16:01:05 #startmeeting F25-blocker-review 16:01:05 Meeting started Mon Oct 31 16:01:05 2016 UTC. The chair is adamw. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:01:05 Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:01:05 The meeting name has been set to 'f25-blocker-review' 16:01:05 #meetingname F25-blocker-review 16:01:05 #topic Roll Call 16:01:05 The meeting name has been set to 'f25-blocker-review' 16:01:11 ahoy mateys 16:01:14 who's around for some blocker review? 16:01:25 Kampuchea 16:01:35 .hello jkurik 16:01:36 jkurik: jkurik 'Jan Kurik' 16:01:44 * satellit listening 16:01:46 cmurf: i don't think that exists any more. 16:02:08 * coremodule is here. Good morning! 16:02:12 * kparal is here 16:02:28 * kparal punches pschindl garretraziel 16:02:40 * pschindl is here 16:02:45 * garretraziel will be lurking around 16:02:46 hey now, violence is reserved for the meeting leader. 16:03:06 adamw: pretty sure it both does and does not, depending on the point of view 16:03:08 * kparal punches them from the other side to negate the first blow 16:03:16 that seems reasonable. 16:03:26 * adamw punches them from both sides just to make sure 16:03:44 wow you guys are violent before blocker review 16:04:16 this is cyber-bullying! 16:04:21 Please direct your violence towards the blockers instead. ;-) 16:04:34 U.S.S. Just Ship It 16:04:49 hi, everyone 16:04:56 #chair cmurf garretraziel 16:04:56 Current chairs: adamw cmurf garretraziel 16:05:01 * adamw throws chairs around, too 16:05:06 you have a rowdy bunch today 16:05:08 hi 16:05:12 hi fale 16:05:12 #topic Introduction 16:05:12 Why are we here? 16:05:12 #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. 16:05:13 (pschindl would also like to sit) 16:05:13 #info We'll be following the process outlined at: 16:05:13 #link https://fedoraproject.org/wiki/QA:SOP_Blocker_Bug_Meeting 16:05:16 #info The bugs up for review today are available at: 16:05:17 #link http://qa.fedoraproject.org/blockerbugs/current 16:05:19 #info The criteria for release blocking bugs can be found at: 16:05:23 #link https://fedoraproject.org/wiki/Fedora_25_Alpha_Release_Criteria 16:05:25 #link https://fedoraproject.org/wiki/Fedora_25_Beta_Release_Criteria 16:05:27 #link https://fedoraproject.org/wiki/Fedora_25_Final_Release_Criteria 16:07:53 who wants to be the secretary? 16:08:01 I'll do it. 16:08:10 thanks coremodule! 16:08:16 #info coremodule will be the secretary 16:08:25 Not a problemo! 16:10:43 we have: 16:10:44 #info 4 Proposed Blockers 16:10:44 #info 5 Accepted Blockers 16:10:48 #info 4 Proposed Freeze Exceptions 16:10:48 #info 7 Accepted Freeze Exceptions 16:11:05 so, starting with the proposed blockers... 16:11:08 #topic (1389762) systemd presets request - switcheroo-control.service 16:11:08 #link https://bugzilla.redhat.com/show_bug.cgi?id=1389762 16:11:08 #info Proposed Blocker, fedora-release, POST 16:11:53 I didn't really understand what this is about. a missing systemd service? 16:12:01 installed by default? 16:12:05 Well it's an interesting question if you block release on an announced new feature not being present. 16:12:28 how is this related to fedora-release package? 16:13:12 kparal: The fedora-release package includes the preset files 16:13:17 In any case I'm -1 blocker, as there is no criterion for it. 16:13:26 It's not a missing service, it's the configuration to allow it to start by default 16:13:31 +1 FE though 16:13:43 I'm -1 blocker and +1 FE too 16:13:46 sgallagh: why is it that all those lines start with "enable" or "disable", just the last one (switcheroo) doesn't? https://pagure.io/fork/sgallagh/fedora-release/blob/f25/f/90-default.preset 16:13:50 looks like -1 blocker, +1 FE to me 16:14:10 kparal: Because I screwed up. Fixing. 