2025-03-24 16:02:10 <@adamwill:fedora.im> !startmeeting F42-blocker-review
2025-03-24 16:02:11 <@meetbot:fedora.im> Meeting started at 2025-03-24 16:02:10 UTC
2025-03-24 16:02:12 <@meetbot:fedora.im> The Meeting name is 'F42-blocker-review'
2025-03-24 16:02:13 <@derekenz:fedora.im> sorry lol
2025-03-24 16:02:14 <@adamwill:fedora.im> i should edit the sop
2025-03-24 16:02:17 <@adamwill:fedora.im> !topic Roll Call
2025-03-24 16:02:22 <@adamwill:fedora.im> everyone say hi again!
2025-03-24 16:02:29 <@boniboyblue:fedora.im> !hi
2025-03-24 16:02:30 <@zodbot:fedora.im> Christopher Boni (boniboyblue)
2025-03-24 16:02:33 <@nhanlon:beeper.com> !hi
2025-03-24 16:02:33 <@derekenz:fedora.im> !hi
2025-03-24 16:02:33 <@nielsenb:fedora.im> !hi
2025-03-24 16:02:34 <@zodbot:fedora.im> Neil Hanlon (neil) - he / him / his
2025-03-24 16:02:34 <@zodbot:fedora.im> Derek Enz (derekenz)
2025-03-24 16:02:34 <@zodbot:fedora.im> Brandon Nielsen (nielsenb)
2025-03-24 16:04:20 <@adamwill:fedora.im> hi hi hi
2025-03-24 16:04:35 <@adamwill:fedora.im> how's everybody doing
2025-03-24 16:04:47 <@conan_kudo:matrix.org> !hi
2025-03-24 16:04:50 <@zodbot:fedora.im> Neal Gompa (ngompa) - he / him / his
2025-03-24 16:05:05 <@lruzicka:matrix.org> We are fine, refined and fined.
2025-03-24 16:05:06 <@derekenz:fedora.im> Fine thanks
2025-03-24 16:05:09 <@nhanlon:beeper.com> doing alright, how about you Adam?
2025-03-24 16:05:21 <@adamwill:fedora.im> oh, the usual
2025-03-24 16:06:26 <@kparal:matrix.org> is blockerbugs completely down?
2025-03-24 16:06:47 <@adamwill:fedora.im> it was working five minutes ago
2025-03-24 16:06:48 <@kparal:matrix.org> ah, seems to work now
2025-03-24 16:06:56 <@nielsenb:fedora.im> Loads for me
2025-03-24 16:07:07 <@adamwill:fedora.im> alrighty, let's get going with some exciting boilerplate!
2025-03-24 16:07:16 <@adamwill:fedora.im> !info The criteria for release blocking bugs can be found at:
2025-03-24 16:07:16 <@adamwill:fedora.im> !link https://fedoraproject.org/wiki/Basic_Release_Criteria
2025-03-24 16:07:16 <@adamwill:fedora.im> !link https://fedoraproject.org/wiki/Fedora_42_Beta_Release_Criteria
2025-03-24 16:07:16 <@adamwill:fedora.im> !link https://fedoraproject.org/wiki/Fedora_42_Final_Release_Criteria
2025-03-24 16:07:16 <@adamwill:fedora.im> !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.
2025-03-24 16:07:16 <@adamwill:fedora.im> Why are we here?
2025-03-24 16:07:16 <@adamwill:fedora.im> !link https://fedoraproject.org/wiki/QA:SOP_Blocker_Bug_Meeting
2025-03-24 16:07:16 <@adamwill:fedora.im> !info The bugs up for review today are available at:
2025-03-24 16:07:16 <@adamwill:fedora.im> !info We'll be following the process outlined at:
2025-03-24 16:07:16 <@adamwill:fedora.im> !topic Introduction
2025-03-24 16:07:16 <@adamwill:fedora.im> !link http://qa.fedoraproject.org/blockerbugs/current
2025-03-24 16:07:29 <@adamwill:fedora.im> !info for Final, we have:
2025-03-24 16:07:30 <@adamwill:fedora.im> !info 3 Proposed Blockers
2025-03-24 16:07:30 <@adamwill:fedora.im> !info 7 Accepted Blockers
2025-03-24 16:07:35 <@adamwill:fedora.im> !info 2 Accepted Freeze Exceptions
2025-03-24 16:07:35 <@adamwill:fedora.im> !info 1 Proposed Freeze Exceptions
2025-03-24 16:07:42 <@adamwill:fedora.im> who wants to secretarialize?
2025-03-24 16:08:42 <@kparal:matrix.org> do I see Lukas Brabec waiving his hand?
2025-03-24 16:09:22 <@kparal:matrix.org> I don't think I'm able to do it today after the meeting, btw
2025-03-24 16:09:23 <@adamwill:fedora.im> that's funny, i see that too!
2025-03-24 16:10:13 <@lruzicka:matrix.org> I can do try doing it. I do not see any hands of Lukas Brabec here.
2025-03-24 16:10:23 <@lruzicka:matrix.org> I can try doing it. I do not see any hands of Lukas Brabec here.
2025-03-24 16:10:52 <@adamwill:fedora.im> i'll do it if kamil and lukas aren't around
2025-03-24 16:11:07 <@adamwill:fedora.im> !info adamw will secretarialize
2025-03-24 16:11:15 <@adamwill:fedora.im> let's get started with:
2025-03-24 16:11:30 <@adamwill:fedora.im> !topic Proposed Final blockers
2025-03-24 16:11:38 <@adamwill:fedora.im> !info Ticket vote: FinalBlocker (+0,0,-1) (-lruzicka)
2025-03-24 16:11:38 <@adamwill:fedora.im> !topic (2353002) biosboot required even on MBR disks, not just GPT, but can't be created
2025-03-24 16:11:38 <@adamwill:fedora.im> !link https://bugzilla.redhat.com/show_bug.cgi?id=2353002
2025-03-24 16:11:38 <@adamwill:fedora.im> !link https://pagure.io/fedora-qa/blocker-review/issue/1804
2025-03-24 16:11:38 <@adamwill:fedora.im> !info Proposed Blocker, anaconda-webui, NEW
2025-03-24 16:11:38 <@adamwill:fedora.im> !info Ticket vote: FinalFreezeException (+0,1,-2) (geraldosimiao, -nielsenb, -derekenz)
2025-03-24 16:13:10 <@kparal:matrix.org> read at least https://bugzilla.redhat.com/show_bug.cgi?id=2353002#c21
2025-03-24 16:13:23 <@kparal:matrix.org> we're exploring new grounds here, with webui violating some criteria with our approval 🙂
2025-03-24 16:13:46 <@kparal:matrix.org> I'm not sure which one will be more visible, whether missing bootloader selection or MBR-related issues
2025-03-24 16:13:48 <@conan_kudo:matrix.org> since it's part of a blocking desktop, it counts for finalblocker for me
2025-03-24 16:14:02 <@conan_kudo:matrix.org> webui bugs can't be bypassed when workstation is shipping it
2025-03-24 16:14:09 <@conan_kudo:matrix.org> +1 FinalBlocker
2025-03-24 16:14:44 <@adamwill:fedora.im> yeah, if webui does not want to deal with mbr disks, it should at least do this consistently (not offer options that don't work)
2025-03-24 16:14:50 <@kparal:matrix.org> well, as I wrote, our job is to make sure included features work and are not missing by accident. But if feature is missing intentionally, it's not really QA job. The question is about communication and user expectations
2025-03-24 16:15:18 <@kparal:matrix.org> from what I tested, this is not broken in any way - it won't people's data. But the error message is quite cryptic for most people.
