17:00:14 #startmeeting F26 Alpha Go/No-Go meeting 17:00:14 Meeting started Thu Mar 16 17:00:14 2017 UTC. The chair is jkurik. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:14 Useful Commands: #action #agreed #halp #info #idea #link #topic. 17:00:14 The meeting name has been set to 'f26_alpha_go/no-go_meeting' 17:00:15 #meetingname F26-Alpha-Go-No-Go-meeting 17:00:15 The meeting name has been set to 'f26-alpha-go-no-go-meeting' 17:00:16 #topic Roll Call 17:00:18 .hello jkurik 17:00:19 jkurik: jkurik 'Jan Kurik' 17:00:25 .hello sgallagh 17:00:26 sgallagh: sgallagh 'Stephen Gallagher' 17:00:29 .hello roshi 17:00:30 roshi: roshi 'Mike Ruckman' 17:00:31 #chair dgilmore nirik adamw sgallagh roshi mboddu 17:00:31 Current chairs: adamw dgilmore jkurik mboddu nirik roshi sgallagh 17:00:31 .hello coremodule 17:00:32 coremodule: coremodule 'Geoffrey Marr' 17:00:47 Hi every one 17:00:51 morning 17:01:01 evening :) 17:02:02 do we have someone from releng ? dgilmore, mboddu ? 17:02:07 afternoon :) 17:02:49 allo 17:02:52 * langdon lurks 17:04:12 lets move on, hopefully someone from releng appears during the intro ... 17:04:21 #topic Purpose of this meeting 17:04:22 #info Purpose of this meeting is to check whether or not F26 Alpha is ready for shipment, according to the release criteria. 17:04:32 #info This is determined in a few ways: 17:04:33 #info No remaining blocker bugs 17:04:34 #info Release candidate compose is available 17:04:36 #info Test matrices for Alpha are fully completed 17:04:40 #topic Current status 17:05:13 As far as I am aware, the RC for F26 Alpha is not yet ready ( https://pagure.io/releng/issue/6701 ) 17:05:15 As such, we do not have test matrices for the RC 17:05:16 The last F26 nightly compose is available at https://kojipkgs.fedoraproject.org/compose/branched/Fedora-26-20170315.n.0/compose/ 17:05:22 Am I correct ? 17:06:06 yep 17:06:07 jkurik: correct, we kicked it off (mboddu was teaching me since I'm new) and it is running as we speak 17:06:14 jkurik: yep. 17:06:23 makes this meeting easy :) 17:06:27 jkurik: we will have to slip 17:06:37 jkurik: but we should do a mini blocker review 17:06:40 yep. We are no go 17:06:43 #info RC for F26 Alpha is not yet ready 17:06:45 #link https://pagure.io/releng/issue/6701 - ticket for the F26 Alha RC 17:06:46 #link https://kojipkgs.fedoraproject.org/compose/branched/Fedora-26-20170315.n.0/compose/ - the last F26 nightly compose 17:06:56 Let's start with Mini-blocker review 17:06:56 jkurik: thats a bit of an unfair statement 17:06:57 Roshi, adamw: may I ask you please to chair the mini-blocker review ? 17:07:11 sure thing 17:07:29 #topic F26 Alpha Mini Blocker Review 17:07:33 dgilmore: you mean the RC compose status ? 17:07:33 jkurik: the request for RC came in 16 hours ago, when we were all in bed 17:08:05 #topic (1431879) Pre-GDM gnome-initial-setup fails to run (when no user created during install), with log WARNING: Unable to find required component 'gnome-settings-daemon' 17:08:08 jkurik: the way you worded it makes it sound like Releng is dragging their feet. which is not the case 17:08:09 #link https://bugzilla.redhat.com/show_bug.cgi?id=1431879 17:08:12 #info Proposed Blocker, gnome-initial-setup, ON_QA 17:08:28 yeah, it's not a releng dragging feet issue at all 17:08:39 sure, so lets rephrase it 17:08:53 roshi, I will roll-back wait a second 17:09:10 #undo 17:09:10 Removing item from minutes: INFO by roshi at 17:08:12 : Proposed Blocker, gnome-initial-setup, ON_QA 17:09:12 #undo 17:09:12 Removing item from minutes: 17:09:15 "The request for a release candidate was not made in time for a release this week"? 