17:00:25 #startmeeting F29 Beta Go/No-Go meeting 17:00:25 Meeting started Thu Sep 20 17:00:25 2018 UTC. 17:00:25 This meeting is logged and archived in a public location. 17:00:25 The chair is bcotton. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:25 Useful Commands: #action #agreed #halp #info #idea #link #topic. 17:00:25 The meeting name has been set to 'f29_beta_go/no-go_meeting' 17:00:27 #meetingname F29-Beta-Go_No_Go-meeting 17:00:27 The meeting name has been set to 'f29-beta-go_no_go-meeting' 17:00:36 #topic Roll Call 17:00:39 .hello2 17:00:40 .hello psabata 17:00:44 .hello2 17:00:45 bcotton: bcotton 'Ben Cotton' 17:00:48 contyk: psabata 'Petr Šabata' 17:00:51 asamalik: asamalik 'Adam Samalik' 17:00:54 .hello2 17:01:06 frantisekz: frantisekz 'František Zatloukal' 17:01:09 * asamalik waves instead of yawning (looks at contyk) 17:01:10 .hello mohanboddu 17:01:18 .hello2 17:01:24 mboddu: mohanboddu 'Mohan Boddu' 17:01:27 coremodule: coremodule 'Geoffrey Marr' 17:02:00 * satellit listening 17:02:07 .hello2 17:02:09 sgallagh: sgallagh 'Stephen Gallagher' 17:02:19 .hello2 17:02:20 mattdm: mattdm 'Matthew Miller' 17:02:24 .hello2 17:02:25 x3mboy: x3mboy 'Eduard Lucena' 17:02:31 .hello adamwill 17:02:32 adamw: adamwill 'Adam Williamson' 17:03:32 okay, looks like we have a good crowd. let's do this 17:03:40 #topic Purpose of this meeting 17:03:48 #info Purpose of this meeting is to check whether or not F29 Beta is ready for shipment, according to the release criteria. 17:03:49 #info This is determined in a few ways: 17:03:56 #info 1. No remaining blocker bugs 17:03:57 #info 2. Release candidate compose is available 17:03:57 morning 17:03:59 #info 3. Test matrices for Beta are fully completed 17:04:08 #link https://qa.fedoraproject.org/blockerbugs/milestone/29/beta/buglist 17:04:10 #link https://fedoraproject.org/wiki/Category:Fedora_29_Test_Results 17:04:17 .hello lruzicka 17:04:20 lruzicka: lruzicka 'Lukáš Růžička' 17:04:26 #topic Current Status -- blocker bugs 17:04:31 #link https://qa.fedoraproject.org/blockerbugs/milestone/29/beta/buglist 17:04:36 for good measure 17:05:11 okay, time for another Blocker Review lightning round? 17:05:17 * contyk nods 17:05:34 #info 2 Proposed Blockers 17:05:35 #info 2 Accepted Blockers 17:05:44 #topic (1630134) gnome-shell stuck when I move and drop favorite icon 17:05:46 #link https://bugzilla.redhat.com/show_bug.cgi?id=1630134 17:05:47 #info Proposed Blocker, gnome-shell, VERIFIED 17:05:53 #chair adamw 17:05:53 Current chairs: adamw bcotton 17:06:20 welp, i seem to have just discovered a shell crasher... 17:06:22 this is actually fixed in RC5 17:06:24 but we should still discuss whether it's a blocker, as if it is, we cannot release rc4 17:06:32 The effect of this one is that if it’s ... yeah 17:06:54 #info BZ 1630134 is fixed in RC5 17:06:58 * nirik leans toward -1 blocker +1 FE 17:07:08 -1 blocker +1 FE 17:07:19 -1 blocker +1 FE 17:07:30 it's really pretty bad, but maybe just -1 17:07:33 it's a beta, and this is an adorble bug you'd almost expect to get in a beta 17:07:35 so -1/+1 also 17:07:36 I’m slightly +1 blocker, honestly. 17:07:52 meh i'll get fixed next week in an update :P 17:07:55 But I’m not going to argue loudly. 