2025-03-24 16:15:52 <@lruzicka:matrix.org> I have found out that one can also use Workstation ISO to workaround this as I am describing in the discussion in BlockerBugs.
2025-03-24 16:16:04 <@kparal:matrix.org> also note that this is not just related to windows dual boot. That's just once common use case where this will be encountered.
2025-03-24 16:16:35 <@kparal:matrix.org> But any old Fedora install will suffer the same issue - can't be reinstalled, can't be manually mounted
2025-03-24 16:16:36 <@adamwill:fedora.im> yeah, there are many ways you might have an MBR disk
2025-03-24 16:16:52 <@adamwill:fedora.im> if you've been reusing the same disk layout since some older fedora (I think before 37?) for instance
2025-03-24 16:17:15 <@nielsenb:fedora.im> Do we have a dual boot criteria for anything but Windows and MacOS?
2025-03-24 16:17:29 <@kparal:matrix.org> I believe that we should comment on this behaviour to let people know and we should perhaps revisit in the future, if Anaconda GTK goes away and there will be no way to perform this. 
2025-03-24 16:17:29 <@kparal:matrix.org> Lukas has an interesting workaround:
2025-03-24 16:17:29 <@kparal:matrix.org> > This can be workarounded by using Anaconda GTK, it is also possible to remove the anaconda-webui package from the Live image before running Anaconda, so the system can still be reinstalled with Fedora 42 no matter which ISO one uses.
2025-03-24 16:17:29 <@kparal:matrix.org> 
2025-03-24 16:17:29 <@kparal:matrix.org> However, I'd probably advise against this, because we haven't tested this approach *at all*.
2025-03-24 16:18:18 <@lruzicka:matrix.org> We don't afaik. 
2025-03-24 16:18:37 <@lruzicka:matrix.org> I tested it. It works.
2025-03-24 16:18:55 <@lbrabec:matrix.org> Sorry I'm late, but sure, I can do it again
2025-03-24 16:19:05 <@kparal:matrix.org> that's nice, but it hasn't been *extensively* tested
2025-03-24 16:19:23 <@zodbot:fedora.im> kparal has already given cookies to lbrabec during the F41 timeframe
2025-03-24 16:19:36 <@lruzicka:matrix.org> Well, the method to switch to GTK by erasing that package has been advised to me by mkolman 
2025-03-24 16:19:49 <@adamwill:fedora.im> !info in a late substitution, lbrabec will secretarialize
2025-03-24 16:20:27 <@lruzicka:matrix.org> So, I think this is how it is supposed to work :D
2025-03-24 16:20:34 <@adamwill:fedora.im> i'd be relatively confident in that 'workaround'
2025-03-24 16:20:50 <@kparal:matrix.org> all in all, I think this is -1 blocker from me, if Anaconda really intends to release it this way. However, I'd really like to see the error message clearer (as a user, what can I do about it?)
2025-03-24 16:20:53 <@adamwill:fedora.im> since it really just gives you the same setup you'd have on any other live image
2025-03-24 16:21:01 <@adamwill:fedora.im> but, i'm not sure it's really sufficient for this
2025-03-24 16:21:21 <@adamwill:fedora.im> it seems like a long road from "the installer is giving my cryptic errors about BIOS boot partitions, wtf?" to "aha, i should uninstall anaconda-webui!"
2025-03-24 16:21:26 <@kparal:matrix.org> the problem is that this is a universal error which means "don't forget to create this partition", but with MBR, you can't
2025-03-24 16:22:00 <@conan_kudo:matrix.org> I don't think this is a viable choice given that we're literally offering a "reinstall fedora" option
2025-03-24 16:22:05 <@adamwill:fedora.im> yeah, i would really prefer to see the paths that don't work greyed out with a 'can't do this coz MBR' message, or just hidden
2025-03-24 16:22:13 <@conan_kudo:matrix.org> (that is, not making it a blocker)
2025-03-24 16:22:21 <@lruzicka:matrix.org> Therefore I suggest documenting it in CommonBugs. Also, we should know if Anaconda plans to do something about it and if not, we should probably make sure people convert their installations.
2025-03-24 16:23:00 <@lruzicka:matrix.org> Well, you cannot reinstall what ceased to be compatible.
2025-03-24 16:23:02 <@adamwill:fedora.im> i think personally i'd vote +1 with a note that this can be resolved by improving the UX, we don't require MBR support to be implemented
2025-03-24 16:23:18 <@adamwill:fedora.im> common issues is fine but not everyone reads it
2025-03-24 16:23:27 <@kparal:matrix.org> Conan Kudo: "reinstall fedora" isn't required to work, it's new and we don't have any criteria for it
2025-03-24 16:23:36 <@conan_kudo:matrix.org> sigh
2025-03-24 16:23:38 <@adamwill:fedora.im> if anaconda think that even fixing the UX isn't practical we can re-discuss or talk about waiving it
2025-03-24 16:23:40 <@kparal:matrix.org> or am I wrong, Adam?
2025-03-24 16:23:40 <@conan_kudo:matrix.org> that's probably bad
2025-03-24 16:23:59 <@conan_kudo:matrix.org> I thought we had discussed criteria for the webui
2025-03-24 16:24:08 <@adamwill:fedora.im> Kamil Páral well, we have the final criterion "The installer must be able to create and install to any workable partition layout using any file system and/or container format combination offered in a default installer configuration"
2025-03-24 16:24:12 <@adamwill:fedora.im> which is, ahem...broad
2025-03-24 16:24:17 <@adamwill:fedora.im> but could at least be *argued* to cover this
2025-03-24 16:24:17 <@jkonecny:fedora.im> I would say that we want to improve the situation but we just started the conversation about how, so it is too early to tell.
2025-03-24 16:24:45 <@kparal:matrix.org> yes, that's for custom install, we don't specifically cover "reinstall fedora while keeping home" in the guided form
2025-03-24 16:24:47 <@adamwill:fedora.im> Conan Kudo we did, i did update the criteria to some extent, but didn't entirely cover this
2025-03-24 16:24:58 <@adamwill:fedora.im> no it isn't
2025-03-24 16:25:07 <@adamwill:fedora.im> nothing in that criteria specifies 'custom' and it is not in a 'custom' section
2025-03-24 16:25:15 <@adamwill:fedora.im> nothing in that criterion specifies 'custom' and it is not in a 'custom' section
2025-03-24 16:25:34 <@conan_kudo:matrix.org> it's a direct guided option, which I assumed we had extended things to cover it
2025-03-24 16:25:43 <@adamwill:fedora.im> and the footnote says "Broadly what it's 'meant to mean' is that you should be able to do anything sane that the Installation Destination spoke attempts to let you do, without the installer exploding or failing", which implies that it is *not* specific to custom
2025-03-24 16:25:50 <@kparal:matrix.org> guided is covered here: https://fedoraproject.org/wiki/Fedora_42_Beta_Release_Criteria#Guided_partitioning
2025-03-24 16:25:56 <@adamwill:fedora.im> this is in Final
2025-03-24 16:26:06 <@adamwill:fedora.im> Final doesn't have separate sections and is intentionally broader than Beta
2025-03-24 16:26:19 <@conan_kudo:matrix.org> it also explicitly covers MBR
2025-03-24 16:26:25 <@conan_kudo:matrix.org> so this bug qualifies
2025-03-24 16:26:42 <@adamwill:fedora.im> based on the wording of the criterion and the footnote, i'd personally say we should read that criterion as covering the guided options in webui - to me that's clearly within what we intended
2025-03-24 16:26:57 <@conan_kudo:matrix.org> same
2025-03-24 16:28:06 <@nielsenb:fedora.im> I agree
2025-03-24 16:28:10 <@adamwill:fedora.im> we should update the footnote for the existence of webui, but for me the intent's pretty clear. in gtkui, "the Installation Destination spoke" contains *all* partitioning options, guided and custom
2025-03-24 16:29:04 <@kparal:matrix.org> so, we seem to agree that this can be fixed with a better UI/error message?
