<@kparal:matrix.org>
16:00:55
!startmeeting F44-blocker-review
<@meetbot:fedora.im>
16:01:00
Meeting started at 2026-03-23 16:00:55 UTC
<@meetbot:fedora.im>
16:01:01
The Meeting name is 'F44-blocker-review'
<@jlinton:fedora.im>
16:01:09
.hi
<@osama-albahrani:matrix.org>
16:01:13
!hi
<@zodbot:fedora.im>
16:01:16
Osama Albahrani (osalbahr)
<@ngompa:fedora.im>
16:01:18
!hi
<@zodbot:fedora.im>
16:01:18
Neal Gompa (ngompa) - he / him / his
<@jlinton:fedora.im>
16:01:23
!hi
<@zodbot:fedora.im>
16:01:24
Jeremy Linton (jlinton)
<@derekenz:fedora.im>
16:01:31
!hi
<@zodbot:fedora.im>
16:01:32
Derek Enz (derekenz)
<@kparal:matrix.org>
16:01:32
!topic Roll Call
<@nielsenb:fedora.im>
16:01:41
Don't we hold our !hi's until roll call?
<@jgroman:fedora.im>
16:01:43
!hi
<@zodbot:fedora.im>
16:01:44
Jaroslav Groman (jgroman)
<@ngompa:fedora.im>
16:01:50
!hi
<@zodbot:fedora.im>
16:01:50
Neal Gompa (ngompa) - he / him / his
<@kparal:matrix.org>
16:01:51
the bot seems delayed a lot
<@derekenz:fedora.im>
16:01:56
Hi Brandon
<@psklenar:fedora.im>
16:01:56
!hi
<@zodbot:fedora.im>
16:01:58
Petr Sklenar (psklenar)
<@kparal:matrix.org>
16:01:59
or I'm delayed a lot 🙂
<@nielsenb:fedora.im>
16:02:01
!hi
<@zodbot:fedora.im>
16:02:01
Brandon Nielsen (nielsenb)
<@kparal:matrix.org>
16:02:12
!hi
<@zodbot:fedora.im>
16:02:15
Kamil Páral (kparal) - he / him / his
<@derekenz:fedora.im>
16:02:18
Just blame AI
<@lruzicka:fedora.im>
16:02:38
!hi
<@zodbot:fedora.im>
16:02:39
Lukáš Růžička (lruzicka)
<@kparal:matrix.org>
16:02:42
we'll have to make do without Adam this time, so please bear with me leading the meeting
<@kparal:matrix.org>
16:05:15
I sent it already, but I'm not sure if the bot saw it
<@kparal:matrix.org>
16:06:14
ok, let's continue
<@kparal:matrix.org>
16:06:16
!topic Introduction
<@kparal:matrix.org>
16:06:16
!info The bugs up for review today are available at:
<@kparal:matrix.org>
16:06:16
!link https://fedoraproject.org/wiki/Fedora_44_Final_Release_Criteria
<@kparal:matrix.org>
16:06:16
!link https://fedoraproject.org/wiki/Fedora_44_Beta_Release_Criteria
<@kparal:matrix.org>
16:06:16
!link http://qa.fedoraproject.org/blockerbugs/current
<@kparal:matrix.org>
16:06:16
!info The criteria for release blocking bugs can be found at:
<@kparal:matrix.org>
16:06:16
!link https://fedoraproject.org/wiki/QA:SOP_Blocker_Bug_Meeting
<@kparal:matrix.org>
16:06:16
!link https://fedoraproject.org/wiki/Basic_Release_Criteria
<@kparal:matrix.org>
16:06:16
!info We'll be following the process outlined at:
<@kparal:matrix.org>
16:06:16
!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.
<@kparal:matrix.org>
16:06:16
Why are we here?
<@kparal:matrix.org>
16:06:36
!info 5 Accepted Freeze Exceptions
<@kparal:matrix.org>
16:06:36
!info 5 Accepted Blockers
<@kparal:matrix.org>
16:06:36
!info 0 Accepted 0-day Blockers
<@kparal:matrix.org>
16:06:36
!info 0 Accepted Previous Release Blockers
<@kparal:matrix.org>
16:06:36
!info 1 Proposed Freeze Exceptions
<@kparal:matrix.org>
16:06:36
today we have:
<@kparal:matrix.org>
16:06:36
!info 3 Proposed Blockers
<@kparal:matrix.org>
16:06:54
anyone signing up to be the meeting secretary?
<@nielsenb:fedora.im>
16:09:19
Is anyone else even here?