17:07:59 s/i'll/it'll 17:08:11 * sgallagh puts the knife back down 17:08:12 -1 blocker, +1 fe 17:08:17 * nirik has never ever done the action that causes this.. :) 17:08:18 haha 17:08:22 no opinion 17:08:26 that is the same bug that Petr has found 17:08:42 i never actually do this either 17:08:46 but some people obviously do 17:08:48 me3 17:08:50 My main thought is that this is the sort of action that mostly happens on a new install 17:08:59 Once people have stuff set up, it’s infrequent 17:09:11 common bugs it 17:09:26 don't need to if we release rc5 17:09:26 -1 Blocker, +1 FE 17:09:27 anyway 17:09:31 bcotton: are you running this or am I? 17:09:32 * mboddu never done that before 17:09:49 adamw i'll run it until i do it wrong and then you push me out of the way? 17:10:02 "push", sure 17:10:02 adamw, we have tried hard to test rc5, why should not we release it? 17:10:04 * adamw borrows sgallagh's knife 17:10:19 so i count +1/-5 on this as a blocker and +6/-0 on FE 17:10:20 lruzicka: i'm not saying that, but that part of the meeting is later :) 17:10:41 I’m driving for the next ten minutes. mattdm has my proxy 17:11:01 proposed #agreed BZ 1630134 is rejected as a Beta Blocker as no blocking criteria have been identified 17:11:18 ack 17:11:34 wait 17:11:37 waiting 17:11:38 patch a bit 17:12:01 haha what blocking criteria? 17:12:03 actually, there is a criteria about the panel ... is not that part of the panel? 17:12:15 proposed #agreed 1630134 - RejectedBlocker (Beta) AcceptedFreezeException (Beta) - this is rejected as a blocker as it does not violate any beta criteria, but it is an easily-encounterable GNOME crasher so accepted as an FE 17:12:24 ack 17:12:28 lruzicka: the beta criterion says something like 'cannot crash on startup or be entirely non-functional' 17:12:28 ack 17:12:33 ack 17:12:34 this doesn't meet that, imo 17:12:42 definitely not, it's a cute bug 17:12:49 the overview does not crash on startup and it's not 'entirely non-functional', it just has a big bug. :P 17:13:01 a cute big bug 17:13:04 cute big bug 17:13:06 this would definitely be a *final* blocker, but it's not going to be around long enough for us to care. 17:13:31 ok, so ... no more objections 17:13:45 sgallagh says he is fine with that 17:13:49 ack 17:13:50 =) 17:14:04 ack 17:14:41 #agreed 1630134 - RejectedBlocker (Beta) AcceptedFreezeException (Beta) - this is rejected as a blocker as it does not violate any beta criteria, but it is an easily-encounterable GNOME crasher so accepted as an FE 17:14:56 #topic (1628495) In UEFI mode, Fedora 29 cannot be installed in Safe Graphics Mode 17:14:57 #link https://bugzilla.redhat.com/show_bug.cgi?id=1628495 17:14:59 #info Proposed Blocker, xorg-x11-server, NEW 17:15:44 strictly speaking it's a blocker because it violates a basic criterion, but it also isn't clear how widespread it is 17:15:49 it might only be UEFI 17:16:01 and itmight only be on hardware where NOT using nomodeset works 17:16:12 thereby making nomodeset not necessary 17:16:14 i think it's pretty clearly only uefi 17:16:17 well, it still violates the criteria 17:16:24 I have found it on UEFI 17:16:26 the fun thing is, it actually doesn't 17:16:35 the criterion is quite carefully worded, if you look at it 17:16:35 and even more, ajax says this is not basic graphics, it's pure desperation graphics 17:17:11 it says "This mechanism should work correctly, launching the installer or desktop and attempting to use the generic driver." 17:17:17 and the mechanism does indeed work correctly. 17:17:26 i.e. it boots with 'nomodeset'. 