2025-03-24 16:29:16 <@kparal:matrix.org> I don't think we want to mandate MBR support forever
2025-03-24 16:29:41 <@adamwill:fedora.im> on the wording of the criterion a better "error message" might not strictly cover it, but certainly if the UX didn't allow you to *try* the affected paths on an MBR disk that'd be OK
2025-03-24 16:30:00 <@conan_kudo:matrix.org> I would probably expect a Change to declare MBR unsupported
2025-03-24 16:30:03 <@conan_kudo:matrix.org> that hasn't happened yet
2025-03-24 16:30:09 <@conan_kudo:matrix.org> so at least for now, I _do_ expect MBR to work
2025-03-24 16:30:11 <@adamwill:fedora.im> if the "error message" was something like "nope you can't do that, try again" but it wasn't *fatal*, we could probably argue that was also sufficient
2025-03-24 16:30:24 <@adamwill:fedora.im> if the "error message" was something like "nope you can't do that, try again" but it wasn't *fatal* - didn't crash or end the installer -  we could probably argue that was also sufficient
2025-03-24 16:30:27 <@kparal:matrix.org> yes, the Change messaging didn't include this
2025-03-24 16:30:37 <@conan_kudo:matrix.org> and MBR can't be unsupported in Anaconda as long as we still support Raspberry Pi 3
2025-03-24 16:31:02 <@adamwill:fedora.im> meh, practically speaking i think i'm ok if webui doesn't support MBR but gtkui does. that's consistent with what we decided on the other blocker where webui wasn't going to support a thing
2025-03-24 16:31:21 <@adamwill:fedora.im> i really hope nobody's trying to install workstation live on the pi 3 :P
2025-03-24 16:31:23 <@lruzicka:matrix.org> +1
2025-03-24 16:31:35 <@conan_kudo:matrix.org> it's supported currently
2025-03-24 16:31:54 <@conan_kudo:matrix.org> and people do it with the rpi3 edk thingy
2025-03-24 16:31:55 <@adamwill:fedora.im> not the live iso, surely? i didn't think pi 3 could do generic UEFI install?
2025-03-24 16:31:59 <@adamwill:fedora.im> oh yikes
2025-03-24 16:32:02 <@adamwill:fedora.im> those people must love pain
2025-03-24 16:32:07 <@conan_kudo:matrix.org> yup
2025-03-24 16:32:12 <@adamwill:fedora.im> anyhoo
2025-03-24 16:32:21 <@conan_kudo:matrix.org> and because of restrictions in rpi3 firmware, you _must_ use MBR
2025-03-24 16:32:45 <@adamwill:fedora.im> i think i'll say formally i'm +1, with the note that we aren't mandating MBR support, a better 'failure path' would be OK
2025-03-24 16:33:39 <@kparal:matrix.org> sounds good
2025-03-24 16:33:49 <@adamwill:fedora.im> other votes?
2025-03-24 16:34:11 <@derekenz:fedora.im> Sounds ok +1
2025-03-24 16:34:40 <@lruzicka:matrix.org> If blocker than do not force Anaconda to implement it in WebUI -> that could cause more harm. I am fine with better error message.
2025-03-24 16:34:42 <@nielsenb:fedora.im> I would be fine with that
2025-03-24 16:35:22 <@lruzicka:matrix.org> If blocker then do not force Anaconda to implement it in WebUI -> that could cause more harm. I am fine with better error message.
2025-03-24 16:36:24 <@adamwill:fedora.im> proposed !agreed 2353002 - AcceptedBlocker (Final) - this is accepted as a violation of "The installer must be able to create and install to any workable partition layout using any file system and/or container format combination offered in a default installer configuration". We note that resolving this does not require implementation of MBR support; this could be sufficiently addressed by not allowing the user to attempt unsupported operations on an MBR-labelled disk.
2025-03-24 16:36:48 <@lruzicka:matrix.org> ack
2025-03-24 16:36:51 <@derekenz:fedora.im> ack
2025-03-24 16:36:53 <@boniboyblue:fedora.im> ack
2025-03-24 16:37:00 <@nielsenb:fedora.im> ack
2025-03-24 16:37:23 <@adamwill:fedora.im> jkonecny if it turns out to be impossible to deal with this, let us know and we'll consider options
2025-03-24 16:37:37 <@adamwill:fedora.im> !agreed 2353002 - AcceptedBlocker (Final) - this is accepted as a violation of "The installer must be able to create and install to any workable partition layout using any file system and/or container format combination offered in a default installer configuration". We note that resolving this does not require implementation of MBR support; this could be sufficiently addressed by not allowing the user to attempt unsupported operations on an MBR-labelled disk.
2025-03-24 16:37:52 <@adamwill:fedora.im> !link https://pagure.io/fedora-qa/blocker-review/issue/1806
2025-03-24 16:37:52 <@adamwill:fedora.im> !info Proposed Blocker, cockpit, NEW
2025-03-24 16:37:52 <@adamwill:fedora.im> !topic (2354497) re-format an encrypt partition make it pretty hard to continue the installation
2025-03-24 16:37:52 <@adamwill:fedora.im> !link https://bugzilla.redhat.com/show_bug.cgi?id=2354497
2025-03-24 16:39:23 <@adamwill:fedora.im> i guess i'd say +1 under the same criterion
2025-03-24 16:39:36 <@conan_kudo:matrix.org> +1 FinalBlocker
2025-03-24 16:41:18 <@boniboyblue:fedora.im> +1 FinalBlocker for me as well.
2025-03-24 16:41:28 <@derekenz:fedora.im> +1 FinalBlocker
2025-03-24 16:41:31 <@nhanlon:beeper.com> ditto, +1 FB
2025-03-24 16:41:49 <@nielsenb:fedora.im> It's not made explicit in the criteria, but I agree it should work as the installer offers it (and I would expect it to work, as the installer also explicitly asks if you want to encrypt on a fresh install)
2025-03-24 16:41:53 <@nielsenb:fedora.im> FinalBlocker +1
2025-03-24 16:42:15 <@adamwill:fedora.im> proposed !agreed 2354497 - AcceptedBlocker (Final) - this is accepted as a violation of "The installer must be able to create and install to any workable partition layout using any file system and/or container format combination offered in a default installer configuration" (can also be argued to violate the Beta custom partitioning criteria)
2025-03-24 16:42:38 <@nielsenb:fedora.im> ack
2025-03-24 16:42:42 <@derekenz:fedora.im> ack
2025-03-24 16:42:48 <@conan_kudo:matrix.org> ack
2025-03-24 16:42:51 <@lruzicka:matrix.org> I do not understand, however, why would someone reformat a disk and wanted the same LUKS? Why can't they create a new LUKS with the same password if it is reformatted anyway?
