17:00:08 #startmeeting F36-blocker-review 17:00:08 Meeting started Mon Feb 28 17:00:08 2022 UTC. 17:00:08 This meeting is logged and archived in a public location. 17:00:08 The chair is adamw. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions. 17:00:08 Useful Commands: #action #agreed #halp #info #idea #link #topic. 17:00:08 The meeting name has been set to 'f36-blocker-review' 17:00:10 #meetingname F36-blocker-review 17:00:10 The meeting name has been set to 'f36-blocker-review' 17:00:13 .hello2 17:00:14 #topic Roll Call 17:00:14 frantisekz: frantisekz 'František Zatloukal' 17:00:15 .hello 17:00:16 cmurf: (hello ) -- Alias for "hellomynameis $1". 17:00:24 morning folks, who's around to review some blockers? 17:00:28 .hello cmurf 17:00:29 cmurf: cmurf 'Chris Murphy' 17:00:40 i exist! 17:01:09 .hello2 17:01:10 coremodule: coremodule 'Geoffrey Marr' 17:01:14 * coremodule willing to act as secretary. 17:01:15 .hello geraldosimiao 17:01:16 geraldosimiao: geraldosimiao 'Geraldo S. Simião Kutz' 17:01:23 cmurf: [citation needed] 17:01:32 .hello2 17:01:33 lruzicka: lruzicka 'Lukáš Růžička' 17:02:48 I'm not canucking today 17:03:35 .hello2 17:03:36 bcotton: bcotton 'Ben Cotton' 17:04:03 ahoyhoy everyone 17:04:11 hoy 17:04:40 #chair lruzicka bcotton 17:04:40 Current chairs: adamw bcotton lruzicka 17:04:47 impending boilerplate alert! 17:04:55 Why are we here? 17:04:55 #topic Introduction 17:04:57 #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. 17:05:01 #info We'll be following the process outlined at: 17:05:05 #link https://fedoraproject.org/wiki/QA:SOP_Blocker_Bug_Meeting 17:05:08 #info The bugs up for review today are available at: 17:05:10 #link http://qa.fedoraproject.org/blockerbugs/current 17:05:13 #info The criteria for release blocking bugs can be found at: 17:05:16 #link https://fedoraproject.org/wiki/Basic_Release_Criteria 17:05:19 #link https://fedoraproject.org/wiki/Fedora_36_Beta_Release_Criteria 17:05:23 #link https://fedoraproject.org/wiki/Fedora_36_Final_Release_Criteria 17:05:33 #info for Beta, we have: 17:05:42 #info 2 Proposed Blockers 17:05:42 #info 8 Accepted Blockers 17:05:44 #info 6 Proposed Freeze Exceptions 17:05:48 #info 3 Accepted Freeze Exceptions 17:05:59 #info for Final, we have: 17:06:00 #info 7 Proposed Blockers 17:06:09 #info 4 Accepted Blockers 17:06:09 #info 1 Proposed Freeze Exceptions 17:06:18 #info coremodule will secretarialize, thanks! 17:06:30 so, let's get started with: 17:06:34 #topic Proposed Beta blockers 17:06:41 #topic (2057193) f36 composes still have firefox 96, f34 f35 have firefox 97 17:06:45 #link https://bugzilla.redhat.com/show_bug.cgi?id=2057193 17:06:49 #link https://pagure.io/fedora-qa/blocker-review/issue/624 17:06:52 #info Proposed Blocker, firefox, NEW, depends on other bugs 17:06:57 #info Ticket vote: BetaBlocker (+2,0,-0) (+bcotton, +nielsenb) 17:06:59 #info Ticket vote: BetaFreezeException (+3,0,-0) (+churchyard, +imsedgar, +nielsenb) 17:08:12 so, per the last two comments on the bug, downgrading from 97 to 96 is known to cause major problems 17:08:17 +1 BetaBlocker 17:08:32 since an upgrade from f35 to f36 would cause that to happen by default, i think, i'd say +1 blocker to this 17:09:05 oh, wait, hm 17:09:08 i think it's clearly a final blocker 17:09:11 not sure about beta 17:09:37 i think we could plausibly argue it for Beta on "All known bugs that can cause corruption of user data must be fixed or documented at Common F36 bugs. " 17:10:04 a firefox user profile is pretty significant 'user data', if mine was rendered unusable by an upgrade 35->36 i'd be narked off 17:11:11 there's also a basic criterion "It must be possible to run the default web browser and a terminal application from all release-blocking desktop environments" 17:11:23 which may be violated by the broken state, by the looks of Chris' mozilla report 17:11:32 adamw, yes, that sounds good ... we should make upgrade as reliable as possible 17:11:33 +1 BetaBlocker then if we find suitable criterion, otherwise BetaFE and FinalBlocker 17:12:01 yeah, and cmurf proposed it with the "hinders execution of test plans or dramatically reduces test coverage" criterion, and i think "destroys my default browser's profile" would drive away a lot of testers 17:12:15 ok, so i think i'm +1 beta blocker and we can cite all three of those 17:12:17 so there are several plausible justifications for accepting this 17:12:55 yeah there's no reasonable way to disambiguate whether behaviors are bugs or due to this 17:13:43 at one point 1 in 3 websites were doing some weird shit and i kept thinking this is one f'd up build of firefox - i was super confused as you can see from the upstream bug report 17:13:59 blow away the profile, voila, fixed 17:15:06 ok, other votes? 17:15:14 there's a test@ rfc i sent to make this a release criteria so we can just skip over this question - it's come up before when we even shipped a beta with a downgraded firefox 17:15:35 * this question in the future - it's 17:15:59 +1 FE 17:15:59 +1 Final Blocker 17:16:08 +1 beta blocker 17:17:38 Ben Cotton (he/him): frantisekz are you +1? 17:17:49 aye 17:17:51 i.e. do you think the plausible justifications are plausible enough? :D 17:17:56 FF +1 BB I(If my vote accepted) 17:18:15 s/I// 17:18:33 all votes are accepted! 