16:14:11 /me sighs 16:14:22 yeah, I am -1 blocker and +1 FE too 16:14:28 yeah, by the criteria this is -1 / +1 16:14:51 i'd actually be sad to ship without changing this, but since we've got lots of time to get it in, let's not worry too much 16:14:52 -1/+1. not having a new feature working is not a blocker 16:15:29 and I guess I'll note, just in case anyone forgot, that FESCo has the power to effectively declare this kind of bug (Change incompletion) a blocker 16:15:38 so if FESCo cares a lot, they can do that. 16:16:49 proposed #agreed 1389762 - RejectedBlocker AcceptedFreezeException - per the criteria and blocker process there is nothing that makes this a blocker, incomplete Changes are not inherently release blocking. However, it's obviously worth a freeze exception. Note FESCo has the power to declare this a blocking issue if it so desires. 16:17:03 ack 16:17:06 ack 16:17:07 ack 16:17:08 kparal: Thanks for catching that. 16:17:16 +1 kparal 16:17:18 ack 16:17:26 ack 16:18:04 #agreed 1389762 - RejectedBlocker AcceptedFreezeException - per the criteria and blocker process there is nothing that makes this a blocker, incomplete Changes are not inherently release blocking. However, it's obviously worth a freeze exception. Note FESCo has the power to declare this a blocking issue if it so desires. 16:18:20 #topic (1384508) [abrt] gnome-session: _gtk_style_provider_private_get_settings(): gnome-session-failed killed by SIGSEGV 16:18:20 #link https://bugzilla.redhat.com/show_bug.cgi?id=1384508 16:18:21 #info Proposed Blocker, gnome-session, NEW 16:19:09 this affects X only, it seems 16:19:19 also the OP says he doesn't know how to reproduce it 16:19:22 It is not clear when this happens 16:19:30 hmm 16:19:31 he doesn't state even frequency 16:19:41 the traceback seems to match https://bugzilla.redhat.com/show_bug.cgi?id=1271837 16:19:48 which is F23 16:20:27 dunno, kinda hard to block if there isn't a reproducer? 16:20:37 kparal: i'm not sure we can conclude this is X only. 16:21:18 kparal: only that when the bug happens when running under X, an X crash is recorded. 16:21:21 ok, you're right, I got mistaken 16:21:54 still, if this were widely encountered I'd *probably* expect more dupes / abrt dupe comments. 16:22:44 FAF report has 110 crashes 16:23:05 oof 16:23:05 but it doesn't seem to say how many users that includes 16:23:48 if it's coming from one person that's a lot of dedication! 16:24:14 cmurf: FAF minireports are sent automatically after each crash 16:24:38 I believe 16:25:15 hmm 16:25:16 however, the number of crashes went down over time 16:26:51 for the moment, I'd say this affects a small portion of the user base and the reproducer or even frequency is unknown, so it's not a blocker material 16:27:07 I haven't seen this personally 16:27:44 but if there's a reproducer or more people hitting this, we can of course re-evaluate 16:27:50 yeah, I pretty much agree with kparal 16:27:52 tentative -1 16:27:54 kparal: I see it in the same way. -1 to block on this 16:28:05 my 'hmm' was because openqa does hit a crash in GNOME sometimes, but i just checked and it's not this one 16:28:29 this might be affects by gnome-contacts contents 16:28:35 since it crashes in its search provider 16:28:56 even though it crashes in css, shrug 16:29:07 hmm, good catch 16:29:12 * adamw bets it's unicode 16:29:15 it's always f**king unicode 16:29:21 :) 16:29:27 i <3 unicode 16:29:36 .fas linuxmodder 16:29:36 linuxmodder: linuxmodder 'Corey W Sheldon' 16:29:44 i wonder if anyone's proposed a 'unicode sucks' unicode code point 16:30:08 anyone +1 at this point? 