20:00:22 #startmeeting horizon 20:00:23 Meeting started Wed Sep 28 20:00:22 2016 UTC and is due to finish in 60 minutes. The chair is r1chardj0n3s. Information about MeetBot at http://wiki.debian.org/MeetBot. 20:00:24 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 20:00:27 The meeting name has been set to 'horizon' 20:00:29 o/ 20:00:31 oh, it worked 20:00:33 :-D 20:00:51 o/ 20:00:55 o/ 20:01:51 #topic Notices 20:01:58 ooh 20:02:15 so a couple of things y'all should know about 20:02:44 most importantly is that the integration tests have been made non-voting while we try to sort out the instability issues 20:03:10 sort out as in get rid of? 20:03:19 when I looked yesterday we had a very large number of patches stuck because of rechecking 20:03:34 o/ 20:03:34 david-lyle: well, that's a question we can discuss, yes 20:03:58 r1chardj0n3s: I think it's a summit topic if robcresswell knows his stuff 20:04:13 another gate change, though this one hasn't happened yet, is that we want to get the Django 1.10 tests voting 20:04:24 the patch for that just needs some tweaking https://review.openstack.org/#/c/378426 20:04:29 o/ 20:04:31 Its a proposed topic :) 20:04:33 david-lyle: oh it's definitely a topic :-) 20:04:50 * robcresswell knows his stuff 20:04:57 so that's the only other thing I wanted to mention this morning was that robcresswell knows his stuff 20:05:10 no, wait, I wanted to mention the summit session topics 20:05:11 \o/ 20:05:11 * david-lyle is not convinced 20:05:23 #link https://etherpad.openstack.org/p/horizon-ocata-summit 20:05:25 r1chardj0n3s: you should nominate him for ptl next time ;) 20:05:34 ducttape_: that's a fine idea!! 20:06:05 we have 5 working group sessions, and it looks like we're getting plenty in the etherpad to talk about 20:07:22 Sounds good 20:07:54 so, that's all the notices I have, we can talk about the summit sessions or even the integration tests 20:08:08 I had a wacky idea on that too 20:08:29 It seems like a lot of our issues come from dodgy IPC from Python through to the Selenium browser 20:08:53 maybe we should use Javascript in-browser tests... 20:09:12 hehe 20:09:14 but Timur's selenium expert might have something to say about that 20:09:22 So the suggestion is "rewrite it in JS" :p 20:09:36 Oh man! 20:09:52 yeah schipiga had some thoughts on rewrites previously 20:09:54 ¯\_(ツ)_/¯ 20:09:56 I don't know where that went 20:09:57 it's an option, especially if we want to otherwise radically change the scope of our integration tests 20:10:33 Yeah, I'm just poking fun. I'll be doing a bit of research around it and seeing what options we have, because atm its a huge pain 20:10:45 Basically held up RC2 by a week just hitting recheck. 20:11:30 yeah, and this seems to happen regularly too - every 6 months or so, something will change in the selenium/firefox mix and suddenly our test suite is unstable again 20:11:50 so alternatives to selenium itself should be considered, if there are any 20:12:03 Yeah, even 6 months is generous IMO. We;ve had total blocks on the gate several times this cycle. 20:12:23 don't limit it to this cycle 20:12:29 since their inception 20:12:43 yep, but we weren't really paying attention back then 20:12:46 :-P 20:12:55 uh, you weren't 20:12:56 that was some other guy's problem 20:13:03 the little monster has become an adult now 20:13:33 Is richard the little monster? I'm confused 20:13:42 LOL 20:13:50 I've set aside gasoline and a match when you're ready to make the call 20:13:50 integration tests 20:13:56 Ah I see 20:14:04 * ducttape_ is a big smoker 20:14:43 I'm glad my final move was to disable all non unit tests. 20:15:05 and then run out the building laughing. 20:15:10 ooh, there's an alternative to selenium written in the node.js programming language, we should use that 20:15:24 * david-lyle drops his stuff and walks out 20:15:29 I hope you've got a lot of gasoline david-lyle 20:15:42 hey, you lot voted for me!! 20:15:55 technically, no 20:15:58 LOL 20:16:03 :P 20:16:19 technically whatever :-P 20:16:38 I was part of the never david-lyle camp 20:16:45 me too 20:17:40 nodejs for stability 20:17:41 SO, on a positive note, RC2 was tagged today and dj10 should be voting soon 20:17:45 lolwut 20:17:45 So thats good. 20:17:45 the other elephant in the room: ditch the whole thing and switch to an iOS app. that'd solve the selenium issue 20:18:00 +2 20:18:06 and introduce 6 others 20:18:08 ducttape_: I think that was the proposal 20:18:10 oh, thanks robcresswell, I forgot to mention the RC2 tagging 20:18:18 I'm wondering how many summits we'll have to say no to the app idea 20:18:21 probably forever 20:18:39 I'm totally not stopping someone else from doing it ;-) 20:18:40 david-lyle: no I get it. more of "so we've come to this - that nodejs is the stable / mature adult in the room?" 20:18:55 yikes 20:18:56 r1chardj0n3s: agreed 20:19:04 hopefully one person out those will provide the motivation other than they can 20:19:24 random companies do have their own mobile apps 20:19:31 for openstack? 20:19:35 yes 20:19:41 you can produce all the iOS / android apps you want. no one is stoping that 20:19:47 I've never seen one in person, but I've heard about them 20:19:54 what are you doing in openstack that you want to use your phone for? 20:20:08 push notifications would be cool 20:20:16 suckering your management into letting you work on a mobile app? 20:20:19 oh! a wild tqtran appears! :-) 20:20:25 * david-lyle faints 20:20:27 catch him! 20:20:34 -_- omg im not a snorlax guys 20:20:39 he's kinda like a pokemon 20:20:51 ;) 20:21:37 Anyone have anything else they'd like to bring up? 20:21:42 My impression is that most people basically use emails from their customers as push notifications 20:21:54 "hey we have a bug" 20:22:01 I thought it was pager duty 20:22:13 icinga / pagerduty / monit / etc 20:23:50 ah okay 20:24:44 r1chardj0n3s: When do you need to finalise the summit sessions by? I don't know if you've had that email yet 20:25:05 Can't remember the dates from last cycle. 20:25:25 robcresswell: I've not had an email about that 20:25:36 oh okay well 20:25:39 *surprise!* 20:26:07 heh 20:26:22 I think there is likely another week or so 20:26:46 but voting or whatever mechanism you want to employ for choosing should happen relatively soon 20:27:08 yup 20:27:23 I was thinking about putting a proposal in email early next week to get feedback from attendees. 20:27:28 My internet is giving up -.- 20:27:41 r1chardj0n3s: sounds good. 20:28:12 OK, I think we're done. Thanks everyone! 20:28:19 would you like people to start voting/commenting or are we still just gathering? 20:28:30 oh 20:29:04 or it's your right to just pick too 20:29:29 rule with the iron fist 20:29:42 please do add your thoughts to the summit session topics etherpad, including voting (+1 / -1) on topics 20:29:49 that link again https://etherpad.openstack.org/p/horizon-ocata-summit 20:30:50 unless some other topic comes along, it does look like we've got about 5 rough topics already in the etherpad 20:31:14 so it'd be good to know what folks' interest is in the topics already proposed 20:32:41 so finally today, I'd like to thank robcresswell for his great work PTL'ing the Newton release 20:33:53 and with that, thanks for coming 20:33:59 #endmeeting