.org>
17:15:21
I cannot find any abrt settings, the `~/.config/abrt/settings` is empty
<@adamwill:fedora.im>
17:15:27
lruzicka it
<@adamwill:fedora.im>
17:15:33
lruzicka it's probably in gsettings ?
<@kparal:matrix.org>
17:15:36
as a workaround, the apikey can be adjusted using seahorse
<@adamwill:fedora.im>
17:15:44
oh, right
<@adamwill:fedora.im>
17:15:48
the key store. since it's a secret.
<@adamwill:fedora.im>
17:15:59
that's a painful workaround, though
<@adamwill:fedora.im>
17:16:07
on the whole i'm maybe a +0.37
<@geraldosimiao:matrix.org>
17:16:24
considering this -> "any specific criteria for crash reporting outside of the installer (which doesn't use gnome-abrt)" I'll probably vote -1 on this.
<@kparal:matrix.org>
17:17:20
geraldosimiao: the criterion I used says that a basic functionality must work for that app. So it depends whether you consider apikey refresh/fixup a basic functionality.
<@conan_kudo:matrix.org>
17:17:35
I personally would consider it basic functionality
<@kparal:matrix.org>
17:17:36
for a bug report, I think it's included
<@kparal:matrix.org>
17:17:41
for a bug reporter, I think it's included
<@conan_kudo:matrix.org>
17:17:52
in the past, I think we would have considered setting user credentials the same way
<@geraldosimiao:matrix.org>
17:18:46
yeah, ok, one must be able to change our API from the UI. agreed
<@geraldosimiao:matrix.org>
17:19:13
+1 Blocker
<@adamwill:fedora.im>
17:21:05
does that mean you're +1 ?
<@conan_kudo:matrix.org>
17:21:14
yup
<@conan_kudo:matrix.org>
17:21:18
+1 FinalBlocker
<@adamwill:fedora.im>
17:21:22
so i think we're at +3.37
<@adamwill:fedora.im>
17:21:24
any other votes?
<@derekenz:fedora.im>
17:21:33
+1 FinalBlocker
<@lruzicka:matrix.org>
17:21:46
+1
<@adamwill:fedora.im>
17:21:48
(after this meeting is done i need to go file a patent on fractional votes)
<@boniboyblue:fedora.im>
17:21:53
Aye - I'll go +1 as well.
<@lruzicka:matrix.org>
17:22:19
I think seahorse is not installed by default, so you are stuck with nothing.
<@farribeiro:matrix.org>
17:22:46
+1 FB
<@adamwill:fedora.im>
17:24:28
proposed !agreed 2356257 - AcceptedBlocker (Final) - this is accepted as a conditional violation of the Final criterion requiring preinstalled apps on Workstation to "start successfully and withstand a basic functionality test", with the Beta upgrade criterion also relevant. We agreed this bug renders the app effectively useless if an invalid API key is set, and there are enough scenarios where that may be the case - e.g. a typo or a previously-valid key on an upgraded system which has now expired - to accept this as a blocker
<@conan_kudo:matrix.org>
17:24:46
ack
<@derekenz:fedora.im>
17:24:48
ack
<@farribeiro:matrix.org>
17:24:49
ack
<@kparal:matrix.org>
17:25:00
ack
<@adamwill:fedora.im>
17:25:03
!agreed 2356257 - AcceptedBlocker (Final) - this is accepted as a conditional violation of the Final criterion requiring preinstalled apps on Workstation to "start successfully and withstand a basic functionality test", with the Beta upgrade criterion also relevant. We agreed this bug renders the app effectively useless if an invalid API key is set, and there are enough scenarios where that may be the case - e.g. a typo or a previously-valid key on an upgraded system which has now expired - to accept this as a blocker
<@kparal:matrix.org>
17:25:23
a 🍪 for adamw for a nice description
<@adamwill:fedora.im>
17:25:28
!link https://bugzilla.redhat.com/show_bug.cgi?id=2355033
<@adamwill:fedora.im>
17:25:28
!link https://pagure.io/fedora-qa/blocker-review/issue/1813
<@adamwill:fedora.im>
17:25:28
!topic (2355033) Fedora 42 beta unable to shutdown despite "No inhibitors"
<@adamwill:fedora.im>
17:25:28
!info Ticket vote: FinalBlocker (+1,0,-5) (+augenauf, -boniboyblue, -derekenz, -geraldosimiao, -nielsenb, -lruzicka)
<@adamwill:fedora.im>
17:25:28
!info Proposed Blocker, gnome-shell, NEW
<@adamwill:fedora.im>
17:25:28
!info Ticket vote: FinalFreezeException (+4,0,-0) (+boniboyblue, +asciiwolf, +nielsenb, +lruzicka)
<@adamwill:fedora.im>
17:25:42
so a lot of people voted -1 on this because they 'could not reproduce', but it does actually seem like a pretty clear bug
<@conan_kudo:matrix.org>
17:25:57
yeah this seems pretty bad
<@zodbot:fedora.im>
17:25:58
farribeiro has already given cookies to adamwill during the F41 timeframe
<@conan_kudo:matrix.org>
17:26:05
is this verified with gnome-shell 48 final though?
