15:00:59 #startmeeting Fedora QA Meeting 15:00:59 Meeting started Mon Oct 1 15:00:59 2018 UTC. 15:00:59 This meeting is logged and archived in a public location. 15:00:59 The chair is adamw. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:59 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00:59 The meeting name has been set to 'fedora_qa_meeting' 15:01:06 #meetingname fedora-qa 15:01:06 The meeting name has been set to 'fedora-qa' 15:01:10 #topic Roll call 15:01:16 morning folks, who's around for QA meeting fun? 15:02:30 * kparal is here 15:02:55 * jlanda is semiafk 15:04:25 * coremodule is here. Good morning everyone. :) 15:04:36 ahoy to the oy 15:04:43 * adamw scratches 15:05:53 kparal: where's the rest of brno? out celebrating promotions? :P 15:06:09 * satellit listening 15:06:17 .hello2 15:06:18 frantisekz: frantisekz 'FrantiĊĦek Zatloukal' 15:06:23 I'm in the process of prodding them 15:07:12 .hell2 15:07:15 .hello2 15:07:16 Southern_Gentlem: Sorry, but you don't exist 15:07:26 * adamw hands kparal the high-voltage prod 15:07:32 .hello jbwillia 15:07:33 Southern_Gentlem: jbwillia 'Ben Williams' 15:08:46 * sumantro is here 15:11:06 alrighty then 15:11:11 guess that's everyone! 15:11:17 #topic Previous meeting follow-up 15:12:37 so, just one thing here: 15:13:45 thanks to jskladan & kparal for attending PRs, and adamw for making it possible ;) 15:14:30 #info "adamw to make sure jlanda's PRs on fedora-qa get reviewed" - I passed those along and they're all getting responses 15:16:51 #topic Fedora 29 status 15:17:07 so, the big note *I* have here is I'm still trying to sort out the plan for updating DNF 15:20:00 we've had a successful Rawhide compose with DNF 3.6 now, which is...something...but there's an outstanding question of whether we're gonna send 3.6 to F29, or do something about https://github.com/rpm-software-management/dnf/pull/1226 first. i'm still waiting to hear back from dmach on that. 15:21:52 adamw dnf test day is day after tomorrow with 3.5 ... not sure if we should then do it with 3.6 15:22:01 sumantro: uff :( 15:22:09 i will try very hard to come up with something sane for that, sorry 15:23:37 its okay :) 15:24:33 #info plans for updating DNF in F29 are still not 100% clear, adamw is trying to find out the plan in relation to https://github.com/rpm-software-management/dnf/pull/1226 15:25:00 that mostly ate my week last week, but at least nightly composes are appearing and mostly passing openQA. anyone have any thoughts/notes on current f29? major issues? 15:25:52 nop here, woprking weel on my tests. I can't either reproduce the big Pictures/ issue :S 15:26:03 I saw some troubles rebooting my VMs, waiting two minutes before a reboot 15:26:24 probably a race 15:28:20 i was seeing that on my desktop for a bit, seemed to stop happening recently... 15:28:26 some service was waiting for 90 seconds 15:28:29 (timeout) 15:28:42 regarding the grub issues with windows dualboot, it seems somebody else is also affected by that, so this is not ideal too 15:28:44 * nirik notes fesco is talking dnf now... 15:29:48 frantisekz: i was just talking to pjones about that 15:29:58 frantisekz: he thinks you two likely don't actually have the same bug 15:30:06 frantisekz: the other person's bug is likely due to them having two ESPs 15:30:16 i asked in-bug for you to check if your affected system also has two ESPs... 15:30:39 ok, that's interesting, i'll handle the needinfo later on, it is on my home pc which I don't have anywhere near atm 15:31:06 rgr 15:32:55 so, just to be sure folks are aware of the schedule again, go/no-go for Final is in 17 days, that's not long 15:33:01 so we *really* need to get all the validation tests run 15:33:16 please do use https://www.happyassassin.net/testcase_stats/29/ to find ones which have not been run and run them 15:33:46 in particular https://fedoraproject.org/wiki/QA:Testcase_desktop_menus is a big test that commonly turns up bugs and it has not yet been run in this cycle afaics 15:33:54 we really need that run for GNOME and KDE 15:34:24 #info we must make sure all validation tests are run ASAP to find all blockers in good time, please use https://www.happyassassin.net/testcase_stats/29/ to find not-yet-run tests and run them 15:35:01 adamw we have test day coming up on 5th 15:35:11 for Gnome.. will make sure to run them 15:35:30 thanks a lot 15:35:48 np np :) 15:36:00 sumantro: please make sure any showstoppers or near-showstoppers in default-installed apps get proposed as blockers 15:36:13 thanks :) 15:36:28 adamw rgr :) 15:37:53 anything else f29-y? 15:39:03 alrighty then 15:39:16 #topic Proposed release criteria discussion 15:39:26 so, once again we don't have a whole lot of people here...but at least some *different* people from last week :) 15:39:38 so i guess we could kick this around a bit more, then i'll try to get some concrete movement on the proposals this week 15:40:54 once again we have three outstanding criteria proposals: 1) stop requiring release-blocking ISOs to boot from physical optical discs, 2) block on printing in some way, 3) move firmware RAID criterion from Beta to Final 15:41:00 #info once again we have three outstanding criteria proposals: 1) stop requiring release-blocking ISOs to boot from physical optical discs, 2) block on printing in some way, 3) move firmware RAID criterion from Beta to Final 15:42:03 1) and 2) spawned quite a lot of discussion, 3) seems to have mostly got support, apart from one note that it may just result in us finding the issues later 15:42:23 anyone have thoughts on any of those? 