21:00:39 #startmeeting networking 21:00:40 Meeting started Mon Oct 1 21:00:39 2018 UTC and is due to finish in 60 minutes. The chair is mlavalle. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:00:41 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:00:43 The meeting name has been set to 'networking' 21:01:34 o/ 21:02:17 o/ 21:02:21 let's wait a minute for others to show up 21:02:31 hi 21:02:37 o/ 21:02:40 hi 21:02:54 there they are :-) 21:03:27 As usual, our agenda is here: 21:03:36 #link https://wiki.openstack.org/wiki/Network/Meetings 21:03:48 #topic Announcements 21:04:37 Firt of all, Stein-1 is only three weeks away, Octover 22nd to 26th 21:04:45 as usual, time flies 21:06:43 I also want to highlight that the weekend right before the Summit in Berlin, there will be a edge computing Hackathon: 21:06:56 #link https://www.openstack.org/summit/berlin-2018/summit-schedule/#day=2018-11-10 21:07:09 It will be Saturday and Sunday 21:07:15 and I plan on attending 21:07:29 I would be delighted if some of you are there :-) 21:07:41 * mlavalle forgot to add it to the wiki. I'll do it afterwards 21:09:05 Per agreement during the PTG, I sent a message to the ML asking Stadium and networking projects for their proposals of jobs to run against Neutron patches 21:09:12 #link http://lists.openstack.org/pipermail/openstack-dev/2018-September/135273.html 21:09:58 I also followed up with requesting volunteers for the neutron-lib effort: 21:09:59 I was wondering, do we want to suggest a common topic for those? 21:10:08 #link http://lists.openstack.org/pipermail/openstack-dev/2018-September/135279.html 21:10:47 njohnston: my initial opinion is that we should let the projects to choose something that they think will protect them from our patches 21:10:56 but I'm open to suggestions 21:10:59 ok 21:11:55 Finally I proposed a patch to start having a meeting for a performance sub-team: 21:12:02 #link https://review.openstack.org/#/c/606870/ 21:12:22 Yes, we originally said in the PTG that it was going to be monthly.... 21:12:52 but yesterday, when putting together the patch, I couldn't find a way to specify monthly frequency 21:13:06 and that led me to believe that biweekly is not such a bad idea 21:13:28 :) 21:13:39 so let's shhot for b-weekly and I'll address your comments in the patch. Thanks for the feedback 21:13:49 speaking of slaweq.... 21:14:04 ok. I know yaml2ical supports quadweekly, which is close to monthly 21:14:11 https://github.com/openstack-infra/yaml2ical#frequencies 21:14:29 thx mlavalle 21:14:48 slaweq: would 1600UTC work for you on Mondays? 21:15:38 mlavalle: yes, that should be fine 21:16:05 njohnston, haleyb, hongbin, manjeets: how about you?^^^^ 21:16:31 no problems here 21:17:15 mlavalle: sorry, just got here 21:17:49 haleyb: taking about https://review.openstack.org/#/c/606870 21:18:05 would 1600UTC every other Monday work for you? 21:18:46 mlavalle: 1600UTC is my lunch time but i am not heavy participant of this meeting and i can eat and meeting :) 21:19:01 ok 21:19:01 yes, nothing else then (which i could assume from nate's response :) 21:19:12 haleyb: LOL 21:19:22 mlavalle, 1600 UTC works for me 21:19:23 I'll update the patch after the meeting 21:19:24 i'm going to need a double cappuccino on monday mornings :) 21:19:39 it's only bi=weekly, haleyb 21:19:50 ok, moving on 21:20:02 #topic Blueprints 21:20:31 This is what we are working on for Stein-1: 21:20:38 #link https://launchpad.net/neutron/+milestone/stein-1 21:21:14 Does anybode have an update on any blueprints? 21:22:00 I have a few updates 21:22:21 nothing for now, I am looking to make a push on bulk port creations this week 21:22:58 njohnston: thanks for the uodate 21:23:05 in regards to https://blueprints.launchpad.net/neutron/+spec/port-mirroring-sriov-vf 21:23:30 I just went again over the spec earlier today: https://review.openstack.org/#/c/606870/ 21:24:21 I left some comments and asked yamahata to do the same, to minimize iterations for munimeha1 21:24:24 finally we have an agreement on design at least 21:24:34 In my opinion, it is not far to be ready 21:24:57 * yamahata is reviewing it now. 21:25:15 so if you have time and an opinion on this subject, please leave your comments, so it can be moved forward 21:25:22 yamahata: Thank you :-) 21:26:22 I also took a quick look over the two smartnics related specs and the StarlingX ones. They haven't been updated since the feedback we gave them during the PTG 21:26:48 but let's keep an eye on them. They are all listed in the PTG summary in their corresponding sections 21:27:33 anything else on blueprints? 21:27:54 ok, moving on 21:28:08 #topic Community goals 21:28:34 Any updates on community goals this week amotoki_, njohnston, slaweq? 21:28:48 not from me yet :/ 21:28:55 slaweq: np 21:28:57 but I plan to start working on it this week 21:29:09 so I hope I will have something on next meeting maybe 21:29:15 Thank you 21:30:03 I started messing with a few of the jobs, transitioning to zuul v3 as part of python 3.6, but I put it on the shelf until this week when I concluded that I needed to understand a little better how some of the templates work. I hope to get back to it this week, and transition uwsgi jobs to python 3.6 and move them into the check queue as we discussed in the CI meeting 21:30:46 that's cool 21:31:10 just keep in mind you will face El Comandante tomorrow during the CI meeting 21:31:27 :-) 21:31:41 * mlavalle shudders at the thought of not having made my homework 21:32:05 ok, let's move on 21:32:05 LOL 21:32:26 #topic Bugs 21:32:54 Our bugs deputy last week was njohnston. 