17:17:37 X then does not actually work, but that's not what the criterion says. :P 17:18:02 uhhh 17:18:17 "...use a generic, highly compatible video driver (such as 'vesa'). This mechanism should work correctly, launching the installer or desktop and attempting to use the generic driver" 17:18:23 adamw: i admire your lawyering 17:18:40 it's not just lawyering, that's literally what it's meant to cover, from the last time we went over it 17:18:40 the mechanism doesn't work, the installer nor desktop are launched 17:19:00 so, we don't have any more idea how widespread this is? 17:19:01 but I'm still a -1 because it doesn't seem broad enough, more hardware specfic 17:19:09 that sounds like wording around ... it still cannot be installed using the safe graphics mode 17:19:24 and the only hardware I think we're hitting it on is hardware that works fine without nomodeset 17:19:27 nirik: so far we know it affects every UEFI system we've tried it on, and no BIOS ones. 17:19:35 oh ok, didn't know that 17:19:35 I am not pushing we block on that, it can be workarounded but I believe it breaks the criteria 17:19:43 but we have not tried it on any system that would actually *need* it, i.e. one which can't reach a graphical installer without it. 17:19:57 lruzicka: as the person who wrote the criteria, i'm telling you, it really is meant to mean what i said. 17:20:01 adamw, yes, thats right 17:20:08 LOL 17:20:11 the criterion is actually really about the boot menu 17:20:24 and rc5 has a 'fix' for this? or unclear what the fix is ? 17:20:28 unclear 17:20:30 the context is that we wanted to require the boot menu show up, and have the things on it that it's supposed to have on it 17:20:34 rc5 does not have a fix. 17:20:46 i've been testing ajax's attempts to fix it and it's confusing why it doesn't work 17:20:46 there is a fix, but it's not in rc5. 17:20:50 ok, so lets not block and put it in the common bugs 17:20:54 works fine for me 17:21:03 lruzicka: +1 17:21:12 ok so even less hardware that this is hitting 17:21:14 those people willing to install it can use the text mode 17:21:14 i think you're getting tripped up by gnome, fwiw. 17:21:18 lruzicka: +1 17:21:20 I'm inclined to -1 blocker, +1 FE again... for it not clearly affecting cases where we need it to work.... 17:21:31 -1 blocker, +1 FE 17:21:38 -1 blocker both because it's restricted in impact and doesn't actually violate the criterion 17:21:50 -1 blocker 17:22:32 -1 Blocker, +1 FE 17:22:44 -1 blocker +1 FE 17:22:49 same 17:23:02 proposed #agreed 1628495 - RejectedBlocker (Beta) AcceptedFreezeException (Beta) - this seems to be limited in impact and has not yet been demonstrated to render any real system incapable of reaching the installer. also it may not technically violate the criterion, as the boot menu does function as intended. It is accepted as an FE issue as improving the success rate of basic mode would at least possibly be helpful 17:23:06 ack 17:23:17 ackitty ack 17:23:18 ack 17:23:20 ack 17:23:29 frantisekz: oh that's a good one 17:23:38 ack 17:24:30 ack 17:24:59 #agreed 1628495 - RejectedBlocker (Beta) AcceptedFreezeException (Beta) - this seems to be limited in impact and has not yet been demonstrated to render any real system incapable of reaching the installer. also it may not technically violate the criterion, as the boot menu does function as intended. It is accepted as an FE issue as improving the success rate of basic mode would at least possibly be helpful 17:25:02 \o/ 17:25:22 okay, well we've managed to avoid adding any new blockers. let's check in on the existing 17:25:31 #topic (1629378) Trying to start dnfdragora causes an error in dnfdaemon and the program cannot continue 17:25:33 #link https://bugzilla.redhat.com/show_bug.cgi?id=1629378 17:25:34 #info Accepted Blocker, dnfdaemon, VERIFIED 17:26:07 This is fixed in RC5 17:26:20 super 17:26:30 #info Fixed in RC5 17:27:22 adamw: do we need to do anything to formalize this, or just note it's fixed and move on 17:27:22 but not in rc4? 