15:00:22 <adamw> #startmeeting Fedora QA Meeting 15:00:22 <zodbot> Meeting started Mon Jun 4 15:00:22 2018 UTC. 15:00:22 <zodbot> This meeting is logged and archived in a public location. 15:00:22 <zodbot> The chair is adamw. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:22 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00:22 <zodbot> The meeting name has been set to 'fedora_qa_meeting' 15:00:31 <adamw> #meetingname fedora-qa 15:00:31 <zodbot> The meeting name has been set to 'fedora-qa' 15:00:35 <adamw> #topic Roll call 15:00:40 <lruzicka> .hello 15:00:41 * kparal is here 15:00:41 <zodbot> lruzicka: (hello <an alias, 1 argument>) -- Alias for "hellomynameis $1". 15:00:41 <adamw> morning folks, who's around for qa fun? 15:00:46 * satellit listening 15:00:54 <frantisekz> .hello2 15:00:55 <zodbot> frantisekz: frantisekz 'František Zatloukal' <fzatlouk@redhat.com> 15:00:59 <lruzicka> .hello2 15:01:00 <zodbot> lruzicka: lruzicka 'None' <lruzicka@redhat.com> 15:01:03 <adamw> (provision of 'fun' has not been confirmed by the food and drug agency) 15:01:12 * coremodule is here! 15:01:15 * sumantrom is here 15:02:11 <Lailah> Hi folks 15:02:23 <Lailah> .fas lailah 15:02:24 <zodbot> Lailah: lailah 'Sylvia Sánchez' <BHKohane@gmail.com> 15:02:25 <sumantrom> hi Lailah 15:03:07 <lruzicka> Hello everybody 15:03:14 <adamw> hi everyone 15:03:23 <adamw> how are you all doing? 15:03:45 <Lailah> Fine. Suffering summer. 15:03:52 <tflink> .hello2 15:03:53 <zodbot> tflink: tflink 'Tim Flink' <tflink@redhat.com> 15:05:02 <sumantrom> awesome adamw , how are you doing ? 15:05:34 <adamw> Lailah: allergies? 15:05:41 <adamw> sumantro: as well as can be expected for 8am on a monday :P 15:06:36 <sumantrom> adamw :P 15:07:08 <adamw> alrighty then, let's get at it 15:07:21 <adamw> #topic Previous meeting follow-up 15:07:34 <adamw> welp, we didn't assign any action items at the last meeting, which is always the sign of a highly productive meeting :P 15:07:40 <adamw> #info no action items to follow up on 15:08:03 <kparal> let's keep the tradition 15:08:27 <x3mboy> .hello2 15:08:27 <sumantrom> adamw there is one, since I wasn't there.. just a follow up, kernel test day is on 2018-06-12 15:08:27 <zodbot> x3mboy: x3mboy 'Eduard Lucena' <eduardlucena@gmail.com> 15:08:42 <adamw> good news to report, folks - especially lruzicka - have been following up on last week's Ticket O' The Week, so that's great 15:08:46 <sumantrom> and Storage test day will follow next 15:08:51 <adamw> sumantro: i mean formal #action's, but thanks 15:09:13 <sumantrom> adamw: np :) 15:09:16 <adamw> #info following up on test day status, sumantrom informs us the next kernel test day will be 2018-06-12, and storage test day will be after that 15:09:28 <lruzicka> adamw: Will need some more time to bite through that :) 15:09:36 <adamw> #info thanks to lruzicka and others for following up on the Ticket O' The Week, https://pagure.io/fedora-qa/issue/397 , we already have some proposals under discussion 15:09:42 <adamw> lruzicka: yup, it's an ongoing task for sure :) 15:11:26 <Kohane> Am I online or offline? Can anyone tell me? I'm tethering from my phone and it's really complicated. 15:11:37 <Kohane> Very unstable and limited. 15:11:42 <adamw> Kohane: we can hear you now :) 15:11:49 <Kohane> Excellent 15:11:55 <Kohane> adamw: thanks 15:12:01 <adamw> alright, anyone have anything else to follow up on from last week? 15:12:22 <Kohane> I wasn't here so I'm only reading here. 15:12:34 <Kohane> On this part, I mean. 15:12:59 <adamw> Kohane: we just noted a couple of test days and that there's activity on https://pagure.io/fedora-qa/issue/397 (the 'ticket of the week' from last week) 15:13:14 <adamw> you can catch up from the minutes later 15:13:21 <Kohane> I can't check the link adamw 15:13:34 <Kohane> I'm on a metered connection, every kb and mb countrs 15:13:36 <Kohane> counts 15:13:51 <adamw> it's the ticket about updating test cases to match release criteria. 