17:00:16 #startmeeting F27 Beta Go/No-Go meeting - 2nd round 17:00:16 Meeting started Thu Sep 21 17:00:16 2017 UTC. The chair is jkurik. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:16 Useful Commands: #action #agreed #halp #info #idea #link #topic. 17:00:16 The meeting name has been set to 'f27_beta_go/no-go_meeting_-_2nd_round' 17:00:18 #meetingname F27-Beta-Go-No-Go-meeting-2nd 17:00:18 The meeting name has been set to 'f27-beta-go-no-go-meeting-2nd' 17:00:26 #chair nirik pschindl sgallagh mboddu kparal 17:00:26 Current chairs: jkurik kparal mboddu nirik pschindl sgallagh 17:00:30 .hello2 17:00:31 bowlofoutage: Sorry, but you don't exist 17:00:38 #topic Roll Call 17:00:41 * pschindl is here 17:00:45 .hello2 17:00:45 morning 17:00:45 sgallagh: sgallagh 'Stephen Gallagher' 17:00:46 .hello2 17:00:47 haha wut 17:00:48 jkurik: jkurik 'Jan Kurik' 17:00:54 .hello2 17:00:55 frantisekz: frantisekz 'František Zatloukal' 17:01:05 oh right 17:01:05 .hello2 17:01:06 bowlofeggs: bowlofeggs 'Randy Barlow' 17:01:09 bowlofoutage: you need to use FAS name 17:01:30 .hello2 17:01:31 kparal: kparal 'Kamil Páral' 17:01:43 .hello mohanboddu 17:01:44 mboddu: mohanboddu 'Mohan Boddu' 17:01:45 hi everybody 17:01:49 #topic Purpose of this meeting 17:01:58 #info Purpose of this meeting is to check whether or not F27 Beta is ready for shipment, according to the release criteria. 17:02:04 #info This is determined in a few ways: 17:02:12 #info * No remaining blocker bugs 17:02:20 #info * Release candidate compose is available 17:02:27 #info * Test matrices for Beta are fully completed 17:02:34 #topic Current status 17:02:40 As far as I am aware, the RC for F27 Beta is not yet ready. 17:02:46 As such, we do not have test matrices for the RC. 17:02:54 There is request to build RC: https://pagure.io/releng/issue/7064 17:03:03 The lastest F27 nightly compose is available at https://kojipkgs.fedoraproject.org/compose//branched/Fedora-27-20170920.n.0/ 17:03:08 it's composing now. 17:03:10 And latest test matrices for a nightly build: 17:03:19 https://fedoraproject.org/wiki/Test_Results:Fedora_27_Branched_20170912.n.0_Base 17:03:56 nirik: Fedora-27-20170920.n.0 should be already composed; Fedora-27-20170921.n.0 is in progress IMO 17:04:07 here's a summary of all the latest test matrices: https://fedoraproject.org/wiki/Test_Results:Fedora_27_Branched_20170912.n.0_Summary 17:04:15 jkurik: I meant the RC is in progress 17:04:26 ah, ok 17:04:55 #info The RC for F27 Beta is not yet ready 17:04:56 #link https://pagure.io/releng/issue/7064 - request for RC compose 17:05:07 #info the compose is currently in progress 17:05:22 #link https://qa.fedoraproject.org/blockerbugs/milestone/27/beta/buglist - F27 Blockers 17:05:35 anyone wants to add something ? 17:06:24 #info As we have no RC there are subsequently no Test Matrices for the F27 Beta RC 17:06:35 Let's do at least Mini-blocker review 17:06:44 pschindl: may I ask you please to chair the mini-blocker review ? 17:06:56 #topic Mini-Blocker Review 17:07:10 jkurik: yep, let's do it. 17:07:13 pschindl: thanks 17:07:23 #topic Introduction 17:07:25 Why are we here? 17:07:33 #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. 17:07:34 #info We'll be following the process outlined at: 17:07:42 #link https://fedoraproject.org/wiki/QA:SOP_Blocker_Bug_Meeting 17:07:44 #info The bugs up for review today are available at: 17:07:46 #link http://qa.fedoraproject.org/blockerbugs/current 17:07:53 #info The criteria for release blocking bugs can be found at: 17:07:55 #link https://fedoraproject.org/wiki/Fedora_27_Alpha_Release_Criteria 17:07:57 #link https://fedoraproject.org/wiki/Fedora_27_Beta_Release_Criteria 17:07:59 #link https://fedoraproject.org/wiki/Fedora_27_Final_Release_Criteria 17:08:13 Currently there is 8 proposed blockers, so let's start with them 17:08:23 #topic (1491119) server/workstation netinst in text mode crashes with pyanaconda.payload.NoSuchGroup: 3 17:08:25 #link https://bugzilla.redhat.com/show_bug.cgi?id=1491119 17:08:27 #info Proposed Blocker, anaconda, ASSIGNED 17:09:15 so, provided traditional server media is atm non-blocking, this only affects Workstation netinst 17:09:31 text mode is completely broken 17:10:04 yeah, completely broken is pretty blockery... 