<@adamwill:fedora.im>
15:00:34
!startmeeting Quality
<@meetbot:fedora.im>
15:00:36
Meeting started at 2026-08-17 15:00:34 UTC
<@meetbot:fedora.im>
15:00:36
The Meeting name is 'Quality'
<@adamwill:fedora.im>
15:00:38
!topic Roll Call
<@adamwill:fedora.im>
15:00:44
hi hi everyone, who's here for the quality meeting?
<@nielsenb:fedora.im>
15:00:59
!hi
<@jgroman:fedora.im>
15:01:01
!hi
<@zodbot:fedora.im>
15:01:08
Brandon Nielsen: Brandon Nielsen (nielsenb)
<@zodbot:fedora.im>
15:01:10
Jaroslav Groman: Jaroslav Groman (jgroman)
<@ephmo:fedora.im>
15:01:23
!hi
<@zodbot:fedora.im>
15:01:25
Ephraim Kaov: Ephraim Kaov (ephmo) - he / him / his
<@derekenz:fedora.im>
15:01:37
!hi
<@zodbot:fedora.im>
15:01:38
Derek Enz: Derek Enz (derekenz)
<@boniboyblue:fedora.im>
15:02:11
!hi
<@zodbot:fedora.im>
15:02:14
Christopher Boni: Christopher Boni (boniboyblue)
<@adamwill:fedora.im>
15:02:51
hi hi everyone, thanks for coming
<@adamwill:fedora.im>
15:03:10
!topic Previous meeting follow-up
<@adamwill:fedora.im>
15:04:49
!info we started the manual validation testing and found several interesting bugs, see https://forge.fedoraproject.org/quality/tickets/issues/926 (we still have more to do, though)
<@adamwill:fedora.im>
15:05:11
!info looks like nobody has any data visualization ideas yet, a reminder to put them in https://forge.fedoraproject.org/quality/tickets/issues/923 if you have any
<@adamwill:fedora.im>
15:06:09
!info planning of several test days is ongoing at https://forge.fedoraproject.org/quality/tickets/issues?q=&type=all&state=open&labels=1920&milestone=0&assignee=0&poster=0 , we should get the ones with dates into the calendar and testdays ASAP
<@adamwill:fedora.im>
15:06:46
!info discussion about replacing mcelog was started at https://discussion.fedoraproject.org/t/should-fedora-continue-installing-mcelog-by-default/198422 (also on devel@, but nobody replied there)
<@adamwill:fedora.im>
15:06:55
anything else to follow up on?
<@nielsenb:fedora.im>
15:10:06
https://bugzilla.redhat.com/show_bug.cgi?id=2484542
<@nielsenb:fedora.im>
15:10:15
Does that also impact Server?
<@nielsenb:fedora.im>
15:11:09
I no longer have any arm64 HW around capable of running a DE, but noticed Server didn't seem to have an initial setup anymore. I didn't have a lot of time to see if I'd missed some change, or if it was documented, or what.
<@adamwill:fedora.im>
15:11:30
i think it *can*, but *typically* you're going to make it out of a server install with a user and/or root password so it's not as bad
<@adamwill:fedora.im>
15:11:43
oh, and yes, it'll absolutely affect the server disk image
<@adamwill:fedora.im>
15:11:50
it affects all non-graphical disk images
<@jgroman:fedora.im>
15:11:52
It breaks Minimal for sure
<@korora:fedora.im>
15:12:03
!hi
<@zodbot:fedora.im>
15:12:04
Jocelyn Gould (UTC-4): Jocelyn Gould (korora) - she / her / hers
<@nielsenb:fedora.im>
15:12:10
Got it, thanks
<@korora:fedora.im>
15:12:32
I’ll make a note of this and bring it to the server WG at our meeting on Wednesday.
<@adamwill:fedora.im>
15:14:09
alrighty
<@ephmo:fedora.im>
15:14:18
I installed Fedora-45-20260817.n.0 Workstation from the Live image onto an external HDD.