17:09:22 #undo 17:09:22 Removing item from minutes: 17:09:24 #undo 17:09:24 Removing item from minutes: 17:09:32 #undo 17:09:32 Removing item from minutes: 17:09:33 #undo 17:09:33 Removing item from minutes: 17:09:37 #undo 17:09:37 Removing item from minutes: INFO by jkurik at 17:06:43 : RC for F26 Alpha is not yet ready 17:10:21 /me notes that he has proposed in the past that if an RC is not available by 23:59 UTC on the Tuesday before Go/No-Go, it should just be an automatic No-Go. 17:10:33 sgallagh: I agree 17:11:17 dgilmore: would you like to make the propper wording of the statement ? 17:12:50 something like #info due to the late resoultion of blocker bugs we are currently without a completed RC compose 17:13:39 #info Due to the late resoultion of blocker bugs we are currently without a completed RC compose 17:13:44 dgilmore: thanks 17:13:50 * adamw here, sorry 17:13:55 #link https://pagure.io/releng/issue/6701 - ticket for the F26 Alha RC 17:14:07 * mboddu here, sorry for the delay 17:14:14 mboddu: no problem 17:14:20 let move on ... 17:14:23 #topic F26 Alpha Mini Blocker Review 17:14:27 sgallagh: we have almost all the alpha testing automated now, so if an rc had been kicked off last night we might actually have got the testing done. but yeah, we got to rc request stage late. 17:15:03 #link https://qa.fedoraproject.org/blockerbugs/milestone/26/alpha/buglist 17:15:10 It happens. We'll be in good shape next week :) 17:15:15 roshi: may I ask you please ? 17:15:20 sure thing 17:15:32 #topic (1431879) Pre-GDM gnome-initial-setup fails to run (when no user created during install), with log WARNING: Unable to find required component 'gnome-settings-daemon' 17:15:36 #link https://bugzilla.redhat.com/show_bug.cgi?id=1431879 17:15:38 #info Proposed Blocker, gnome-initial-setup, ON_QA 17:16:31 Unless there is more information, I am standing with my assertion that this wouldn't pass the "last blocker at Go/No-Go" test. 17:16:33 I'm +1 blocker on this, I don't think the wording is meant to be one or the other, but rather either but not both 17:16:36 That said, I'm +1 FE 17:16:38 so this basically means that if you install Workstation without creating a user during install, the installed system appears to hang during boot 17:17:02 And given that we now have a week to do it, I'd encourage a fix ASAP 17:17:28 there's already a fix for the initial bug, i ran into something else after that 17:17:31 need to re-test a few times 17:18:40 roshi: There's enough wiggle room here that if we had an RC without this fixed in it, I think we'd argue about it quite a bit. 17:19:18 i honestly don't completely remember what we 'really' meant by the relevant criteria any more. i'd have to go back and look at the archives. 17:19:58 we can always revise it later, but it is ambiguous now 17:20:18 blocker votes? (FE doesn't matter since it already has the votes itneeds for that) 17:20:22 I strongly suspect that the intent was that whichever one you picked had to work... but I'd argue that "at least one of them works" would be good enough for Alpha and "both approaches must work" is a Beta criterion 17:20:30 Of course, that's going to be moot in F27 anyway 17:20:49 roshi: it's unavoidably ambiguous, i do remember we've been through like four wordings already. 17:20:50 true 17:20:55 it's complicated because we have a lot of different first boot scenarios. 17:21:01 -1 blocker for the reasons above. 17:21:04 sgallagh: i don't have the same recollection as you, as i said on the bug. 17:21:15 i am +1 blocker 17:21:20 i recall the 'or' wording being there just because there are cases where no first boot screen appears and we didn't want the criteria to sound like they required one to appear. 17:21:39 I am +1 to block 17:21:47 at the keast anyone running workstation on armhfp will hit this 17:21:54 /me shrugs 17:21:55 so we're +4/-2 counting nirik in the bug 17:22:01 or did I miscount? 17:22:11 I am +1 blocker 17:22:58 I'd probably argue louder if this was the only reason we weren't shipping today. 