2025-03-24 16:43:15 <@nielsenb:fedora.im> I reformat / which is under luks, but preserve home
2025-03-24 16:43:28 <@nielsenb:fedora.im> As a common installer flow, at least for me
2025-03-24 16:43:30 <@conan_kudo:matrix.org> if there are multiple security keys, it may be desirable to preserve that setup
2025-03-24 16:43:46 <@lruzicka:matrix.org> hmm, ok
2025-03-24 16:44:04 <@adamwill:fedora.im> anyway, the installer offers it, so...
2025-03-24 16:44:12 <@nielsenb:fedora.im> Yeah, it's probably not what most people do
2025-03-24 16:44:27 <@nielsenb:fedora.im> But I can see where people would find themselves doing it
2025-03-24 16:44:36 <@adamwill:fedora.im> !agreed 2354497 - AcceptedBlocker (Final) - this is accepted as a violation of "The installer must be able to create and install to any workable partition layout using any file system and/or container format combination offered in a default installer configuration" (can also be argued to violate the Beta custom partitioning criteria)
2025-03-24 16:45:07 <@nielsenb:fedora.im> If someone told me it was going to be unsupported tomorrow, I don't think it would be world ending, but then the installer shouldn't let you do it
2025-03-24 16:45:15 <@nielsenb:fedora.im> Kinda like the last two bugs I guess
2025-03-24 16:45:23 <@adamwill:fedora.im> right, again, just not supporting it is an option. but it looks like a real fix is posted anyway
2025-03-24 16:45:43 <@adamwill:fedora.im> !info Proposed Blocker, kernel, NEW
2025-03-24 16:45:43 <@adamwill:fedora.im> !link https://bugzilla.redhat.com/show_bug.cgi?id=2353148
2025-03-24 16:45:43 <@adamwill:fedora.im> !topic (2353148) Fedora 42 crashes while booting on IPU6 laptops with an ivsc chip
2025-03-24 16:45:43 <@adamwill:fedora.im> !info Ticket vote: FinalFreezeException (+1,0,-0) (+adamwill)
2025-03-24 16:45:43 <@adamwill:fedora.im> !info Ticket vote: FinalBlocker (+3,0,-4) (+geraldosimiao, +nielsenb, +eischmann, -kparal, -derekenz, -lruzicka, -boniboyblue)
2025-03-24 16:45:43 <@adamwill:fedora.im> !link https://pagure.io/fedora-qa/blocker-review/issue/1801
2025-03-24 16:45:48 <@adamwill:fedora.im> whee
2025-03-24 16:45:57 <@conan_kudo:matrix.org> oof, that's pretty bad
2025-03-24 16:46:04 <@adamwill:fedora.im> i am currently trying to figure out a kernel arg workaround for this...blocklisting enough kernel modules *should* do it
2025-03-24 16:46:16 <@adamwill:fedora.im> but yeah, if you have an affected system it's bad.
2025-03-24 16:46:22 <@conan_kudo:matrix.org> FinalBlocker +1 on the basis this is actually a F42 change too
2025-03-24 16:47:10 <@kparal:matrix.org> the fix is ready, just not released
2025-03-24 16:47:19 <@kparal:matrix.org> it's even merged to the kernel tree
2025-03-24 16:47:21 <@adamwill:fedora.im> i'm super on the fence about blocking
2025-03-24 16:47:51 <@adamwill:fedora.im> obviously a fix existing makes it easier to say 'sure, +1' but if i'm strict on myself and apply the good ol' "would we really not release if this was the last blocker on thursday?" test it's tricky
2025-03-24 16:48:08 <@kparal:matrix.org> there's no such release criterion, sorry to inform you 🙂
2025-03-24 16:48:26 <@kparal:matrix.org> "all Changes must work" would be super nice, but we'd never release
2025-03-24 16:48:39 <@conan_kudo:matrix.org> I dunno man, crashing the whole system is a pretty valid criterion :)
2025-03-24 16:48:47 <@nielsenb:fedora.im> This passes the final blocker test for me at least, just because you can't easily "fix it in post"
2025-03-24 16:48:53 <@conan_kudo:matrix.org> I would be less about making sure it worked if we didn't advertise about it
2025-03-24 16:49:24 <@kparal:matrix.org> I covered that in the blocker ticket. It is and it is not, depends on the scale.
2025-03-24 16:49:47 <@conan_kudo:matrix.org> stop making this hard kamil :P
2025-03-24 16:50:38 <@adamwill:fedora.im> i think it is legit to consider it as a factor when making these subjective calls though
2025-03-24 16:51:18 <@conan_kudo:matrix.org> right, I'd be more on adamw's side about being on the fence if it wasn't a directly advertised feature for this release
2025-03-24 16:51:29 <@conan_kudo:matrix.org> that makes this much more painful
2025-03-24 16:51:57 <@kparal:matrix.org> punt and release the new kernel, so that we can avoid this decision? 🙂
2025-03-24 16:52:22 <@conan_kudo:matrix.org> if the new kernel releases before freeze again, sure :)
2025-03-24 16:52:48 <@kparal:matrix.org> this will surely get +1 FE
2025-03-24 16:52:51 <@adamwill:fedora.im> ok, got a module_blacklist recipe that works...it's a pretty icky workaround for folks who aren't used to kernel args i guess, but it works
2025-03-24 16:53:02 <@conan_kudo:matrix.org> the only time we waived such a similar condition was when we literally couldn't get shim signed
2025-03-24 16:53:13 <@conan_kudo:matrix.org> and we just had to accept fedora didn't work on newer computers
2025-03-24 16:53:21 <@adamwill:fedora.im> that one definitely had wider impact than this, i'd say
2025-03-24 16:53:39 <@adamwill:fedora.im> this really does just appear to be post-~2023 dell laptops, afaict
2025-03-24 16:53:44 <@conan_kudo:matrix.org> it's going to be messy as that hardware spreads though, esp with lenovo and framework preloads
2025-03-24 16:53:46 <@adamwill:fedora.im> nobody else seems to be using the vsc chip
2025-03-24 16:53:53 <@conan_kudo:matrix.org> it's going to be messy as that hardware spreads though, esp with lenovo preloads
2025-03-24 16:54:05 <@adamwill:fedora.im> Conan Kudo are you not aware of the vsc wrinkle?
2025-03-24 16:54:11 <@conan_kudo:matrix.org> no?
2025-03-24 16:54:16 <@adamwill:fedora.im> this isn't affecting *all* systems with IPU6 cameras
2025-03-24 16:54:25 <@adamwill:fedora.im> only ones with some extra chip, which in practice seems to be dells only
2025-03-24 16:55:23 <@adamwill:fedora.im> https://patchwork.kernel.org/project/linux-media/cover/1690631575-15124-1-git-send-email-wentong.wu@intel.com/
2025-03-24 16:55:24 <@conan_kudo:matrix.org> my thing is that is the case _right now_, but we have no idea about future hardware being prepped right now
2025-03-24 16:55:48 <@adamwill:fedora.im> says it's "available in existing commercial platforms from multiple OEMs" but no idea who else aside from dell that'd be
2025-03-24 16:55:58 <@adamwill:fedora.im> yeah, that's a valid concern
2025-03-24 16:56:13 <@nielsenb:fedora.im> In a world where Dell is the largest non-Apple vendor, I'm not sure "Dell only" has much weight to me
2025-03-24 16:56:13 <@conan_kudo:matrix.org> Dell is an early adopter for Intel stuff because $reasons
2025-03-24 16:56:21 <@adamwill:fedora.im> people could potentially be installing f42 on hardware all the way up till mid next year, and it'll be the "current" version till november
2025-03-24 16:56:40 <@conan_kudo:matrix.org> so it's a good signal for future hardware platforms from other OEMs
2025-03-24 16:56:50 <@adamwill:fedora.im> Conan Kudo other vendors are already using IPU6 cameras, but they *haven't* picked up ivsc it seems. no idea why/why not.