<@ephmo:fedora.im>
15:14:18
<@ephmo:fedora.im>
15:14:18
After rebooting into the installed Fedora system, reboot/poweroff can completely freeze the machine. During the failure I saw:
<@ephmo:fedora.im>
15:14:18
<@ephmo:fedora.im>
15:14:18
usb 1-3: hub failed to enable device, error -108
<@ephmo:fedora.im>
15:14:18
<@ephmo:fedora.im>
15:14:18
usb usb1-port3: couldn't allocate usb_device
<@ephmo:fedora.im>
15:14:18
<@ephmo:fedora.im>
15:14:18
The machine was completely frozen: Ctrl+Alt+Delete, Ctrl+Alt+Backspace and switching to TTYs with F1-F7 did not work, so I had to power it off using the physical power button.
<@ephmo:fedora.im>
15:14:18
<@ephmo:fedora.im>
15:14:18
I moved the external HDD to another USB port and Fedora booted, but reboot/poweroff still freezes.
<@ephmo:fedora.im>
15:14:18
<@ephmo:fedora.im>
15:14:18
As a comparison, I booted Aurora with the same external HDD attached/mounted and reboot/poweroff works normally.
<@ephmo:fedora.im>
15:14:18
<@ephmo:fedora.im>
15:14:18
I haven't determined yet whether this is a GNOME, systemd, kernel, or USB-storage issue. I can reproduce it and can collect the previous-boot journal/kernel logs.
<@adamwill:fedora.im>
15:15:19
most likely kernel. try different kernels
<@ephmo:fedora.im>
15:15:30
I’ll also install Fedora KDE to check whether the issue is GNOME-related or happens with Fedora more generally. Alternatively, I can install the GNOME image on another machine, although that will take some time to prepare. I’ll report back with the results.
<@adamwill:fedora.im>
15:15:41
those types of issues are *usually* hardware-specific too, but if anyone else is having trouble shutting down f45, yell...
<@adamwill:fedora.im>
15:15:55
thanks for testing
<@derekenz:fedora.im>
15:15:57
Works for me
<@niknikovsky:fedora.im>
15:16:15
Famous last words somewhere
<@ephmo:fedora.im>
15:16:21
I will also try another kernel
<@derekenz:fedora.im>
15:16:39
Famous yeah lol
<@adamwill:fedora.im>
15:17:18
!topic Fedora 45 status
<@adamwill:fedora.im>
15:17:55
!info F45 branching is more or less complete at this point, it was complicated substantially by https://fedoraproject.org/wiki/Changes/RelocateRpmRepoConfigsToUsr
<@adamwill:fedora.im>
15:18:26
!info on that note...yes, https://fedoraproject.org/wiki/Changes/RelocateRpmRepoConfigsToUsr is a thing. please be aware of it. repo configs and GPG keys have been in their current/previous locations for many years and there may be many things that presume they will always be there
<@adamwill:fedora.im>
15:19:40
!info there have been various issues with the openqa silverblue build/install tests lately which are making the openqa results messy, I'm hoping to sort that out soon
<@adamwill:fedora.im>
15:22:30
!info we have one proposed blocker and two accepted blockers so I didn't schedule a blocker review meeting today, but please do vote on the proposed blocker and propose any bugs you know of that meet the criteria - https://qa.fedoraproject.org/blockerbugs/milestone/45/beta/buglist
<@adamwill:fedora.im>
15:22:37
anything else for f45?
<@derekenz:fedora.im>
15:23:21
Nothing earth shattering that I can think of
<@derekenz:fedora.im>
15:23:45
So far
<@adamwill:fedora.im>
15:25:33
yay
<@adamwill:fedora.im>
15:25:37
we like our earth...unshattered?
<@derekenz:fedora.im>
15:25:44
Sure lol
<@adamwill:fedora.im>
15:25:50
!topic Test Day / community event status
<@adamwill:fedora.im>
15:26:32
!info GNOME 51 test days are apparently starting today - https://testdays.fedoraproject.org/testday/24 , https://fedoraproject.org/wiki/Test_Day:2026-08-17_GNOME_51_Desktop
<@adamwill:fedora.im>
15:26:59
!info internationalization test week is scheduled for Sept 7 to 13 - https://testdays.fedoraproject.org/testday/23
<@adamwill:fedora.im>
15:27:57
i think that's all we have finalized ATM, righT?
