16:00:35 #startmeeting Fedora QA meeting 16:00:35 Meeting started Mon Nov 21 16:00:35 2016 UTC. The chair is adamw. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:35 Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:00:35 The meeting name has been set to 'fedora_qa_meeting' 16:00:39 #meetingname fedora-qa 16:00:39 The meeting name has been set to 'fedora-qa' 16:00:42 #topic Roll call 16:00:47 morning folks - who's around for a QA meeting? 16:01:02 * Sumantrom is here 16:01:08 * a2batic is here 16:01:10 * satellit listening 16:01:13 maybe 16:01:38 * tflink is here 16:01:42 * kparal is here 16:02:38 * kparal pokes pschindl 16:02:57 * coremodule is here. 16:03:19 * pschindl is here 16:04:22 * garretraziel is lurking around, cleaning 16:04:41 morning everyone 16:04:44 how are you all today? 16:04:52 #chair sumantrom pschindl 16:04:52 Current chairs: adamw pschindl sumantrom 16:04:56 Morning adamw ! 16:05:41 morning adamw :) 16:06:24 #topic Previous meeting follow-up 16:06:30 nice and light here, just a couple of action items for me: 16:06:53 #info "adamw to come up with revised version of installation validation test change proposal this week" - that was the combining of USB and 'default install' tests, we got that revised and implemented 16:07:29 #info "adamw and sumantro to discuss state of Wayland with hybrid graphics systems with hansg, desktop team etc. to see if we need to worry about it for F25 release" - well, we had a test day on it and we're releasing it tomorrow :) everyone was pretty happy with the current state I think 16:07:53 there aren't any worrying cases where we know people will have a bad experience out of the box and have to switch to X manually or anything (besides isolated bugs) 16:08:21 * satellit add place to say how USB tests were done dd gnome-software or mediawriter? 16:09:04 satellit: yeah, we do need to take another look at the usb stuff 16:09:31 also f25 updates for mediawriter? 16:09:41 i think we had a rough idea to add a new table for usb tool coverage 16:10:03 kparal: should one of us take an action item for that? 16:10:45 I thought we agreed we expect people using mediawriter, but we also expect that from time to time someone will use dd or something else dd-based and therefore we're covered 16:11:10 kparal: well, we don't have anywhere to indicate status of mediawriter on f23/f24/f25 (for this cycle) 16:11:27 ruhroh 16:11:44 i.e. to make sure we've tested on all supported releasesi 16:11:49 the current table is focused on whether 'the image boots' 16:12:20 we could add "mediawriter tested: [ blank cell ]", yes, and also "dd tested: [ blank cell ]" 16:12:37 is that what you think? 16:12:46 i was thinking of a little extra table with columns as fedora releases (maybe also os x / windows) and rows as tools, mediawriter definitely blocking, the other rows can be livecd-iso-to-disk and anything else we want to test, maybe as Optional 16:13:08 so in the end we're back to two tables, but with a different focus 16:13:20 also mac or windows mediawriter? 16:13:25 ...i wrote that. 16:13:44 * adamw should look back at the old table to check if we're going in circles here. 16:13:47 I'm not completely clear on your idea, but please submit a draft :) 16:13:51 anyhow...seems like we should look into it 16:13:56 and i draw the short straw 16:13:57 *sigh* 16:13:58 it sounds sane in general 16:14:07 * adamw throws it on the giant burning tire dump that is his todo list 16:14:22 well I don't mind doing it 16:14:24 #action adamw to look into USB writing tool coverage in the validation matrices 16:14:29 #undo 16:14:29 Removing item from minutes: ACTION by adamw at 16:14:24 : adamw to look into USB writing tool coverage in the validation matrices 16:14:32 :) 16:14:37 #action adamw to look into USB writing tool coverage in the validation matrices, draft something up 16:14:46 nah, i can do it when i'm sick of perl for the afternoon or something. 