<@adamwill:fedora.im>
17:26:30
i mean, per the upstream description i can't imagine why it wouldn't still happen
<@kparal:matrix.org>
17:26:56
adamw: your gedit comment confuses me, how is it related?
<@adamwill:fedora.im>
17:26:58
the upstream explanation is that this is caused by a change in systemd and gnome-shell needs to change *something* to adapt to systemd's new behaviour, and I don't see any indication that that already happened
<@adamwill:fedora.im>
17:27:38
Kamil Páral well, gedit will inhibit shutdown so long as any unsaved change is present, and you can't shutdown even if you tell gnome to ignore it and shutdown anyway...
<@adamwill:fedora.im>
17:27:47
you have to quit gedit first
<@adamwill:fedora.im>
17:28:03
upstream seems to think this is all the same bug, as far as I can make out
<@conan_kudo:matrix.org>
17:28:11
I wonder why this doesn't happen on KDE Plasma
<@conan_kudo:matrix.org>
17:28:17
by all rights the same problems should be there too
<@kparal:matrix.org>
17:28:31
so any app that would inhibit shutdown would cause this, right?
<@adamwill:fedora.im>
17:28:38
yeah, afaict
<@boniboyblue:fedora.im>
17:28:43
So this has been an intended change?
<@adamwill:fedora.im>
17:29:00
but it seems like some of the reporters don't have any app obviously inhibiting shutdown, but upstream looked at some debug output and confidently declared it's the same bug
<@conan_kudo:matrix.org>
17:29:13
"broken as intended"
<@kparal:matrix.org>
17:29:36
I can confirm this with gedit. However, from the dialog, it's pretty clear which apps holds the lock.
<@adamwill:fedora.im>
17:29:37
Christopher Boni well, sort of. the change in systemd was intended by systemd devs. the consequence for gnome was not exactly 'intended', but systemd is not going to revert, gnome has to change to adapt to the new systemd behavior. again aiui
<@kparal:matrix.org>
17:30:33
s/gedit/gnome text editor
<@adamwill:fedora.im>
17:31:22
i think sometimes that's the case and sometimes it isn't
<@kparal:matrix.org>
17:31:37
btw, logout is not blocked
<@adamwill:fedora.im>
17:31:42
https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/8296#note_2397718 has the bit where skeller looked at robatino's dbus traffic and decided there's definitely an inhibition
<@salimma:fedora.im>
17:31:58
if it's like the text editor screenshot above - that's similar to the macOS behavior it seems
<@salimma:fedora.im>
17:31:58
<@salimma:fedora.im>
17:31:58
if the user can't tell what application causes this that definitely sounds bad
<@kparal:matrix.org>
17:32:05
if we can find a case where it doesn't identify the app, and where it happens regularly, I would consider this as a blocker
<@adamwill:fedora.im>
17:32:27
Michel Lind UTC-6 well, part of the bug is that if you click "Power Off" there, it doesn't work (it's supposed to override the inhibit but it doesn't)
<@adamwill:fedora.im>
17:32:46
but seems there are other cases where gnome won't tell you what's causing the inhibition but it won't shutdown
<@adamwill:fedora.im>
17:33:16
i guess maybe that's when something like dnf is trying to inhibit shutdown...
<@boniboyblue:fedora.im>
17:33:27
Do you get the same indication if you use systemctl poweroff?
<@adamwill:fedora.im>
17:33:29
maybe test running a `dnf update` on a vt and trying to shutdown while it's happening?