<@nielsenb:fedora.im>
16:09:46
Like, in the receiving messages sense.
<@kparal:matrix.org>
16:10:18
it's a cold and empty room here. Alright, let's continue, we can deal with this later.
<@kparal:matrix.org>
16:10:29
!topic Proposed Blockers
<@nielsenb:fedora.im>
16:10:35
I can do it
<@sgallagh:fedora.im>
16:10:44
!hi
<@zodbot:fedora.im>
16:10:45
Stephen Gallagher (sgallagh) - he / him / his
<@nielsenb:fedora.im>
16:10:59
Like, I don't know what I'm doing, but I've read the SOP and I'm sure I can handle it
<@kparal:matrix.org>
16:11:00
awesome! Do you know how, Brandon?
<@xariann:fedora.im>
16:11:26
!hi
<@zodbot:fedora.im>
16:11:27
Xariann Cat (xariann) - she / her / hers
<@lruzicka:fedora.im>
16:11:31
I can do it, too. And I am here, just trying to do some coreos stuff
<@kparal:matrix.org>
16:11:42
!info Brandon Nielsen will secretarialize
<@kparal:matrix.org>
16:11:42
Great 🙂
<@lruzicka:fedora.im>
16:11:51
But if Brandon wants to try, I do not object
<@nielsenb:fedora.im>
16:12:01
What could possibly go wrong.
<@kparal:matrix.org>
16:12:04
Brandon Nielsen: if you have questions, ping Lukáš Růžička
<@kparal:matrix.org>
16:12:09
!topic (2448283) Selecting a non-ASCII capable keyboard layout should automatically also select US English as a second layout
<@kparal:matrix.org>
16:12:09
!info Proposed Blocker, plasma-setup, NEW
<@kparal:matrix.org>
16:12:09
!link https://pagure.io/fedora-qa/blocker-review/issue/2078
<@kparal:matrix.org>
16:12:09
!link https://bugzilla.redhat.com/show_bug.cgi?id=2448283
<@kparal:matrix.org>
16:12:09
!info Ticket vote: FinalBlocker (+0,0,-5) (-adamwill, -ngompa, -derekenz, -nielsenb, -psklenar)
<@ngompa:fedora.im>
16:12:48
as much as I don't like it, "equally broken" is valid enough to disqualify this
<@kparal:matrix.org>
16:12:55
so, people seem to agree that this is not a blocker. Even though it seems to be a good match for our release criteria.
<@ngompa:fedora.im>
16:13:17
it is very hard to fix in the timeframe we have, and it's broken in both GNOME and KDE
<@ngompa:fedora.im>
16:13:27
the only place where this works as intended is Anaconda initial-setup
<@kparal:matrix.org>
16:13:40
so why wouldn't we accept it as a blocker, and then use the hard-to-fix waiver?
<@ngompa:fedora.im>
16:13:46
we could
<@ngompa:fedora.im>
16:13:49
I'm fine with that too
<@ngompa:fedora.im>
16:13:57
it would also give us a door to prioritize it for F45
<@nielsenb:fedora.im>
16:14:12
Do we apply the waiver *now*, or closer to release?
<@ngompa:fedora.im>
16:14:26
it would be nice if we could make it an automatic blocker for F45
<@kparal:matrix.org>
16:14:29
that will imply it's going to be an F45 Beta blocker, per our processes
<@kparal:matrix.org>
16:14:43
the waiver gets applied during go/nogo
<@ngompa:fedora.im>
16:15:08
I can basically guarantee that we'd need to waive it
<@kparal:matrix.org>
16:15:08
a hard-to-fix waiver is an automatic blocker for the next release
<@nielsenb:fedora.im>
16:15:09
Okay
<@ngompa:fedora.im>
16:15:15
there's no way it's getting fixed in 4 weeks
<@derekenz:fedora.im>
16:15:25
Good to know
<@kparal:matrix.org>
16:15:51
I think this is a better approach, honestly, because it sends a message to developers that this is important to us
<@ngompa:fedora.im>
16:15:56
yes, I agree
<@ngompa:fedora.im>
16:16:03
I have forgotten this is a tool we have
<@ngompa:fedora.im>
16:16:08
we so rarely use it
<@sgallagh:fedora.im>
16:16:15
I'll admit, I'm having a bit of trouble parsing the description in the BZ.