2025-03-24 16:57:08 <@conan_kudo:matrix.org> I dunno either, but I've seen Dell be the start for many things
2025-03-24 16:57:17 <@nielsenb:fedora.im> Their Intel checks haven't cleared yet
2025-03-24 16:57:26 <@conan_kudo:matrix.org> so I have to assume it will spread this year because Dell used it for a couple of years
2025-03-24 16:57:42 <@conan_kudo:matrix.org> LPCAMM was the same way, as well as a bunch of other things
2025-03-24 16:57:45 <@boniboyblue:fedora.im> Does today's new kernel fix this or will it be a future release?
2025-03-24 16:58:21 <@adamwill:fedora.im> there's no f42 kernel today
2025-03-24 16:58:31 <@adamwill:fedora.im> f40 and f41 aren't affected as the bug only appears with gcc 15
2025-03-24 16:58:50 <@adamwill:fedora.im> kernel-6.14.0-0.rc7.20250321gitb3ee1e460951.60.fc43 has the fix for rawhide, it looks like
2025-03-24 16:58:54 <@adamwill:fedora.im> i don't see an f42 build with the fix yet
2025-03-24 16:59:01 <@boniboyblue:fedora.im> My bad - though someone told me 6.14 stable was pushed.
2025-03-24 16:59:18 <@adamwill:fedora.im> the next f42 build should be for the next 'milestone' (rc8 or final)
2025-03-24 16:59:24 <@boniboyblue:fedora.im> My bad - thought someone told me 6.14 stable was pushed.
2025-03-24 16:59:45 <@conan_kudo:matrix.org> with 6.14 final tagged this morning, maybe we'll see 6.14 final land for f42 today
2025-03-24 17:00:09 <@adamwill:fedora.im> so hmm
2025-03-24 17:00:14 <@adamwill:fedora.im> let's at least do the easy one:
2025-03-24 17:00:16 <@adamwill:fedora.im> +1 FE
2025-03-24 17:00:24 <@nielsenb:fedora.im> FinalFE +1
2025-03-24 17:00:34 <@boniboyblue:fedora.im> FinalFE +1
2025-03-24 17:00:39 <@derekenz:fedora.im> FinalFE +1
2025-03-24 17:00:42 <@conan_kudo:matrix.org> FinalFE +1
2025-03-24 17:01:12 <@kparal:matrix.org> +1 FE
2025-03-24 17:01:41 <@adamwill:fedora.im> ok, so that's clear
2025-03-24 17:01:45 <@adamwill:fedora.im> now...blocker votes
2025-03-24 17:01:50 <@adamwill:fedora.im> i think i'm kinda at +0.1?
2025-03-24 17:02:03 <@adamwill:fedora.im> if we don't have a clear blocker vote we can punt it and hope it goes away with the FE. :P
2025-03-24 17:02:05 <@conan_kudo:matrix.org> +1 FinalBlocker
2025-03-24 17:02:08 <@nielsenb:fedora.im> 12%-ish percent of the market, can't fix release media
2025-03-24 17:02:12 <@nielsenb:fedora.im> FinalBlocker +1
2025-03-24 17:03:05 <@kparal:matrix.org> I'm more towards -1, but I'm happy to withdraw it if most people want to block on it
2025-03-24 17:03:17 <@adamwill:fedora.im> too late, i counted it!
2025-03-24 17:03:22 <@derekenz:fedora.im> Same for me
2025-03-24 17:03:40 <@kparal:matrix.org> Pitty we don't have something like "delay 1 week most, then go"
2025-03-24 17:03:41 <@boniboyblue:fedora.im> I'm still FinalBlocker -1 on this.
2025-03-24 17:04:03 <@adamwill:fedora.im> Derek Enz same as who? is that a -1 or a +1?
2025-03-24 17:04:12 <@derekenz:fedora.im> -1
2025-03-24 17:04:53 <@derekenz:fedora.im> Agreed
2025-03-24 17:05:16 <@lruzicka:matrix.org> Same as Kamil
2025-03-24 17:05:19 <@nielsenb:fedora.im> I'm kinda surprised people are okay with unusable install media for relatively common new hardware, but the votes are the votes
2025-03-24 17:05:35 <@geraldosimiao:matrix.org> +1 FinalBlocker
2025-03-24 17:06:19 <@boniboyblue:fedora.im> If it was ALL Dell machines or ALL lenovo machines I would have a different opinion.
2025-03-24 17:06:23 <@adamwill:fedora.im> ok, so we're at +3.1 / -3 , i think
2025-03-24 17:06:27 <@adamwill:fedora.im> that sure smells like a punt!
2025-03-24 17:07:06 <@conan_kudo:matrix.org> I'm pretty disappointed too.
