15:01:15 #startmeeting Fedora QA meeting 15:01:15 Meeting started Mon Jul 7 15:01:15 2014 UTC. The chair is adamw. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:01:15 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:01:18 #meetingname fedora-qa 15:01:18 The meeting name has been set to 'fedora-qa' 15:01:21 #topic Roll call 15:01:25 ahoyhoy folks, who's around? 15:01:29 * roshi is here 15:01:30 o/ 15:01:35 * satellit listening 15:02:00 * tflink 15:02:16 * brunowolff is here 15:03:20 here o/ 15:03:22 * kparal is here 15:04:23 * cmurf is lurking 15:04:26 pschindl and mkrizek send word that they won't come today 15:04:51 * pwhalen is here 15:05:42 kparal: did they decide any independence day is worth a few beers? :) 15:05:44 * handsome_pirate is nomming tasty bacon 15:06:20 * danofsatx-work is here, surprisingly 15:06:50 adamw: their today's priority was football (/soccer) I heard 15:07:35 there is apparently some red hat organized football team here in brno 15:08:01 oh right, go team! 15:08:07 * adamw cheerleads 15:08:42 * handsome_pirate pictures adamw in a cheerleader outfit and decides he's not drunk enough 15:09:03 You or adamw? 15:09:10 both 15:11:15 handsome_pirate: i'll have you know my thighs are the envy of the squad 15:11:22 and i never cheerlead sober 15:11:30 heh 15:11:43 allllrighty then 15:11:45 adamw: Now, imagine *me* in a cheerleader outfit 15:11:51 #chair tflink brunowolff 15:11:51 Current chairs: adamw brunowolff tflink 15:12:04 handsome_pirate: oh, i've always pictured you more in a sailor fuku 15:12:12 lol 15:13:07 #topic Fedora 21 test planning status 15:14:00 so, in case anyone was unaware, we are supposed to do F21 Alpha TC1 tomorrow 15:14:08 yeah, that kinda crept up on us 15:14:18 ouch 15:14:19 * kparal is surprised and unaware 15:14:23 #info tomorrow (2014-07-08) is the date for branching and Alpha TC1 15:14:37 it's been on http://fedorapeople.org/groups/schedule/f-21/f-21-quality-tasks.html all along ;) 15:14:43 * satellit no boot.iso or lives atm ? 15:14:54 satellit: Not yet 15:15:26 so, i think everyone's really a bit behind on this 15:15:50 I think you'd be right on that :) 15:16:08 which makes the alpha release date that much more strange, in my mind 15:16:13 i asked dgilmore for releng's position; he says there isn't an agreed list of deliverables (between fesco/the WGs/releng) yet, and koji does not yet have the ability to build Docker base images, which are apparently desired for F21 15:17:08 of the products, Server hasn't really brought any product-specific change to the point where it'd actually show up in a spin yet, Workstation is probably the best positioned as they can plausibly throw the 'workstation' name on whatever pops out of the workstation kickstart, and i'm not sure of the position of cloud...roshi? 15:17:47 it's pretty cloudy :p 15:17:49 aren't we still waiting on server deliverables before it can be built? 15:18:14 * tflink doesn't quite grok the roles bit that got an extension on the feature change freeze 15:18:18 tflink: this is fedora! no-one really knows! 15:18:29 aiui, cloud is still waaiting on the docker builds 15:19:03 So, in other words, this is going to be interesting 15:19:04 but the freeze/branch hasn't been discussed too much 15:19:07 tflink: i mean, we can spin something with no server images and call it 'alpha tc1'. should we? will we? iunno. 15:19:39 Branch is tomorrow. 15:19:40 yeah, I'm not quite sure what's going on, either 15:19:46 tflink: the server role API is the big new thing for Server. change freeze deadline is independent of alpha tc1 date, i believe - one got moved, not the other 15:19:47 it should be workstation-alpha-tc1 15:19:53 cloud WG meetings have had a hard time getting enough people the last couple times 15:19:58 danofsatx: workstation-and-cloud-tc1? :P 15:20:13 Except docker 15:20:42 adamw: are we going with separate test matrices and RC? 15:20:51 we have three products for F21. workstation-alpha-tc1, cloud-alpha-tc1, server-alpha-tc1 - iow, it should be three separate releases. 15:20:54 * roshi doesn't recall that being Decided (TM) 15:22:06 it probably hasn't been, but that's my take on the subject 15:22:08 anyway, i just wanted to call attention to the general lack of readiness for context and information 15:22:10 our job is QA, so... 15:22:20 roshi: that is indeed a thing we should decide :P 15:22:22 what's QA again? 