<@kparal:matrix.org>
16:16:19
also, we're not then violating our own release criteria principles
<@ngompa:fedora.im>
16:16:49
Stephen Gallagher: the underlying issue is that when choosing a non-Latin language, there is no secondary Latin language input mode available
<@ngompa:fedora.im>
16:17:02
you need both for things to function properly
<@kparal:matrix.org>
16:17:20
the problem is that if you need to use a keyboard layout that can't print latin characters, you should be able to configure it in anaconda, which uses a us layout as a secondary layout, and then the installed system should inherit it
<@kparal:matrix.org>
16:17:23
that doesn't happen
<@sgallagh:fedora.im>
16:17:47
Got it, thanks
<@ngompa:fedora.im>
16:17:47
Anaconda Initial Setup does this, as does Anaconda itself (obviously), but neither g-i-s nor p-s do
<@lruzicka:fedora.im>
16:18:05
it does not happe automatically, you can configure two layouts
<@sgallagh:fedora.im>
16:18:12
Yeah, this sounds like a clear violation of the cited criterion, then.
<@lruzicka:fedora.im>
16:18:18
it does not happen automatically, you can configure two layouts
<@sgallagh:fedora.im>
16:18:21
+1 FinalBlocker
<@ngompa:fedora.im>
16:18:27
+1 FinalBlocker
<@nielsenb:fedora.im>
16:18:36
FinalBlocker +1
<@lruzicka:fedora.im>
16:18:39
+1 FinalBlocker
<@derekenz:fedora.im>
16:18:41
Already voted
<@psklenar:fedora.im>
16:18:41
+1 FinalBlocker
<@derekenz:fedora.im>
16:18:46
FinalBlocker 0
<@sgallagh:fedora.im>
16:19:12
I'll defer to the experts on the "hard to fix" classification
<@kparal:matrix.org>
16:19:13
yes, there are manual workarounds, which Adam described in the bug, they seem quite annoying to use, though
<@sgallagh:fedora.im>
16:19:24
But 4 weeks is a fair amount of time.
<@kparal:matrix.org>
16:19:46
FB +1 from me as well
<@kparal:matrix.org>
16:20:14
Derek Enz: you voted -1 in the ticket, are you keeping that vote?
<@ngompa:fedora.im>
16:20:24
Stephen Gallagher: the actual problem is that rebuilding the locale and keyboard input settings module from scratch is not simple
<@derekenz:fedora.im>
16:20:27
FinalBlocker +1
<@ngompa:fedora.im>
16:20:36
the entire way that module works is wrong for this
<@ngompa:fedora.im>
16:20:58
and realistically a fix isn't landing in Plasma 6.6, but 6.7 instead
<@kparal:matrix.org>
16:21:00
keyboard layouts are hard 🙂 but it would be nice if it was fixed, yet
<@sgallagh:fedora.im>
16:21:11
Neal Gompa (Fedora): I said I'd defer to the experts :)
<@ngompa:fedora.im>
16:21:20
yeah, I'm just explaining for the benefit of logs :)
<@kparal:matrix.org>
16:21:51
ok, so if I count correctly, we have -1 blocker from Adam (not here), and +7 blocker from this meeting
<@kparal:matrix.org>
16:22:17
including people who adjusted their vote previously in the ticket
<@kparal:matrix.org>
16:22:44
so I think we can override Adam here and accept this
<@jlinton:fedora.im>
16:24:23
So, whats the solution for non english speakers, they just avoid using fedora?
<@kparal:matrix.org>
16:24:27
proposed !agreed 2448283 - accepted as a final blocker - This is a violation of "keyboard layout configured in installer must be used in initial setup" criterion, when non-latin layouts are configured (and a secondary latin layout is added by the installer).
<@sgallagh:fedora.im>
16:24:36
(Also, before anyone asks, absent a "hard-to-fix" waiver, I'd consider this as passing the "last-blocker-at-go-nogo" litmus test, for the sake of better international usage)
<@sgallagh:fedora.im>
16:25:07
Ack
<@derekenz:fedora.im>
16:25:11
ack
<@lruzicka:fedora.im>
16:25:15
They can add a second layout in Anaconda webUI during installation
<@lruzicka:fedora.im>
16:25:18
ack
<@nielsenb:fedora.im>
16:25:20
ack
<@psklenar:fedora.im>
16:25:22
ack
<@kparal:matrix.org>
16:25:56
Jeremy Linton: install with us keyboard and then later change it, or make some manual workaround steps to pass through initial setup and then change it in the system. We will document it in common bugs.
<@jlinton:fedora.im>
16:26:05
Second layout being a keyboard they don't understand, from what I understand of the defect.