17:18:42 vote early and vote often, folks 17:18:45 +1 BB 17:21:41 proposed #agreed 2057193 - AcceptedBlocker (Beta) - upgrade criterion footnote requires upgraded system to meet other release criteria, we consider this to violate Basic criterion footnote "The web browser must be able to download files, load extensions (if applicable), and log into FAS" and Beta criterion "All known bugs that can cause corruption of user data must be fixed or documented at Common F36 bugs." if Firefox is downgraded 17:21:41 from 97 to 96 on upgrade to F36 17:21:49 did that make it on one line, IRC folks? 17:22:31 not quite 17:22:51 grr 17:23:06 "from 97 to 96 on upgrade to F36" on next line 17:23:30 proposed #agreed 2057193 - AcceptedBlocker (Beta) - upgrade criterion footnote requires upgraded system to meet other release criteria, we consider this to violate Basic criterion footnote "The web browser must be able to download files, load extensions (if applicable), and log into FAS" and Beta criterion "All known bugs that can cause corruption of user data must be fixed or documented at Common F36 bugs" if Firefox downgraded on 17:23:30 upgrade to F36 17:23:49 uhh i don't hate it 17:23:53 better? if it missed by less than 9 characters we're good because of "proposed" 17:23:56 but thing is, the browser mostly works 17:24:04 it's about 1/3 of cases that don't 17:24:13 cmurf: don't complicate things! it's easy this way. :D 17:24:19 ack 17:24:24 i think the stronger argument is the beta criterion catch all for inhibiting testing 17:24:39 but i'll General Ackbar this one because otherwise it's a trap 17:24:43 i'd like to include all three but there isn't space 17:24:48 we can reference all three on the bug or ticket 17:24:57 i'll add a supplementary #info too 17:26:18 ack 17:26:27 ACK 17:26:41 #agreed 2057193 - AcceptedBlocker (Beta) - upgrade criterion footnote requires upgraded system to meet other release criteria, we consider this to violate Basic criterion footnote "The web browser must be able to download files, load extensions (if applicable), and log into FAS" and Beta criterion "All known bugs that can cause corruption of user data must be fixed or documented at Common F36 bugs" if Firefox downgraded on upgrade to 17:26:41 F36 17:26:44 ack 17:28:02 #info for 2057193 we also cite the Beta blocker catch-all wording, that "A bug is considered a Beta blocker bug if ... Bug hinders execution of required Beta test plans or dramatically reduces test coverage" - we believe this bug would have that effect, as it would prevent people who would otherwise upgrade to the Beta and help test from doing so 17:28:19 ok, next bug! 17:28:20 #topic (2057719) After upgrade gnome-control-center to 42~beta-1.fc37 unable to configure VPN connection 17:28:26 #link https://bugzilla.redhat.com/show_bug.cgi?id=2057719 17:28:32 #link https://pagure.io/fedora-qa/blocker-review/issue/630 17:28:33 #info Proposed Blocker, gnome-control-center, NEW 17:28:43 #info Ticket vote: FinalBlocker (+1,0,-0) (+bcotton) 17:28:45 #info Ticket vote: BetaFreezeException (+1,0,-0) (+bcotton) 17:28:56 btw, bcotton gets the gold star for ticket voting this week, everyone else gets a lump of coal 17:29:32 🎉 17:29:32 😶 17:30:14 umm fc37? 17:30:45 cmurf: comments suggest it affects 42~beta in f36 17:30:58 which means it doesn't affect stable right now, but, we have an FE for 42~beta to be included and the update is pending stable 17:31:10 so i could push it into f36 right now (and mclasen asked me to do so). so we should probably consider this 17:31:13 got it 17:31:21 i feel like we discussed specific VPN requirements before? 17:31:28 but i don't see any in the criteria. hmm. let me go digging 17:31:42 well actually the FE is for gnome 42 RC being in f36 beta 17:31:50 i think it was with the systemd-resolved change a release or two or three ago (what even is time?) 17:32:23 cmurf: oh. yes. we should probably re-title that then. 17:32:39 pretty sure i titled it rc 17:32:55 yes 17:33:02 so we could change that. maybe it should be vaguer. :P 17:33:06 i'll, uh, throw that on the agenda for later 17:33:29 ship GNOME 42.rc in Fedora 36 beta ? 17:33:55 i think what makes me do a double take is we just got 42.beta into stable 17:34:04 cmurf: no we didn't 17:34:06 that's the point 17:34:09 42~beta is not in stable yet 17:34:19 🤦 17:34:37 ok, i guess i got it from u-t then 17:34:45 anyhoo, yeah, there's a thread from august 2020 titled "Release criteria proposal: networking requirements" 17:35:21 i sent a second draft on 2020-08-28 17:36:05 after that there was some argy bargy about mDNS and kamil said the draft looked good, and then we forgot about it 17:36:20 business as usual at fedora hq, then 17:37:16 * adamw resurrects the thread 17:37:22 +1 FE 17:37:22 +1 FinalBlocker 17:38:14 fwiw, the wording i proposed was "Using the default network configuration tools for the console and for 17:38:14 release-blocking desktops, it must be possible to establish a working 17:38:14 connection to common OpenVPN, openconnect-supported and vpnc-supported 17:38:14 VNC servers with typical configurations." as a Basic criterion. 17:38:39 I would support this criterion. 17:38:39 we could punt this for a week to resurrect the criterion discussion, i guess? 