<@adamwill:fedora.im>
17:35:15
i dunno, i feel like i don't entirely understand this, but it seems bad. :P
<@kparal:matrix.org>
17:37:27
Alternatively, ignore inhibitors and users with 'systemctl poweroff -i'.
<@kparal:matrix.org>
17:37:27
```
<@kparal:matrix.org>
17:37:27
kparal@localhost-live:~$ systemctl poweroff
<@kparal:matrix.org>
17:37:27
Operation inhibited by "kparal" (PID 5394 "gnome-session-b", user kparal), reason is "user session inhibited".
<@kparal:matrix.org>
17:37:27
Please retry operation after closing inhibitors and logging out other users.
<@kparal:matrix.org>
17:37:27
'systemd-inhibit' can be used to list active inhibitors.
<@kparal:matrix.org>
17:37:27
```
<@kparal:matrix.org>
17:37:45
doesn't identify the app
<@adamwill:fedora.im>
17:38:34
what happens graphically when you try to shut down?
<@adamwill:fedora.im>
17:38:37
it just does nothing?
<@geraldosimiao:matrix.org>
17:39:20
Fedora-Workstation-Live-42-20250331.n.0.x86_64.iso identify the app
<@kparal:matrix.org>
17:39:31
I can confirm that dnf blocks poweroff/reboot, but the dialog doesn't say anything
<@kparal:matrix.org>
17:39:40
bingo
<@geraldosimiao:matrix.org>
17:39:43
me too
<@adamwill:fedora.im>
17:39:43
ok
<@adamwill:fedora.im>
17:40:02
so you just get a 'normal' shutdown dialog, but it doesn't actually work?
<@lruzicka:matrix.org>
17:40:12
Seems to me like identification: /etc/libreport/events/report_Bugzilla.conf
<@kparal:matrix.org>
17:40:20
with dnf operation happening, yes
<@adamwill:fedora.im>
17:40:24
alright
<@adamwill:fedora.im>
17:40:38
so, we pretty much understand what's going on now...do we consider this bad enough to be a blocker?
<@lruzicka:matrix.org>
17:40:51
Seems to me like identification: Operation inhibited by "kparal" (PID 5394 "gnome-session-b", user kparal), reason is "user session inhibited".
<@kparal:matrix.org>
17:41:01
on the other hand, with a dnf operation, perhaps forcing poweroff shouldn't be really easy, I don't know 🙂
<@matthiasc:gnome.org>
17:41:07
either way, this is systemd changing behavior, breaking users
<@adamwill:fedora.im>
17:41:37
i guess it's a conditional violation of beta "Shutting down, rebooting, logging in and logging out must work using standard console commands and the mechanisms offered (if any) by all release-blocking desktops", the condition being 'something is inhibiting shutdown', and it's especially bad if the 'something' is not tracked by the gnome session so it doesn't tell you what it is
<@adamwill:fedora.im>
17:42:09
Matthias Clasen i do kinda grok why systemd would change it, though. it's not like it was *good* that it just let you shut down while dnf was running on a VT before...
<@adamwill:fedora.im>
17:42:55
what we really want here is something that didn't exist before (consistently warn you about all inhibitions, but let you override them if you really want to). aiui
<@matthiasc:gnome.org>
17:43:10
yes, but I see a lot of discussion that sounds like gnome is to blame, when really, this is badly executed changes on the system side
<@matthiasc:gnome.org>
17:43:18
*systemd*
<@adamwill:fedora.im>
17:44:16
anyhow, the practical question is, do we think it's a bad enough violation of the criteria to block on, per the framing above
<@adamwill:fedora.im>
17:44:27
voting time...
<@matthiasc:gnome.org>
17:44:50
if you think the systemd changes make sense, the you ought to vote no, I think
<@kparal:matrix.org>
17:44:58
can you imagine more real-world scenarios? Because particularly with dnf, this is actually good...