15:13:59 <Kohane> .fas lailah 15:13:59 <zodbot> Kohane: lailah 'Sylvia Sánchez' <BHKohane@gmail.com> 15:14:08 <adamw> SWITCH TO TELEGRAM LANGUAGE STOP SAVE KOHANE MONEY STOP 15:14:09 <Kohane> Ah, okay 15:14:26 <Kohane> LMAO 15:14:34 <adamw> #topic Fedora 29 status 15:14:42 <adamw> FEDORA 29 ALL GREEN STOP 15:14:47 <adamw> ...ok i'll stop now. :P 15:16:07 <adamw> so i've run into some interesting bugs running rawhide lately, mostly got them fixed. 15:16:49 <adamw> openqa testing of composes up to 20180531.n.0 was looking good, almost all tests passing; 19 tests failed on the last compose (20180603.n.1), so i'll have to take a look at what's going on there 15:17:10 <adamw> #info automated testing of composes up to 20180531.n.0 looked great, 20180603.n.1 has 19 failures so will need to look into that 15:17:48 <adamw> there's a couple of controversial Changes under discussion ATM... 15:17:58 <Kohane> Which ones? 15:18:08 <adamw> I'm thinking of "Hide the grub menu" in particular 15:18:23 <Kohane> Oh... that one... 15:18:28 <adamw> but i also see "i686 Is For x86-64" which showed up this morning, that looks like it might provoke some debate 15:18:31 <Kohane> What's the point of that? 15:18:37 <adamw> polish, i think. 15:18:45 <adamw> and there's "Strong crypto settings: phase 2" 15:19:12 <adamw> have folks looked at these? anyone think we should take a position on them as a team, or look into them more extensively? what do we think about the test plans? 15:19:54 <adamw> some links: 15:19:54 <adamw> https://fedoraproject.org/wiki/Changes/StrongCryptoSettings2 15:19:58 <Kohane> I dislike hide the grub. It makes it harder to switch between kernels 15:20:07 <adamw> https://fedoraproject.org/wiki/Changes/HiddenGrubMenu 15:20:30 <adamw> https://fedoraproject.org/wiki/Changes/i686_Is_For_x86-64 15:20:55 <adamw> Kohane: it is easy to change the configuration back again if that's your personal preference 15:21:03 <satellit> will this apply to spins also or just wks? hidden grub menu 15:21:29 <adamw> satellit: good question...i suspect it would 15:21:32 <adamw> let's see if the change says so 15:22:05 <coremodule> I think hiding the grub menu makes Fedora more polished for the average, non-technical end-user. But how many of those do we cater to? 15:22:20 <adamw> "Write patches for anaconda to set GRUB_HIDDEN_TIMEOUT=1 and GRUB_HIDDEN_TIMEOUT_QUIET="true" in /etc/default/grub on non multi-boot systems" 15:22:26 <adamw> so as proposed, it'd affect all installs. 15:22:46 <satellit> ugh 15:23:01 <lruzicka> adamw: The prop says that it will be on systems with one OS installed. 15:23:28 <lruzicka> If I start with one OS only, and later add another OS, will the setting change automatically? 15:23:41 <adamw> coremodule: what i'm primarily after is whether we see sufficient reason to take an official stance on these changes as a team - if it's more a case of evaluating the proposal on its merits you can do that on a personal basis on devel@ 15:23:47 <lruzicka> because if not, it will be quite confusing for someone inexperienced 15:24:02 <adamw> lruzicka: probably not, no 15:24:16 <adamw> of course, it's generally the case that if you install Fedora first then Windows second you're gonna not be able to get into Fedora any more 15:24:26 <adamw> but we could consider the case of other Linux distributions