17:27:56 it was fixed in rc4 too. 17:28:00 er 17:28:02 wait 17:28:04 let me check that 17:28:27 yeah. fixes were in rc4 too. 17:28:33 adamw: I think its fixed in rc4 as well 17:28:34 #undo 17:28:34 Removing item from minutes: INFO by bcotton at 17:26:30 : Fixed in RC5 17:28:40 #info Fixed in RC4 17:28:55 (also i fixed a bonus bug at the same time, just because i'm awesome.) 17:28:56 * sgallagh takes back his proxy from mattdm 17:29:02 adamw++ 17:29:42 adamw++ 17:29:44 x3mboy: Karma for adamwill changed to 6 (for the current release cycle): https://badges.fedoraproject.org/tags/cookie/any 17:29:58 woohoo karma fishing 17:30:42 bcotton: that's all you need, you can move on 17:30:46 rock 17:30:55 #topic (1629340) PackageKit update crashes at end of transaction with "TransactionItem state is not set: grub2-tools-1:2.02-57.fc29.x86_64" 17:30:56 #link https://bugzilla.redhat.com/show_bug.cgi?id=1629340 17:30:58 #info Accepted Blocker, libdnf, ON_QA 17:31:28 also fixed in rc4. 17:31:41 #info Fixed in RC4 17:32:00 okay, so there are no outstanding blocker bugs 17:32:29 Always good 17:32:30 #topic Current status - Release candidate compose is available 17:32:56 hey, we aced this one 17:32:59 we have *two*! 17:33:06 adamw: Haha :D 17:33:09 :-D 17:33:12 we're twice as good! 17:33:14 Now I wonder which one we will pick 17:33:17 lets ship em both! 17:33:28 mix the contents a little 17:33:35 lets ship RC4 as Beta and RC5 as final. 17:33:40 lruzicka++ 17:33:41 sgallagh: Karma for lruzicka changed to 2 (for the current release cycle): https://badges.fedoraproject.org/tags/cookie/any 17:33:41 :D 17:33:44 lruzicka++ 17:33:45 adamw: Karma for lruzicka changed to 3 (for the current release cycle): https://badges.fedoraproject.org/tags/cookie/any 17:33:45 so are there reasons we shouldn't call RC5 our Beta RC? 17:33:47 * adamw goes on vacation 17:33:52 lruzicka++ 17:33:53 bcotton: Karma for lruzicka changed to 4 (for the current release cycle): https://badges.fedoraproject.org/tags/cookie/any 17:33:58 (since everyone else is handing out karma) 17:34:08 bcotton: at this point i'd prefer we say we have both RC4 and RC5 as options 17:34:13 * mboddu also goes on vacation, no need to run final rc's :D 17:34:13 then go on to test coverage 17:34:18 ok 17:34:21 as that's kinda what should help determine which we ship 17:34:26 #info RC4 and RC5 are available as options 17:34:36 adamw: Can you highlight the differences between them? 17:34:40 #info neither has outstanding blockers 17:34:41 sure 17:35:20 #info RC5 includes a Silverblue installer image, RC4 does not 17:36:13 is that the only difference? 17:36:15 #info RC5 includes fixes for several GNOME bugs that RC4 does not 17:36:26 #info RC5 fixes https://bugzilla.redhat.com/show_bug.cgi?id=1628462, https://bugzilla.redhat.com/show_bug.cgi?id=1631068 and https://bugzilla.redhat.com/show_bug.cgi?id=1630134 17:37:15 * contyk votes for RC5 17:37:38 contyk: let's check out the test coverage first (that's the next #topic) 17:37:40 https://lists.fedoraproject.org/archives/list/test@lists.fedoraproject.org/thread/2GASMQHGCWWGNDG34D3XDB4RSLAZR6UV/ has a list of the bugs fixed 17:37:44 * nirik also is with contyk 17:37:52 one of the fixed issues was a proposed blocker today, even 17:37:54 * lruzicka goes kontiki 17:38:02 from a quick look, rc5 is not missing any images that rc4 had 17:38:10 you can compare https://fedoraproject.org/wiki/Template:Fedora_29_Beta_1.5_Download vs. https://fedoraproject.org/wiki/Template:Fedora_29_Beta_1.4_Download 17:38:41 seems like we picked up a couple of s390 cloud base images too 17:38:48 so for everyone running on s390 clouds, that'll be great? 