17:10:12 +1 to block 17:10:38 current workaround is to use graphical mode or everything netinst 17:10:45 * nirik looks for critera... just all install methods must work? 17:11:18 Lili linked https://fedoraproject.org/wiki/Fedora_27_Alpha_Release_Criteria#Installation_interfaces 17:11:18 nirik: When using a dedicated installer image, the installer must be able to complete an installation using the text, graphical and VNC installation interfaces. 17:11:22 "When using a dedicated installer image, the installer must be able to complete an installation using the text, graphical and VNC installation interfaces. " 17:11:52 right. thanks. 17:11:57 +1 blocker here 17:12:07 Yeah, +1 blocker 17:13:21 I actually was expecting people saying text mode is not that important for workstation image, or that everything netinst is fine :) 17:13:44 +1 blocker is the correct way here, yes 17:14:03 proposed #agreed - 1491119 - AcceptedBlocker (beta) - This bug violates the alpha criterion: ""When using a dedicated installer image, the installer must be able to complete an installation using the text, graphical and VNC installation interfaces." 17:14:14 ack 17:14:16 ack 17:14:37 ack 17:14:38 ack 17:14:50 #agreed - 1491119 - AcceptedBlocker (beta) - This bug violates the alpha criterion: ""When using a dedicated installer image, the installer must be able to complete an installation using the text, graphical and VNC installation interfaces." 17:15:18 #topic (1490832) dnf system-upgrade: dnf.exceptions.MarkingError: no package matched 17:15:20 #link https://bugzilla.redhat.com/show_bug.cgi?id=1490832 17:15:22 #info Proposed Blocker, dnf-plugins-extras, POST 17:15:49 this affects only upgrade attempts which use --enablerepo option 17:16:03 and there is an easy workaround available - just enable the repo permanently before running system-upgrade 17:16:22 and the failure on --enablerepo just doesn't do the update? ie, it doesn't mess up the system? 17:16:27 therefore I believe this is -1 Beta blocker, perhaps +1 Final blocker (we don't need to discuss that now) 17:16:41 nirik: it just doesn't do the update 17:16:43 nirik: no, it doesn't break anything, just reboots back almost immediately 17:16:48 right. just confirming. 17:17:06 -1 blocker, I guess I could be +1 FE... 17:17:16 I am +1FE 17:17:17 we don't need FE, it's for F26 17:17:20 the fix 17:17:40 ah ha. 17:17:47 ok, cool. 17:17:50 true 17:18:09 -1 beta blocker from me 17:18:13 I'm -1 beta totally, there is an easy workaround and no data are lost. 17:18:22 ack 17:18:34 -1 blocker 17:18:35 -1 beta blocker 17:18:47 BTW, for 1491119 - I'll be doing an Anaconda build with a fix later today 17:19:07 mkolman: great, thanks 17:19:27 mkolman: thanks 17:19:34 proposed #agreed - 1490832 - RejectedBlocker (beta) - there is an easy workaround - enable repo before running update. No data are lost. We will repropose as Final blocker 17:19:45 ack 17:19:46 ack 17:19:47 ack 17:19:50 ack 17:20:07 #agreed - 1490832 - RejectedBlocker (beta) - there is an easy workaround - enable repo before running update. No data are lost. We will repropose as Final blocker 17:20:18 #topic (1492036) system-upgrade tried to connect to online mirrors during upgrade 17:20:20 #link https://bugzilla.redhat.com/show_bug.cgi?id=1492036 17:20:22 #info Proposed Blocker, dnf-plugin-system-upgrade, NEW 17:21:24 we didn't have much time to test this due to the other dnf bug, but we haven't hit this, at least I'm not aware of anyone 17:21:38 so this is a single person reporting this issue 17:21:41 well, the reporter... 17:22:03 I think this is safe to reject as beta blocker 17:22:04 I wonder if it's related to the other one, seems similar symptoms. 17:22:27 I pinged dnf folks about it, I hope they'll look at it once they finish fixing the other issue 17:22:44 Who will do the secretary stuff? 17:22:47 yeah, if it's not widespread/reproducable, then -1 blocker. 