<@ephmo:fedora.im>
15:30:19
https://bugzilla.redhat.com/show_bug.cgi?id=2517178
<@ephmo:fedora.im>
15:30:19
<@ephmo:fedora.im>
15:30:19
This is also reproducible to baremetal, but is harder to reproduce that in a VM
<@adamwill:fedora.im>
15:31:32
i'd definitely report that upstream at https://gitlab.gnome.org/GNOME/gnome-shell/
<@ephmo:fedora.im>
15:31:44
OK
<@adamwill:fedora.im>
15:32:06
it's almost certainly an upstream issue and the relevant devs just don't look at rhbz :(
<@derekenz:fedora.im>
15:32:26
Speaking of Gnome I need to verify that user switching bug too
<@ephmo:fedora.im>
15:32:48
For me is not reproducible
<@nielsenb:fedora.im>
15:33:05
https://bugzilla.redhat.com/show_bug.cgi?id=2501223
<@nielsenb:fedora.im>
15:33:33
Not to derail, but I literally just hit this again, is it just me somehow?
<@adamwill:fedora.im>
15:34:28
well, er, for a start, you shouldn't get a login screen in the first place
<@adamwill:fedora.im>
15:34:32
lives are supposed to boot to a desktop
<@nielsenb:fedora.im>
15:34:46
That's what I thought
<@nielsenb:fedora.im>
15:35:01
Does not work that way for me on my BIOS hardware
<@adamwill:fedora.im>
15:35:11
huh. that's weird. it appears ok in openqa
<@adamwill:fedora.im>
15:35:43
maybe take a look in the journal and see if you can see anything incriminating?
<@adamwill:fedora.im>
15:36:13
if other folks can test recent rawhide lives on bare metal too...
<@derekenz:fedora.im>
15:36:50
The KDE image worked as expected for me
<@derekenz:fedora.im>
15:37:21
Will verify WS too
<@nielsenb:fedora.im>
15:37:38
Definitely works right in Boxes
<@derekenz:fedora.im>
15:38:00
WS?
<@nielsenb:fedora.im>
15:38:28
Yeah, 45 Branched 20260817.n.0
<@adamwill:fedora.im>
15:38:29
since we appear to be on it already:
<@adamwill:fedora.im>
15:38:33
!topic Open floor
<@ephmo:fedora.im>
15:38:55
What do you think would be the best next step regarding the mcelog discussion?
<@ephmo:fedora.im>
15:38:55
<@ephmo:fedora.im>
15:38:55
https://discussion.fedoraproject.org/t/should-fedora-continue-installing-mcelog-by-default/198422
<@ephmo:fedora.im>
15:38:55
<@ephmo:fedora.im>
15:38:55
Should I wait for more feedback there first, or would it make sense to start a thread on devel@? If there is enough interest in replacing mcelog with rasdaemon, should I then propose a more formal change/vote somewhere?
<@nielsenb:fedora.im>
15:40:14
Whatever gets us "switch the rasdaemon" fastest. :D
<@adamwill:fedora.im>
15:44:47
i mean, there's three hearts on the 'switch to rasdaemon' post and nobody objected
<@adamwill:fedora.im>
15:44:59
i'd say that's enough support to propose a Change, if you want to go through that effort
<@adamwill:fedora.im>
15:45:32
https://docs.fedoraproject.org/en-US/operations/changes_guide/
<@adamwill:fedora.im>
15:49:07
anything else, folks?
<@derekenz:fedora.im>
15:49:28
Good for now
<@nielsenb:fedora.im>
15:49:34
Not from me
<@adamwill:fedora.im>
15:50:35
alrighty, thanks for coming, everyone
<@ephmo:fedora.im>
15:50:57
Thanks everyone
<@adamwill:fedora.im>
15:51:29
!endmeeting