<@ngompa:fedora.im>
16:26:06
ack
<@ngompa:fedora.im>
16:26:19
this is not possible anymore
<@ngompa:fedora.im>
16:26:33
at least it didn't work last I tried
<@kparal:matrix.org>
16:26:52
I think the assumption is that they understand latin, because they need to use it to create user accounts anyway
<@ngompa:fedora.im>
16:27:05
do we have the OpenQA test that verifies this works from anaconda itself?
<@lruzicka:fedora.im>
16:27:07
The installer lets you choose two layouts (maybe more but I have not tested). If one is English, you are fine.
<@lruzicka:fedora.im>
16:27:39
I tried several times for Beta and it was possible.
<@ngompa:fedora.im>
16:27:44
oh maybe I need to deliberately choose something that forces a second input layout, since I usually test with Japanese, and that uses QWERTY already
<@jlinton:fedora.im>
16:28:02
The defect talks about the primary keyboard layout then being the only one that works later in the setup, where the non latin keyboards can't actually create usable user ids.
<@kparal:matrix.org>
16:28:19
!agreed 2448283 - accepted as a final blocker - This is a violation of "keyboard layout configured in installer must be used in initial setup" criterion, when non-latin layouts are configured (and a secondary latin layout is added by the installer).
<@jlinton:fedora.im>
16:28:33
this defect is basically "the only thing that is usable is understanding enough english/latin to use the machine that way"
<@lruzicka:fedora.im>
16:28:44
When I start WebUI, I get Czech as a default which results in no English layout, but I am fine with that. However, I can easily add English to the setup.
<@lruzicka:fedora.im>
16:28:58
And then have two layouts.
<@ngompa:fedora.im>
16:29:40
that makes sense
<@kparal:matrix.org>
16:29:48
Czech is not affected by this bug, though, because it can natively input basic latin characters
<@ngompa:fedora.im>
16:29:50
I've just been testing it wrong 😓
<@ngompa:fedora.im>
16:29:54
I've just been testing it wrong 😅
<@kparal:matrix.org>
16:30:02
we're talking mostly about cyrillic and similar
<@ngompa:fedora.im>
16:30:13
Japanese isn't affected either because its base layout is Latin QWERTY
<@kparal:matrix.org>
16:30:17
ok, let's move to the next bug
<@lruzicka:fedora.im>
16:30:30
I am going to test right away.
<@kparal:matrix.org>
16:30:34
!link https://bugzilla.redhat.com/show_bug.cgi?id=2446745
<@kparal:matrix.org>
16:30:34
!info Ticket vote: FinalFreezeException (+4,0,-0) (+asciiwolf, +catanzaro, +nielsenb, +kparal)
<@kparal:matrix.org>
16:30:34
!info Proposed Blocker, selinux-policy, ON_QA
<@kparal:matrix.org>
16:30:34
!info Ticket vote: FinalBlocker (+1,0,-1) (+nielsenb, -catanzaro)
<@kparal:matrix.org>
16:30:34
!topic (2446745) gnome-initial-setup does not enable third party repos
<@kparal:matrix.org>
16:30:34
!link https://pagure.io/fedora-qa/blocker-review/issue/2080
<@ngompa:fedora.im>
16:31:10
I would consider this a blocker since default flow stuff should work
<@ngompa:fedora.im>
16:31:23
and it's rather surprising that pushing the button does nothing
<@ngompa:fedora.im>
16:31:51
+1 FB
<@kparal:matrix.org>
16:31:54
this again seems to be a direct criterion violation where people, but Michael Catanzaro voted -1 because it doesn't seem serious enough to him
<@nielsenb:fedora.im>
16:32:05
I agree it's not really serious.
<@kparal:matrix.org>
16:32:10
note that there is a pretty easy workaround - just start gnome-software and it asks about third party repos again
<@ngompa:fedora.im>
16:32:12
but also it shouldn't be hard to fix
<@ngompa:fedora.im>
16:32:29
it's not like the earlier one where we're talking about a rewrite of the whole logic
<@nielsenb:fedora.im>
16:32:30
And truthfully, I feel like "No" has been ignored for ages. You say no in initial setup, and the first time to launch Gnome Software is asks again anyway.
<@kparal:matrix.org>
16:32:42
Neal Gompa (Fedora): I suppose it's selinux again
<@mcatanzaro:gnome.org>
16:32:45
The primary risk here is embarassment. Basic things not worknig looks bad.
<@nielsenb:fedora.im>
16:33:00
But I feel like the first 5 minutes of user experience is the most important. Stuff presented in initial setup should just work.
<@ngompa:fedora.im>
16:33:08
I agree too.