17:22:51 pschindl: I'm already doing that 17:22:58 kparal: thank you 17:23:46 I am -1 to block. People might re-propose in case this will be more offten issue 17:24:37 -1 17:25:20 proposed #agreed - 1492036 - RejectedBlocker (beta) - Just single person (reporter) was able to reproduce. If there will be more reports, it can be re-proposed 17:25:38 ack 17:25:39 ack 17:26:53 what about others? ack/nack/edit? 17:26:55 ack 17:27:04 ack 17:27:22 #agreed - 1492036 - RejectedBlocker (beta) - Just single person (reporter) was able to reproduce. If there will be more reports, it can be re-proposed 17:27:42 #topic (1494061) gnome-software doesn't show F25->F27 upgrade even though it should 17:27:44 #link https://bugzilla.redhat.com/show_bug.cgi?id=1494061 17:27:46 #info Proposed Blocker, gnome-software, NEW 17:28:06 kalev: have you had chance to look at this? 17:28:34 I discovered this today 17:29:40 again, it's a direct violation. the workaround is to do 2 upgrades, or use dnf system-upgrade 17:30:06 should not this bug be somehow related to this: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/thread/X62V2QS6RGT7TKS25O6RF2DSUB7YXAXA/#X62V2QS6RGT7TKS25O6RF2DSUB7YXAXA ? 17:30:28 yes, perhaps there's just something wrong in pkgdb 17:30:39 I can't say, the json retrieved looks quite ok 17:30:47 I looks to me like so 17:31:06 I don't think it is. 17:31:33 if you set that you want to see prereleases, it's just showing you the prerelease... 17:31:48 it's showing F27 correctly when you're on F26 17:32:00 just for F25, it's showing F26 even if it should show F27 17:33:05 anyway, we have an workaround for this bug 17:33:35 so I am -1 to block on this 17:33:49 well we had workaround even for workstation netinst text mode 17:34:06 kparal: do you mean graphical install ? 17:34:12 use a different tool seems similar to use a different image :) 17:34:14 so, the fix here is likely in f25's gnome-software? 17:34:25 nirik: yes, it might 17:34:38 kparal: well, you can do f26 then f27 via the tool right? 17:34:53 nirik: yes, it's just not 'direct' as required in criterion 17:36:07 I don't have a strong opinion here. it should be +1 blocker, but sufficient other ways exist, so if you want to reject it, I won't object 17:36:08 yeah, I think thats the workaround jkurik was indicating... 17:37:15 voting ? I am still -1 to block 17:37:33 I guess I am -1 blocker as well... +1 FE, and if we can get a fix in f25 before we release f27 beta all the better. 17:37:41 -1 blocker 17:37:43 again, FE is not needed here :) 17:38:01 kparal: just in case :) 17:38:57 sure, likely the fix is in f25 side 17:38:57 -1 blocker 17:41:12 kparal: you did set it to show prerelease right? (just making sure) 17:41:22 nirik: I double checked now 17:41:23 proposed #agreed - 1494061 - RejectedBlocker - There is an workaround to update to F26 or to use another tool. 17:41:29 ok. :) 17:41:32 ack 17:41:51 ack 17:42:03 patch 17:42:07 RejectedBlocker(beta) 17:42:27 proposed #agreed - 1494061 - RejectedBlocker (beta) - There is an workaround to update to F26 or to use another tool. 17:42:56 ack 17:43:14 ack 17:43:26 #agreed - 1494061 - RejectedBlocker (beta) - There is an workaround to update to F26 or to use another tool. 17:43:44 #topic (1494138) crashes in wayland/Xorg on vc4 with 4.13 with the Raspberry Pi 17:43:46 #link https://bugzilla.redhat.com/show_bug.cgi?id=1494138 17:43:48 #info Proposed Blocker, kernel, ON_QA 17:44:08 this one is very fresh 17:44:18 I haven't seen it yet 17:44:57 I believe only XFCE is release blocking on armhfp 17:45:39 yeah, but not sure what desktop(s) were hitting this... possibly all? 17:45:48 pbrobinson: you happen to be around? 17:46:01 I wonder where this is documented, I again can't find it 17:47:01 it is XFCE, but the kernel issue I put as a blocker was all desktops, worse on workstation, but seen actively on XFCE too 17:47:10 * pbrobinson is here for about 2 mins 17:47:37 pbrobinson: is it this image: Spins/armhfp/images/Fedora-Xfce-armhfp-_RELEASE_MILESTONE_-sda.raw.xz 17:47:40 ? 