17:23:33 proposed #agreed - AcceptedBlocker - RHBZ#1431879 - This bug is a clear violation of the following 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." 17:23:45 i'd say 'violation' not 'clear violation' since we're not unanimous 17:23:46 ack 17:23:49 otherwise ack 17:24:10 what adamw said 17:24:34 proposed #agreed - AcceptedBlocker - RHBZ#1431879 - This bug is a violation of the following 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." 17:24:57 #agreed - AcceptedBlocker - RHBZ#1431879 - This bug is a violation of the following 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." 17:25:08 #topic (1432667) kernel 4.11.0-0.rc2.git0.1.fc26 oops on Allwinner SoCs 17:25:11 #link https://bugzilla.redhat.com/show_bug.cgi?id=1432667 17:25:13 #info Proposed Blocker, kernel, ON_QA 17:26:09 I need to investigate some, but the docker base images are getting an opps also 17:26:14 not sure it is the same 17:26:17 Are Allwinner SoCs blocking? 17:26:28 sgallagh: yes 17:27:02 https://fedoraproject.org/wiki/Architectures/ARM/Supported_Platforms 17:27:11 though its not been updated for 26 17:27:22 this is a funny case 17:27:28 Right, Spins/armhfp/images/Fedora-Minimal-armhfp-_RELEASE_MILESTONE_-sda.raw.xz is release-blocking 17:27:29 ok 17:27:34 because we were only pulling in the kernel that broke as an FE 17:27:41 didn't the oops happen on a different kernel than we had in the RC request? 17:28:06 no, we had the kernel that crashed in the original RC request to fix a different FE 17:28:22 ah 17:28:50 so it's kind of a weird situation, but it can't really hurt to take it as a blocker just on the basis of ensuring whatever kernel we wind up with, it's not one that's subject to this bug 17:28:55 so on that basis i'm +1 blocker 17:29:08 +1 blocker 17:29:11 I can buy that. +1 blocker 17:29:50 +1 to block 17:30:11 proposed #agreed - AcceptedBlocker - RHBZ#1432667 - This bug is a violation of the following criterion: "All release-blocking images must boot in their supported configurations." 17:30:27 ack 17:30:32 ack 17:30:37 #agreed - AcceptedBlocker - RHBZ#1432667 - This bug is a violation of the following criterion: "All release-blocking images must boot in their supported configurations." 17:30:39 ack 17:30:52 that's all the proposed blockers, now to look at progress on the accepted blockers 17:31:04 we have 3 17:31:06 #topic (1430250) bind-pkcs11 keeps failing to connect to LDAP server during FreeIPA server deployment on current Rawhide 17:31:09 #link https://bugzilla.redhat.com/show_bug.cgi?id=1430250 17:31:11 #info Accepted Blocker, 389-ds, ON_QA 17:32:08 all the accepted blockers are ON_QA, so really nothing to do here except test the RC when it comes down 17:32:09 so this is fixed for new installs, but we need another change to handle existing freeipa servers on upgrade 17:32:44 oh wait, sorry 17:32:45 wrong bug 17:33:15 this one is straightforward, it's fixed now and listed in the RC request, we just need to psuh the update stable. 17:33:39 yup 17:33:42 right 17:34:03 #info this bug is fixed and listed in the RC request. Please test when the RC lands. 17:34:15 #topic (1432582) authconfig 7.0.0 breaks nsswitch.conf (empty passwd, shadow and group lines) 17:34:18 #link https://bugzilla.redhat.com/show_bug.cgi?id=1432582 17:34:20 #info Accepted Blocker, authconfig, ON_QA 17:35:02 same deal as before, I think 17:35:31 yeah, it's a one-man show so far (I found the bug and wrote the patch and confirmed the fix) so it'd be good if anyone else can verify 17:35:35 but works so far as i'm concerned 17:35:54 adamw: I'll review it after the meeting. I didn't realize it hadn't been getting attention 17:36:07 oh, tmraz +1ed the update, another confirmation would be good. 17:36:25 #info this bug is fixed and listed in the RC. Please test after the RC lands. 