16:30:13 I ❤ unicde 16:30:20 I'm -1 also 16:30:28 no but it'd be nice to offer some advice how to gather more information 16:30:38 to unicode sucking or the code point ? adamw 16:30:56 or if this should be taken upstream for fixing? some way to kick it in the butt 16:31:16 -1 to suck on unicode, -1 to block on 1384508 16:31:23 I seem to be able to search in my contacts containing some unicode chars 16:31:25 -1 block 16:31:43 cmurf: an upstream report would probably be a good idea, yeah 16:32:57 unicode is awesome the problem is how various things implement it that sucks 16:32:59 proposed #agreed 1384508 - RejectedBlocker - As things stand, there is not sufficient indication that this affects more than a handful of users, so it's hard to consider it a blocker (not a wide enough impact). If more details emerge that indicate this could affect many users, we may re-consider the decision. 16:33:10 linuxmodder: haha, vote on both 16:33:23 ack 16:33:27 ack 16:33:39 ack 16:33:45 ack 16:34:00 adamw, im rather indefiffernt on unicode 16:34:20 it has use cases not that I like or use them all 16:34:39 #agreed 1384508 - RejectedBlocker - As things stand, there is not sufficient indication that this affects more than a handful of users, so it's hard to consider it a blocker (not a wide enough impact). If more details emerge that indicate this could affect many users, we may re-consider the decision. 16:37:03 ǝɯɐlq oʇ sᴉ ǝpoɔᴉun ʇɐɥʇ sɯǝǝs ʇᴉ 'oslɐ 16:37:04 #topic (1389885) gnome-shell freeze when holding F11 key in gnome-terminal 16:37:04 #link https://bugzilla.redhat.com/show_bug.cgi?id=1389885 16:37:04 #info Proposed Blocker, gnome-shell, NEW 16:37:18 * adamw turns the internet upside down 16:38:19 why someone would like to hold F11 key ? 16:38:20 did you ever read about that guy who found out his neighbour was stealing his wifi, so he put an http proxy inline which turned all the images upside down? 16:38:27 jkurik: i was wondering that too, what's it supposed to do? 16:38:39 this one is X only :) 16:39:01 jkurik: I think it's just a "mistake" - you want to make it fullscreen and hold it longer than needed 16:39:53 anyway, not a basic functionality 16:40:07 might be a FE, but not a blocker IMO 16:40:08 and not broken on wayland, which is the default 16:40:09 yes, and x only -1 16:40:32 -1 blocker, not sure about FE...it'd kinda depend on how invasive the fix is, for me 16:40:49 -1 blocker 16:41:08 Agreed. Not a blocker, but worth it to fix if the ramifications of the fix aren't bad. 16:41:10 -1 blocker. 16:41:39 On the fence for FE until there's something from a dev about it all, and side effects. 16:42:24 -1, I wouldn't vote on FE at this point 16:42:39 it's kindof a bad bug to hit but I dn't see it as a blocker 16:42:52 proposed #agreed 1389885 - RejectedBlocker - as this isn't really in basic functionality and seems to affect X only, it doesn't meet the blocker criteria. We may be willing to grant an FE for a fix if it's simple and non-invasive, but would like to see more details before voting on that 16:42:57 ack 16:43:10 *affect 16:43:21 My bad, I can't read today. 16:43:38 ack 16:43:40 ack 16:43:42 ack 16:43:57 ack 16:44:41 #agreed 1389885 - RejectedBlocker - as this isn't really in basic functionality and seems to affect X only, it doesn't meet the blocker criteria. We may be willing to grant an FE for a fix if it's simple and non-invasive, but would like to see more details before voting on that 16:44:46 grr 16:45:03 hexchat doesn't seem to have typing history, or it doesn't when i hit 'up', at least, so i'm stuck copy pasting... 16:45:12 #agreed 1389885 - RejectedBlocker - as this isn't really in basic functionality and seems to affect X only, it doesn't meet the blocker criteria. We may be willing to grant an FE for a fix if it's simple and non-invasive, but would like to see more details before voting on that 16:45:24 #topic (1383471) [abrt] WARNING: CPU: 3 PID: 12176 at ./include/linux/swap.h:276 page_cache_tree_insert+0x1cc/0x1e0 16:45:24 #link https://bugzilla.redhat.com/show_bug.cgi?id=1383471 16:45:24 #info Proposed Blocker, kernel, NEW 16:45:50 we have any kernel devs available for comment? 