17:38:44 well, accept it as FE and punt on blocker 17:38:45 let's do it 17:38:51 ack 17:38:53 cos i'm definitely +1 FE 17:39:35 +1 FE 17:39:58 i'm not convince this should be a beta blocker, but i think accepting an FE and punting on blocker for criterion discussion is a good approach 17:40:42 +1fe 17:40:44 +1 FE 17:41:27 ACK 17:42:14 i think punt is enough, we're not at final freeze yet 17:42:55 proposed #agreed 2057719 - punt (delay decision) on blocker status, AcceptedFreezeException (Beta) - we do not have specific criteria for VPN connections, and trying to stretch existing criteria to fix this is difficult. we do have proposed specific criteria from 2020, so we will resurrect the discussion on those criteria and decide on blocker status when that's done. This is definitely serious enough to grant an FE, though 17:43:05 ack 17:43:23 ack 17:44:46 ack 17:45:38 #agreed 2057719 - punt (delay decision) on blocker status, AcceptedFreezeException (Beta) - we do not have specific criteria for VPN connections, and trying to stretch existing criteria to fix this is difficult. we do have proposed specific criteria from 2020, so we will resurrect the discussion on those criteria and decide on blocker status when that's done. This is definitely serious enough to grant an FE, though 17:45:58 OK, since we're in Beta freeze, next let's do: 17:46:02 #topic Proposed Beta freeze exceptions 17:46:13 #topic (2058502) libyui-mga-gtk won't currently install on fc36 or Rawhide 17:46:18 👋 17:46:18 #link https://bugzilla.redhat.com/show_bug.cgi?id=2058502 17:46:23 #link https://pagure.io/fedora-qa/blocker-review/issue/634 17:46:28 #info Proposed Freeze Exceptions, libyui-mga-gtk, MODIFIED 17:46:34 yes, Conan Kudo ? did you bring enough gum for the whole class? 17:46:50 nope 17:46:59 just hangry and wanting to get my blockers/FEs through :) 17:47:15 okay, so three of the proposed FEs are fails-to-install issues 17:47:38 which we typically grant FEs on the basis that they cause upgrades to require --allowerasing if you have the package installed, and we want to avoid that if possible 17:47:54 so i'm gonna propose a blanket vote on those 17:48:20 +1 blanket BetaFE for FTI bugs 17:48:35 +1 blanket BetaFE for FTI/FTBFS bugs 17:48:41 proposal is we accept 2058502 , 2046220 and 2044286 as FE as they are fails-to-install bugs 17:48:43 +1 from me 17:48:51 ack 17:48:54 +1 17:48:57 i should probably write something for the automatic FE page or something here. 17:49:01 +1 blanket beta fe for fti/ftbfs bugs 17:49:34 yeah, I think FTI/FTBFS bugs should get automatic FEs 17:49:50 it's not like we can't identify them from the blocker tickets associated with them 17:49:50 proposed #agreed 2058502 , 2046220 and 2044286 are all: AcceptedFreezeException (Beta) as they are fails-to-install bugs. We typically grant FEs for these to avoid upgrades of systems with these packages installed requiring the use of `--allowerasing`, as that can be dangerous and we want to avoid it where possible 17:50:04 ack 17:50:08 ack 17:50:09 ack 17:50:24 #agreed 2058502 , 2046220 and 2044286 are all: AcceptedFreezeException (Beta) as they are fails-to-install bugs. We typically grant FEs for these to avoid upgrades of systems with these packages installed requiring the use of --allowerasing, as that can be dangerous and we want to avoid it where possible 17:50:33 okey dokey 17:50:34 #topic (2054921) dnf system-upgrade 35 to 36 fails with various pipewire wireplumber conflicts 17:50:41 #link https://bugzilla.redhat.com/show_bug.cgi?id=2054921 17:50:43 #link https://pagure.io/fedora-qa/blocker-review/issue/612 17:50:48 #info Proposed Freeze Exceptions, pipewire, NEW 17:50:54 #info Ticket vote: BetaFreezeException (+1,0,-0) (+bcotton) 17:51:08 +1 BetaFE 17:51:15 did we actually figure out any way to fix this yet? i'm not entirely sure it's possible 17:51:19 though I'd consider that blockery honestly 17:51:50 well, i guess we still didn't figure out exactly what's going on. 17:51:53 this is the first time I've ever seen something like this actually 17:52:03 it's already rejected as a blocker. 17:52:03 last night I did such an upgrade F35kde - F36 and pipewire did well 17:52:05 oh, I see what's happening 17:52:11 +1 BetaFE 17:52:31 it's because someone explicitly swapped from wireplumber back to pipewire-media-session 17:52:39 and dnf wants to swap them back on upgrade 17:52:46 because that's what comps says we should do 17:52:49 Eighth_Doctor, this is what I believe, too 17:52:51 Conan Kudo: no, not necessarily. there was also a time during f35 cycle where you got pipewire-media-session on upgrade. 17:52:59 only problems here was plocate and libyui-mga-gtk 17:53:00 so the swappable one-and-only-one setup blocks it until the user resolves the problem 17:53:08 but the question is why dnf wants to pull in wireplumber now. 17:53:42 it should always pull wireplumber since f35 17:53:52 adamw, I think it wanted to pull it for F35 already 17:54:01 if it pulled pipewire-media-session ever, something messed up 17:54:16 but some people experienced troubles at some point so they reverted media-session 17:54:45 Conan Kudo: it did. we know it did. it was a blocker bug for f35. 17:55:04 that bug should just be closed IMO, and i'm the one who filed it 17:55:05 lruzicka: true 17:55:26 but the point is, if you have a system with pipewire-media-session installed right now, that satisfies deps and what we want to happen on upgrade is you keep it. we don't want the system to try and switch to wireplumber. 