<@mcatanzaro:gnome.org>
16:33:12
Reportedly there is a second problem after selinux is fixed, but we can evaluate that after the update reaches stable.
<@xariann:fedora.im>
16:33:13
I would feel like a new user would get very confused if it didn't work
<@mcatanzaro:gnome.org>
16:33:23
The primary risk here is embarassment. Basic things not working looks bad.
<@ngompa:fedora.im>
16:33:25
we should care about front door experiences, because they set the impression of the OS forever
<@nielsenb:fedora.im>
16:33:38
Right, it's hard to build a good reputation, easy to lose it.
<@derekenz:fedora.im>
16:33:45
Saw that. Is that the expected behavior?
<@kparal:matrix.org>
16:33:46
I don't have a strong opinion here, it's a direct violation, but the consequences are quite minor for most users (just another dialog prompt)
<@nielsenb:fedora.im>
16:34:11
It's been doing it for as long as I can remember if you say "No" during initial setup.
<@kparal:matrix.org>
16:34:32
if gnome-software didn't ask, I would be +1 blocker
<@derekenz:fedora.im>
16:34:35
Hmmm dont recall seeing two messages
<@derekenz:fedora.im>
16:34:45
Before
<@kparal:matrix.org>
16:35:24
I can confirm it asks simply every time (maybe except if you approve it in initial setup and it actually work, then I guess it doesn't).
<@kparal:matrix.org>
16:35:39
so it's likely expected, or at least not a regression
<@derekenz:fedora.im>
16:35:54
Ok so the initial message isnt working
<@lruzicka:fedora.im>
16:36:41
I never understood why GIS asks about it and Software again. But if it asks, it should work.
<@kparal:matrix.org>
16:36:47
currently we're at +2/-1
<@lruzicka:fedora.im>
16:36:55
FinalBlocker +1
<@derekenz:fedora.im>
16:36:56
FB +1
<@nielsenb:fedora.im>
16:37:00
Agreed, I wish it just didn't.
<@xariann:fedora.im>
16:37:01
FB +1
<@sgallagh:fedora.im>
16:37:34
+1 FinalBlocker
<@nielsenb:fedora.im>
16:37:40
FinalBlocker +1
<@kparal:matrix.org>
16:38:17
I'm somewhere between +0 and a weak +1
<@kparal:matrix.org>
16:38:48
I see +6.5/-1
<@kparal:matrix.org>
16:40:06
proposed !agreed 2446745 - accepted as a final blocker - This is a violation of "each page or panel of the initial setup utility should withstand a basic functionality test" criterion.
<@derekenz:fedora.im>
16:40:15
ack
<@jgroman:fedora.im>
16:40:23
ack
<@nielsenb:fedora.im>
16:40:25
ack
<@lruzicka:fedora.im>
16:40:41
ack
<@psklenar:fedora.im>
16:40:50
ack
<@sgallagh:fedora.im>
16:40:53
ack
<@kparal:matrix.org>
16:41:54
!agreed 2446745 - accepted as a final blocker - This is a violation of "each page or panel of the initial setup utility should withstand a basic functionality test" criterion.
<@kparal:matrix.org>
16:42:07
!topic (2448365) rpi4 fails to boot from usb drive after upgrade to RC3
<@kparal:matrix.org>
16:42:07
!info Ticket vote: FinalFreezeException (+5,0,-0) (+adamwill, +jgroman, +derekenz, +nielsenb, +kparal)
<@kparal:matrix.org>
16:42:07
!info Proposed Blocker, uboot-tools, ASSIGNED
<@kparal:matrix.org>
16:42:07
!link https://pagure.io/fedora-qa/blocker-review/issue/2079
<@kparal:matrix.org>
16:42:07
!link https://bugzilla.redhat.com/show_bug.cgi?id=2448365
<@kparal:matrix.org>
16:42:19
this is the last proposed blocker
<@kparal:matrix.org>
16:42:37
I nominated it directly because I have no idea whether it should be 😄
<@kparal:matrix.org>
16:42:51
i.e. how important is USB boot for RPi devices
<@ngompa:fedora.im>
16:43:13
my understanding is that RPi devices are blocking devices
<@ngompa:fedora.im>
16:43:17
so it probably should be
<@kparal:matrix.org>
16:43:26
do we have Peter Robinson ?
<@derekenz:fedora.im>
16:43:38
I dont but some folks use USB it seems
<@derekenz:fedora.im>
16:43:45
For the Rpi
<@ngompa:fedora.im>
16:44:05
I use usb for the rpi400 I do testing with, so it'd be nice if this worked 😅
<@kparal:matrix.org>
16:44:19
Jeremy Linton: could also help us here?
