20:00:50 #startmeeting Octavia 20:00:51 Meeting started Wed Jan 20 20:00:50 2016 UTC and is due to finish in 60 minutes. The chair is johnsom_. Information about MeetBot at http://wiki.debian.org/MeetBot. 20:00:52 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 20:00:53 Howdy folks! 20:00:54 o/ 20:00:55 The meeting name has been set to 'octavia' 20:00:59 Hi there 20:01:02 Hi 20:01:08 #topic Announcements 20:01:18 O/ 20:01:24 o/ 20:01:29 o/ 20:01:38 dougwig you are missing out on a ton of redbull 20:01:38 So, last week we had a successful mid-cycle in San Antonio. 20:01:49 We have notes on the etherpad 20:01:57 #link https://etherpad.openstack.org/p/lbaas-mitaka-midcycle 20:01:59 Yep! It was great. 20:02:16 hi 20:02:25 This week is the Kosmos mid-cycle in Seattle. We just started today. 20:02:25 lots of BBQ 20:02:32 #link https://etherpad.openstack.org/p/kosmos_2016_winter_midcycle_meetup 20:02:46 Thank you to everyone the participated last week! 20:03:06 Out of that week we have a few high priority patches that need reviews: 20:03:18 Add spec for active-active: 20:03:28 #link https://review.openstack.org/234639 20:03:30 and 20:03:36 (Which specifically needs reviews from non-IBM folks) 20:03:39 Shared pools support: 20:03:43 Yay! 20:03:53 #link https://review.openstack.org/256369 20:03:58 Which needs *everyone* to review it. 20:04:00 :D 20:04:08 These are blocking work, so please take the time to review. 20:04:28 Any other reviews that folks consider high-priority? 20:04:44 Yes... um... 20:04:55 There's a corresponding neutron-lbaas shared pools patch as well. 20:05:02 o/ 20:05:04 hi 20:05:05 It's probably less priority than the Octavia patch. 20:05:07 Let me link it... 20:05:21 #link https://review.openstack.org/#/c/218560/ 20:05:23 Ok. 20:05:37 (It's only slightly less of a priority than the Octavia shared pools patch) 20:05:47 #topic Mitaka milestone 2 20:06:03 This week is the M2 week 20:06:09 o/ 20:06:12 #link http://docs.openstack.org/releases/schedules/mitaka.html 20:06:38 We did not cut an M1 release for Octavia anticipating the active/standby work to finish up. 20:06:56 My question for the team is do we want/need an M2 release cut? 20:07:05 I don't think so. Not yet, anyway. 20:07:23 active-standby is great and as far as I can tell, usable... so I might be wrong about thta. 20:07:25 that. 20:07:30 Any other comments/thoughts? 20:07:59 I only want to cut it if someone has a need/want. 20:08:17 Is a2hill the only RAXer here today? 20:08:28 Ok, so I'm not hearing a request for it, so I will move on... 20:08:33 apparently they need a week to recover form the midcycle 20:08:40 Heh! 20:08:40 blogan: is out sick i htink 20:08:44 and TrevorV is here now 20:08:45 #topic Discuss L7 feature (and review) timeline re: Mitaka (sbalukoff) 20:08:45 Yay, Trevor is here, too! 20:08:49 Ok! 20:08:50 sbalukoff: i just goined 20:08:51 and other two are on internal stuff 20:08:53 joined 20:08:56 Yeah, sorry, my client was bein crazy 20:08:59 Oh, good! 20:09:01 is rm_work somewhere? 20:09:04 No worries... 20:09:16 xgerman he's been on internal stuff mostly of late 20:09:33 Ok, so! If I understand the mitaka timeline correctly, we have a little over 5 weeks until feature freeze, correct? 20:09:34 oh, ok, I had some Barbican questions... 20:09:34 just got back from vacation - getting caught up with stuffs 20:09:36 yea, fun times :/ 20:09:50 fanaal we missed you at the mid cycle 20:09:54 fnaval 20:10:02 Week of Feb 29th FF 20:10:04 xgerman: yes - i see lots of tempest related stuff to do 20:10:07 from the etherpad 20:10:09 xgerman you'll have to ping him later. He should be online this afternoon/evening 20:10:23 If that's the case, I think that adding L7 support is doable, but it's a fairly aggressive timeline, given the volume of code this is going to entail. 