16:46:01 adamw: up arrow works for me 16:46:04 hrm 16:46:10 maybe i'm missing a setting 16:46:14 ooh, X or wayland? 16:46:26 adamw: wayland 16:46:29 actually this is supposed to be fixed in 4.8.4 16:46:31 hm. 16:46:47 so, in this issue, it seems they actually identified kernel commit that should fix this 16:46:48 and I guess no one wanted to try it 16:47:00 the question is how often this happens 16:47:07 I can't tell if it's only hit on live media, or once installed 16:47:17 or if that doesn't matter 16:47:25 and yeah how often it happens 16:47:34 anyway 4.8.4 is stable for F25 16:47:59 could just punt on it for now and see if it's actually fixed, ask people to use a nightly and test? 16:48:32 since 2016-10-28. 16:49:19 cmurf: sounds good 16:49:27 yeah...seems reasonable 16:50:59 yeah, lets wait for the next nightly with the 4.8.4 16:51:04 does 20161029 have it? 16:51:13 proposed #agreed 1383471 - punt (delay decision) - we believe this may already have been resolved, so we will ask for feedback from reporters and close the bug if appropriate 16:51:19 cmurf: should do, i haven't check for sure. 16:51:21 20161031 should 16:51:32 ack 16:51:34 ack 16:51:36 ack 16:51:37 ack 16:51:53 #agreed 1383471 - punt (delay decision) - we believe this may already have been resolved, so we will ask for feedback from reporters and close the bug if appropriate 16:51:54 ack 16:51:59 ok, that's all the proposed blockers 16:52:10 #info moving on to proposed freeze exceptions 16:52:16 Admiral ACKbar says, it's a trap. 16:52:20 #topic (1389959) Switching keyboard layout with key combo does not work in Wayland 16:52:20 #link https://bugzilla.redhat.com/show_bug.cgi?id=1389959 16:52:21 #info Proposed Freeze Exceptions, anaconda, NEW 16:52:24 yeah, but that guy always says that 16:52:34 it's even a famous children's story, "the ackbar who cried trap" 16:52:44 haha 16:52:49 question is when it is NOT a trap 16:53:11 flip it and is around though 16:53:47 so yeah, after i read hedayat's beta review I went and tested layout switching on lives and it's kinda busted 16:54:00 ick 16:54:00 there's this and 1331382 (we'll come to that in a minute) 16:54:11 adamw: can you use win+space to switch layouts? 16:54:19 the standard gnome shortcut 16:54:29 dunno. didn' 16:54:33 didn't try. let me see. 16:54:52 kparal: windows key + space bar? nothing happens for me 16:55:00 uh, can't tell in a vm, because the host machine eats it. 16:55:06 * adamw writes a USB stick 16:56:02 In any case I'm +1FE 16:56:31 I'd be nice to solve the planet's language issues today, in this channel, though. 16:57:34 it doesn't work 16:58:06 because the layouts are not added to gnome (live) 16:58:12 that makes sense, yeah. 16:58:20 if they were, it would work 16:58:54 +1 FE 16:59:16 this is going to be more of a problem with wayland, apparently, as there's no wayland-level input method configuration bits it seems. the compositor has to handle it. so we're gonna have to teach anaconda how to tell GNOME and KDE and any other wayland-y desktops we care about how to do it... 16:59:34 there's more about that in the other bug. 16:59:34 * kparal wonders if we require mouse for functionality or block for issues not workaroundable without a mouse 16:59:37 hmm... does it means it is a "design issue" or it is by a purpose 16:59:47 jkurik: somewhere in between. 17:00:18 this is how wayland's supposed to work, but anaconda doesn't know about it. essentially. 17:00:38 (what anaconda tries to do is set the X level keyboard layout switch key combo, which in Wayland, doesn't exist.) 