<@derekenz:fedora.im>
16:44:40
I probably should test that more lol
<@nielsenb:fedora.im>
16:44:58
I literally always forget running off USB is an option
<@kparal:matrix.org>
16:45:01
we no longer have a list of blocking devices and their features, we decided on case by case basis based on how aarch64 hardware is popular, similar to x86_64
<@nielsenb:fedora.im>
16:45:06
I'm just so used to uSD for Pis
<@lruzicka:fedora.im>
16:45:28
Jeremy Linton: So, I have now installed F44 Beta using both English and Russian keyboard layout and the Russian language. When I started Anaconda webUI, it selected Czech (based on my GEO) but left keyboard to EN which would end up in installation of a Czech system with English layout (probably not good), anyway, I added a Russian layout instead and switched to Russian, installed Fedora, did not change anything in GIS and ended up with English/Russian keyboard layout, working normally. So maybe the bug sounds ambiguous?
<@kparal:matrix.org>
16:45:32
we no longer have a list of blocking devices and their features, we decide on case by case basis based on how aarch64 hardware is popular, similar to x86\_64
<@kparal:matrix.org>
16:45:59
Lukáš Růžička: can we get back to it in the open floor perhaps?
<@lruzicka:fedora.im>
16:46:16
We can, if there is interest.
<@kparal:matrix.org>
16:46:32
or we can switch back to it once the rpi bug is done
<@jlinton:fedora.im>
16:46:34
I have a couple pi's but they are edk2... I guess its possible to reinstall one of them. sigh.
<@nielsenb:fedora.im>
16:46:54
I keep wanting to try edk2, never get to it
<@ngompa:fedora.im>
16:47:15
the experience is nicer :)
<@kparal:matrix.org>
16:47:15
let's not discuss two bugs simmultaneously
<@jlinton:fedora.im>
16:48:15
I think pbrobinson has talked about this kind of stuff in the past because of the uboot release cycle means it doesn't release until late in the fedora release cycle so it needs a last minuite update nearly every cycle.
<@lruzicka:fedora.im>
16:48:16
RPI 400 can be installed using USB? I only installed with an image burnt to a card.
<@kparal:matrix.org>
16:49:35
Jaroslav Groman: you said you used usb boot for F44 Beta and it worked ok, right?
<@jgroman:fedora.im>
16:49:54
yep, beta worked fine
<@kparal:matrix.org>
16:49:58
anyone has any idea whether this is so important feature to block F44 Final on it?
<@ngompa:fedora.im>
16:50:32
yes it can, all RPi4 series devices can with a correctly working uboot
<@ngompa:fedora.im>
16:50:34
or with edk2
<@kashyapc:fedora.im>
16:50:53
You can try in a VM :)
<@jlinton:fedora.im>
16:50:55
USB storage is almost required for fedora, i've killed a bunch of SD cards trying to run it on them, whatever magic disk pattern fedora uses for journal/etc its not kind.
<@sgallagh:fedora.im>
16:51:01
Most Pi installs are done via burning an image which can be updated after GA, correct?
<@ngompa:fedora.im>
16:51:43
we have no process for updating ARM images
<@ngompa:fedora.im>
16:51:49
that's what makes it a problem
<@kparal:matrix.org>
16:52:02
ah, so this is not just about booting the installer from a flash drive? This is actually about installing the system to a usb disk and then running the installed system from it permanently?
<@ngompa:fedora.im>
16:52:07
yes
<@kparal:matrix.org>
16:52:25
ok, I had no idea!
<@ngompa:fedora.im>
16:52:36
using the iso requires edk2, booting the disk image from usb is what we're talking about
<@kparal:matrix.org>
16:52:38
well, then this is bit more serious than I thought it was
<@sgallagh:fedora.im>
16:53:00
Yeah, with that information, I'm a +1 FinalBlocker here
<@derekenz:fedora.im>
16:53:03
Yeah
<@ngompa:fedora.im>
16:53:15
+1 FinalBlocker
<@derekenz:fedora.im>
16:53:15
FB +1
<@nielsenb:fedora.im>
16:53:29
FinalBlocker +1
<@jgroman:fedora.im>
16:53:35
FB +1
<@kparal:matrix.org>
16:54:05
I guess we wouldn't block x86 on booting the installed system (not the installer) from usb, would we? OTOH this use case seems much more useful for aarch64 than x86
<@sgallagh:fedora.im>
16:54:36
Yes we would
<@nielsenb:fedora.im>
16:54:43
Yeah, this is the most common way to boot a Pi as far as I'm aware
<@ngompa:fedora.im>
16:54:49
we definitely would
<@nielsenb:fedora.im>
16:54:54
And I'm guessing the Pi is still one of the most common aarch64 platforms
<@ngompa:fedora.im>
16:55:31
people install fedora to removable drives on x86 too, it's just not commonly done in fedora testing
<@lruzicka:fedora.im>
16:55:40
We are booting a LIVE system, which looks similar to me. I would block on it.