17:55:43 the root I had been using was not the root I thought it was, I thought it was an f35 ga clean install, but it was an f34 clean install upgraded to f36 after beta freeze and before beta GA 17:56:09 adamw: fair enough 17:56:23 i'm 99% certain adamw has it exactly right, there was a period of ambiguity with wireplumber during that time frame and i just got sucked into the vortex 17:56:32 well, the way to fix that is to make dnf warn instead of error when something is not satisfiable from comps 17:56:46 like it does when no package is found 17:56:54 cmurf, adamw: this is also a problem, because there might be people who would upgrade from 34 so they will experience this with this release 17:56:57 virtually certain that only testers would get hit by this 17:57:04 lruzicka: no, they won't 17:57:10 there is no pipewire-media-session in f34 17:57:19 if you upgrade from clean f34 you'll get switched to wireplumber, which is what we want 17:57:44 adamw, hah ... then no big deal, we should advice people to leave media-session and switch over 17:57:56 yes and somehow i got both and it's not worth trying to figure out why that happened for that 2-3 week period between f35 beta freeze and GA 17:58:42 lruzicka: the case we care about now is people who specifically switched to pipewire-media-session because wireplumber is broken for them. the best outcome for them is if the upgrade keeps them on pipewire-media-session, rather than failing or forcing a switch to wireplumber. 17:58:53 however, it may be tricky to actually achieve that... 17:59:07 adamw, but do we know it IS STILL broken for them? 17:59:08 gotcha, that's a valid concern 17:59:11 you can --exclude=wireplumber and that'll work 17:59:44 but rather than those kinds of work arounds being used regularly in the wild, we need to escalate bug fixes for those users 17:59:45 lruzicka: we don't, but on principle we should probably let them choose to try switching back. i guess. 17:59:46 because then dnf falls through to the "no package" case for comps 18:00:15 hmm, i wonder what would happen if we made it 'default' in comps. right now the type isn't specified which i think means it gets assumed to be 'mandatory'. 18:00:40 well, it's assumed to be default, I think 18:00:48 hence, well... default? 18:01:10 wireplumber should be the default 18:01:12 but I think we explicitly set mandatory 18:01:21 Conan Kudo: nope. it's unspecified. 18:01:30 mmmm 18:01:31 wireplumber 18:01:36 (in the multimedia group) 18:01:59 looking at libcomps, it actually has a specific case for this, it calls it COMPS_PACKAGE_UNKNOWN . dunno what things down the stack do with that. 18:02:28 aaanyhoo 18:02:33 let's try and get back on track 18:02:43 assuming there is something that can be done to make this case Better(tm), do we think it merits an FE? 18:03:05 * Eighth_Doctor shrugs 18:03:05 i'm gonna say 'potentially yes', so +1 but would review any update that claims to address this bug carefully 18:03:20 I suppose so... +1 BetaFE 18:03:28 but I suspect there's nothing we can really do here 18:03:29 yes, if we can fix it for those who need that, let's do it, I say 18:03:35 +1 BetaFE 18:03:43 this is basically comps weirdness 18:04:05 +1 BetaFE 18:04:12 well, comps and dnf. i guess it's possibly the case that we could change how dnf handles this case without breaking the world 18:07:10 proposed #agreed 2054921 - AcceptedFreezeException (Beta) - this seems like a tricky case to handle, but if there is in fact some way we can make it better, we think it's worth a freeze exception so that it reaches people upgrading before the Beta release 18:07:22 ack 18:07:29 ack 18:07:31 ack 18:07:39 ack 18:07:52 ack 18:08:49 #agreed 2054921 - AcceptedFreezeException (Beta) - this seems like a tricky case to handle, but if there is in fact some way we can make it better, we think it's worth a freeze exception so that it reaches people upgrading before the Beta release 18:09:07 #topic (2057184) cgroupify: Error moving pid into new cgroup (11) 18:09:10 #link https://bugzilla.redhat.com/show_bug.cgi?id=2057184 18:09:15 #link https://pagure.io/fedora-qa/blocker-review/issue/623 18:09:18 #info Proposed Freeze Exceptions, uresourced, NEW 18:10:21 so i was a 0 because i have no idea what the effect is here. it seems like it's just a spurious log message, which is a pretty meh reason to grant a freeze exception 18:11:50 I'm fine with +1 BetaFE here 18:12:27 yeah, i can't quite parse the consequences of this from the upstream report 18:12:34 but, i mean, my system didn't explode yet, so. 18:12:44 kinda -1 on this unless someone can explain it's more serious than a log message. 18:13:05 it means that cgroupify failed to move processes into a cgroup 18:13:26 which means that the resource control enforced by things like systemd-oomd would be broken 18:14:12 iirc, cgroupify is used to handle this for terminal processes, chrome tabs, etc. which normally don't get put into a single cgroup 18:14:12 did it actually fail to do that, though? 18:14:13 or did it do it, then try to do it again by mistake? 18:14:32 or did it not fail, but think it failed? 18:14:39 (sort of looks a bit like the latter to me) 18:15:41 it looks like 1 18:15:49 where it failed to put it into the scope 18:15:56 no, i think it's 3. 18:16:46 the fix changes what `move_to_subgroup` returns. it does not change what it *does*. 18:16:53 hmm, I see 18:16:57 and the consequence of what it returns is...an error gets logged if what it returns is not 0. 18:18:23 so, yeah, if I'm understanding that right, i'm -1 on principle. wouldn't likely hurt to take this, but that's what we always think. :D 18:18:59 lol 18:19:51 okay, i'm convinced to get off the fence and also go -1 18:21:04 we can always revote if i'm wrong 18:22:13 sure 18:22:24 any other votes? 