17:47:51 looks about right 17:48:20 thanks, I am asking because this one is a blocking one 17:48:32 yup, that's what I thought 17:49:18 this doesn't affect installation nor live, and the kernel can be easily updated after installation 17:49:29 also, this is Beta. so I don't see this as critical 17:49:50 but rpi is popular, that's true 17:49:52 kparal: it affects disk images which is similar to lives 17:50:06 I mean it can't prevent installation 17:50:08 since that's just dd 17:50:16 it affects you after you boot 17:50:18 it can be updated post install 17:50:32 post boot even 17:51:05 if it doesn't crash while you're doing initial setup so you can actually log in 17:51:29 if we can update it post install/boot I would be -1 to block on this 17:52:41 we already have a fix, so I am about to give it +1 FE 17:52:47 I've seen it crash during graphical initial-setup which prevents login and hence updating 17:53:58 :( thats no good. 17:54:21 I'm not too happy about bringing yet another broken kernel in, but if it can prevent logging in, I guess I'd be more convinced 17:54:48 I guess +1 blocker... I don't see much way around it unless we want to hope people can get in and upgrade (and they may not know to even) 17:54:53 another? what's the NVR of the "broken" kernel, and why was it broken? 17:55:06 kparal: ^^ that was for you 17:55:58 the latest kernel updates broke qxl VMs 17:56:14 which means almost all VMs, since it's default in Fedora 17:56:25 kparal: if you look at the changelog that apparently fixes that too 17:56:31 that was rejected as Beta blocker, which in comparison can be considered worse than rpi problems 17:57:12 kparal: but the RPi is a blocking ARM device and it's affecting a blocking desktop for ARM 17:58:14 * kparal shrugs 17:58:19 I'm fine with +1 blocker 17:58:29 this update also has a bunch of other fixes too... 17:58:40 4.13.3 stable stuff. 17:58:52 kparal: https://koji.fedoraproject.org/koji/buildinfo?buildID=973169 "Fixes for QXL (rhbz 1462381) " 17:59:08 plus a bunch of CVEs too I believe 17:59:11 * jkurik is changing his mind 17:59:19 yeah, some cves also. 17:59:20 +1 to block as we can not log-in 17:59:24 seems a big pile of change. 17:59:42 pbrobinson: according to https://bugzilla.redhat.com/show_bug.cgi?id=1462381#c52 that doesn't sound so easy, but I'll be hoping 17:59:52 nirik: yeah, a source of more potentional blockers 17:59:57 the 4.13 stable series haven't been too large in terms of upstream changes 18:00:37 well, anyhow, I am +1 blocker 18:00:38 like upstream for 4.13.3 the biggest was XFS https://lwn.net/Articles/734243/ 18:01:02 but IMHO we will need a fair bit of testing to everything to make sure there's no gotchas in this. 18:01:39 I've tested this kernel on a x86 VM and around a dozen ARM devices today 18:01:58 Do someone has some criterion which it violates? 18:02:01 since we're slipping, we should be able to test new kernel properly 18:02:24 but we should push it stable asap, so that it gets testing in composes 18:03:16 pschindl: you can use " A working mechanism to create a user account must be clearly presented during installation and/or first boot of the installed system. " 18:03:18 https://fedoraproject.org/wiki/Fedora_27_Alpha_Release_Criteria#Expected_installed_system_boot_behavior 18:03:29 that breaks when that kernel panic occurs 18:03:36 kparal: Thanks. I was just thinking about this one :) 18:05:09 proposed #agreed - 1494138 - AcceptedBlocker (beta) - This bug can prevent initial-setup to finish. This violates the Alpha criterion: "A working mechanism to create a user account must be clearly presented during installation and/or first boot of the installed system." 18:05:26 ack 18:05:45 ack 18:06:12 ack 18:06:48 pbrobinson, frantisekz: ack/nack? 18:07:20 ack, sorry :) 18:07:31 #agreed - 1494138 - AcceptedBlocker (beta) - This bug can prevent initial-setup to finish. This violates the Alpha criterion: "A working mechanism to create a user account must be clearly presented during installation and/or first boot of the installed system." 