15:22:24 :p 15:22:30 How is testing fedup going to work for this release? 15:22:38 Quadbike Adventuring. let's say it's Quadbike Adventuring. 15:22:42 I was just going to ask that brunowolff 15:22:42 * adamw roars off into the distance 15:22:48 adamw: +1 15:22:52 +1 15:22:59 * roshi is good with adventuring 15:23:20 OK, so, i should really start making a list of Good Questions We Can't Answer Ourselves 15:23:21 * handsome_pirate is a hipster in his giant redneck truck 15:23:31 #info GQWCAA: how is fedup supposed to work for F21? 15:23:45 oh good lord, i can't spell. 15:23:46 #undo 15:23:46 Removing item from minutes: INFO by adamw at 15:23:31 : GQWCAA: how is fedup supposed to work for F21? 15:23:52 handsome_pirate: there's a bit of cognitive dissonance there with "redneck hipster" 15:23:52 #info GQWCAO: how is fedup supposed to work for F21? 15:24:00 There's also a lot of discussion on moving from one product to another 15:24:06 what's life without some cognitive dissonance? 15:24:16 roshi: I wouldn't know. 15:24:17 ie, if you install Server and want to move to Workstation 15:24:24 me either :) 15:24:31 handsome_pirate: IIRC, that's not supported for now 15:24:42 Aye 15:24:47 and it's unclear if that would ever be supported 15:24:54 but I could be mis-remembering 15:24:58 #info GQWCAO: Are the Products separate 'streams' for releng/testing/delivery purposes, or must they be built together? 15:25:00 yeah - I was thinking it was an F22+ thing to think about/discuss 15:25:47 release blocking products are what, all products? 15:26:12 From what I understand 15:27:04 I fear cloud won't have enough people testing it (at least enough for solid validation) 15:27:21 business as usual, unfortunately 15:27:29 yeah 15:27:47 It just so happens that I just sold off the hardware that I'd have used to build a private cloud 15:27:52 I don't have a cloud instance to test it on, or I would. 15:27:58 there haven't been a lot of answers to my questions or calls for more test cases to be written by cloudy people 15:28:14 Also business as usual 15:28:19 i believe there is a test fedora openstack instance you can run things on, which releng can grant access to (and will, for testing purposes) 15:28:27 I'm working on some documentation for running the images standalone locally with qemu 15:28:34 in fact, my free year of AWS is set to expire soon :( 15:28:34 so people can test the images themselves 15:28:43 roshi: I can help with that, if you'd like 15:28:59 I'm down for all the help I can get :) 15:29:10 Bang heads after the meeting? 15:29:16 sure 15:30:11 so in terms of actual validation testing...i'd say what we need at minimum is a plausible set of release criteria for each product (we can bear with some messy layout for a while) and decent test case coverage for them 15:30:43 of course, i sent out that nice 'fedora 21 test plan' saying the WGs should be involved in building all that out and everyone nodded and said 'sure, that sounds good' and none of them actually *did* anything 15:30:49 somewhere, viking-ice will be reading this log and smirking 15:30:58 lol 15:31:09 for sure 15:31:31 so Workstation again probably involves the least work over Fedora.last 15:31:39 (i just invented that, see what I did there?) 15:32:14 #info adamw threw together Workstation-specific criteria and new test cases on Friday: https://lists.fedoraproject.org/pipermail/desktop/2014-July/009985.html 15:32:52 #info Server-specific criteria draft is at https://fedoraproject.org/wiki/User:Adamwill/Draft_server_release_criteria (still), new test cases have not yet been written 15:33:11 note though that none of the server stuff has actually been *built* yet, so it's kind of a dead letter at this point in time 15:33:23 roshi: what's the status on cloud criteria / test cases? 15:33:36 I'll work today to get my cloud stuff to roughly match yours 15:34:01 well, largely the "yeah, that sounds good. Someone work on it." and then it's just me 15:34:19 is it possible/practical to have one set of criteria for all three produces, like "base" 15:34:29 and then each product has a list of unique criteria? 