and how that interaction works 15:24:33 <adamw> probably a point worth raising 15:24:38 <lruzicka> adamw: I meant two different linux distros 15:24:44 <lruzicka> for example 15:25:08 <lruzicka> the Windows case is pretty obvious, you cannot in fact, install windows later without reinstalling grub 15:25:14 <adamw> #info discussion of changes "i686 Is For x86-64", "Hide the grub menu" and "Strong crypto settings: phase 2" 15:25:39 <adamw> #info lruzicka points out that "Hide the grub menu" might have issues if installing Fedora first and a different Linux distribution second 15:26:04 <adamw> the "How to Test" section is fairly bare bones and does not consider this case at all 15:26:17 <adamw> it doesn't even consider what is presumably their expected multi-OS case, installing Fedora after Windows 15:26:21 <adamw> so we should probably point that out 15:26:35 <lruzicka> thanks 15:27:01 <adamw> #info "Hide the grub menu" Change's 'how to test' section is very bare and does not even cover checking that "Windows then Fedora" case does *not* have hidden boot menu 15:27:25 <sumantrom> adamw will grub-customizer still work and if yes where will the output be shown ? 15:27:51 <adamw> sumantro: i don't see why it wouldn't - the implementation of this Change is just to have anaconda change the default grub config a bit 15:27:58 <adamw> there's no fundamental messing around with how grub works 15:28:29 <sumantrom> got it! 15:28:34 <adamw> the idea behind the Change is you have to press a key to see the boot menu, of course you'd see your changes then 15:30:41 <lruzicka> adamw: Also, what comes to my mind, the testing machine in Brno shows the motherboard's boot menu. 15:31:12 <lruzicka> adamw: If GRUB will also take F8, will it not interfere? 15:31:20 <adamw> there's some discussion about that on devel@ 15:31:30 <lruzicka> I see 15:31:45 <adamw> i think there should always be a window where it would trigger grub, not the system firmware, but on a system where f8 triggers the system firmware it could be hard to hit 15:32:02 <adamw> there's a suggestion on list that grub should be set up to allow you to hit any key to get into the menu 15:33:33 * sumantrom is going through @devel thread to understand this, maybe a test day sometime for this changeset 15:33:52 * lruzicka will read that thread, too 15:33:54 <adamw> note these are still under review, they could be rejected or postponed 15:34:04 <adamw> but yes, if the crypto or grub changes are accepted, we should definitely have test days, i think 15:34:34 <sumantrom> adamw: sure :) 15:36:04 <adamw> i think the 'how to test' for the crypto change is also pretty lacking 15:36:47 <adamw> i think we could put some pressure on the change proposer to do a much more substantial evaluation of what release-critical paths are potentially affected 15:38:14 <sumantrom> I will soon have a mail thread going for both and come up with test cases and test day plans 15:38:16 <adamw> proposal - i'll reply to the change proposal threads with our concerns about the 'how to test' coverage 15:38:32 <adamw> sumantro: it might be best to wait for the changes to be approved (or not) at least 15:38:43 <adamw> no need to organize test days if they get rejected 15:38:51 <sumantrom> adamw +1 15:39:01 <lruzicka> adamw: That's a great idea. We should at least get enough info, so that we could develop tests 15:39:23 <lruzicka> adamw: Or they could develop them :) 15:39:38 <adamw> alright 15:39:54 <adamw> #action adamw to follow up on crypto and grub Change proposals with concerns about 'how to test' sections 15:40:13 <adamw> any other notes on this topic before we move on? 15:42:16 <adamw> #topic Release criteria discussions 15:42:30 <adamw> so last week i noticed a few release criteria discussions came up 15:43:04 <adamw> figured we could kick them around here a bit 15:43:33 <Kohane> Sure. But can I ask you a bit of context, please? I couldn't read any emails. 