18:07:44 #topic (1483331) Couldn't start mariadb.service 18:07:46 #link https://bugzilla.redhat.com/show_bug.cgi?id=1483331 18:07:48 #info Proposed Blocker, selinux-policy, POST 18:08:25 workaround is in comment 4 18:08:34 fix is ready: https://github.com/fedora-selinux/selinux-policy-contrib/commit/542b8bfcda1e26524d735ec09016abd3278ae4e0 18:08:49 but i don't see any koji build with the fix inc 18:08:54 the justification is in comment 3 18:09:27 since the criteria only talk about postgress, I think a blocker is not justified here 18:09:38 -1 blocker, +1 FE 18:09:44 this will be available as an update, of course. and we can give it FE 18:09:57 -1 blocker; +1 FE 18:10:03 -1 blocker, +1 FE too 18:10:24 -1 blocker, +1 FE 18:11:16 +1 FE 18:12:16 -1 blocker, +1 FE 18:12:30 * mattdm apologizes -- firefighting right now 18:12:52 proposed #agreed - 1483331 - RejectedBlocker (beta) AcceptedFreezeException - mariadb isn't blocking for Server Database role, but it is important enough so the fixing update will be considered as Freeze Exception 18:12:58 mattdm: Et tu, Brute? 18:13:12 ack 18:13:18 ack 18:13:48 ack 18:13:51 ack 18:14:02 #agreed - 1483331 - RejectedBlocker (beta) AcceptedFreezeException - mariadb isn't blocking for Server Database role, but it is important enough so the fixing update will be considered as Freeze Exception 18:14:14 #topic (1494108) On KDE, selinux prevents log in with newly created user 18:14:16 #link https://bugzilla.redhat.com/show_bug.cgi?id=1494108 18:14:18 #info Proposed Blocker, selinux-policy, NEW 18:14:58 What about just saying that KDE isn't primary DE :) 18:15:11 pschindl: +1 :) 18:15:16 pschindl: but it is blocking anyway 18:15:21 lbrabec references https://fedoraproject.org/wiki/Fedora_27_Beta_Release_Criteria#Shutdown.2C_reboot.2C_logout 18:15:29 which doesn't exactly cover this I think 18:15:47 but it's quite severe that newly created users can't log in 18:16:16 it could be under https://fedoraproject.org/wiki/Fedora_27_Final_Release_Criteria#Default_application_functionality but that's Final 18:16:30 what about the Alpha criterion: "A working mechanism to create a user account must be clearly presented during installation and/or first boot of the installed system." 18:16:39 Isn't this just problem of the tool which is used to create the user? 18:16:56 jkurik: that's just about the first user account, and that works 18:17:02 true 18:17:08 even though it was untested with graphical initial setup for KDE, it might not 18:17:34 it's unfortunate lbrabec didn't also test useradd 18:17:54 but according comment 5 that might be the case 18:18:19 ok, so -1 beta blocker, +1 final blocker ? 18:18:21 in that case this would be Final 18:18:40 pschindl: I don't have KDE installed, do you? 18:18:47 or frantisekz 18:19:01 I have, but F26 only 18:19:31 ok, so let's settle for -1 beta +1 final. it seems like kuser issue 18:19:54 I have KDE 18:20:08 should I try useradd? 18:20:13 frantisekz: can you quickly test useradd foo, passwd foo, login? 18:20:21 yep 18:21:10 BTW, in some cases Initial Setup might be needed in case you don't have root on the machine 18:21:25 as IS runs under root & can create a user 18:22:05 a pretty specific use case though (pre-prepared/OEM installs) 18:23:32 I was able to log in just fine after creating the account with useradd 18:23:42 ok, -1 beta +1 final stands 18:23:53 ...just for the record, I have KDE only in VM... 18:24:01 the criterion would be "All applications that can be launched using the standard graphical mechanism of a release-blocking desktop after a default installation of that desktop must start successfully and withstand a basic functionality test. " 18:24:13 frantisekz: should not affect anything 18:24:54 proposed #agreed - 1494108 - RejectedBlocker (beta) AcceptedBlocker (final) - This bug violates the final criterion: "All applications that can be launched using the standard graphical mechanism of a release-blocking desktop after a default installation of that desktop must start successfully and withstand a basic functionality test. 18:24:59 " 18:25:03 ack 18:25:04 ack 18:25:24 ack 18:26:00 #agreed - 1494108 - RejectedBlocker (beta) AcceptedBlocker (final) - This bug violates the final criterion: "All applications that can be launched using the standard graphical mechanism of a release-blocking desktop after a default installation of that desktop must start successfully and withstand a basic functionality test." 18:26:05 And the last one 18:26:15 #topic (1492981) pyanaconda.ui.gui.xkl_wrapper.XklWrapperError: Failed to replace/add layout with: 'in (eng)' 18:26:17 #link https://bugzilla.redhat.com/show_bug.cgi?id=1492981 18:26:19 #info Proposed Blocker, xkeyboard-config, ON_QA 18:27:02 this is already fixed, verified just now 18:27:35 so I'll close it and we can skip it 18:27:48 fixed in a stable package? 