17:01:11 * adamw brb, call of nature 17:05:06 so, any more votes? 17:05:26 +1FE 17:05:44 +1FE as well 17:06:55 proposed #agreed 1389959 - AcceptedFreezeException (Final) - this is an installer issue and cannot be fixed with an update, and is sufficiently serious to warrant a freeze exception 17:07:33 ack 17:07:40 ack 17:08:19 ack 17:08:48 ack 17:09:23 #agreed 1389959 - AcceptedFreezeException (Final) - this is an installer issue and cannot be fixed with an update, and is sufficiently serious to warrant a freeze exception 17:10:16 #info we already discussed the next bug on the list (#1389762) as a proposed blocker, and granted it a freeze exception. 17:10:21 #topic (1188774) [anaconda] white screen on external monitor attached to a Lenovo ThinkPad T400 when docked and lid closed 17:10:22 #link https://bugzilla.redhat.com/show_bug.cgi?id=1188774 17:10:22 #info Proposed Freeze Exceptions, metacity, NEW 17:11:59 this is not going to get fixed, let's not kid ourselves 17:12:08 let's vote when there's a patch actually ready 17:12:40 same as last week, i guess 17:12:42 yeah' 17:12:59 proposed #agreed 1188774 - delay decision (punt) - same conditions apply as last week 17:13:12 if it sticks around much longer i guess we can reject it with a 'propose when there's a fix' note 17:14:15 ack 17:14:16 it is almost 2 years old bug with no fix yet, I am +1 to reject it 17:14:30 (that as well) 17:14:35 okay, pivot 17:14:41 not reject, just unpropose 17:14:46 right 17:14:47 yeah, i agree 17:14:55 everyone OK with that? 17:14:58 yes 17:15:00 yes 17:15:26 Agreed. 17:17:09 #agreed 1188774 - unpropose (drop Blocks:, but do not apply RejectedFreezeException tag) - this bug has been around a long time and we are not willing to grant an FE without an actual code fix, so there's no point punting every week until that's the case 17:17:22 coremodule: can you un-propose it and add a note to re-propose if there is actually a fix? 17:17:58 adamw, Gotcha, I can do that. 17:18:03 thanks 17:18:11 #topic (1331382) Anaconda layout indicator does not switch the keyboard layout when clicked when running in Workstation live 17:18:11 #link https://bugzilla.redhat.com/show_bug.cgi?id=1331382 17:18:11 #info Proposed Freeze Exceptions, mutter, ASSIGNED 17:18:22 so this is the other layout switch bug 17:18:22 this affects X, rather than Wayland 17:18:30 i'm pretty sure it's the same issue we granted an FE to last cycle 17:18:53 it's worse because you can't switch layout *at all*, but of course X is no longer the default. 17:19:09 ruhroh 17:19:34 Is there a fix on the way? 17:19:43 well it is the default if you boot in basic graphics mode, right? 17:20:33 cmurf: just got POSTed, I think. 17:20:41 kparal: er, probably? i don't know. 17:20:56 mclasen_: exactly what circumstances are people gonna wind up with X when booting the 25 Workstation live? 17:20:57 it seems to be fixed now, can we wait for the next compose and re-test ? 17:21:17 halfline: ^ care to answer that ? 17:21:18 * kparal trying 17:22:11 basic graphics boots X 17:22:13 in a VM 17:22:25 well the new case where people will end up with X is if they have dual gpus that have connectors spread across both 17:22:59 since otherwise some of the connectors wouldn't work out of the box 17:23:07 until we fix mutter 17:23:13 so I think this is definitely +1 FE. not sure about blocker, just putting into common bugs would be enough I guess 17:23:36 for F24 we decided FE not blocker 17:23:51 so logically speaking it would be weird to make it a blocker for F25 when you're *less* likely to hit iot 17:23:59 ok, that answers that 17:24:22 ok, so +1 FE 17:24:30 halfline: any other cases aside from that case and 'basic graphics'? (i'm curious for other reasons) 17:25:00 well if there's a