21:01:00 #startmeeting networking 21:01:01 Meeting started Mon Jul 23 21:01:00 2018 UTC and is due to finish in 60 minutes. The chair is mlavalle. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:01:02 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:01:04 The meeting name has been set to 'networking' 21:01:11 Hi there! 21:01:19 o/ 21:01:38 hi 21:02:09 #topic Announcements 21:02:57 Since this is the week of Rocky-3 and we all have lot's of things to do and code to review, let's see if we can do this meeting short and sweet 21:03:09 This week is Rocky-3 21:03:54 We are going to to cut / freeze Wednesday night (Thursday morning amotoki's time) 21:04:02 amotoki: ok with that? 21:04:09 mlavalle: fine with that 21:04:37 I will prepare a release patch around wed (or thu) and then update it accordingly 21:04:54 perfect, thanks you very much! 21:05:37 Any other big annoucements? 21:06:42 ok, let's move on to go over the Blueprints 21:06:48 PTL election time? http://lists.openstack.org/pipermail/openstack-dev/2018-July/132402.html 21:06:50 #topic Blueprints 21:07:32 o/ 21:07:45 This is what we are targeting for Rocky-3: https://launchpad.net/neutron/+milestone/rocky-3 21:08:23 I would encourage everybody to concentrate your reviewing efforts on what is in this dashboard 21:09:20 does anyone want to highlight something on this blueprints? 21:09:37 and bugs? 21:10:25 amotoki: any update on https://blueprints.launchpad.net/neutron/+spec/get-policy-from-neutron-lib? 21:10:32 (although it is too late to land in Rocky-3) it turns out policy-in-code can be supported only by the neutron repo. 21:11:04 I just posted a working code https://review.openstack.org/585037 21:11:05 patch 585037 - neutron - Convert policy.json into policy-in-code 21:11:09 so are we looking at Stein-1 for it? 21:11:36 Stein-1 is a good candidate. 21:11:51 Looking at the code maturity and risk, we can discuss it as FFE 21:12:28 one tricky point is neutron-lib side is that we need a small snippet like https://review.openstack.org/585036/ 21:12:29 patch 585036 - neutron-lib - WIP: Minor fix for policy-in-code support 21:12:35 ok, let's make a final determination towards the end of the week. We can use the drivers meeting on Friday to make a final decision on FFEs 21:13:19 amotoki, what do you mean by "policy-in-code can be supported only by the neutron repo" - is this a statement about stadium projects? 21:13:48 njohnston: I mean policy-in-code can be supported without neutron-lib in 99% 21:14:22 njohnston: stadium/neutron-related projects which implements API extensions need works. 21:14:44 amotoki: you mean their own work, right? 21:14:49 ok, just making sure I understand :-) 21:15:47 Moving on.... 21:15:54 mlavalle: yes. I will post an example using neutron-fwaas in a few hours 21:16:08 Thanks! 21:16:20 With https://blueprints.launchpad.net/neutron/+spec/live-migration-portbinding... 21:16:32 We have 6 of 8 patches merged 21:16:55 and the other 2 are lined up on final approach, so I think this blueprint looks good 21:17:40 Thanks to all the reviewers: slaweq, njohnston, manjeets, haleyb 21:17:46 Much approciated 21:17:49 :) 21:18:43 With https://blueprints.launchpad.net/neutron/+spec/port-forwarding 21:19:44 I and the other usual suspects (mentioned above) have spent a good deal of time reviewing the series of patches: https://review.openstack.org/#/q/topic:bp/port_forwarding+(status:open+OR+status:merged) 21:19:57 We landed already the DB and objects layer 21:20:28 The service plugin patch has some minor problems 21:20:47 But works well in my test environment, where I deployed it 21:21:25 mlavalle, did you use curl to create resources for pf ? 21:21:47 manjeets: yes, I have a few curl scripts 21:21:57 one for each CRUD operation 21:22:10 can you share ? I need to clear some doubts on other patches ? 21:23:09 manjeets: sure, I will do that. The easisest thing to do is to add them to https://github.com/miguellavalle/tunnelssovslbvagrant 21:23:17 I'll add them to the utils folder 21:23:43 thanks, I'll check later 21:24:02 so this blueprint looks like a toss up to me right now 21:24:14 let's see how much progress we make over the next few days 21:24:45 As for https://blueprints.launchpad.net/neutron/+spec/extend-logging-to-fwassv2.0 21:25:07 They already requested a FFE last week. I will attend the FWaaS meeting this week 21:25:20 anything else on blueprints? 21:25:32 mlavalle, I have a question w 21:25:38 FYI: if you need to talk neutron API directly, https://pypi.org/project/oscurl/ might help you. 21:26:00 when the bp for next cycle will open up ? I mean what is good time to start marking proposal for next cycle in neutron-specs ? 