<@jlinton:fedora.im>
16:55:42
well normal pi4, not the p400 likely.
<@ngompa:fedora.im>
16:56:10
and of course, we _could_ make x86 disk images like we do for aarch64
<@nielsenb:fedora.im>
16:56:10
I actually don't see an install to USB test case?
<@jlinton:fedora.im>
16:56:19
There is one,
<@kparal:matrix.org>
16:56:44
proposed !agreed 2448365 - accepted as a final blocker - This is a violation of "the installed system must boot" criterion (in "Expected installed system boot behavior"), when booting a RPi system installed to a USB device.
<@sgallagh:fedora.im>
16:56:52
ack
<@derekenz:fedora.im>
16:56:53
ack
<@lruzicka:fedora.im>
16:56:57
ack
<@nielsenb:fedora.im>
16:56:59
ack
<@ngompa:fedora.im>
16:57:08
ack
<@psklenar:fedora.im>
16:57:25
ack
<@kparal:matrix.org>
16:57:31
!agreed 2448365 - accepted as a final blocker - This is a violation of "the installed system must boot" criterion (in "Expected installed system boot behavior"), when booting a RPi system installed to a USB device.
<@kparal:matrix.org>
16:57:59
ok, do you want to go back to https://bugzilla.redhat.com/show_bug.cgi?id=2446745 Lukáš Růžička ?
<@kparal:matrix.org>
16:58:04
errr
<@kparal:matrix.org>
16:58:10
back to https://bugzilla.redhat.com/show_bug.cgi?id=2448283
<@kparal:matrix.org>
16:58:42
or we can debug in bugzilla, to clarify when it breaks exactly. If it turns out to be different than we expected, we can re-do the blocker vote.
<@kparal:matrix.org>
16:59:14
I suggest trying the Adam's steps first
<@lruzicka:fedora.im>
16:59:57
Well, in Bugzilla this only describes KDE. I tried on Workstation. I will have to retry on KDE to know more.
<@kparal:matrix.org>
17:00:33
the gnome one is already upstream, linked from that bugzilla
<@ngompa:fedora.im>
17:01:00
we will probably want to make a g-i-s counterpart rhbz for f45
<@kparal:matrix.org>
17:01:04
alright, let's debug this one after the meeting/tomorrow, and move on
<@kparal:matrix.org>
17:01:30
yeah, probably
<@lruzicka:fedora.im>
17:01:39
For me, it works in Gnome as described above, so I believe that the situation does not stand like "nobody is able to use Fedora in Cyrillic countries."
<@kparal:matrix.org>
17:01:50
!topic Proposed Freeze Exceptions
<@lruzicka:fedora.im>
17:02:07
I still support to block on this not being created automatically.
<@kparal:matrix.org>
17:02:09
!link https://bugzilla.redhat.com/show_bug.cgi?id=2373699
<@kparal:matrix.org>
17:02:09
!link https://pagure.io/fedora-qa/blocker-review/issue/2007
<@kparal:matrix.org>
17:02:09
!info Proposed Freeze Exceptions, fedora-obsolete-packages, ASSIGNED
<@kparal:matrix.org>
17:02:09
!info Ticket vote: FinalFreezeException (+2,0,-0) (+kparal, +psklenar)
<@kparal:matrix.org>
17:02:09
!topic (2373699) Obsolete packages that used to require Python 3.13 but are gone in Fedora 43 and 44
<@kparal:matrix.org>
17:02:09
!info Ticket vote: BetaFreezeException (+5,0,-0) (+nielsenb, +derekenz, +lruzicka, +boniboyblue, +psklenar)
<@kparal:matrix.org>
17:02:59
we approved this for Beta, I think it's a thumbs up for Final as well
<@kparal:matrix.org>
17:03:23
Miro takes care about Python dependencies each cycle
<@kparal:matrix.org>
17:03:30
Miro takes care of Python dependencies each cycle
<@kparal:matrix.org>
17:04:52
votes?