<@adamwill:fedora.im>
17:46:55
well, the gedit case is annoying
<@adamwill:fedora.im>
17:47:10
i often want to shutdown and know that the sixty tabs i have in gedit that are 'unsaved' because i did a git pull don't matter
<@adamwill:fedora.im>
17:47:31
i would like to able to just ignore them and shut down, instead of having to close gedit first before it will shut down
<@kparal:matrix.org>
17:48:05
yes, but that case doesn't sound blockery enough to me
<@adamwill:fedora.im>
17:48:14
if there are things that trigger the 'you can't shut down but we won't tell you why' behaviour which are *less important* than 'there's actually a system update running', that's quite bad i think
<@kparal:matrix.org>
17:48:33
I'm more concerned with the case where there's no app displayed, but you still can't shut it down
<@lruzicka:matrix.org>
17:48:36
it is definitely annoying if you cannot ignore
<@adamwill:fedora.im>
17:48:38
yeah
<@adamwill:fedora.im>
17:48:56
but we don't know what the actual inhibiting...thing was in those cases yet, i don't think?
<@kparal:matrix.org>
17:49:14
for example, a firefox download might block it? not sure
<@adamwill:fedora.im>
17:49:17
we *could* punt this to try and identify what things set inhibitions and can trigger that flow, i guess
<@lbrabec:matrix.org>
17:49:28
how does it behave with gnome-text-editor?
<@adamwill:fedora.im>
17:49:33
same as gedit i think
<@lruzicka:matrix.org>
17:49:41
cant switch off
<@adamwill:fedora.im>
17:49:51
it triggers the problem but the dialog does at least tell you what's inhibiting (though it doesn't tell you why overriding it isn't working)
<@derekenz:fedora.im>
17:50:27
Yes maybe we should punt
<@kparal:matrix.org>
17:53:08
Perhaps make a call so that people can help us cases when this is misbehaving, using `systemd-inhibit` to identify processes which are not visible in the dialog
<@lruzicka:matrix.org>
17:53:15
Can't gnome force it to systemd?
<@geraldosimiao:matrix.org>
17:53:45
I'm not sure about it...
<@boniboyblue:fedora.im>
17:53:50
Well, I'm defo still FinalFE +1
<@geraldosimiao:matrix.org>
17:53:53
+1 Punt
<@adamwill:fedora.im>
17:54:00
is systemd-inhibit output actually enough?
<@geraldosimiao:matrix.org>
17:54:10
yeah, FE +1 for sure
<@adamwill:fedora.im>
17:54:18
robatino's output in https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/8296#note_2397689 doesn't show anything, to my eyes anyway
<@adamwill:fedora.im>
17:54:32
all those things are only inhibiting sleep, not shutdown
<@adamwill:fedora.im>
17:54:41
oh, yeah, good call, finalfe +1 for sure
<@lruzicka:matrix.org>
17:54:44
The `systemctl poweroff -i` does work.
<@kparal:matrix.org>
17:55:27
+1 FE
<@adamwill:fedora.im>
17:56:01
see https://gitlab.gnome.org/GNOME/gnome-session/-/issues/142#note_2340847 for the discussion of potential fixes
<@farribeiro:matrix.org>
17:57:49
+1 FE
<@adamwill:fedora.im>
17:58:06
proposed !agreed 2355033 - punt (delay decision) - after some discussion of the details behind this bug, we're not sure whether to block on it. The worst case is when something not tracked by GNOME is inhibiting shutdown; in this case, shutdown fails without explanation. But that will not usually be the case, and in at least some cases, the new behaviour is probably better than the old behaviour (e.g. it wasn't *good* that you could happily shut down while a package install operation was happening outside of GNOME). we will punt this to try and investigate specific triggers to aid in deciding whether to block
<@adamwill:fedora.im>
17:58:12
er, nm, patching that
<@derekenz:fedora.im>
17:58:30
ack
<@boniboyblue:fedora.im>
17:58:44
ack
<@geraldosimiao:matrix.org>
17:58:49
ack
<@adamwill:fedora.im>
17:58:55
proposed !agreed 2355033 - punt (delay decision) on blocker status, AcceptedFreezeException (Final) - after some discussion of the details behind this bug, we're not sure whether to block on it. The worst case is when something not tracked by GNOME is inhibiting shutdown; in this case, shutdown fails without explanation. But that will not usually be the case, and in at least some cases, the new behaviour is probably better than the old behaviour (e.g. it wasn't *good* that you could happily shut down while a package install operation was happening outside of GNOME). we will punt this to try and investigate specific triggers to aid in deciding whether to block. It's clearly desirable to improve this behaviour during freeze if we can, though, so this is accepted FE.