16:15:04 aaanyhoo, that brings us to: 16:15:18 I plan to submit a proposal for reducing optical media coverage 16:15:21 which is somewhat related 16:15:32 yeah, i think you mentioned that 16:15:52 #topic Fedora 25 final check-in and retrospective setup 16:16:07 so, first of all, big thanks to everyone for all their work on Fedora 25 testing! 16:17:35 we missed a couple of things but overall we had great coverage and managed to help only slip the release by a couple of weeks while keeping the standard high, great job everyone 16:17:52 #info thanks to all who worked on validation, updates and test day testing for Fedora 25! 16:18:03 we had a good test day cycle also, and it's awesome that we're shipping with Wayland now. 16:19:06 * kparal sends out his thanks as well 16:19:42 * Sumantrom a big shout out to all the new testers! 16:19:45 yay! 16:19:49 indeed 16:19:54 #info particular thanks to all the new testers 16:20:24 so in terms of cleaning up f25 loose ends...we have the common bugs to write 16:20:38 kparal: have you done anything on that yet? i know you've been working on it for the last few milestones, which is awesome 16:20:45 done 16:20:49 yaay 16:20:54 #info kparal is the best 16:21:02 * kparal spent the day with common bugs 16:21:42 we also haven't found the obligatory brown paper bag bug yet (1170803 doesn't quite count), which is worrying, but never mind :P 16:22:48 we'll see about https://fedoraproject.org/wiki/Common_F25_bugs#wayland-frozen-login-after-upgrade 16:23:00 * satellit note: https://getfedora.org/fmw/FedoraMediaWriter-osx-3.97.1-17-g480c143.dmg not updated to 4.0.4 yet 16:23:28 kparal: btw, i have a wiki twitch: we can argue it if you disagree :), but I like using {{code|somecodehre}} over somecodehere on the wiki 16:23:47 adamw: is there a difference? 16:23:50 kparal: yeah, that might be it 16:24:13 kparal: well, wiki templates are a bit more amenable to being tweaked / migrated in future, i think 16:24:17 the difference for me is that I don't remember the second syntax :) 16:24:18 it may just be a prejudice 16:24:20 heh :) 16:24:37 there are way too many templates and no useful list of them anywhere 16:24:39 somehow i just hate seeing html tags in wiki pages and want to set them on fire whenever i do. i should probably see an analyst. 16:24:47 sure there is, it's in my head 16:25:19 here, take this hacksaw and have a look 16:25:27 in that case you need to plug in an ethernet cable and run a web server in your head 16:25:56 well, i tried, but it seems my brain was written in terrible php and i kept getting infected with ads for porn websites. it was rather uncomfortable 16:26:20 we can discuss this later. but we should probably contact someone about the outdated mediawriter on our websites? 16:26:25 =) 16:26:57 yeah, that does seem important, i was gonna bring up mediawriter status but satellit beat me to it 16:26:57 let's see, who to poke 16:27:03 mattdm: ahoy? 16:27:52 also, 4.0.4 are not stable for f23 and f24 yet 16:27:56 we should up-karma those updates 16:28:32 https://bodhi.fedoraproject.org/updates/FEDORA-2016-9659f7b138 (F23) needs one more karma point, https://bodhi.fedoraproject.org/updates/FEDORA-2016-1a54b4389a (F24) needs two 16:28:34 sigh, yep 16:28:43 #info we need folks to test and karma the F23 and F24 mediawriter updates 16:30:03 I will do it by tomorrow :) 16:30:54 thanks sumantro 16:31:00 i'll do f24 as well, and that should be enough to hit +3 on both 16:31:04 more testing always welcome, though! 16:31:30 I have 4.0.0 here on f24...need to test updates 16:34:36 satellit: yeah, you can get 4.0.4 from updates-testing 16:34:37 okay, moving on! 16:34:52 well, does anyone have anything else we need to clean up for f25, first? 16:35:29 nope 16:36:04 After release tomorrow, the HoF 25 Final post will be submitted to CommOps and posted to the Fedora Blog. A good ETA is by the start of next week. 16:36:43 coremodule, i have a question on that 16:36:44 awesome 16:36:45 Coremodule awesome! :) 16:37:01 jwboyer, Shoot! 