17:38:57 =) 17:39:03 nice 17:39:23 I know the people who put work into that will appreciate it 17:39:24 More the merrier :D 17:39:52 okay, then let's talk test coverage 17:40:05 #topic Current status - Test coverage 17:40:13 #link https://fedoraproject.org/wiki/Category:Fedora_29_Test_Results 17:40:46 #link https://www.happyassassin.net/testcase_stats/29/ 17:42:07 i'm not seeing anything missing for rc4 and rc5 combined 17:42:26 except the 'two release ago' gnome upgrade test which we already know has a problem that we decided isn't beta blocking 17:42:29 someone even tested SAS... 17:42:54 oh, the AD tests 17:42:57 sgallagh, did you do those? 17:43:02 I did those on 1.1 17:43:16 We haven't pulled anything in subsequently that would alarm me 17:43:36 okay, i think that's reasonable 17:43:58 i guess you didn't put them in the matrix as testcase_stats shows they've never been done...could you update that when you get a minute? 17:44:06 Yeah, will do 17:44:16 the only thing that concerns me i guess is we don't seem to have any tests of physical CD media for 1.4 or 1.5 17:44:36 i was going to do one while the meeting was going on...only i can't seem to get gnome to see a blank disc to burn to :/ 17:44:42 OMG PROPOSAL DROP CD MEDIA 17:44:43 (and i managed to crash gnome-shell one time while trying) 17:45:02 adam, I would have done it but I though that CDs were for final release 17:45:05 mattdm: haha 17:45:08 lruzicka: they're only *required* for final 17:45:13 but it's nice to test at least one or two for beta 17:45:33 i can live with it if we haven't, though. at least no-one's reported any problems on list or anything. 17:45:42 adamw, ok ... I will remember that for next time 17:45:45 thanks 17:46:03 just test, like, the workstation x64 live and server netinst or something like that 17:46:17 I see 17:46:57 ok, that's all i got for test coverage 17:47:00 anyone else? 17:47:15 oh, i lied 17:47:24 i'd say rc5 coverage is sufficient to consider it a reasonable candidate for shipping 17:47:41 I agree with Adam 17:47:41 we got all the workstation tests re-run with it, and quite a lot of other tests too, enough to say there's no reason to believe some weird, unexpected regression showed up 17:47:52 thanks a lot for all the work to do that, lruzicka+co 17:48:09 rather co+lruzicka :D 17:48:27 lruzicka++ 17:49:01 okay, anything else on test coverage or objections to RC5? 17:49:29 shall we #info ? 17:49:36 yes! 17:49:49 (that was an enthusiastic ! not a demanding one) 17:50:23 #info test coverage between RC4 and RC5 is effectively complete (one upgrade test is from RC3 with a known bug that is rejected as blocker, AD join tests are from RC1, we are happy that these have not changed in RC4/RC5) 17:50:30 * x3mboy is hoping we have at least 1 minute of Open Floor 17:50:35 x3mboy: ack 17:50:44 #info RC5 test coverage seems sufficient to be quite confident it has no unexpected regressions compared to RC4 17:50:54 x3mboy: anything that will affect the go/no-go decision, before we make it? 17:51:07 adamw, nope 17:51:10 okok 17:51:20 It's a mktg thing related I want to bring up 17:51:24 last call for objections to RC5 before we do this whole decision thing 17:51:33 speak now or forever hold your peace 17:51:43 * lruzicka is shushing 17:51:51 3 17:51:53 2 17:51:56 1 17:52:06 #topic Go/No-Go decision 17:52:07 I will poll each team. Please reply “go” or “no-go” 17:52:10 FESCo 17:52:38 maybe 'go rc5' or 'go rc4' or 'no go'? 