<@adamwill:fedora.im>
17:59:01
patched, added acceptedFE wording
<@geraldosimiao:matrix.org>
17:59:30
ack
<@kparal:matrix.org>
17:59:41
ack
<@lruzicka:matrix.org>
18:00:24
ack
<@adamwill:fedora.im>
18:04:09
!agreed 2355033 - punt (delay decision) on blocker status, AcceptedFreezeException (Final) - after some discussion of the details behind this bug, we're not sure whether to block on it. The worst case is when something not tracked by GNOME is inhibiting shutdown; in this case, shutdown fails without explanation. But that will not usually be the case, and in at least some cases, the new behaviour is probably better than the old behaviour (e.g. it wasn't good that you could happily shut down while a package install operation was happening outside of GNOME). we will punt this to try and investigate specific triggers to aid in deciding whether to block. It's clearly desirable to improve this behaviour during freeze if we can, though, so this is accepted FE.
<@adamwill:fedora.im>
18:04:39
!info that's all the proposals (that last one was the only outstanding proposed FE)
<@adamwill:fedora.im>
18:04:47
let's take a look at:
<@adamwill:fedora.im>
18:04:51
!topic Accepted Final blockers
<@adamwill:fedora.im>
18:04:59
as a reminder, we're just checking status here, not revoting unless there's reason to
<@adamwill:fedora.im>
18:05:15
!info Accepted Blocker, anaconda, POST
<@adamwill:fedora.im>
18:05:15
!topic (2354805) putting /boot on MDRAID0 is accepted and results in a non-booting system
<@adamwill:fedora.im>
18:05:15
!link https://bugzilla.redhat.com/show_bug.cgi?id=2354805
<@adamwill:fedora.im>
18:05:15
!link https://pagure.io/fedora-qa/blocker-review/issue/1810
<@adamwill:fedora.im>
18:05:27
!info this is in POST so we're waiting on a new anaconda build to test
<@adamwill:fedora.im>
18:05:44
!topic (2353002) webui does not support MBR disks, but doesn't properly communicate this, instead e.g. requires biosboot partition even though that is impossible
<@adamwill:fedora.im>
18:05:44
!link https://bugzilla.redhat.com/show_bug.cgi?id=2353002
<@adamwill:fedora.im>
18:05:44
!link https://pagure.io/fedora-qa/blocker-review/issue/1804
<@adamwill:fedora.im>
18:05:44
!info Accepted Blocker, anaconda-webui, ASSIGNED
<@adamwill:fedora.im>
18:06:04
so...i think there was an attempt to fix this but it didn't work and got pinged back to ASSIGNED? is that right?
<@kparal:matrix.org>
18:06:16
yes
<@kparal:matrix.org>
18:06:23
there's a new patch ready
<@adamwill:fedora.im>
18:06:27
whee
<@adamwill:fedora.im>
18:06:31
so, set it back to POST i guess
<@adamwill:fedora.im>
18:06:52
i don't see a note about the new patch though?
<@kparal:matrix.org>
18:06:54
ugh, sorry
<@kparal:matrix.org>
18:07:02
I'm still looking at the previous bug
<@kparal:matrix.org>
18:07:05
you're too fast!
<@adamwill:fedora.im>
18:07:35
hah, yeah
<@kparal:matrix.org>
18:07:44
I moved this back to assigned because multiple people confirmed it's not fixed
<@kparal:matrix.org>
18:08:02
and I have no more background info on this 🙂
<@adamwill:fedora.im>
18:09:37
ok
<@adamwill:fedora.im>
18:09:54
!info a fix for this was attempted, but testing seems to indicate it's still broken, we will wait to hear back from anaconda team
<@adamwill:fedora.im>
18:10:02
!link https://bugzilla.redhat.com/show_bug.cgi?id=2352679
<@adamwill:fedora.im>
18:10:02
!info Accepted Blocker, distribution, ASSIGNED
<@adamwill:fedora.im>
18:10:02
!link https://pagure.io/fedora-qa/blocker-review/issue/1794
<@adamwill:fedora.im>
18:10:02
!topic (2352679) Fedora 42: Server boot aarch64 image exceeds maximum size
<@adamwill:fedora.im>
18:10:17
!info adamw sent a patch for dracut which would get this *nearly* back under size but not quite
<@adamwill:fedora.im>
18:10:25
!info we can try to trim a bit more, or just bump the target size
<@adamwill:fedora.im>
18:10:43
!info adamw sent a patch for lorax which would get this *nearly* back under size but not quite
<@adamwill:fedora.im>
18:11:03
still, not much to worry about
<@lruzicka:matrix.org>
18:11:34
I need to be going.