15:34:31 so - in leiu of someone stepping up and saying they aren't complete - no one has complained so they must be good enough 15:34:58 so the base matrix/criteria can be met with testing any of the products? 15:34:59 though I feel that might be a dangerous position to take :p 15:35:54 cmurf: that makes too much sense, so of course we won't be doing it that way. 15:36:37 cmurf: we still didn't really figure out the best presentation for the criteria for the products 15:36:44 cmurf: but as i see it that's priority #2 15:36:49 that also presumes that there's a significant similarity between all the products 15:36:54 priority #1 is at least making sure all the necessary criteria are actually *written* 15:37:54 that might mean we have a sort of chaotic mix of the Fedora.last criteria plus some pages with product-specific criteria on them, which is more or less what we have right now, but we would at least have everything actually on paper and could work with it. 15:38:34 wait, I have to write this *on paper?* 15:38:43 * handsome_pirate wonders how arm is going to fit in the criteria 15:38:56 cmurf: and yeah, so far as testing goes...my rough read on it so far is there will be some criteria that apply to all products, some that apply to only workstation and server, and some that apply to only a single product 15:38:59 roshi: hehe 15:39:18 if you tell me it has to be in cursive, you're going to have a bad time 15:39:32 submitted to central admin in triplicate 15:39:57 I have the handwriting of a doctor :) poor people in central admin 15:39:58 roshi: No, you have to print everything with capital letters 15:40:19 capital cursive 15:40:21 You also have to fill in all the little bubbles 15:40:24 roshi: as long as you follow all the rules, they're happy 15:40:37 ah, ok - good 15:40:58 because "technically correct" is the best kind of correct :) 15:41:03 haha 15:41:38 it's TECHNICAL! 15:41:44 and all responses should start with "Actually..." 15:42:03 Actually, I like bacon 15:42:26 roshi: do you have any test cases for the cloud criteria yet? 15:42:57 not wiki testcases 15:43:16 going to work on it today 15:43:37 since no one from cloud did it 15:44:06 roger, thanks 15:44:43 #info Cloud criteria drafts are at https://fedoraproject.org/wiki/User:Roshi/QA/Cloud_Docs/Cloud_Alpha_Release_Criteria , https://fedoraproject.org/wiki/User:Roshi/QA/Cloud_Docs/Cloud_Beta_Release_Criteria , https://fedoraproject.org/wiki/User:Roshi/QA/Cloud_Docs/Cloud_Final_Release_Criteria (in a different form to server/workstation drafts) 15:44:43 http://s1.tsuki-board.net/pics/figure/big/99523.jpg?t=1329556068 15:44:47 oops 15:45:01 handsome_pirate: are you playing reddit One Of These Things Is Not Like The Others again? 15:45:17 #info Cloud-specific test cases are not yet written, roshi will work on it today 15:45:34 adamw: Gel banana :) 15:46:11 handsome_pirate: If you have access to a time machine, I could really use a "gemini event" right about now... 15:46:24 #action adamw to look at how the Install, Base and Desktop matrices should be adjusted for Fedora.next 15:46:29 sgallagh++ 15:46:38 I just want a hyperbolic time chamber 15:46:44 waaay easier than a time machine 15:46:53 #action adamw to, uh, completely revise / write all the backing process documentation for Fedora.next testing, that'll be easy 15:47:39 since i'm a process nerd we actually have quite a lot of stuff like https://fedoraproject.org/wiki/QA:Desktop_validation_testing 15:47:50 all of it needs adjusting for .next, and we need some new stuff written. 15:48:24 * handsome_pirate wonders if we should just roll an old-style tc1 just to get something out 15:49:02 An Hey, does F21 at least install/boot type of image 15:49:06 we could do something hybrid-ish, like a Workstation live, cloud images, and a network install image 15:49:26 but as mentioned earlier...no-one's got releng, WGs and FESCo on the same page about that 15:49:41 i guess we'll just have to file a TC request with some suggestions and see what happens 15:49:49 oh, someone asked about ARM, that is another fun question 15:50:01 given that ARM stands at a slightly weird angle to the Products 15:50:32 i asked on the arm list a few weeks back if they're still expecting to ship their crop of platform-specific disk images for F21, if so what status those would have, and if they're planning to ship any Product-y images 15:51:16 * handsome_pirate admits to having fallen out of the arm thing for a while 15:51:47 * handsome_pirate will poke that ant hill a bit with a nice long stick, if y'all want 15:51:59 adamw, we discussed this at the meeting last week, and plan on having the same deliverable (minus vfat) 15:52:01 https://lists.fedoraproject.org/pipermail/arm/2014-May/007779.html was the ARM thread 15:52:10 pwhalen: same...as F20? 