21:27:03 manjeets: you can start now. You won't get much attention over the next few days 21:27:24 But will get to it soon enough 21:27:36 mlavalle, ohk, I have one request from internal team, so I propose it by next week 21:27:39 thanks 21:27:44 cool 21:27:55 neutron-specs repo already has 'stein' directory, so it is already open 21:28:26 (though you won't get much attention as mentioned) 21:28:34 LOL 21:29:04 #topic Community goals 21:29:31 wsgi support is in the dashboard we already discussed 21:29:41 please keep an eye on those patches 21:29:54 njohnston: anything to say about Python 3? 21:31:36 I take that as a "nope" 21:31:42 #topic Bugs 21:32:36 Our bugs deputy last week was lujinluo. She put together this summary for us http://lists.openstack.org/pipermail/openstack-dev/2018-July/132405.html 21:33:19 I went through the list and IMO it is pretty much under control 21:33:32 The only concern I have is https://bugs.launchpad.net/neutron/+bug/1782421 21:33:32 Launchpad bug 1782421 in neutron "Port creation fails during concurrent deploys" [Undecided,New] 21:33:47 I would like the team to look and chime in 21:34:24 There are some patches somehow related to it: https://review.openstack.org/#/c/583527 21:34:24 patch 583527 - neutron-lib - Make the DB retry mechanism more robust 21:34:30 That needs some eyes 21:35:05 any otherr bugs we should highlight today? 21:35:20 mlavalle for https://review.openstack.org/#/c/583527 21:35:21 patch 583527 - neutron-lib - Make the DB retry mechanism more robust 21:35:46 does that mean we may have to create another lib release for it and re-cut the rocky branch? 21:37:02 boden: yeah, maybe. would that be too much problem? 21:37:09 even if we cut another release we don't need to recreate the rocky branch. we can just backport it to rocky 21:37:52 backporting maybe safer, but I havent’ looked at what we’ve landed since 1.18.0 21:38:23 boden: why don't you take a look at what we have landed and we talk tomorrow morning? 21:38:34 with the caveat that backporting is safer 21:38:43 sure 21:38:48 cool 21:38:53 thanks! 21:39:24 The bug deputy this week is yamamoto. I already sent him a gentle reminder earlier today 21:39:37 so I think we are covered this week :-) 21:39:48 #topic neutron-lib 21:40:11 boden: anyhing else on neutron-lib? 21:40:43 just an FYI regarding the neutron-lib version for rocky that I sent to ML earlier http://lists.openstack.org/pipermail/openstack-dev/2018-July/132412.html 21:41:21 as it turns out most of the failures I’m seeing in those mentioned patches https://review.openstack.org/#/q/topic:rocky-neutronlib are due to invalid zuul setup 21:42:05 vpnaas, dynamic-routing, sfc are failing... 21:42:07 so you are our involuntary zuul setup expert :-) 21:42:17 so either I will have to fix the zuul setup for all these failures, or projects will have to spend the time to do so…. this is again related to the zuul setup ML I sent awhiole back 21:42:59 I already fixed lbaas, but I haven’t had time for the rest.. I will see if I can get to them in the next week or so… 21:43:05 boden: how is it important to bump the min version of neutron-lib? 21:43:28 most projects depend on neutron and if neutron requires neutron-lib >=1.18.0 I think 1.18.0 will be used. 21:43:31 amotoki well if your project depends on neutron for stable/rocky, won’t you get 1.18.0 anyway 21:44:12 regardless, it’s still proper for them to be at the same level IMO 21:44:30 and I don’t want to keep sweeping this zuul issue under the covers… it’s been a consistent issue for the lib work 21:44:33 so I will pick at it 21:45:03 boden: fair enough. it sounds nice to bump the min version but I think it won't be a blocker 21:45:14 anyway let's try to fix them 21:45:18 I think that’s enough on neutron-lib from me… 21:45:42 amotoki: the vpnaas guys can take care of this themselves. would you alert them, since they are closer to you time zone wise? 21:46:11 mlavalle: yes, I will take care of it. it seems straight-forward 21:46:18 as only docs and l-c job fail. 21:46:25 Thanks 21:46:35 #topic Open Agenda 21:46:47 I skipped a few topics today 21:47:04 But if you have anything to dicusss today, please bring it up now 21:47:33 for neutronclient release, I looked thru open patches but there is nothing to highlight. there seems no need to cut another release of neutronclient. 21:47:57 that's good. one less thing :-) 21:48:40 ok team. Thanks for showing up 21:48:51 Let's go nail this release 21:48:55 Sorry I missed the section earlier about python3, I had a power outage at my house 21:49:05 LOL 21:49:13 anything to say about it? 21:49:45 The thing I am chasing right now is a failure with the functional job, what fails is "subunit.parser", so I am trying to delve into what is going on there 21:49:52 http://logs.openstack.org/50/580450/2/experimental/neutron-functional-python35/4d72ae4/job-output.txt.gz#_2018-07-11_20_25_11_125540 21:50:35 that is it for now 21:51:07 thanks njohnston :-) 21:51:11 ok guys 21:51:17 Have a great week! 21:51:22 #endmeeting