15:43:46 <adamw> yup, was working on it :0 15:44:10 <adamw> so, lukas proposed a test case for https://fedoraproject.org/wiki/Fedora_29_Final_Release_Criteria#Kickstarts criterion 15:44:17 <adamw> which made me think again about whether we should really have that criterion at all 15:45:28 <adamw> i think the reason we have that criterion at all is sort of about reproducibility, we wanted it to be possible to reproduce the artifacts from a fedora release using only the contents of that release 15:46:09 <adamw> i.e. you can take the livecd-creator or livemedia-creator from a release, the kickstarts from the fedora-kickstarts package, and the packages and comps from the release repo, and reproduce the live images 15:47:08 <adamw> i'm just...not totally sure we still need to jump through those hoops, or whether we can just do something much easier like having tags in the fedora-kickstarts and fedora-comps repos 15:47:37 <adamw> #info one topic for discussion is whether we still need/want the https://fedoraproject.org/wiki/Fedora_29_Final_Release_Criteria#Kickstarts criterion or whether git repo tags would be sufficient and much easier 15:47:53 <adamw> nirik: mboddu: any thoughts on this one? ^^^ 15:47:58 <adamw> mattdm: also 15:48:39 <adamw> note the requirement is kind of a pain in the ass, as it requires someone to jump through the hoops of cutting a 'release' of the kickstarts 'project' and doing a package build and submitting an update and then we have to file and approve a blocker bug to get it through the freeze 15:49:00 <nirik> I'm fine dropping it... 15:49:43 <lruzicka> Where will we keep the kickstarts then? In a separate dnf repo, or in a pagure repo? 15:50:25 <mboddu> adamw: +1 dropping it since 1 less thing to worry if we merge something after the GOLD RC 15:50:33 <lruzicka> If in a dnf repo, then the repo configuration file should be added to the system, so that people can easily switch that on, and then it will also need a release criterion 15:52:01 <adamw> lruiz: they're in a git repo 15:52:09 <adamw> lruzicka: they're in a git repo 15:52:10 <nirik> I think he was suggesting they just be in the git repo... but not sure what that means for the package... does it gets dropped? or just update in updates repo? 15:52:17 <adamw> the actual distro compose does not use the package, it just uses the git repo 15:52:33 <Kohane> adamw: About the kickstarts.... 15:52:59 <adamw> nirik: i mean, we could do various things. we could just make it a text file that says 'check out this git repo and here's the tag scheme'. or we could just not sweat updating it. or we could kill it. the other part of the puzzle is livemedia-creator docs and the wiki, of course 15:53:16 <lruzicka> In that case, I believe it will be enough if they live in a git repo, because you can obtain them from there, when you need them. 15:53:30 <adamw> Kohane: uh huh? 15:53:45 * nirik nods. We just need to do _something_ before next beta... 15:54:18 <Kohane> I don't know, I don't think there's any need for that criterion. 15:54:39 <Kohane> At least, simplify it somehow? Looks like a lot of hassle... 