2025-03-24 17:07:18 <@conan_kudo:matrix.org> especially since this year is a refresh year for most companies
2025-03-24 17:07:34 <@conan_kudo:matrix.org> but in practice we'll hopefully have a fix anyway
2025-03-24 17:07:37 <@adamwill:fedora.im> proposed !agreed 2353148 - punt (delay decision) on blocker status, AcceptedFE (Final) - the vote on blocker status for this is split, but there is clear consensus for a freeze exception. As the fix for this is already lined up, we expect it will land before deciding the blocker status becomes crucial
2025-03-24 17:08:05 <@geraldosimiao:matrix.org> ack
2025-03-24 17:08:14 <@boniboyblue:fedora.im> ack
2025-03-24 17:08:19 <@conan_kudo:matrix.org> ack
2025-03-24 17:08:20 <@kparal:matrix.org> ack
2025-03-24 17:08:22 <@derekenz:fedora.im> ack
2025-03-24 17:08:32 <@nielsenb:fedora.im> Yeah, even if it's a relatively small portion of Dell laptops, that's still a lot of units
2025-03-24 17:08:33 <@nielsenb:fedora.im> ack
2025-03-24 17:08:35 <@adamwill:fedora.im> proposed !agreed 2353148 - punt (delay decision) on blocker status, AcceptedFreezeException (Final) - the vote on blocker status for this is split, but there is clear consensus for a freeze exception. As the fix for this is already lined up, we expect it will land before deciding the blocker status becomes crucial
2025-03-24 17:08:47 <@adamwill:fedora.im> !agreed 2353148 - punt (delay decision) on blocker status, AcceptedFreezeException (Final) - the vote on blocker status for this is split, but there is clear consensus for a freeze exception. As the fix for this is already lined up, we expect it will land before deciding the blocker status becomes crucial
2025-03-24 17:08:57 <@conan_kudo:matrix.org> we're going to have to figure out how to handle "new hardware borked" as a criterion
2025-03-24 17:09:40 <@conan_kudo:matrix.org> I'm not sure it's a straightforward one either
2025-03-24 17:09:42 <@kparal:matrix.org> it could work as a criterion, if we had a fix ready. I think I would support that
2025-03-24 17:09:53 <@adamwill:fedora.im> i think the existing criteria and process docs cover it fine, tbh
2025-03-24 17:10:08 <@adamwill:fedora.im> nothing in them to prevent us considering future hw
2025-03-24 17:10:15 <@adamwill:fedora.im> anyhoo
2025-03-24 17:10:21 <@nielsenb:fedora.im> It's just so hard to judge "specific hardware"
2025-03-24 17:10:31 <@adamwill:fedora.im> !info the sole proposed FE was already accepted as a blocker
2025-03-24 17:10:37 <@adamwill:fedora.im> so let's move on to
2025-03-24 17:10:37 <@nielsenb:fedora.im> Like, literally any Dell laptop will be more common than a DIY desktop config
2025-03-24 17:10:44 <@adamwill:fedora.im> !topic Accepted Final blockers
2025-03-24 17:11:12 <@adamwill:fedora.im> !info all accepted blockers are in some kind of 'fix on the way' state except:
2025-03-24 17:11:19 <@adamwill:fedora.im> !link https://pagure.io/fedora-qa/blocker-review/issue/1794
2025-03-24 17:11:19 <@adamwill:fedora.im> !topic (2352679) Fedora 42: Server boot aarch64 image exceeds maximum size
2025-03-24 17:11:19 <@adamwill:fedora.im> !link https://bugzilla.redhat.com/show_bug.cgi?id=2352679
2025-03-24 17:11:19 <@adamwill:fedora.im> !info Accepted Blocker, distribution, ASSIGNED
2025-03-24 17:11:27 <@adamwill:fedora.im> i think we did poke server WG about this
2025-03-24 17:11:47 <@adamwill:fedora.im> and it was on the agenda for the meeting last week which i didn't make it to
2025-03-24 17:11:49 <@adamwill:fedora.im> let me check the notes
2025-03-24 17:13:20 <@adamwill:fedora.im> https://meetbot-raw.fedoraproject.org/meeting_matrix_fedoraproject-org/2025-03-19/fedora-server.2025-03-19-17.00.log.txt
2025-03-24 17:13:27 <@adamwill:fedora.im> ```
2025-03-24 17:13:27 <@adamwill:fedora.im> ```
2025-03-24 17:13:27 <@adamwill:fedora.im> 2025-03-19 18:06:41 <@pboy:fedora.im> Anything to discuss? Otherwise I would close.
2025-03-24 17:13:27 <@adamwill:fedora.im> 2025-03-19 18:06:25 <@pboy:fedora.im> Let's discuss this problem on the mailing list.
2025-03-24 17:13:27 <@adamwill:fedora.im> 2025-03-19 18:05:56 <@pboy:fedora.im> I see that we have already exceeded our time. 
2025-03-24 17:13:27 <@adamwill:fedora.im> 2025-03-19 18:05:18 <@pboy:fedora.im> Any ideas? I'd hate to just change the permissible size. We deliberately limited it to take into account the conditions in critical areas of our world.
2025-03-24 17:13:36 <@adamwill:fedora.im> so it seems like pboy wanted to look into it and potentially fix it
2025-03-24 17:13:40 <@adamwill:fedora.im> i guess i can start an ml thread
2025-03-24 17:14:22 <@conan_kudo:matrix.org> I don't think he understood how the image grows because of linux-firmware
2025-03-24 17:17:22 <@adamwill:fedora.im> yeah, i'm gonna mention that
2025-03-24 17:17:32 <@adamwill:fedora.im> i haven't looked into it specifically yet but it's always my first suspect these days :/
2025-03-24 17:17:45 <@adamwill:fedora.im> we can't do much about OEMs gleefully dumping another 50MB into it upstream every week
2025-03-24 17:18:03 <@adamwill:fedora.im> besides continuing to try and split things off where we can
2025-03-24 17:18:08 <@conan_kudo:matrix.org> there was a lot of arm specific firmware dumped in this cycle
2025-03-24 17:18:25 <@conan_kudo:matrix.org> specifically with Qualcomm X1 Extreme SoCs
2025-03-24 17:18:51 <@adamwill:fedora.im> fun
2025-03-24 17:18:53 <@adamwill:fedora.im> that's probably it, then
2025-03-24 17:18:57 <@conan_kudo:matrix.org> and there's probably more, but that one I paid attention to
2025-03-24 17:19:00 <@adamwill:fedora.im> i'll run through my usual analysis later if i get time
2025-03-24 17:19:15 <@conan_kudo:matrix.org> since Fedora KDE has been trying to get a WoA device supported
2025-03-24 17:19:48 <@adamwill:fedora.im> !info server WG is aware of this and has had an initial discussion, adamw has posted a mailing list thread about it now too. we'll make sure either the image is shrunk or the limit gets bumped
2025-03-24 17:20:13 <@farchord:fedora.im> I lack enough knowledge to properly boot my laptop on Fedora KDE lol I just tried, I guess using the dtb isnt enough lol
2025-03-24 17:20:38 <@farchord:fedora.im> On the Asus Vivobook S15 btw (Snapdragon X1 Elite)
2025-03-24 17:20:41 <@adamwill:fedora.im> !topic Open floor
2025-03-24 17:20:46 <@adamwill:fedora.im> any other business, folks?
2025-03-24 17:21:12 <@boniboyblue:fedora.im> Nothing from me at this time.
2025-03-24 17:21:18 <@nielsenb:fedora.im> Not from me
2025-03-24 17:21:27 <@derekenz:fedora.im> Nothing here
2025-03-24 17:21:32 <@conan_kudo:matrix.org> nothing from me
2025-03-24 17:24:12 <@kparal:matrix.org> I have something
2025-03-24 17:24:18 <@adamwill:fedora.im> go for it
2025-03-24 17:24:33 <@kparal:matrix.org> https://discussion.fedoraproject.org/t/136593/4
2025-03-24 17:24:56 <@kparal:matrix.org> https://discussion.fedoraproject.org/t/cant-log-in-to-gnome-when-mouse-keys-accessibility-option-is-enabled/135380
2025-03-24 17:24:56 <@kparal:matrix.org> the user is proposing this as a blocker:
2025-03-24 17:25:12 <@nielsenb:fedora.im> Well that's not great
2025-03-24 17:25:20 <@kparal:matrix.org> and it is honestly quite humiliating that we keep releasing like that, because the consequences are dire
2025-03-24 17:25:35 <@adamwill:fedora.im> oh good grief
2025-03-24 17:25:52 <@kparal:matrix.org> but it's already present in several releases, so my assessment is that there's low chance of having it as a blocker
2025-03-24 17:26:03 <@kparal:matrix.org> but I want to hear your thoughts
2025-03-24 17:26:22 <@conan_kudo:matrix.org> that's freaking awful
2025-03-24 17:26:31 <@nielsenb:fedora.im> I don't see a bug? So there was a low chance of it ever becoming blocking.
2025-03-24 17:26:43 <@conan_kudo:matrix.org> that's easily remediated
2025-03-24 17:26:49 <@kparal:matrix.org> there's a gnome bug link
2025-03-24 17:26:51 <@nielsenb:fedora.im> I would almost certainly be FinalBlocker +1 today.
2025-03-24 17:26:51 <@adamwill:fedora.im> Matthias Clasen around? any hope of this getting fixed?