16:37:17 #action adamw to find someone to poke to make sure the mediawriter downloads for os x and windows on getfedora.org are up to date for release 16:37:24 coremodule, i honestly have no idea, so please forgive me. does the HoF take into account alternative architectures? 16:38:02 It should, so far as validation testing and Bodhi updates are concerned. adamw, can you confirm here? 16:38:18 i don't believe secondary arches (or whatever we call them now) go through bodhi for updates, do they? 16:38:24 they don't 16:38:27 we also do not have validation testing for secondary updates 16:38:27 for wiki matrices, we only have x86_64 and i686 tables available, but it should count both 16:38:30 so for those two things, no 16:38:34 kparal: also arm. 16:38:36 that's kind of a shame to be honest 16:38:41 for bugzilla it should count everything I think 16:38:46 right 16:39:00 adamw: yes, and armhfp 16:39:05 there are fewer number of people working on alternative architectures, but they're still being tested and shipped at the same time 16:39:10 the issue isn't really the stats scripts, it's more just that two of the three things it's covering don't involve secondary arches in the first place. 16:39:10 what would it take to add them? 16:39:38 well, it'd require there to be secondary arches in bodhi and the release validation system. 16:39:59 what would it take to add them? 16:39:59 those stats are basically QA stats, and since we most of our work on primary arches, it's mainly primary stats 16:40:20 there has been some interest in the past about having something like the release validation system for secondary arches, but it'd be a lot of work (oh god, i'm already dreading the wikitext) and i've never gotten around to it. we could perhaps include it in the 'release validation ng' planning. 16:40:35 that would be nice 16:40:46 it'd probably be easier, practically speaking, to just write some new stats queries which somehow captured secondary arch activity 16:40:53 but what is there we could capture besides bugzilla? 16:41:09 is there secondary arch testing activity going on somewhere else which we *could* easily query, but aren't? 16:41:18 would it be possible to do something in the interim? perhaps look at the wiki pages for the alternative architectures for validation testing and add the people that filled in the most results to the HoF? 16:41:25 jwb: where are those pages? 16:41:42 * jwboyer digs 16:41:42 jwboyer: it'll depend a lot on how consistently formatted and hence parseable they are 16:41:51 the release validation pages are in a pretty rigid format 16:42:15 if the pages consistently use the {{result}} template for e.g. it should be possible 16:42:40 or at a bare minimum, [[User:]] wikilinks 16:42:52 oh damn, I have a blocker for F24 mediawriter - the custom image file open dialog doesn't work at all 16:43:04 kparal: yaay 16:43:08 I'll try downgrading 16:43:31 jwb: the pages would also have to be discoverable, which means being consistently named or consistently in predictable categories 16:44:02 they are, i'm just having trouble finding them (ironically) 16:44:16 heh 16:44:35 kaparal works in 4.0.0 f24 here 16:44:40 oh, and while we're on this - i'll be opening up a Retrospective page again for Fedora 25, trying to get back into that system 16:45:22 for new folks, we usually have a wiki page after each release which is intended to let people write down things that they thought went well and things that they thought we could improve on, from the last cycle 16:45:32 then we can use that for planning improvements for the next cycle 16:45:34 (sorry, in meeting; reads scrollback) 16:45:41 i'll put that up this week and send out a mail about it 16:45:57 (or maybe i'll ask sumantro or someone to do it and help them with it) 16:46:05 #action adamw to put up Retrospective page, or find someone else to do it 16:46:06 