18:27:59 yes 18:28:07 cool 18:28:51 are we going now through accepted blocker or what's the usual practice, pschindl ? 18:29:15 #info This bug is already verified and in stable. 18:29:53 We probably should go through them. If there isn't someone against. 18:29:57 we definitely need to decide which accepted blockers no longer block common fedora release and are modular server specific. but that seems pretty clear 18:30:21 pschindl: so let's do that 18:30:38 rather quickly, before everybody dies of boredom 18:30:53 #topic (1491333) kickstart installations using autopart fail with 'Kickstart insufficient' 18:30:55 #link https://bugzilla.redhat.com/show_bug.cgi?id=1491333 18:30:57 #info Accepted Blocker, anaconda, ON_QA 18:31:34 This still needs testing. 18:31:43 We are waiting for today's compose 18:31:46 correct 18:32:18 #info We are waiting for build with fixed anaconda to test. 18:32:26 #topic (1489164) Fedora 27 Beta backgrounds must be different from Fedora 26 18:32:28 #link https://bugzilla.redhat.com/show_bug.cgi?id=1489164 18:32:30 #info Accepted Blocker, distribution, VERIFIED 18:32:53 this also needs verification with latest compose 18:32:58 This is already verified and fixing updates should be in RC 18:32:58 we verified with just a package update 18:33:06 ok 18:33:48 #info We verified that update fixes this but we are still waiting for build with updated packages to test it. 18:33:56 #topic (1490762) Ipa-server-install update dse.ldif with wrong SELinux context 18:33:58 #link https://bugzilla.redhat.com/show_bug.cgi?id=1490762 18:34:00 #info Accepted Blocker, freeipa, ON_QA 18:34:32 this one is moved to modular server beta blocker queue, correct? 18:34:47 kparal: I would say so 18:34:59 we hope a next compose will be tested by openqa and can confirm the fix 18:35:15 otherwise we'll need to pester server sig to confirm this, since adamw is away 18:35:19 sgallagh: ^^ 18:35:25 This should be on today's compose 18:35:29 * sgallagh looks 18:35:30 (the same applies for all freeipa bugs) 18:35:34 Sorry, juggling a lot today 18:35:35 and in the rc 18:35:47 true 18:36:07 I'll add a prefix to subject to all modular server bugs so that it's clear 18:36:17 unfortunately we can't separate them in blockerbugs app easily 18:36:21 I already tried 18:36:37 * nirik has no idea which are modular, I haven't looked at the proposed critera 18:37:14 nirik: I'm just assuming everything that would block server normally now blocks modular server 18:37:18 nirik: all bugs affecting only the server variant 18:37:22 we can deal with details later 18:37:30 jkurik: right 18:37:38 I don't think that this one is covered with openqa, so it would be good to have someone who is able to test it 18:37:42 well, I am sure there will be details... and I know some things won't make sense. 18:37:58 anything calling rolekit for example 18:38:28 anyhow, don't mean to sidetrack. This should be addressed, so lets move on. ;) 18:38:31 ok, we'll bug server sig about this 18:39:04 #info the fix will be on RC and today's compose and needs testing 18:39:15 #topic (1491056) FreeIPA enrolment via kickstart fails 18:39:17 #link https://bugzilla.redhat.com/show_bug.cgi?id=1491056 18:39:19 #info Accepted Blocker, freeipa, ON_QA 18:39:21 the same thing here 18:39:37 This is covered with openqa, so it will be easier to test. 18:40:20 #info fixed freeipa will be on today's compose and RC. Needs testing. 