20:11:02 So! Keeping in mind that I'm probably going to be writing most of the Octavia-side code and will be working with Evgeny on the neutron-lbaas side code... 20:11:08 Yes, that seems aggressive 20:11:15 Are people still interested in pushing toward getting L7 in by mitaka? 20:11:20 (I'm hoping the answer is yes...) 20:11:31 Yes, I would really like to see L7 in mitaka 20:11:35 TravorV thanks 20:11:36 I was under the impression that wasn't on the plate, but I'm not AGAINST it 20:11:36 Ok, so.... 20:11:40 We are for it, just runnin short on resources and have lots of internal things to attend to 20:11:42 Yes, we are 20:11:58 I wrote something up about this thta I'd like to copy-paste in here... let me see if the IRC server boots me as a result... 20:12:14 Following are ready for review / (hopefully close to) merging. Note that these make up the entirety of the shared pools functionality. Shared pools touches a *lot* of things in the code, so these changes are likely to conflict with other in-develoment features or bugfixes (meaning, the sooner we can get it merged, the better): 20:12:14 1. Octavia shared pools: 20:12:14 #link https://review.openstack.org/#/c/256369/ 20:12:14 2. Neutron LBaaS independent pools: 20:12:15 what about a gist? 20:12:15 #link https://review.openstack.org/#/c/218560/ 20:12:17 3. neutron client adding shared pools CLI support: 20:12:19 #link https://review.openstack.org/#/c/218563/ 20:12:25 We can always hope... 20:12:29 xgerman: I'd like to get this in the minutes more directly. 20:12:34 But I'll gist it later as well. 20:12:35 k 20:12:43 Let's setup an etherpad for this. 20:12:46 Following are WIP, but depend on the above patches. Also, the remainder of this list is purely L7 support, which is almost entirely "additive" and should not conflict with other patches in progress or bugfixes. Most of these patches will need to be merged at about the same time for full L7 support: 20:12:46 1. Octavia L7 schema update: 20:12:46 #link https://review.openstack.org/#/c/265430/ 20:12:46 2. Octavia L7 repo updates: 20:12:47 #link https://review.openstack.org/#/c/265529/ 20:12:49 3. Octavia L7 API update: 20:12:51 #link https://review.openstack.org/#/c/265690/ 20:12:53 4. Octavia L7 API tests: 20:12:55 #link https://review.openstack.org/#/c/265718/ 20:12:57 5. Neutron LBaaS L7 support (by Evgeny Fedoruk): 20:12:59 #link https://review.openstack.org/#/c/148232/ 20:13:01 johnsom_: Ok I will! 20:13:07 Beyond the above, the following are "in progress" but have yet to have anything committed. But these are also necessary for L7 support, and given the directive to try to keep patches < 750 lines, will probably end up in separate patch sets (all chained together--- yay!) 20:13:07 * Additional L7 Rule sanity checking 20:13:11 * Tests for the L7 Rules 20:13:13 * Jinja template updates 20:13:15 * Tests for jinja template updates 20:13:17 * Updates to neutronclient CLI for L7 manipulation 20:13:19 * Documentation updates 20:13:21 WHAT I NEED FROM Y'ALL: 20:13:23 * I can write the code, but I can't independently review or merge it. With only 6 weeks until the feature freeze, and a LOT to do yet for L7... and since it's such a pain in the ass to have to keep on rebasing the root patch set (Octavia shared pools and Neutron LBaaS independent pools) whenever even minor changes happen, the best way that you can help make sure L7 lands in Mitaka is to review the shared pools fu 20:13:28 nctionality ASAP and get that merged. I will be working as fast as I can after I get feedback to fix any problems you may find so we can get these patch sets merged ASAP. But I need your feedback in order to do that! 20:13:32 * The rest of the L7 stuff can lag a bit because it will mostly not conflict with anything being developed in parallel, and will all have to be merged close to the same time. 20:13:35 Woo-hoo! I didn't get kicked! 20:13:36 johnsom_: I am curious about your patch for failover updates, i need to refactor some stuff once that gets in, so i have an eye on it 20:14:11 So... that's a lot to digest all at once. I'll get it added to an etherpad after this meeting. 