18:23:36 okay, well, don't all rush at once! 18:24:20 proposed #agreed 2057184 - RejectedFreezeException (Beta) - as we understand it, this is only an incorrect log message, the group move actually worked fine. on that basis it's not serious enough to be worth a freeze exception and can be addressed with a 0-day update. 18:24:50 ack 18:24:53 ack 18:25:16 ack 18:25:19 +1 beta FE 18:25:21 ack 18:25:31 it'll go in after beta freeze is lifted anyway 18:25:59 it does mean cgroupify does not put firefox tabs into their own cgroup 18:26:58 but yeah, it's not terrible that it doesn't work correctly on the media - probably a low chance of oomd triggering there anyway i guess 18:27:27 and the update isn't cut yet upstream anyway, and i found another bug that's still being looked into 18:27:56 cmurf: see above. my understanding is that it doesn't stop anything happening. 18:28:08 it just causes an error message to be logged saying it didn't work, but it actually did work. 18:28:08 it does though 18:28:14 it does not actually work 18:28:17 that's not what the fix looks like 18:28:47 shrug all i know is the tabbed processes are not put in their own cgroup 18:29:06 adamw: ack 18:29:52 well, hmm 18:30:00 i guess as well as logging the failure we break out of a loop 18:30:24 so if other things were gonna happen later in that loop, they don't. so, hmm. it could have practical effects 18:30:39 so, um, i think i'm changing my vote to "iunno" 18:30:53 it's suboptimal to ship it in the beta this way but not "bad" 18:30:58 it'll get fixed when freeze is lifted 18:31:27 i wouldn't worry too much about it, it was really a "nice to have" per the pre-freeze exception language 18:31:53 well, "freeze exception" is what we changed the name to because we didn't like "nice to have". :P 18:32:04 i know 18:32:04 ok, new proposal 18:32:20 punt on it and ask ben how much he thinks it ought to be in beta 18:32:26 lol 18:32:31 i'll circle back with benzea 18:32:46 punt to next week is plenty soon enough 18:33:53 proposed #agreed 2057184 - punt (delay decision) - we're not entirely sure how serious of a problem this is, and hence whether it's worth a freeze exception. We'll ask the maintainer for his opinion in the bug report and discuss this again next week or in the ticket 18:34:05 ack 18:34:31 ack 18:34:41 ack 18:35:01 ack 18:35:03 #agreed 2057184 - punt (delay decision) - we're not entirely sure how serious of a problem this is, and hence whether it's worth a freeze exception. We'll ask the maintainer for his opinion in the bug report and discuss this again next week or in the ticket 18:35:23 okay, let's dive into... 18:35:27 ok just talked to ben berg, it's not a big deal to reject this 18:35:28 #topic Proposed Final blockers 18:35:33 cmurf: TOO LATE 18:35:41 i wish to get off this goddamn bug before nuclear war destroys us all 18:35:48 haha 18:35:48 #topic (2055908) bluetooth, cannot remove a bluetooth device 18:35:54 yikes 18:35:55 #link https://bugzilla.redhat.com/show_bug.cgi?id=2055908 18:36:00 #link https://pagure.io/fedora-qa/blocker-review/issue/617 18:36:04 #info Proposed Blocker, gnome-control-center, NEW 18:36:08 #info Ticket vote: FinalBlocker (+4,0,-0) (+adamwill, +lruzicka, +catanzaro, +nielsenb) 18:36:21 oh, look, i missed a ticket vote. bad me. 18:36:27 so, we have +4 here, anyone wanna vote -1? 18:37:06 +1 finalbocker 18:37:30 +1 FinalBlocker 18:37:59 +1 fb 18:38:32 it'd be good to have more people test, of course, there could be some hw-specific element here 18:38:38 my bluetooth is completely screwed with all 5.17 kernels so far 18:38:59 oh, from the upstream report looks like it's a known issue 18:41:03 and 18:41:15 +1 FB 18:41:35 looks like the 42~beta update should actually fix this 18:41:40 anyhoo 18:42:01 which already has an approved FE 18:42:04 proposed #agreed 2055908 - AcceptedBlocker (Final) - accepted as a violation of "All elements of the default panel (or equivalent) configuration in all release-blocking desktops must function correctly in typical use" 18:42:11 ack 18:42:15 ack 18:42:31 ack 18:42:38 ack 18:42:45 cmurf: can you test if it works with the packages from the 42~beta update? specifically gnome-bluetooth-42~beta.2-1.fc36 18:43:27 ack 18:43:33 this is fixed 18:44:12 #agreed 2055908 - AcceptedBlocker (Final) - accepted as a violation of "All elements of the default panel (or equivalent) configuration in all release-blocking desktops must function correctly in typical use" 18:44:35 #topic (2058920) [abrt] gnome-control-center: cc_object_storage_create_dbus_proxy_finish(): gnome-control-center killed by SIGABRT 18:44:41 #link https://bugzilla.redhat.com/show_bug.cgi?id=2058920 18:44:42 #link https://pagure.io/fedora-qa/blocker-review/issue/632 18:44:46 #info Proposed Blocker, gnome-control-center, NEW 18:47:28 this will definitely need an upstream report i think 18:47:30 so yeah, i'm getting some abrt notifications of a crash in gnome-control-center bluetooth and i don't know why or how to reproduce it 18:47:47 ok i'll do that now and we can punt until next week 18:48:33 i don't see this one on my system 18:48:41 be good if anyone else can check 18:49:25 Bluetooth seems to be recurrent source of bugs. Just a thought. 