crash immediately at startup with the wayland session it will fall back to X 17:25:02 +1FE 17:25:10 even if the crash isn't directly related to wayland support 17:25:29 proposed #agreed 1331382 - AcceptedFreezeException (Final) - this is somewhat less likely to be encountered than on F24 as Wayland is now the default, but there are still enough cases where we fall back to X that it is worth fixing, and cannot be fixed with an update 17:25:31 +1 FE 17:25:37 halfline: rgr. thanks 17:25:40 ack 17:25:41 ack 17:25:48 ack 17:25:49 ack 17:26:04 halfline: thanks for the explanation 17:26:09 #agreed 1331382 - AcceptedFreezeException (Final) - this is somewhat less likely to be encountered than on F24 as Wayland is now the default, but there are still enough cases where we fall back to X that it is worth fixing, and cannot be fixed with an update 17:26:10 yup, thanks 17:26:43 looking at the accepted blockers... 17:27:04 Where's sddm at? 17:27:25 #info that's all proposals, now taking a look at selected accepted blockers 17:27:25 Direct fix or pull out user switching UI? 17:27:33 cmurf: i wanted to discuss them in order, thanks. 17:27:37 mclasen_: halfline: where do we stand on backporting the fix for https://bugzilla.redhat.com/show_bug.cgi?id=1376471 ? 17:27:50 that is, the patches posted in https://bugzilla.gnome.org/show_bug.cgi?id=768531 17:28:11 I'll ask florian 17:28:45 thanks 17:29:12 #info desktop team will look into status of #1331382 17:29:33 #info #1381996 fix should be in composes from today onwards, pschindl will verify it 17:30:08 #info fixes for #1378156 and subsequent iscsi bugs are working through the anaconda/blivet review processes, we should have new builds in a few days 17:30:27 ok, so the other two are the sddm bugs, #1382001 and #1370222 17:30:33 Kevin_Kofler: any info on where we stand there? 17:34:46 or anyone else? 17:35:17 how bout those cubbies, huh? 17:35:55 cubbies? 17:35:57 looks like #1370222 is waiting on a spin-kickstarts PR to work around the bug 17:36:05 cmurf: sports reference. pick up a newspaper, jeez. 17:36:20 A Brit, in Canada, who cares about U.S. baseball? 17:36:31 I suppose there are many stranger things... 17:37:07 cmurf, screw them cubs 17:37:08 :) 17:37:15 adamw: I can send the PR. but I still don't know if people think it's a good idea or not 17:37:24 who can tell? 17:37:39 I took the hammer and "fixed" the issue 17:37:41 cmurf: *ahem* MLB is a *north american* league. 17:37:56 yaaay hammer 17:38:06 #info #1370222 is waiting on a spin-kickstarts PR to work around the bug, kparal has the hammer 17:38:43 there seems to be no progress at all on #1382001 17:38:55 #info #1382001 is showing no progress at all 17:41:12 and...that's the lot 17:41:34 #topic Open floor 17:41:39 so, any other business, folks? 17:41:58 we have the switchable graphics Test Day coming up this week, 2016-11-03 (rescheduled from last week), please come help with that if you can 17:42:02 and we still have validation tests to run 17:43:23 * kparal completely forgot about that 17:43:46 there's also an anaconda update to test: https://bodhi.fedoraproject.org/updates/FEDORA-2016-98dad065d7 - i'll run openqa on it some time today 17:43:56 though there will be at least one more anaconda and blivet build before final, for the iscsi fixes 17:45:39 that's about all I've got 17:45:42 reminder: tomorrow 2016-Nov-01 is Final Freeze 17:45:58 oh, F25 non-live composes should be working again from the next compose on, says dgilmore 17:46:00 No idea about the SDDM bugs, unfortunately. 17:46:04 I don't know of any progress there. 17:47:18 thanks 17:47:24 welp, seems like we're about done 17:47:27 * adamw sets the fuse 17:48:33 thanks for coming, fokks 17:48:36 also folks 17:50:43 thanks adamw 17:50:56 #endmeeting