20:14:33 But mostly, I just wanted to emphasize, in all seriousness, that I just really need reviews. 20:14:42 a2hill it should be good for review. I have mostly just been rebasing it. So, get folks to review it and we can land it 20:14:48 I think we can absolutely deliver L7 in mitaka, but ONLY if people are reviewing. 20:14:51 sbalukoff one might say you should have linked an etherpad to this information instead of splatting it here anyway :P But I read it all. I'll do what I can for reviews 20:15:07 It was waiting on the taskflow fix that has now been released 20:15:11 TrevorV: Yeah, I know. 20:15:15 And thanks! 20:15:26 johnsom_ we have to talk about taskflow. It doesn't quite look like its rolling-back correctly 20:15:26 indeed, good deal johnsom_ Looks like some reviewing is in order 20:15:46 Please setup the etherpad. You can create an empty one now and get the link in the minutes.... 20:16:19 johnsom_: Ok... 20:16:34 If you want to move on with the agenda, I can link it during open discussion. 20:16:46 And thank you all for your patience... and reviews! 20:16:47 Ok, yes, shared pools is my next monster review to do, then I will hit active/standby. 20:17:07 TrevorV let's talk about that in open discussion 20:17:12 sounds good 20:17:25 #topic Brief progress reports 20:17:35 Ooh, do we get another task flow holy war? 20:17:42 Good stuff at the mid-cycle. We closed some bugs 20:17:59 * johnsom_ Throws full redbull cans at dougwig 20:18:12 No, it is likely a bug 20:18:42 dougwig: only shows up for holy wars, he is now gone 20:18:45 adams really likes L7 20:18:52 So, we kind of covered the major work going on. I am in mid-cycle/review mode 20:18:58 * dougwig catches them. 20:19:13 dougwig with your eye? 20:19:25 heh, bullseye 20:19:26 i start working on anti-affinity server group 20:19:29 No love. 20:19:41 who ditched us? 20:20:01 Hmmm? 20:20:03 minwang2 ++ 20:20:07 Oh, before I forget, the LBaaSv2 horizon panels could use some reviews as well. 20:20:09 minwang2: +1 20:20:18 johnsom_: Link? 20:20:19 sbalukoff dougwig didn’t come to SEA 20:20:32 #link https://review.openstack.org/#/q/openstack/neutron-lbaas-dashboard+status:open,n,z 20:20:34 xgerman: And... you're complaining about this!? 20:20:34 Ill be testing reviewing tonight 20:20:35 ;) 20:20:50 Lol 20:20:50 yep 20:20:51 johnsom_: Danke! 20:21:08 Lots of progress there, but not many reviews, so please help out. ajmiller has been working on reviewing that from our side 20:21:30 johnsom_: I will review either later today or tomorrow. 20:21:30 ajmiller ++ 20:21:41 #topic Open Discussion 20:21:45 :) 20:22:01 Ok, here's the link to the L7 todo list etherpad: 20:22:02 https://etherpad.openstack.org/p/lbaas-l7-todo-list 20:22:06 Er.. 20:22:08 Ok, so the TaskFlow revert issue. I think we left it last week that you were going to test with 1.25. How did that go? 20:22:09 #link https://etherpad.openstack.org/p/lbaas-l7-todo-list 20:22:17 sbalukoff Thanks! 20:22:52 Alright, johnsom_ my real question is do you have time tonight/tomorrow to see what I'm lookin at with these errors? 20:23:26 Like, time for a Google Hangout session where I screen-share and show you whats up? 20:23:42 Since mugsie just stepped out of the room, yeah, I can probably carve some time out of the kosmos mid-cycle to take a look. 20:24:05 I'll be busy from 3-4 my time, but I'm good after that and all of tomorrow 20:24:19 Would 2pm pacific work for you? 20:24:33 Cool, so 4pm? 20:24:37 That's what, right at 4 Central? Yeah, should work. 20:25:00 Done. I will IRC you at 4 20:25:08 And we can setup a hangout 20:25:22 perfect. 20:25:36 Will likely have more taskflow stuff to report next meeting then, or just "I was dumb, no big" sort of response ha ha 20:26:01 Haha, ok, any other topics? 20:26:18 * johnsom_ chants "Review, Review, Review" 20:26:29 guess we are done... 20:26:35 Ok, thanks folks. 20:26:36 Sweet! 20:26:38 #endmeeting