15:52:32 right, we would like to have a boot.iso type option, which is being looked at 15:52:43 masta may have an update 15:53:21 okay. well, that gives us a KDE-ish 'square peg in a round hole' issue, but at least it's easy from a test process perspective 15:54:05 so...i guess my meta-plan here is that i'm going to try and pull a plausible validation process which will be a bit rough around the edges out of my ass by tomorrow afternoon 15:54:24 i don't want to be a prima donna, but at this point it's a bit late to split up the work :( 15:54:35 true 15:54:36 (and as noted the WGs really didn't seem to want to do any of the lifting) 15:54:56 perhaps do an etherpad or something? 15:54:56 * handsome_pirate imagines Viking-Ice chuckling 15:54:58 hopefully they are busy engineering 15:55:00 roshi: are the cloud criteria 'drafts' pretty much what you actually want at this point? 15:55:05 have there been any updates on what's happening with the kde spin? 15:55:10 what I want? 15:55:28 roshi: well, i mean, is there any more work to do in terms of writing/tweaking the criteria or are they ready to go for alpha 15:55:30 whether it'll end up a product, part of workstation or a spin? 15:55:36 I just want something that'll do the job :) 15:55:47 tflink: i believe it was decided a while ago that it'll be a sort of special spin for f21 15:55:48 I just asked in cloud if they were good to go as is, and people stopped talking 15:56:00 basically it'll be the same as it was for f20: we'll have a KDE live which is release blocking 15:56:01 I asked mattdm 15:56:18 so again, that's going to be a fun one to try and message, but it kinda makes things a bit easier for us as it's just 'business as usual'. 15:56:49 pwhalen: sees his name called? 15:56:57 since no one took issue with them as I put them up - we're left to assume that they're good to go 15:57:03 masta: Ahoy 15:57:04 roshi: well unless i hear otherwise i'm just gonna go with those as the alpha cloud criteria and pull them in however seems best in a timeframe of, like, 30 hours... 15:57:06 pwhalen: am I on the hook for an iso image? 15:57:11 roshi: ayup 15:57:20 masta, we were talking deliverables - you were looking into a boot.iso option for arm 15:57:30 not sure if you had any updates 15:57:48 roshi: i could put up a kinda meta-etherpad with notes on the process, i guess, sure...mostly i usually keep that in my head but it might be nice 15:58:09 do we have a fedora etherpad? 15:58:14 nope 15:58:35 eh. i don't have one on happyassassin. 15:58:39 packaging issues, AFAIK 15:58:40 we do have a goby tho 15:58:56 https://fedoraproject.org/wiki/Gobby 15:59:18 adamw: I was thinking then people like me might be able to help you work on it in said 30 hours 15:59:26 true 15:59:31 gobby would work 16:00:25 roshi: you can certainly help - write cloud test cases! 16:00:30 that's actually probably the biggest thing we're missing right now 16:00:41 i don't care about missing server test cases so much since there's nothing to damn test 16:00:49 for sure 16:01:18 my top priority right now 16:01:21 we have workstation criteria and test cases (even if they're first draft), so the main thing now is cloud test cases and getting *some* kind of viable coherent set of matrices 16:01:43 cloud test cases are weird 16:01:59 heh 16:02:03 it'd be nice to split the load of that second bit, but i can't see a great way to do it because it's kinda something one person has to visualize and hack up really 16:02:10 on the wiki anyway - since we have a small subset of people who can do EC2 testing and whatnot 16:02:44 roshi: i'd write the test cases in Perfect Theoretical Testing Land, where there are no practical barriers to testing 16:02:44 * handsome_pirate raises his hand at that, btw 16:02:56 * handsome_pirate has amazon instances 16:02:56 then you can deal with the barriers at test plan level or by slapping on advisory boxes later 16:03:06 I asked the cloud people if local testing would suffice to say things are valid and never got an answer 16:03:14 yup 16:03:25 roshi: I'd say go for it 16:03:28 good to know handsome_pirate - we can link up after the meeting 16:03:33 roshi: it works for most things 16:03:40 yeah - if no response, that means yes :) 16:04:00 so...i think we mostly covered things? 