15:42:55 re 1), that's probably not our thing to decide 15:43:10 but of course would be nice 15:43:50 well...the criteria process has always been sort of fuzzy 15:44:08 there isn't actually any very clear formal rule about who decides what the criteria should be, and how 15:44:26 in practice it's always been done just by starting a mailing list thread and making the change once there's a rough consensus 15:44:31 well, +1 from me 15:44:47 * satellit is this because of size requirements for a DVD 1) ? 15:44:47 i've never proposed we change this because the fuzziness actually seems to work quite well :P 15:45:02 satellit: no, it is mainly about the inconvenience of testing combined with a belief that it's just not a common use case any more 15:45:11 k 15:45:21 re 2), makes sense, as long as it's completely broken and not just for a particular printer type/family 15:45:26 we have quite a lot of release blocking ISO media, and technically we are supposed to test each both in BIOS and UEFI, which doubles the number of tests required 15:45:38 and it takes time to burn the media and it takes quite a while to actually run an install from them 15:45:47 and you need to actually have access to a burner, which is becoming less common 15:45:53 re 3), shrug. It's just a bit of shuffling. I'd rather see the criterion gone :P 15:46:05 kparal: the discussion about 2) has mostly been about the detail of exactly what to block on, indeed 15:47:04 and testing the optical boot stuff just in one not all? Testing would be much easyer and those who complain on the mailing will have a tested booting iso so they can continue burning dvds from their ancient coffee mahcines 15:48:33 that's a possible compromise, yeah 15:48:36 the boot thing will block just on netinstall for example 15:49:35 or just say 'we need at least one real-media test on UEFI and one on BIOS'... 15:49:40 might be worth proposing 15:50:41 well we have 2 images under that criterion so it's not much of a difference 15:50:59 but yes, 1 is better than 2 :) 15:51:17 no, we have more than that. 15:51:35 oh, no, you're right. huh 15:51:49 the main complains against this proposal are those who owns ancient machines, testing UEFI worth on this case? they'll be all using BIOS :D 15:52:19 didn't realize server didn't consider optical boot of the DVD to be blocking 15:52:34 and the KDE live... 15:52:43 jlanda: also a reasonable point 15:52:54 is there any uefi machine that can't boot from native usb? 15:53:01 also, that page clearly needs updating for modularity. :P 15:53:13 jlanda: i don't think that's possible. 15:53:22 (though firmware engineers never fail to surprise me.) 15:53:32 https://www.happyassassin.net/2013/05/03/a-day-in-the-life-of-a-firmware-engineer/ 15:53:50 ok, we're running low on time, so... 15:53:57 #info some more discussion of proposed criteria 15:54:13 #action adamw to try and move along outstanding criteria proposals towards resolution 15:54:16 #topic Test Day status 15:54:19 what's coming up, sumantro? 15:55:17 So, we have the regular test day schedule 15:55:31 only addition is Modularity on 12th Oct 15:55:53 Currently the Atomic and Cloud Test Day is underway 15:56:18 and DNF is on Wednesday? 15:56:30 Dnf will follow day after tomorrow and followed by Gnome a day after DNF test day 15:57:56 adamw yep! 15:58:02 alrighty 15:58:15 #info Atomic/Cloud Test Day is underway in #fedora-test-day right now 15:59:02 sumantro: isn't GNOME on Friday 10-05, not Thursday 10-04 ? 15:59:53 Gnome test day is on 5th 16:00:05 DNF is on 3rd 16:00:18 right 16:00:33 #info DNF Test Day is on Wednesday 10-03, GNOME is on Friday 10-05 16:00:35 be there or be square! 16:00:44 blocker review meeting is about to start in #fedora-blocker-review, btw 16:00:48 #topic Open floor 16:00:54 once again, i fail at timing! 16:00:56 any other business? 16:01:36 nothing from my end :) 16:02:24 sumantro: oh, DNF test day seems to be missing frm the calender? 16:02:26 calendar* 16:03:14 * sumantro is sorry.... the test day got confirmed on friday ..then I slept :P 16:03:21 adding it rightaway 16:05:33 adamw added 16:06:33 =) 16:06:34 thanks 16:06:37 alrighty, guess that's all 16:06:40 thanks for coming, everyone! 16:07:14 #endmeeting