🤷‍♂️ 18:50:34 I might need to go, children need to go to bed. 18:50:42 thanks lruzicka 18:50:56 geraldosimiao: as i said i have kernel issues with bluetooth in 5.17, which may play into that, hard to say 18:51:24 proposed #agreed 2058920 - punt (delay decision) - we don't have enough info on the causes, consequences or prevalence of this crash to make a decision yet. Chris will report it upstream and see if that gets us anywhere 18:52:19 ack 18:53:00 ack 18:53:01 ack 18:53:13 ack 18:53:20 upstream bug report in the rhbz now 18:53:25 ack 18:53:46 I'm not even sure this is bluetooth related because even though abrt claims it is, the backtrace has nothing about bluetooth in it 18:54:01 could just be some transient crash of g-c-c 18:55:26 #agreed 2058920 - punt (delay decision) - we don't have enough info on the causes, consequences or prevalence of this crash to make a decision yet. Chris will report it upstream and see if that gets us anywhere 18:55:31 #topic (2056927) Require authselect for use in scriptlets 18:55:36 #link https://bugzilla.redhat.com/show_bug.cgi?id=2056927 18:55:41 #link https://pagure.io/fedora-qa/blocker-review/issue/622 18:55:44 #info Proposed Blocker, nss-mdns, NEW 18:55:47 #info Ticket vote: FinalBlocker (+1,0,-0) (+nielsenb) 18:57:18 +1 finalblocker 18:57:25 * bcotton has to disappear. will return in ~30 minutes if we're still going 18:57:36 +1 FinalBlocker 18:57:51 there's also a related issue where authselect itself fails to install properly because of a missing secriptlet dependency 18:57:57 https://src.fedoraproject.org/rpms/authselect/pull-request/17 18:59:01 ehhh, i dunno if we really meant to cover ipp everywhere here, i think we were more about locally connected printers 18:59:14 +1 FinalBlocker 18:59:16 but it doesn't seem worth fighting too hard about 19:01:03 * Eighth_Doctor shrugs 19:02:31 proposed #agreed 2056927 - AcceptedBlocker (Final) - this is accepted as a violation of "Printing must work in release-blocking desktops on at least one printer using...The generic IPP driver" 19:05:07 echo? echo? is this thing on? 19:05:15 ack 19:06:05 ack 19:08:06 ack 19:08:16 #agreed 2056927 - AcceptedBlocker (Final) - this is accepted as a violation of "Printing must work in release-blocking desktops on at least one printer using...The generic IPP driver" 19:09:24 #topic (2057563) The About button sends Discover into loop and the application stops responding. 19:09:30 #link https://bugzilla.redhat.com/show_bug.cgi?id=2057563 19:09:33 #link https://pagure.io/fedora-qa/blocker-review/issue/627 19:09:38 #info Proposed Blocker, plasma-discover, NEW 19:09:41 #info Ticket vote: FinalBlocker (+2,0,-0) (+lruzicka, +nielsenb) 19:09:49 well, this is clearly silly, but...would it pass the last blocker test? 19:10:11 be honest, who really clicks on About buttons? :D 19:10:12 apparently someone did :o 19:10:28 😆 19:10:58 well yeah, but it was lruzicka 19:11:03 what normal person clicks on an About button 19:11:26 I don't know if I am normal, but occasionally yes 19:11:38 +1 FinalBlocker 19:13:16 adamw to me that the same as all desktop apps should work 19:13:29 well, the criterion is "Basic functionality" 19:13:36 is an about button basic functionality? that's the vote, i guess. 19:13:45 yep 19:14:01 I would think so 19:14:03 +1 FinalBlocker 19:14:17 +1 FB 19:14:41 +1 fb 19:16:00 alrighty 19:17:16 proposed #agreed 2057563 - AcceptedBlocker (Final) - this is accepted as a violation of the default application functionality criterion, which requires that the KDE package manager (that's Discover) "must start successfully and withstand a basic functionality test" 19:18:35 ack 19:18:42 ack 19:19:10 ack 19:19:51 ack 19:19:53 so this review has been going on for 2+ hours 19:20:32 #agreed 2057563 - AcceptedBlocker (Final) - this is accepted as a violation of the default application functionality criterion, which requires that the KDE package manager (that's Discover) "must start successfully and withstand a basic functionality test" 19:20:41 yeah, if more people had done ticket votes it'd be faster 19:20:43 but also there were just a lot of proposals 19:20:48 the cap is 3 hours, but we should be done by then 19:20:57 three more to go 19:21:02 #topic (2053790) Fedora look and feel is not fully applied on new installs 19:21:05 #link https://bugzilla.redhat.com/show_bug.cgi?id=2053790 19:21:09 #link https://pagure.io/fedora-qa/blocker-review/issue/606 19:21:14 #info Proposed Blocker, plasma-workspace, MODIFIED 19:21:17 #info Ticket vote: FinalBlocker (+0,0,-1) (-nielsenb) 19:21:22 I finally fixed this on Sunday 19:21:42 so as this stands i'm kinda -1 on it, the justification was that things could actually be so badly themed as to be hard to see, but i don't see any screenshots supporting that 19:21:56 just screenshots of things not being as expected, but they were perfectly 'okay' for usage purposes 19:22:20 -1 for now 19:22:22 well, there's other bugs around appearance stuff, which I need to deal with upstream 19:22:35 like the fact that none of the theme setting buttons work at all 19:23:06 but it's important to me and the KDE SIG that our default look and feel is used consistently, esp. with OpenQA testing and other things 19:23:15 well the +FB from above should take care of this as well 19:23:49 In retrospect, I should have filed a BetaFE too 19:24:00 but I didn't because I didn't know what the problem was yet 19:24:10 oh, yeah, i'm +1 Beta FE on this if it isn't already 19:24:12 but... I fixed it and I'd like to have this in for Beta 19:24:24 +1 BFE 19:24:40 +1 BetaFE (though not sure I can do that...) 19:25:14 at this point you can have your dog vote if it'll get the meeting done faster 19:25:32 of course the one time i could do with my cat jumping on the keyboard, he isn't here 19:25:44 +1 Beta FE 19:25:54 +1 bfe 19:26:19 -1 final blocker 19:26:38 proposed #agreed 2053790 - RejectedBlocker (Final) AcceptedFreezeException (Beta) - this doesn't really seem to violate any criteria as it stands (the elements are perfectly usable, they just don't look as intended), but we do grant it a Beta freeze exception as obviously we want things to look as intended for Beta if possible 19:26:52 ack 19:26:58 ack 19:27:01 ack 19:27:08 ack 19:27:12 #agreed 2053790 - RejectedBlocker (Final) AcceptedFreezeException (Beta) - this doesn't really seem to violate any criteria as it stands (the elements are perfectly usable, they just don't look as intended), but we do grant it a Beta freeze exception as obviously we want things to look as intended for Beta if possible 19:27:23 #topic (2010595) Cannot install firmware if secureboot is enabled 19:27:27 #link https://bugzilla.redhat.com/show_bug.cgi?id=2010595 19:27:32 #link https://pagure.io/fedora-qa/blocker-review/issue/635 19:27:36 #info Proposed Blocker, shim, NEW 19:27:41 that's a bit of a yikes 19:28:05 again 19:28:34 ugh 19:28:36 +1 Beta FE, +1 FinalBlocker 19:30:04 +1 BFE +1 FB 19:30:06 ehhh. i'm having trouble seeing this as a blocker, honestly 19:30:15 or an FE. it doesn't need to be on media. 19:30:23 it's an important bug, but that's not the same thing as a release blocker or FE. 19:30:27 not being able to update firmware in the default setup is pretty bad 19:30:50 i mean, kinda? but otoh, you can't update system firmware from the OS at all on a lot of other systems, and we still let you install fedora on those, so? 19:31:00 there's also the marketing/political angle here, where Ubuntu is able to do it and we can't 19:31:05 and it been going on for a couple of releases 19:31:06 again, not a release blocker. 19:31:15 which is, frankly, pretty maddening 19:31:17 * bcotton returns 19:31:25 which means we shipped two releases with it and the world didn't end, so, another argument that it isn't release blocking :D 19:31:28 perhaps we should make firmware management part of the release criteria then 19:31:41 because it's a pretty critical aspect of Fedora's security story 19:31:48 it's an option, sure 19:32:04 the problematic part is testing that, since people don't always have updates pending 19:32:12 and dummy firmware updates aren't really a thing :/ 19:32:41 i mean, it sounds like the problem is known and fixed upstream already 19:32:52 and the issue here is just "someone needs to corner pjones and make him send out an update" 19:33:15 or javier, i guess. 19:33:42 +1 Beta FE 19:33:42 +1 FinalBlocker 19:33:58 so i don't love this, but I think I'm in the -1 camp 19:34:32 i'm gonna ask someone who's +1 to come up with a justification for that based on the criteria 19:34:41 because we can't really accept things without one 19:35:13 "so this review has been going on..." <- 🥲 19:35:27 this is the second-to-last bug on the list 19:35:50 IMO, it inhibits the updating software criterion: https://fedoraproject.org/wiki/Fedora_36_Final_Release_Criteria#Installing.2C_removing_and_updating_software 19:36:00 since GNOME Software and Plasma Discover fall under that 19:36:08 and their ability to apply updates is inhibited by this 19:36:26 all updates are inhibited or just firmware updates? 19:36:28 that very clearly specifies "software", though. not "firmware" 19:36:39 if this bug prevents a software update working if a firmware update is available, that might be an argument 19:36:42 but otherwise, i don't see it 19:36:47 well, Software stops you from doing both 19:37:00 if you try to trigger an update, it'll do it, fail, and loop back 19:37:19 are you sure about that? i don't see it stated in the bug report. 19:37:36 the firmware update process and software update process are separate. i would expect the software update portion would go ahead even if the firmware update fails. 19:37:47 that's not how GNOME Software works 19:37:50 (i'm pretty sure i've actually seen that happen on my system before) 19:37:58 it picks one, does it, and then will let you do the other 19:38:02 if they are so compartmentalized, that would be the blocking bug, not the inability to update the firmware 19:38:08 unless you manually pick an update process 19:38:13 Conan Kudo: no, it really is. for a firmware update it sets a flag to tell the firmware to handle an update. gnome software per se does not do it. 19:38:23 ^not so compartmentalized 19:38:48 if the firmware doesn't manage to update itself, i don't think that would prevent fedora itself booting into the special upgrade mode and running the software update. 19:39:11 but if so, that's the bug we could block on 19:39:28 if so, yes. but i'd rather see that confirmed 19:39:36 afaik nobody has claimed it's the case in the bug 19:39:41 agreed 19:39:46 right so needinfo and punt 19:39:58 i love ignoring things! 19:40:04 or -1 final blocker 19:40:11 maybe by next week we'll all just be traces of radiation and this will seem less important 19:40:12 this whole thing is kind of bullshit... having packages that only one person can update is garbage 19:40:17 if that's that controversial: +1 punt 19:40:24 Conan Kudo: well, javier did the last couple of updates 19:40:35 because Peter is basically never around 19:40:38 he wasn't CCed on the bug, though. I CCed him now, which might help. 19:40:44 so I reiterate: having only one person is bad 19:40:49 we have that for both kernels and bootloaders 19:40:58 adamw: 🤯 19:40:59 and nobody else can do anything about it 19:41:08 could also cc rharwood 19:41:19 there are meant to be two kernel maintainers, but they keep leaving :P 19:41:27 for boot chain, we do have javier now. 19:41:47 he's supposed to be moving off the boot stuff :( 19:41:56 and other people do have the privileges, they're just reluctant to touch the packages unless it's absolutely necessary as they're not experts and it's complicated. 