16:04:06 the only ec2-specific stuff is making sure the instance boots into a usable state 16:04:07 lots of work to do, but we should be able to pull *something* out 16:04:17 yup 16:04:23 gonna be a long day methinks 16:06:01 so, we're over time, but just a quick: 16:06:04 #topic open floor 16:06:09 anyone have anything else to bring up? 16:06:14 oh 16:06:21 we should probably do a blocker review meeting wednesday 16:06:29 can we farm most of this out to mechanical turk or something? :p 16:06:41 anyone want to volunteer to run it? taskotron essential personnel do not qualify :P 16:06:41 +1 16:06:55 roshi: subcontract it to canonical 16:07:06 heh 16:07:30 payment in macaroons 16:08:50 so danofsatx-work, you're running the blocker review meeting? 16:09:03 or were you +1ing roshi? 16:09:22 * roshi hadn't considered that 16:09:24 lol 16:09:55 I was agreeing with roshi 16:10:13 * danofsatx-work isn't qualified to run a meeting 16:10:26 no no too late, dos-w is running blocker review 16:10:29 alright, so the plum job of blocker meeting moderator with a salary of $0 is still available 16:10:30 bah - what does "qualified" even mean :P 16:10:51 roshi: i think it's about meeting the required minimum level of alcohol consumption? 16:11:05 * cmurf is disqualified in that case 16:11:10 adamw: Well, you know I can do that :) 16:11:17 what's the minimum on that? 16:11:20 daily or weekly? 16:11:34 roshi: you're in colorado so that's two shots and a puff 16:11:41 I have this thing called a day job that tends to interfere. That's how I lost the election to the Server WG :( 16:12:06 zodbot can run the meeting 16:12:20 can zodbot write the release criteria too? 16:12:21 all hail zodbot! 16:12:28 what's the translation from puff to shot? 16:12:44 roshi: well urban dictionary tells me I definitely should have said toke rather than puff 16:12:44 because two shots isn't much... 16:12:57 haha 16:13:00 lol 16:13:23 * roshi wouldn't have known any better 16:13:26 puff is so 60s, man 16:14:32 urban dictionary has been raided by 60's Brits apparently 16:14:48 anyway roshi, it is a MINIMUM 16:15:12 roshi: i guess one of us gets to take the bullet 16:15:24 I believe the term is "Smoke a bowl" 16:15:34 I can do it 16:15:35 it's about attiude adjustment rather than personality distorting 16:15:52 I guess now I know I'm not "taskotron essential" :p 16:16:07 * roshi was looking forward to a badge 16:16:19 there's a badge for that? 16:16:29 I was hoping 16:16:40 * kparal looks up 16:16:44 badges? 16:16:45 * threebean looks up 16:16:50 badge/bear trap ... same thing? 16:16:55 yeah 16:17:03 bear trap is a badge with a better story :p 16:17:09 handsome_pirate: that would only apply to a pipe. 16:17:16 or, you know, pipe-like apparatus. 16:17:20 or, uh, so this guy told me once. 16:17:30 lol 16:17:48 we *really* need to build a fedora badge flashlight we can project on the night sky any time we need threebean. 16:17:57 lol 16:18:09 +1 to that 16:18:10 adamw: it looks like he has badge highlighted in irc :) 16:18:19 tflink: well SURE, but that's just too dull. 16:18:45 ah crud, i'm supposed to be playing tennis on wednesday morning. 16:18:51 threebean: So, we need a badge for running a blocker bugs meeting, a badge for taskotron, and a badge for smoking bowls 16:18:52 hum, blocker meeting on thursday? :P 16:18:54 Does that cover it? 16:19:09 if you like, I can run it either way 16:19:09 handsome_pirate: a badge for outsourcing to zodbot? 16:19:14 heh 16:19:22 #action roshi to run a blocker review meeting some time this week 16:19:40 aaand we're way over time, so i guess i'll set the quantum fuse 16:20:17 Explosions! 16:23:13 QUANTUM explosions! 16:23:16 thanks for coming, folks 16:23:19 #endmeeting