17:52:45 good call adamw 17:53:02 Go RC5 17:53:12 Releng 17:54:06 mboddu: 17:54:18 Lets ship RC5 17:54:22 QA 17:54:28 * nirik is good with rc5 also 17:54:58 Go RC5 (however, adamw will have his last say) 17:55:09 it's a team thing! 17:55:18 go rc5 (go per qa team policy, rc5 is my opinion) 17:55:20 go rc5 17:55:41 well that's 3-0 17:55:43 #agreed Fedora 29 Beta is GO with RC5 17:56:01 Hoorayyyy and up she rises 17:56:03 thanks & congrats ;) 17:56:08 COMPLETELY ON TIME 17:56:10 #info Fedora 29 Beta will release on 2018-09-25 17:56:16 \o/ 17:56:22 anyone who remembers some kind of meeting last week is a filthy traitor 17:56:24 #action bcotton to announce decision 17:56:30 #topic Open floor 17:56:31 yay!!!! 17:56:38 x3mboy, the floor is yours 17:56:42 #chair x3mboy 17:56:42 Current chairs: adamw bcotton x3mboy 17:56:43 adamw++ 17:56:57 ha. 17:57:20 Hi, I know is painful, I've being done this for 3 releases, but please, we need the Talking Point 17:57:27 Points* 17:57:35 https://fedoraproject.org/wiki/Fedora_29_talking_points 17:57:40 #link https://fedoraproject.org/wiki/Fedora_29_talking_points 17:57:48 * mboddu starts working on rc5 release 17:58:00 mboddu++ 17:58:10 x3mboy++ 17:58:10 mattdm: Karma for x3mboy changed to 14 (for the current release cycle): https://badges.fedoraproject.org/tags/cookie/any 17:58:11 This will help our ambassadors, Advocates and the Marketing team to promote the new release 17:58:11 x3mboy++ 17:58:12 adamw: Karma for x3mboy changed to 15 (for the current release cycle): https://badges.fedoraproject.org/tags/cookie/any 17:58:27 bcotton: can you help x3mboy badger the various teams? 17:58:36 mattdm: you got it 17:58:52 bt0 is also working on an annoyance campaign with them 17:58:52 bcotton++ 17:58:59 x3mboy++ 17:58:59 mboddu: Karma for x3mboy changed to 16 (for the current release cycle): https://badges.fedoraproject.org/tags/cookie/any 17:59:04 bt0 have been tryng to reach the teams by pagure 17:59:23 Hopefully we can have some to work on the Beta Release Announcement 17:59:25 x3mboy: i'd poke mailing lists and individuals on irc, if it was me 17:59:40 adamw, we did that 17:59:46 Also infiltrating the meetings 17:59:46 x3mboy: Server doesn't have a lot to report this release. 17:59:53 It's not easy 17:59:54 But ok 18:00:01 Not much in the way of flashy updates, but I'll see if the Cockpit folks can provide some highlights 18:00:03 sgallagh,good to know 18:00:07 modularity is new on WS 18:00:18 true 18:00:20 sgallagh: we have a whole new motd mechanism! 18:00:23 that nearly works! 18:00:30 lruzicka, yes, that is one of the biggest point this release 18:00:33 * adamw dies of excitement 18:00:37 lruzicka: but we don't have any explicit support in the gui tooling yet 18:00:46 it just doesn't explode anymore 18:00:51 contyk, no ... just CLI 18:00:52 adamw: I need to file a BZ, but the fix for the motd AVC broke the /etc/issue display :-P 18:00:54 RIP adamw 18:01:35 eom 18:01:47 thanks x3mboy 18:02:14 anyone else have a topic or is it time for me to go send some emails? 18:03:30 last call 18:03:37 sgallagh: ...har. 18:03:57 okay, thank you everyone for your hard work. i'll see you again when it's time for final! 18:04:14 Wooohoooooo!!!!!!!! 18:04:16 thanks bcotton 18:04:18 #endmeeting