18:40:29 #topic (1487305) Raspberry Pi 3: run-initial-setup hangs 18:40:31 #link https://bugzilla.redhat.com/show_bug.cgi?id=1487305 18:40:33 #info Accepted Blocker, kernel, ON_QA 18:40:53 this got tested by sumantro, I hope with rpi3 18:40:56 will check 18:42:27 he did 18:42:37 I'll close, we can go on 18:43:04 #info sumantrom[m] verified the fix. We can close this one. 18:43:12 #topic (1170803) calls e2fsck on all ext volumes, provides no status indicator, and hangs indefinitely if e2fsck doesn't exit 18:43:13 #link https://bugzilla.redhat.com/show_bug.cgi?id=1170803 18:43:15 #info Accepted Blocker, python-blivet, ON_QA 18:43:34 there are so many people complaining about this one but no one wants to test the fix 18:43:53 if you happen to have 1PB partition or something, please help out 18:44:08 otherwise we'll just close and assume it's fixed 18:45:02 #info The fix for this bug still needs testing from someone who can reproduce the bug. 18:45:09 can Fedora buy some 1PB drives for people? ;) 18:45:14 #topic (1491508) FreeIPA server deployment fails with SELinux in enforcing mode, despite no obvious denials 18:45:16 #link https://bugzilla.redhat.com/show_bug.cgi?id=1491508 18:45:18 #info Accepted Blocker, selinux-policy, POST 18:45:28 this one is probably still broken 18:45:32 but that only affects server 18:45:57 so not our current concern 18:46:10 look like they have fixes... 18:46:13 but yeah 18:47:20 #info This bug will probably block just modular server. 18:47:28 #topic (1475570) Rescue mode fails while trying to access LVM volumes from existing install 18:47:30 #link https://bugzilla.redhat.com/show_bug.cgi?id=1475570 18:47:32 #info Accepted Blocker, systemd, VERIFIED 18:48:04 we'll test with latest compose, just to be sure 18:48:42 I tested my own boot.iso 18:48:42 #info We are waiting for update, but the build should appear on RC and is verified to work. 18:48:48 And that's it :) 18:49:01 Nothing else on the list. 18:49:06 jkurik: you take over 18:49:17 ok 18:49:18 So wake up 18:49:24 the overall QA answer is "not ready", I believe :) 18:49:34 pschindl: thanks for the blocker review 18:49:45 #topic Test Matrices coverage 18:49:50 #info As there is no RC yet, Test matrices are not ready 18:49:55 #info We are skipping the Test Matrices coverage check 18:50:02 #topic Go/No-Go decision 18:50:13 < QE, FESCo, RelEng - we need you opinion :) 18:50:27 Sadly no go. 18:50:31 QE is no go 18:50:38 I am no-go as well 18:50:41 no go 18:50:50 mboddu: what about releng ? 18:50:53 there's still a slight chance that mkolman will fix anaconda right now and we can create a new RC over night 18:50:56 * nirik isn't sure if we now slip final a week also or what 18:51:04 no go 18:51:04 but I guess it's not very likely 18:51:12 kparal: https://koji.fedoraproject.org/koji/taskinfo?taskID=22005667 18:51:12 nirik: that is what we need to agree 18:51:30 the build should finish in a few minutes :) 18:51:36 ok, so mkolman is the superman here 18:51:37 ha 18:51:42 * kparal checks blockers again 18:51:48 #agreed Due to missing RC for the F27 Beta release and presence of blocker bugs, the decision is “No Go”. 18:51:55 kparal: do we have all blockers actally addressed by some update? 18:51:58 * mkolman notes he is actually the bicycle-repair-man 18:52:02 mattdm: are you here ? 18:52:15 The question is what with server bugs? 18:52:44 jkurik: yep! 18:53:11 mattdm: we need to agree on the slip. you were proposing to slip for one day only (if possible) 18:53:15 one of freeipa bugs isn't probably fixed (https://bugzilla.redhat.com/show_bug.cgi?id=1491508) 18:53:17 I guess we do... so yeah, we could fire an rc2, but I am not sure heroics are good here. 18:53:35 jkurik: that wasn't my proposal; I think kparal suggested that? 18:53:37 pschindl: but that should be non blocking now. 18:53:48 pschinld: I guess we are not going to care about FreeIPA if Server is not blocking 18:53:49 mattdm: ah, ok 18:53:49 so, we're only missing the anaconda and the rpi3 kernel fix 18:53:58 both of them are available or almost available 18:54:01 non modular Server to be correct 18:54:17 if you can start a new RC shortly with both those builds, we could test it tomorrow 18:54:28 We're pushing ourselves pretty crazily hard here anyway. I'd rather slip a whole week and burn out people less if it comes to that. 18:54:45 * kparal just stating options 18:54:59 yeah. 18:55:07 I am not a big fan of slipping one week, we have a day off next Thursday here in Czech Republic :) 18:55:08 * sumantrom[m] sides with kparal 18:55:08 proposed #agreed The Beta release slips for one week. The one week slip is going to affect the Final GA as well. 18:55:17 that also doesn't give much time to retest things with the new kernel... 