19:42:01 whee 19:42:30 okay, anyway, so, we punting? 19:43:06 fine 19:43:07 let's punt 19:43:10 * Eighth_Doctor grumbles 19:43:23 +1 punt 19:43:50 proposed #agreed 2010595 - punt (delay decision) - there was some support in principle for this being a blocker, but also some opposition, and it doesn't obviously violate the existing criteria. we will ask how software updates are affected if a firmware update is queued but fails (if the software update also does not happen in this case, that could be a blocker issue). someone may also choose to propose specific criteria for firmware 19:43:50 updates. 19:44:20 adamw, can you shorten it by 8 characters for IRC? 19:44:20 ack 19:44:23 ack 19:44:25 ack 19:44:26 ack 19:44:30 if it's 8 characters it's fine 19:44:34 "proposed " is 9 :P 19:44:44 wonderful! 19:44:45 ack 19:44:57 acl 19:45:01 ack, too 19:45:08 #agreed 2010595 - punt (delay decision) - there was some support in principle for this being a blocker, but also some opposition, and it doesn't obviously violate the existing criteria. we will ask how software updates are affected if a firmware update is queued but fails (if the software update also does not happen in this case, that could be a blocker issue). someone may also choose to propose specific criteria for firmware updates 19:45:15 #topic (2054594) crash happens when try to set up a google or microsoft online account 19:45:18 #link https://bugzilla.redhat.com/show_bug.cgi?id=2054594 19:45:24 #link https://pagure.io/fedora-qa/blocker-review/issue/609 19:45:26 #info Proposed Blocker, xdg-desktop-portal, NEW 19:45:30 #info Ticket vote: FinalBlocker (+0,0,-2) (-nielsenb, -catanzaro) 19:45:34 okay, last bug, folks 19:46:36 yeah, i think i'm -1 on this for now unless we pin it down to something more concrete that will be commonly encountered 19:47:06 -1 ditto 19:47:27 agreed, -1 Final Blocker 19:47:32 i haven't experienced this 19:47:52 if it were reproducible i'd be +1 final blocker, either fix it or remove the option 19:48:03 and i recently had to use it for some other bug 19:48:50 proposed #agreed 2054594 - RejectedBlocker (Final) - we agreed this would be a blocker issue if it were clearly reproducible, but so far other testers have not been able to reproduce it. We can reconsider this decision if we are able to pin the issue down more precisely and it still seems concerning 19:48:56 ack 19:49:05 ack 19:49:05 ack 19:49:17 ack 19:49:41 ack 19:50:00 #agreed 2054594 - RejectedBlocker (Final) - we agreed this would be a blocker issue if it were clearly reproducible, but so far other testers have not been able to reproduce it. We can reconsider this decision if we are able to pin the issue down more precisely and it still seems concerning 19:50:02 okey dokey 19:50:15 uhhhh, i said earlier i was gonna add something to the agenda later didn't i 19:50:20 and now i've forgotten what that was 19:50:43 also, does anyone know why I came in here and where I left my teeth? 19:51:09 oh yes I remember 19:51:28 Dog ate them ? 19:51:38 lol 19:51:59 proposal: retitle the accepted FE https://bugzilla.redhat.com/show_bug.cgi?id=2049147 from "ship GNOME 42.rc in Fedora 36 beta" to "ship GNOME 42.beta or 42.rc in Fedora 36 beta" 19:52:18 i figured it'd be correct to have a vote on this somewhere instead of just ninja'ing it 19:52:47 i am +1 to the proposal, we should include the latest gnome bits we safely can, even if that's beta not rc. 19:52:58 otherwise we're stuck with alpha, which nobody wants. 19:52:59 +1 19:53:03 +1 19:53:27 ok insofar as it gets beta into stable asap +1 19:53:29 +1 19:53:42 #agreed we will retitle the accepted FE https://bugzilla.redhat.com/show_bug.cgi?id=2049147 from "ship GNOME 42.rc in Fedora 36 beta" to "ship GNOME 42.beta or 42.rc in Fedora 36 beta" to make it clearer that we want to include 42.beta in F36 Beta if rc is not ready in time, beta still better than alpha 19:53:48 ack 19:53:53 ack 19:53:59 ack 19:53:59 ack 19:54:02 #topic Open floor 19:54:13 ok, sorry for the long meeting and thanks for sticking it out, folks 19:54:13 and now back to the continuing coverage of the end of the world as we know it... 19:54:22 please do vote in the tickets in future, it makes these meetings shorter! 19:54:23 * Eighth_Doctor sighs 19:54:36 cmurf: on the plus side, great excuse to bust out some classic 80s REM 19:54:44 haha 19:54:48 that is so corny 19:55:12 i feel fine? 19:55:23 now you can add the i feel fine dog meme to the mix 19:55:29 haha 19:55:51 goodbye people, see you all next time 👋 19:55:56 any other business before i go finally get a shower? 19:56:02 geraldosimiao: fingers crossed! 19:56:29 😉 19:56:35 * bcotton hands adamw a bar of soap 19:57:03 pffft, i use only the finest body shop body wash i'll have you know 19:57:10 oh just got rub yourself with some sand and swim in the ocean 19:57:15 (because bar soap tends to clog up our drains) 19:57:39 body shop soap hahaha, extra pumice 19:57:48 cmurf: it's 7 degrees celsius right now. plan rejected 19:58:09 so warm ... 19:58:20 still not great for ocean swimming 19:58:37 so it'd be brief... 19:59:43 there's still ice on my sidewalks 20:00:20 (2 degrees centigrade) 20:01:20 alrighty, thanks again folks 20:01:29 #endmeeting