<@adamwill:fedora.im>
18:12:30
sure, thanks for coming
<@adamwill:fedora.im>
18:12:36
!info Accepted Blocker, fedora-repos, NEW
<@adamwill:fedora.im>
18:12:36
!link https://pagure.io/fedora-qa/blocker-review/issue/1816
<@adamwill:fedora.im>
18:12:36
!topic (2356173) updates-testing needs to be disabled for F42 GA
<@adamwill:fedora.im>
18:12:36
!link https://bugzilla.redhat.com/show_bug.cgi?id=2356173
<@adamwill:fedora.im>
18:12:51
pretty straightforward, just needs the thingy in the spec file flipping and a new build
<@adamwill:fedora.im>
18:13:01
!info this is pretty straightforward, just waiting for a maintainer to do the new build
<@adamwill:fedora.im>
18:13:10
!info Accepted Blocker, mdadm, POST
<@adamwill:fedora.im>
18:13:10
!link https://pagure.io/fedora-qa/blocker-review/issue/1790
<@adamwill:fedora.im>
18:13:10
!link https://bugzilla.redhat.com/show_bug.cgi?id=2325906
<@adamwill:fedora.im>
18:13:10
!topic (2325906) [live] Can't reuse existing RAID partitioning
<@adamwill:fedora.im>
18:13:47
ugh, this one is a pain. i have a patch that fixes it, but upstream doesn't like it, but isn't sending any alternative. i can just put my patch downstream, and will do that if necessary, but it'd be nice if we could resolve the upstream logjam :/
<@adamwill:fedora.im>
18:14:01
i'm trying to find anyone at RH who has enough clout to make things move upstream
<@adamwill:fedora.im>
18:14:42
!info we have a potential fix for this, but it's encountering pushback upstream, trying to get that resolved before backporting it
<@adamwill:fedora.im>
18:16:13
!link https://pagure.io/fedora-qa/blocker-review/issue/1807
<@adamwill:fedora.im>
18:16:13
!topic (2354592) GNOME crashes on startup if keyboard accessibility features are enabled
<@adamwill:fedora.im>
18:16:13
!link https://bugzilla.redhat.com/show_bug.cgi?id=2354592
<@adamwill:fedora.im>
18:16:13
!info Accepted Blocker, mutter, POST
<@adamwill:fedora.im>
18:16:30
so the fix for this is available and tested, but we're hanging on for a mutter 48.1 release instead of backporting it...
<@adamwill:fedora.im>
18:16:41
any news on that, Matthias Clasen ?
<@matthiasc:gnome.org>
18:16:55
release is incoming
<@matthiasc:gnome.org>
18:17:16
within the next few hours
<@adamwill:fedora.im>
18:18:08
yay
<@adamwill:fedora.im>
18:18:32
!info the fix for this is done and tested and will be part of an imminent mutter 48.1 which should show up in fedora soon after release
<@adamwill:fedora.im>
18:18:44
!topic (2354865) repo states are not synchronized between DNF5 and PackageKit (GUI package managers), because of repo overrides
<@adamwill:fedora.im>
18:18:44
!info Accepted Blocker, PackageKit, NEW
<@adamwill:fedora.im>
18:18:44
!link https://pagure.io/fedora-qa/blocker-review/issue/1812
<@adamwill:fedora.im>
18:18:44
!link https://bugzilla.redhat.com/show_bug.cgi?id=2354865
<@adamwill:fedora.im>
18:18:52
any news on this one, Kamil Páral?