18:55:20 kparal: which is the rpi thing? 18:55:29 mattdm: https://bugzilla.redhat.com/show_bug.cgi?id=1494138 18:55:47 I updated blockerbugs page, you can refresh 18:55:51 https://qa.fedoraproject.org/blockerbugs/milestone/27/beta/buglist 18:56:00 jkurik: so, that means we slip from the sooner date to the rain date? or we push both those milestones out a week? 18:56:01 that's becoming increasingly strategically important 18:56:03 ignore "[Modular Server]" entries for the moment 18:56:14 nirik: both out a week. 18:56:23 we're gonna need that rain date still :) 18:56:43 it's true that for a new kernel a day of testing is not awesome 18:56:49 update with #1491119 fix: https://bodhi.fedoraproject.org/updates/FEDORA-2017-3184b77d1e 18:56:55 nirik: we push both milestones; we already used the "rain day feature" 18:57:32 mkolman: thanks for very fast response 18:57:33 alright, so 31st or 7th. 18:57:53 kparal: I'm glad I can help :) 18:57:54 would be nice to have a halloween release again... but anyhow. 18:58:00 as frantisekz already said, just noting that next thursday is a state holiday in CZ 18:58:34 I guess that also make is more likely people will be on PTO on Friday 18:58:39 *makes 18:58:47 yeah, but that's after next go/nogo :) 18:58:55 so shouldn't matter that much 18:59:00 good point 18:59:00 mattdm & jkurik i think you are saying opposite things.. you may want to use actual dates vs one week/rain date 18:59:03 I'm for just slipping a week and doing things right... but we shouldn't be complacient. I think we should fire rc2 today anyhow to get it more testing... 18:59:07 proposed #agreed The Beta release slips for one week. The one week slip is going to affect the Final GA where the target date is going to move to 2017-10-31 18:59:19 nirik: sounds good 18:59:23 kparal: Let's get it all signed off before the holiday :) 18:59:30 langdon: I think I said the same thing. 18:59:41 nirik, langdon: is it better ^^^ (the proposal) 19:00:07 how about: all futher milestones will move out 1 week also 19:00:11 pschindl: we'll request RC2 with those kernel and anaconda builds on top of RC1 19:00:18 ok.. just checking.. i am not sure.. and don't have a horse in this race :) .. so as long as you (jkurik, mattdm and nirik) are on the same page ... 19:00:18 nirik: sound good 19:00:26 I think so :) 19:00:32 final freeze 17th, scheduled final release 31st, raid date nov 7th 19:00:54 rain. sheesh. ;) everyone add disks to your storage pool on the 7th! 19:00:54 proposed #agreed The Beta release slips for one week. All futher milestones will move out 1 week also. 19:00:55 nirik: is that when you kill the bugs? 19:01:24 they check in, but they don't check out! 19:01:30 nirik: lol 19:01:32 ack 19:01:49 more acks please ? 19:01:54 ack 19:02:01 ack 19:02:09 ack 19:02:16 #agreed The Beta release slips for one week. All futher milestones will move out 1 week also. 19:02:23 langdon: soon we can just merge the modular server schedule back in. ;) 19:02:24 #action jkurik to publish the Go/No-Go result 19:02:26 * nirik runs 19:02:51 pschindl, kparal: are you willing to join the go/no-go the next week (at the same time) ? 19:02:54 jkurik: would you mind pinging me when you update the schedule? cause i am gonna want to check it against my proposed schedule for modular server 19:03:03 nirik: nah.. we are gonna release first :) 19:03:06 langdon: ok 19:03:09 :) 19:03:43 jkurik: I can't promise it right now, but I'll figure it out and let you know. as a back up tflink can do it, I believe 19:04:01 #action jkurik to update the F27 schedule and ping langdon to check proposal of the Modular Server schedule 19:04:12 kparal: ok, thanks 19:04:21 jkurik: thanks.. i probably will catch it anyway.. but "interrupt driven" 19:04:49 #action jkurik to organize third round of Go/No-Go meeting for F27 Beta on Thursday, September 28th at 17:00UTC 19:04:55 #topic Open floor 19:04:59 anything else ? 19:06:07 nothing from me 19:06:11 I am going to close the meeting in 1 minute 19:07:03 thanks jkurik! 19:07:14 thanks; bye all! 19:07:19 #endmeeting