17:36:33 that enough for this one? 17:36:47 sure 17:36:58 #topic (1430511) cloud init doesn't setup ssh keys for access 17:36:59 #link https://bugzilla.redhat.com/show_bug.cgi?id=1430511 17:36:59 #info Accepted Blocker, cloud-init, ON_QA 17:37:56 same kinda deal, requested for RC 17:38:23 #info this bug is fixed and listed in the RC. Please test after the RC lands. 17:38:35 that's it, unless we also want to go through FEs 17:38:48 but since we're slipping, I think we can forego that this time around 17:39:01 thoughts? 17:39:19 yes, I would skip the FE review 17:40:05 are there any proposed FEs? 17:40:19 there's one 17:40:33 #topic (1413387) SELinux is preventing spice-vdagentd from 'getattr' accesses on the filesystem /sys/fs/cgroup/systemd. 17:40:36 #link https://bugzilla.redhat.com/show_bug.cgi?id=1413387 17:40:38 #info Proposed Freeze Exceptions, selinux-policy, NEW 17:41:39 +1 FE 17:41:51 it is nominated as Final blocker 17:42:15 yes, but also as Alpha FE. 17:42:16 jkurik: and alpha FE 17:42:17 :P 17:42:26 ok, +1 FE for Alpha 17:42:29 for alpha I am +1 FE 17:42:39 +1 FE 17:42:49 +1 FE 17:42:51 +1 17:43:15 proposed #agreed - AcceptedFreezeException - RHBZ#1413387 - This would be good to get pulled in for Alpha release if a tested fix is available. 17:43:57 ack 17:43:59 ack 17:44:05 ack 17:44:10 #agreed - AcceptedFreezeException - RHBZ#1413387 - This would be good to get pulled in for Alpha release if a tested fix is available. 17:44:14 that's it 17:44:21 roshi: thanks 17:44:21 back to you jkurik :D 17:44:50 as we do not have the RC, we can skip the "Test Matrices coverage" topic 17:44:54 #topic Go/No-Go decision 17:45:02 releng is no go 17:45:10 engineering is no-go 17:45:10 qa is no go 17:45:17 it's pretty clear no go. ;) 17:45:30 sure, no go from my self as well 17:45:30 * roshi thinks he's allowed to say that 17:45:46 roshi: course you can 17:46:10 let's ship nothing! 17:46:27 it's much easier that way 17:46:31 proposed #agreed Due to missing RC of the F26 Alpha release, the decision is “No Go”. The whole release slips for one week. 17:46:33 * dgilmore ships adamw some whiskey 17:46:36 ack 17:46:40 ack 17:46:41 ack 17:46:48 patch 17:46:56 sgallagh: go on 17:47:32 Actually, never mind. 17:47:38 :) 17:47:42 I was trying to avoid the same phrasing problem we had before. 17:47:45 sgallagh: what were you thinking? 17:48:04 "missing RC" sounds again like we're blaming rel-eng 17:48:08 Which we clearly arent 17:49:05 sgallagh: context is different 17:49:20 proposed #agreed As the RC of F26 Alpha release is missing due to late blockers, the decision is “No Go”. The whole release slips for one week. 17:49:25 No problem, go ahead 17:49:27 sgallagh: sounds better ? 17:49:49 jkurik: I prefer that one, yes 17:49:55 sgallagh: at the start we had 17:49:56 17:06 < jkurik> #info RC for F26 Alpha is not yet ready | 17:49:56 Yes, this sounds a lot better. 17:50:00 17:06 < jkurik> #link https://pagure.io/releng/issue/6701 - ticket for the F26 Alha RC | 17:50:04 17:06 < jkurik> #link https://kojipkgs.fedoraproject.org/compose/branched/Fedora-26-20170315.n.0/compose/ - the last F26 nightly compose 17:50:24 ack 17:50:26 ack 17:50:30 ack 17:50:33 jkurik: nicely worded :) 17:50:37 #agreed As the RC of F26 Alpha release is missing due to late blockers, the decision is “No Go”. The whole release slips for one week. 17:50:41 dgilmore: thanks :) 17:50:51 #action jkurik to publish the Go/No-Go result 17:50:58 #topic Open floor 17:51:00 anything else to discuss today on this meeting ? 17:51:06 nada 17:51:14 * roshi has nothing 17:51:22 Nothing that I can remember. 17:51:35 * roshi also can't believe we're already at this point for F26 - thought we had more time :p 17:52:28 ok, so thanks for your presence on this meeting 17:52:32 #endmeeting