15:55:47 <adamw> roger 15:55:55 <adamw> so, seems like there's solid support for ditching that criterion 15:56:25 <adamw> #info nirik, mboddu, lruzicka, adamw, kohane agreed it's worth looking into dropping the kickstart criterion 15:56:38 <adamw> #action adamw to draft proposal to drop the kickstart criterion 15:56:53 <adamw> (i'd like to take the action as i kinda know where all the bits that go together are there) 15:57:03 <Kohane> Sure. 15:57:10 <adamw> so, we're running out of time now...let's cut it here and we can carry on next time 15:57:20 <adamw> #topic Open floor 15:57:22 <lruzicka> sounds good 15:57:28 <adamw> #info time running short, we will pick up remaining items next week 15:57:38 <adamw> anyone have anything we should definitely deal with now? 15:57:44 <adamw> or quick notes, thoughts etc? 15:58:13 <Kohane> I just wanted to say a quick comment on KDE Spin. 15:58:27 <Kohane> It has a silly yet annoying bug. 15:58:43 <Kohane> Should I file it against KDE upstream or Fedora? 15:58:53 <Kohane> It's related to networks. 15:59:10 <Kohane> Well, not all of them, wifi/hotspots 15:59:32 <Kohane> Well, file the bug when I get connection, but anyway. 15:59:33 <lruzicka> Kohane: Is it a KDE bug or a NetworkManager Bug? 15:59:47 <Kohane> That's a very good question. 16:00:07 <Kohane> I suspect KDE or the KDE bit that connect NM with Plasma. 16:00:13 <Kohane> But it could be NM itself 16:00:21 <lruzicka> Kohane: Are you using KDE as your primary desktop? If so, try with a different DE. 16:00:22 <adamw> what's the bug? 16:01:02 <Kohane> Thing is: If hotspots change, like you turn a router on, or one goes off, you don't see it. 16:01:25 <Kohane> You may keep seeing hotspots that are already off, and not seeing the new ones that are on. 16:01:30 <adamw> so it's kinda stuck with whatever picture it initially gets (on KDE startup) of available APs? 16:01:35 <adamw> it doesn't refresh? 16:01:39 <Kohane> It may take between several seconds to minutes to refresh 16:01:51 <Kohane> Yeah. 16:01:52 <adamw> i'd probably suspect the same thing as you then, KDE->NM integration 16:02:29 <adamw> for a KDE bug...I'd probably file it downstream initially, and maybe ask in #fedora-kde about it 16:02:38 <adamw> then file upstream if they ask you to, or there's no response on the bug... 16:03:09 <Kohane> I don't have another desktop to try right now, but I could try installing a Gnome related one when I get a decent connection. 16:03:15 <Kohane> Maybe Cinnamon. 16:03:17 <Kohane> If it helps 16:03:30 <Kohane> adamw: Okay, I'll do so. 16:03:30 <adamw> it'd be a good cross-comparison, yeah 16:03:38 <adamw> Workstation would be the best to compare with 16:04:06 <Kohane> Uhm. I dislike Gnome Shell. Maybe in VM? 16:04:14 <adamw> Kohane: it's just for comparison 16:04:21 <adamw> you need to test it on the same machine, though 16:04:27 <adamw> maybe just boot the live image? no need to install it 16:04:30 <Kohane> Yeah. 16:04:32 <Kohane> Oh! 16:04:39 <Kohane> Yes, I have it 16:04:58 <Kohane> I have two lives, one with the KDE Spin and another with Workstation. 16:05:26 <Kohane> Okay, I'll boot the live and test it. No need to install by now. 16:05:31 <adamw> right 16:05:33 <Kohane> adamw: Thanks for the idea! 16:05:37 <adamw> no probs :) 16:05:42 <adamw> OK, anything else, folks? we're past time now 16:05:51 <lruzicka> not from me 16:06:36 <sumantrom> nothing from my side 16:06:47 <Kohane> Nothing here. I already said my bit. 16:06:48 <tflink> nothing from me, either 16:07:21 <adamw> alrighty, thanks for coming, folks! 16:07:40 <sumantrom> thanks for hosting adamw :) 16:07:46 <adamw> see you next time 16:07:53 <adamw> note, there's no blocker review meeting today as there are almost no bugs to review 16:07:56 <Kohane> See you! 16:07:59 <adamw> so, everyone go start drinking :P 16:08:09 <Kohane> Eh... No. 16:08:16 <adamw> just me, then? ah well 16:08:17 <adamw> #endmeeting