2025-03-24 17:26:56 <@conan_kudo:matrix.org> the forum post can be transposed into a fedora bug
2025-03-24 17:27:05 <@conan_kudo:matrix.org> it's absolutely FinalBlocker to me
2025-03-24 17:27:12 <@kparal:matrix.org> https://gitlab.gnome.org/GNOME/mutter/-/issues/3708
2025-03-24 17:27:22 <@adamwill:fedora.im> since there's an upstream bug, a downstream bug is only useful for blocker/fe tracking
2025-03-24 17:27:24 <@adamwill:fedora.im> i'd certainly give this an fe
2025-03-24 17:27:27 <@conan_kudo:matrix.org> sigh anubis
2025-03-24 17:27:45 <@adamwill:fedora.im> you have problems with anubis? i haven't had any issues wit hit
2025-03-24 17:27:59 <@adamwill:fedora.im> the fact that we ship this in existing releases makes it a bit tricky as a blocker for me
2025-03-24 17:27:59 <@kparal:matrix.org> funnily enough, I think Lukas Brabec even reinstalled his mother's computer because of this, IIRC. She had it turned on probably by some accident.
2025-03-24 17:28:35 <@kparal:matrix.org> I might have mixed up names, but it doesn't really matter
2025-03-24 17:28:48 <@conan_kudo:matrix.org> apparently this bug kicks in with any a11y stuff turned on: https://gitlab.gnome.org/GNOME/mutter/-/issues/3708#note_2388446
2025-03-24 17:29:05 <@conan_kudo:matrix.org> that's considerably more serious
2025-03-24 17:29:17 <@conan_kudo:matrix.org> apparently this bug kicks in with any a11y stuff turned on in f42: https://gitlab.gnome.org/GNOME/mutter/-/issues/3708#note\_2388446
2025-03-24 17:29:29 <@conan_kudo:matrix.org> sometimes it causes my computer to lock up
2025-03-24 17:29:36 <@kparal:matrix.org> what does `a11y.keyboard` do?
2025-03-24 17:29:40 <@conan_kudo:matrix.org> my daily driver is still a 10 year old Intel MBP
2025-03-24 17:29:53 <@adamwill:fedora.im> there are two MRs for this, both of which seem to be just sitting around
2025-03-24 17:29:58 <@adamwill:fedora.im> https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/4083 and https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/4195
2025-03-24 17:30:21 <@conan_kudo:matrix.org> it's the onscreen keyboard
2025-03-24 17:30:38 <@kparal:matrix.org> I wonder how many a11y users we actually have...
2025-03-24 17:30:55 <@nielsenb:fedora.im> None, they can't sign in
2025-03-24 17:31:00 <@conan_kudo:matrix.org> it's the onscreen keyboard, IIRC
2025-03-24 17:31:01 <@kparal:matrix.org> true
2025-03-24 17:31:08 <@adamwill:fedora.im> it looks like ubuntu backported the earlier MR
2025-03-24 17:31:09 <@kparal:matrix.org> that's what I was hinting at
2025-03-24 17:31:43 <@adamwill:fedora.im> well, there was also the infamous orca/wayland mess that was only recently resolved. but that would be for a different category of folks I guess (visual vs. movement impairment)
2025-03-24 17:32:57 <@matthiasc:gnome.org> adamw: for all the upsetness about this, it seems nobody has gotten around to looking at it in the last 5 months...
2025-03-24 17:33:10 <@adamwill:fedora.im> well, ubuntu backported one of the PRs, so i guess they're happy
2025-03-24 17:33:28 <@matthiasc:gnome.org> could be that mousekeys is just not a very commonly used a11y feature
2025-03-24 17:33:30 <@adamwill:fedora.im> carlos did 'look at it' apparently and filed https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/4195
2025-03-24 17:33:51 <@adamwill:fedora.im> but that seems to have stalled after some initial discussion with jonas
2025-03-24 17:34:13 <@matthiasc:gnome.org> i'll see if they still remember any outcomes
2025-03-24 17:34:36 <@adamwill:fedora.im> Matthias Clasen daniel noted in the earlier MR: "Despite only happening when mouse keys is enabled, we're finding this has been close to the top gnome-shell 47 crashes for Ubuntu 24.10 in the last few days."
2025-03-24 17:35:02 <@matthiasc:gnome.org> curious
2025-03-24 17:35:36 <@adamwill:fedora.im> and per some discussion on the original issue it seems the crash now also happens with `org.gnome.desktop.a11y.keyboard` set (whatever that does)
2025-03-24 17:35:55 <@conan_kudo:matrix.org> rathann isn't in this room, but he is around in other matrix rooms
2025-03-24 17:36:30 <@adamwill:fedora.im> "Enable accessibility keyboard shortcuts", according to gsettings-desktop-schemas
2025-03-24 17:36:37 <@conan_kudo:matrix.org> huh neat
2025-03-24 17:40:49 <@lruzicka:matrix.org> What exactly does mousekeys do? I will try to switch it on so what do I expect?
2025-03-24 17:41:31 <@kparal:matrix.org> control your mouse with numpad, I believe
2025-03-24 17:41:41 <@matthiasc:gnome.org> it lets you move the pointer with the numeric block, indeed
2025-03-24 17:41:56 <@matthiasc:gnome.org> but who has a numpad nowadays...
2025-03-24 17:42:13 <@adamwill:fedora.im> filed https://bugzilla.redhat.com/show_bug.cgi?id=2354592
2025-03-24 17:42:38 <@adamwill:fedora.im> i'm gonna bet on "anyone who needs this accessibility feature", for a start
2025-03-24 17:43:19 <@conan_kudo:matrix.org> ooh that means this is a RHEL 10 bug too (even though RHEL 10 is out of scope here)
2025-03-24 17:43:26 <@conan_kudo:matrix.org> since this affects GNOME 47 and 48
2025-03-24 17:43:51 <@adamwill:fedora.im> probably?
2025-03-24 17:44:41 <@conan_kudo:matrix.org> will test later and find out
2025-03-24 17:45:59 <@lruzicka:matrix.org> I do have it, but I do not want cut myself of the system, spinning a VM
2025-03-24 17:46:04 <@adamwill:fedora.im> i'd guess the bug happens whether your keyboard has one or not
2025-03-24 17:46:10 <@adamwill:fedora.im> if you turn the feature on
2025-03-24 17:46:37 <@kparal:matrix.org> I tested it with F42, the bug is still present
2025-03-24 17:46:43 <@nielsenb:fedora.im> I would think the feature would take non-numpad keyboards into account and give you WASD or something, but maybe not
2025-03-24 17:46:59 <@adamwill:fedora.im> do we want to formally vote this one now, or just note it and vote next week if it's not resolved by then?
2025-03-24 17:47:03 <@kparal:matrix.org> you don't need to have a numpad, just enable it and you won't log in again
2025-03-24 17:47:20 <@adamwill:fedora.im> quick! vote on whether we should vote!
2025-03-24 17:47:41 <@nielsenb:fedora.im> +1 no votes from the open floor
2025-03-24 17:47:42 <@kparal:matrix.org> possibly keep it for later, because some people already left I think
2025-03-24 17:48:27 <@conan_kudo:matrix.org> enough of us are here to vote
2025-03-24 17:50:19 <@lruzicka:matrix.org> We can vote.
2025-03-24 17:50:34 <@kparal:matrix.org> we don't have an accessibility criterion, and this is not "basic functionality" of gnome-settings, most probably. But the outcome is quite severe.