<@kparal:matrix.org>
18:19:38
no news
<@adamwill:fedora.im>
18:20:04
hmm, okay
<@adamwill:fedora.im>
18:20:07
we're getting kinda tight on time
<@kparal:matrix.org>
18:20:48
I don't think this will get fixed
<@kparal:matrix.org>
18:20:53
this cycle
<@adamwill:fedora.im>
18:21:02
let's reassign it to libdnf for now
<@adamwill:fedora.im>
18:21:06
to get dnf devs' input
<@adamwill:fedora.im>
18:21:51
!info we'll get dnf developers' input on this, but it's a high risk not to be fixable within the release time frame, we may have to waive it
<@kparal:matrix.org>
18:22:26
I already had dnf devs feedback somewhere
<@adamwill:fedora.im>
18:22:41
oh, you did? can't have been on the bug, i don't see it there
<@kparal:matrix.org>
18:23:14
well, it was a related bug: https://bugzilla.redhat.com/show_bug.cgi?id=2336535
<@adamwill:fedora.im>
18:23:30
ah
<@kparal:matrix.org>
18:23:39
they're not really saying "we'll fix it"
<@kparal:matrix.org>
18:23:45
but we can ask again
<@adamwill:fedora.im>
18:23:52
hmm, well, doesn't seem quite the same
<@adamwill:fedora.im>
18:23:56
worth asking for the purpose of this bug, anyhow
<@adamwill:fedora.im>
18:24:00
i did the reassignment
<@kparal:matrix.org>
18:24:03
ack
<@adamwill:fedora.im>
18:25:14
!link https://pagure.io/fedora-qa/blocker-review/issue/1814
<@adamwill:fedora.im>
18:25:14
!link https://bugzilla.redhat.com/show_bug.cgi?id=2355207
<@adamwill:fedora.im>
18:25:14
!topic (2355207) Remote install via RDP fails (client either drops connection immediately or hangs at a white screen) since Fedora-42-20250316.n.0
<@adamwill:fedora.im>
18:25:14
!info Accepted Blocker, pipewire, POST
<@adamwill:fedora.im>
18:25:29
so i figured this out and sent a fix to lorax, bcl just wants me to tweak it a bit, i'll do that oday
<@adamwill:fedora.im>
18:25:36
should be able to land it and get rid of this soon
<@adamwill:fedora.im>
18:25:40
so i figured this out and sent a fix to lorax, bcl just wants me to tweak it a bit, i'll do that today
<@adamwill:fedora.im>
18:25:56
!info fix for this is pending, just needs a bit of a tweak on PR review, should land soon
<@adamwill:fedora.im>
18:27:27
aaand that's everything
<@adamwill:fedora.im>
18:27:31
!topic Open floor
<@adamwill:fedora.im>
18:27:45
any other business, folks? thanks for the careful attention to complex bugs this week
<@aggraxis:fedora.im>
18:27:50
Hi!
<@aggraxis:fedora.im>
18:27:58
The server image bug (I missed it)
<@aggraxis:fedora.im>
18:28:22
We talked about it during the server meeting last week, and they were ok with pushing up the limit, just had to figure out how
<@aggraxis:fedora.im>
18:28:23
https://meetbot-raw.fedoraproject.org//meeting_matrix_fedoraproject-org/2025-03-26/fedora-server.2025-03-26-17.00.txt
<@boniboyblue:fedora.im>
18:28:45
The giant fedora logo at login in Workstation should be sorted upstream now.
<@geraldosimiao:matrix.org>
18:29:27
you say about this bug? https://bugzilla.redhat.com/show_bug.cgi?id=2352679
<@aggraxis:fedora.im>
18:29:33
yes
<@adamwill:fedora.im>
18:29:52
Paul Maconi (Aggraxis) thanks. the way to do it is a ticket or pr on https://pagure.io/fedora-pgm/pgm_docs
<@aggraxis:fedora.im>
18:30:01
Thank you
<@adamwill:fedora.im>
18:30:13
although...we should really write a better process for this...
<@aggraxis:fedora.im>
18:31:03
I will relay that back to the Server folks. I can do it if they want, but I don't want to step on anyone's toes. :)
<@adamwill:fedora.im>
18:31:25
https://pagure.io/fedora-pgm/pgm_docs/blob/main/f/releases/modules/ROOT/pages/f42/blocking.adoc is the file to change
<@adamwill:fedora.im>
18:31:45
also tag me on the ticket because i have to update the size data for the relval check when it changes
<@aggraxis:fedora.im>
18:32:08
Will do
<@adamwill:fedora.im>
18:32:37
thanks!
<@adamwill:fedora.im>
18:32:43
alrighty, thanks for coming folks
<@derekenz:fedora.im>
18:32:57
Thanks Adam!
<@adamwill:fedora.im>
18:33:00
!endmeeting