adamw: what was that ahoy for? 16:46:21 mattdm: satellit says the OS X version of mediawriter on getfedora.org is outdated (3.97 vs. 4.04) 16:46:41 mattdm: so i was gonna ping you to check that that's all lined up for tomorrow, for windows as well as os x 16:47:35 adamw: ok. I guess that's for websites team / robyduck ... I'll see what I can do 16:47:40 thanks a lot 16:47:46 sorry to use you as a switchboard :) 16:47:47 adamw, https://fedoraproject.org/wiki/Architectures/AArch64/QA/Fedora-25-20161107.n.0 16:47:51 is an example 16:47:56 adamw: np it's what I'm here for :) 16:48:01 adamw, https://fedoraproject.org/wiki/Category:ARM_QA being the category 16:48:40 hmm, that's interesting, so they're using tweaked versions of the matrices...i'll have to talk to pwhalen about how that works 16:48:45 satellit: yes, 4.0.0 works, 4.0.4 does not 16:48:52 but yes, we could definitely get the scripts to parse that. somehow or other. 16:48:55 * kparal files a bug and gives -1 karma 16:48:55 ouch 16:48:57 adamw, right. i think there's something similar for power and s390x 16:49:05 adamw, satellit -- it looks like 4.0.4 at https://stg.getfedora.org/en/workstation/download/ 16:49:07 what am I missing? 16:49:34 adamw, anyway, i just know there are a number of people working hard on it. it seems a shame not to include them in the HoF considering they're the ones making sure the Alt arches ship right along with primary 16:49:42 jwb: for sure 16:49:56 adamw, we can follow up offline to 16:49:56 unfortunately that's likely one bullet i can't dodge as no-one else has a damn clue how wikitcms works 16:49:58 er, too 16:50:07 * jwboyer needs to run now 16:50:17 thanks for the time and sorry to interrupt your meeting 16:50:17 #action adamw to talk to jwboyer and pwhalen about how secondary arch release validation runs and capturing the results in the HoF stats 16:50:25 jwboyer: not an interruption at all, thanks :) 16:50:41 mattdm: ah, great - satellit may have been checking production not stg 16:51:11 * satellit thanks 16:51:31 alright, so we're a bit behind now, i wasn't expecting this to take as long... 16:51:52 the next topic is fedora 26 initial plans, but maybe we can punt that to next week 16:52:04 or does anyone have something on that topic they really want to discuss this week? 16:52:20 we have some mockups for 'release validation ng' from design team which is great, we'll try and move forward on planning for that this week 16:55:36 sounds like a 'no' 16:55:41 Hold 16:55:45 holding! 16:55:50 (Sorry, just wandered through and have a thought) 16:55:59 oh god, brace yourselves everyone 16:55:59 :P 16:56:13 This may not be strictly QA's responsibility, but since it usually causes you grief, maybe it's the right place. 16:56:18 Here is the ticket -> https://fedorahosted.org/design-team/ticket/483 16:56:28 Every cycle, we have the same problem RE: wallpapers and last-minute branding. 16:56:39 Perhaps we should REALLY aim to solve this before F26 Alpha this time around? 16:57:39 sgallagh: i wrote a mean ticket about that. my work here is done. :P 16:57:57 OK, I just figured I'd bring it up. 16:58:05 sgallagh: but my take here is, it shouldn't be in any way QA's problem (except checking that it all happened, which *ought* to be a boring checkbox item each cycle, not a 'oh god disaster' moment.) 16:58:14 Agreed 16:58:37 and i guess providing karma on the update for final. 16:59:07 #info we'll table the 'f26 plans' topic for next week, as we're running out of time 16:59:11 let's have a quick: 16:59:14 #topic Open floor 16:59:18 any other pressing business? 17:01:10 sounds like a no also! 17:01:18 thanks again for the f25 testing, everyone, and for coming out this morning 17:01:29 let's hope we have a good release day tomorrow :) 17:01:41 please do blog / tweet / g+ / etc etc about the release! 17:01:52 there'll be an official announcement you can link to quite early in the day i think 17:02:34 #endmeeting