2025-03-24 17:50:40 <@derekenz:fedora.im> Lets do this
2025-03-24 17:51:02 <@lruzicka:matrix.org> This would be accessibility, right? Why gnome settings?
2025-03-24 17:51:05 <@kparal:matrix.org> and it's already present in stable releases
2025-03-24 17:51:25 <@kparal:matrix.org> I don't think we can justify +1
2025-03-24 17:51:25 <@adamwill:fedora.im> i'd call it a conditional violation of the 'must boot to a working desktop' criterion
2025-03-24 17:51:33 <@adamwill:fedora.im> the condition being 'you need any affected accesibility feature
2025-03-24 17:51:47 <@kparal:matrix.org> unless there's someone with a high degree of criteria-fu...
2025-03-24 17:51:51 <@adamwill:fedora.im> well, i could topic it if we wanted to vote.
2025-03-24 17:51:51 <@kparal:matrix.org> like Adam
2025-03-24 17:52:27 <@conan_kudo:matrix.org> let's topic it and vote
2025-03-24 17:52:38 <@lruzicka:matrix.org> We did not know about the bug, did we? So basically we should not advocate with "we have released"
2025-03-24 17:52:48 <@conan_kudo:matrix.org> right
2025-03-24 17:53:04 <@adamwill:fedora.im> !topic (2354592) GNOME crashes on startup if keyboard accessibility features are enabled
2025-03-24 17:53:14 <@nielsenb:fedora.im> Also with the "we have released" argument, if it's something that keeps people from using Fedora, how / why would they report it?
2025-03-24 17:53:15 <@adamwill:fedora.im> !link https://bugzilla.redhat.com/show_bug.cgi?id=2354592
2025-03-24 17:53:27 <@adamwill:fedora.im> !info Proposed Blocker, mutter, POST
2025-03-24 17:53:34 <@adamwill:fedora.im> !info this is a late proposed blocker
2025-03-24 17:53:57 <@adamwill:fedora.im> the 'we already released' argument is a *practical* one
2025-03-24 17:54:06 <@adamwill:fedora.im> it more or less runs thus:
2025-03-24 17:54:26 <@adamwill:fedora.im> if we take a bug as a blocker, the *practical effect* of that for users is that when they go to download the "current Fedora", they will still get the old release, not the new one, until we fix the bug
2025-03-24 17:54:40 <@adamwill:fedora.im> however, since the old release *also contains the same bug*, we don't achieve any concrete improvement by blocking on it
2025-03-24 17:54:54 <@adamwill:fedora.im> a user who downloads "the current Fedora" is affected by the bug whether we block on it or not
2025-03-24 17:55:20 <@adamwill:fedora.im> the counterargument i guess is that we look worse if we knew about the problem and said 'eh we don't care we'll ship anyway'
2025-03-24 17:55:39 <@lruzicka:matrix.org> this +1
2025-03-24 17:55:45 <@nielsenb:fedora.im> And improvements to "future" Fedoras should hold weight
2025-03-24 17:55:58 <@conan_kudo:matrix.org> tbh, I am not super-motivated by "well it was always borked so it's fine"
2025-03-24 17:56:23 <@conan_kudo:matrix.org> we didn't _know_ it was broken until now
2025-03-24 17:56:24 <@nielsenb:fedora.im> There are times I can see looking the other way
2025-03-24 17:56:39 <@nielsenb:fedora.im> But when it's "oh hey, a class of people just can't use our marquee product"...
2025-03-24 17:57:02 <@kparal:matrix.org> well I published the common issue article in November
2025-03-24 17:57:18 <@matthiasc:gnome.org> immediate success, carlos merged one of the fixes
2025-03-24 17:57:20 <@adamwill:fedora.im> this is proposed as a violation of Basic criterion "A system installed with a release-blocking desktop must boot to a log in screen where it is possible to log in to a working desktop using a user account created during installation or a 'first boot' utility" in the case that the user requires an affected accessibility option
2025-03-24 17:57:31 <@adamwill:fedora.im> yay, we like success
2025-03-24 17:57:44 <@adamwill:fedora.im> i'm clearly +1 FE on this
2025-03-24 17:57:51 <@nielsenb:fedora.im> Yeah
2025-03-24 17:57:52 <@conan_kudo:matrix.org> success is great
2025-03-24 17:57:55 <@nielsenb:fedora.im> FinalFE +1
2025-03-24 17:58:11 <@kparal:matrix.org> FE +1
2025-03-24 17:58:12 <@derekenz:fedora.im> +1 FE
2025-03-24 17:58:16 <@adamwill:fedora.im> i'll do a quick build with the patch applied so we can be sure it actually works
2025-03-24 17:58:18 <@geraldosimiao:matrix.org> +1 FE
2025-03-24 17:58:57 <@conan_kudo:matrix.org> +1 FE +1 FB
2025-03-24 17:58:59 <@adamwill:fedora.im> ok, so we have clear FE approval
2025-03-24 17:59:07 <@adamwill:fedora.im> blocker votes? i'm kinda +0
2025-03-24 17:59:20 <@adamwill:fedora.im> (which is infinitesimally more blocker-y than -0)
2025-03-24 17:59:20 <@conan_kudo:matrix.org> +1 FinalBlocker
2025-03-24 17:59:23 <@nielsenb:fedora.im> FinalBlocker +1
2025-03-24 17:59:42 <@kparal:matrix.org> +0
2025-03-24 18:00:09 <@derekenz:fedora.im> +1 FinalBlocker
2025-03-24 18:00:40 <@adamwill:fedora.im> that's +3.(2*infinitesimal amount)
2025-03-24 18:01:35 <@adamwill:fedora.im> proposed !agreed 2354592 - AcceptedBlocker (Final) - this is accepted as a conditional violation of "A system installed with a release-blocking desktop must boot to a log in screen where it is possible to log in to a working desktop using a user account created during installation or a 'first boot' utility" in the case that an affected a11y setting is enabled
2025-03-24 18:01:47 <@conan_kudo:matrix.org> ack
2025-03-24 18:01:52 <@derekenz:fedora.im> ack
2025-03-24 18:02:11 <@nielsenb:fedora.im> ack
2025-03-24 18:02:54 <@adamwill:fedora.im> !agreed 2354592 - AcceptedBlocker (Final) - this is accepted as a conditional violation of "A system installed with a release-blocking desktop must boot to a log in screen where it is possible to log in to a working desktop using a user account created during installation or a 'first boot' utility" in the case that an affected a11y setting is enabled
2025-03-24 18:03:00 <@adamwill:fedora.im> alrighty
2025-03-24 18:03:04 <@adamwill:fedora.im> !topic Open floor, take 2
2025-03-24 18:03:07 <@adamwill:fedora.im> any other other business? :D
2025-03-24 18:03:18 <@kparal:matrix.org> nothing more from me 🙂
2025-03-24 18:03:28 <@conan_kudo:matrix.org> nothing more from me
2025-03-24 18:03:38 <@derekenz:fedora.im> nothing from me
2025-03-24 18:03:49 <@lruzicka:matrix.org> nothingham
2025-03-24 18:03:50 <@nielsenb:fedora.im> Nothing from me
2025-03-24 18:04:36 <@adamwill:fedora.im> alrighty, thanks a lot folks!
2025-03-24 18:04:40 <@adamwill:fedora.im> see you next bat time
2025-03-24 18:04:42 <@adamwill:fedora.im> !endmeeting