<@nielsenb:fedora.im>
17:05:11
FinalFE +1
<@lruzicka:fedora.im>
17:05:50
FinalFE +1
<@jgroman:fedora.im>
17:06:06
FinalFE +1
<@kparal:matrix.org>
17:06:21
proposed !agreed 2373699 - accepted as a final freeze exception - This helps fix broken dependencies in the main repo and also for pre-release users.
<@derekenz:fedora.im>
17:06:41
ack
<@nielsenb:fedora.im>
17:06:46
ack
<@jgroman:fedora.im>
17:07:10
ack
<@kparal:matrix.org>
17:07:32
!agreed 2373699 - accepted as a final freeze exception - This helps fix broken dependencies in the main repo and also for pre-release users.
<@kparal:matrix.org>
17:08:02
ok, the most favorite part of the meeting, let's go through accepted blockers
<@kparal:matrix.org>
17:08:07
!topic Accepted Blockers
<@kparal:matrix.org>
17:08:28
I'll skip the oversize images bugs, there's not much to discuss there
<@kparal:matrix.org>
17:08:36
!link https://pagure.io/fedora-qa/blocker-review/issue/2074
<@kparal:matrix.org>
17:08:36
!info Accepted Blocker, mesa, NEW
<@kparal:matrix.org>
17:08:36
!topic (2359799) Inital-setup: VK_ERROR_DEVICE_LOST using nvidia hardware
<@kparal:matrix.org>
17:08:36
!link https://bugzilla.redhat.com/show_bug.cgi?id=2359799
<@kparal:matrix.org>
17:10:36
we still need someone with an nvidia card (who uses those?! the whole world, apparently... ☹️ ) to reproduce it and perhaps provide better logs, or try some workarounds, not sure. And a developer to start looking at it as well.
<@kparal:matrix.org>
17:11:26
Michael Catanzaro: have anyone from the gnome team tried to look at it? Perhaps it's some incorrect call from g-i-s/gtk after all?
<@kparal:matrix.org>
17:13:26
it just occurred to me that on my laptop I could've tried with F43, because it might actually be fixed in latest F44. Or do we have reports from F44 as well?
<@mcatanzaro:gnome.org>
17:13:31
Kamil Páral I'm afraid I don't know much about graphics. I think if anybody had investigated, they would have mentioned it in the bug report....
<@mcatanzaro:gnome.org>
17:13:57
I don't know of any reports from F44 users. F42 and F43 are definitely broken.
<@kparal:matrix.org>
17:14:53
ok, if people have some nvidia cards, please try this with F43, and if it's broken there, also with latest F44, and update bugzilla. Thanks!
<@kparal:matrix.org>
17:15:20
going to the next one
<@kparal:matrix.org>
17:15:22
!link https://bugzilla.redhat.com/show_bug.cgi?id=2391723
<@kparal:matrix.org>
17:15:22
!topic (2391723) `shim-ia32` missing since `shim-15.8-4`
<@kparal:matrix.org>
17:15:22
!link https://pagure.io/fedora-qa/blocker-review/issue/2075
<@kparal:matrix.org>
17:15:22
!info Accepted Blocker, shim, MODIFIED
<@kparal:matrix.org>
17:16:28
it seems we need to get https://bugzilla.redhat.com/show_bug.cgi?id=2391723 pushed
<@kparal:matrix.org>
17:16:40
it seems we need to get https://bodhi.fedoraproject.org/updates/FEDORA-2026-ea1a86a8eb pushed
<@kparal:matrix.org>
17:17:18
I just gave it +1 karma, please do as well
<@kparal:matrix.org>
17:17:42
once it has +3, it will go stable, and then the next image should be built with it, and we'll see if this is fixed
<@kparal:matrix.org>
17:18:25
anything else to note?
<@kparal:matrix.org>
17:19:17
ok, that brings us to...
<@kparal:matrix.org>
17:19:20
!topic Open Floor
<@kparal:matrix.org>
17:19:59
if you think blocker meetings are too short, this is your chance to bump the average length! 😄
<@kparal:matrix.org>
17:20:11
do anyone have any topic?
<@kparal:matrix.org>
17:20:17
does anyone have any topic?
<@derekenz:fedora.im>
17:20:19
Nothing here
<@nielsenb:fedora.im>
17:20:38
Not from me
<@lruzicka:fedora.im>
17:20:42
Nosing
<@kparal:matrix.org>
17:22:18
Alright, let's go home (meaning up the hallway to the kitchen, obviously)
<@kparal:matrix.org>
17:22:24
thanks for attending, everyone
<@kparal:matrix.org>
17:22:28
!endmeeting