21:33:07 LOL 21:33:20 Bug deputy report is here: http://lists.openstack.org/pipermail/openstack-dev/2018-October/135318.html 21:33:21 Is there anything you want to highlight this week? 21:34:06 As I said, if a bug doesn't have any comments under it means it is probably open for working 21:34:47 One of them - https://bugs.launchpad.net/bugs/1795280 "netns deletion on newer kernels fails with errno 16" I have called in some help on with folks working on those newer kernel versions 21:34:47 Launchpad bug 1795280 in neutron "netns deletion on newer kernels fails with errno 16" [High,New] 21:35:06 njohnston: I was going to ask about that one 21:35:42 this one I could not reproduce but perhaps someone else will have more luck: https://bugs.launchpad.net/neutron/+bug/1794991 21:35:43 Launchpad bug 1794991 in neutron "Inconsistent flows with DVR l2pop VxLAN on br-tun" [Undecided,New] 21:37:00 For this one I can try to reproduce first in master 21:37:15 I have an environment where I can test it 21:37:27 excellent 21:37:47 last one came in today, I did not get a chance to confirm it: https://bugs.launchpad.net/neutron/+bug/1795432 21:37:47 Launchpad bug 1795432 in neutron "neutron does not create the necessary iptables rules for dhcp agents when linuxbridge is used " [Undecided,New] 21:37:58 that's the highlights 21:38:17 https://bugs.launchpad.net/neutron/+bug/1795482 is also a contender 21:38:17 Launchpad bug 1795482 in neutron "Deleting network namespaces sometimes fails in check/gate queue with ENOENT" [High,In progress] - Assigned to Brian Haley (brian-haley) 21:38:38 https://review.openstack.org/#/c/607009/ 21:38:55 mlavalle: can you take a look at that, will help with the gate i hope 21:39:05 haleyb: will take a look 21:39:38 thanks 21:39:41 njohnston: would you do me a favor in regards with https://bugs.launchpad.net/neutron/+bug/1795432\ 21:39:41 Launchpad bug 1795432 in neutron "neutron does not create the necessary iptables rules for dhcp agents when linuxbridge is used " [Undecided,New] 21:40:05 njohnston: I talked about this one today with one guy from RH (he reported it) 21:40:19 njohnston: but I didn't have time to confirm it yet 21:40:36 slaweq: do you have a way to verify it? 21:40:57 I will have to deploy some multinode env to verify it 21:41:27 currently I don't have it but if I will have few minutes I can try to do it 21:41:46 ok, if that becomes very difficult, ping me. I have to build an environmet, but I could also do it 21:42:30 mlavalle: thx 21:42:38 :-) 21:42:52 Any other bugs? 21:43:33 no, just to say there are a lot of fixes in review so plenty of changes for folks to take a look at :-) 21:43:58 Great! that's a good problem 21:44:43 Our deputy for this week is Swami. I don't see him around. I will send him an email at the end of the meeting 21:44:50 * mlavalle knows where he lives 21:45:08 #topic neutron-lib 21:45:22 boden is of today 21:45:29 off^^^ 21:45:40 but he sent me an email with his update 21:46:33 A reminder that we are now requiring projects that want neutron-lib patches to opt-in. Before long neutron-lib patches will only go to those who have opted-in 21:47:03 here's the message he sent to the ML: http://lists.openstack.org/pipermail/openstack-dev/2018-September/135063.html 21:48:05 He already started working with the zuul/tox set-ups: https://etherpad.openstack.org/p/neutron-sibling-setup and https://review.openstack.org/#/q/topic:neutronlib-zuulv3 21:49:01 He is Working on a neutron non-voting job for vmware-nsx's UTs as discussed in the PTG. He proposed a patch: https://review.openstack.org/#/c/605861 21:49:33 My feedback there is that in the vmware-nsx case we should talk about a 3rd party CI job 21:50:25 There's still discussion on how to best test with neutron-lib/master + neutron in https://review.openstack.org/#/c/602748/ 21:50:34 he wants us to chime in 21:51:02 and he also needs feedback on https://review.openstack.org/#/c/596959/3/neutron/plugins/ml2/db.py@336 21:51:57 #topic os-ken 21:52:05 hongbin: any updates? 21:52:16 two things 21:52:29 first, we requested to switch to storyboard: https://review.openstack.org/#/c/605224/ 21:52:50 second, we are requesting the first release of os-ken: https://review.openstack.org/#/c/607010/ 21:52:55 yeah, thanks to slaweq for taking the initiative on storyboard 21:53:20 np :) 21:53:29 we will see if it will be fine for us ;) 21:53:58 so right now, we are waiting for those two patches 21:54:16 hongbin: is the failure in the tox-validate job due to this patch? 21:54:31 the failure is because the release job and storyboard is not setup yet 21:54:38 hongbin: shouldn't we add some tests first before we release something? 21:55:05 mlavalle: i am pushing a patch to add a release job for os-ken 21:55:16 mlavalle: after that, it should be fine 21:55:46 slaweq: i did some manual test and it looks fine 21:56:06 slaweq: we need the first release so that we can further test it in the CI 21:56:17 ok 21:56:41 thanks for the update 21:56:41 ok then 21:56:51 #topic On demand agenda 21:57:07 Any other topics we should discuss today? 21:57:44 ok team, thanks for attending 21:57:50 have a great week! 21:57:55 #endmeeting