14:00:18 <slaweq> #startmeeting networking 14:00:19 <openstack> Meeting started Tue Jan 14 14:00:18 2020 UTC and is due to finish in 60 minutes. The chair is slaweq. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:20 <slaweq> hi 14:00:21 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:23 <openstack> The meeting name has been set to 'networking' 14:00:28 <njohnston> o/ 14:00:31 <bcafarel> \o 14:00:44 <davidsha> o/ 14:01:30 <amotoki> hi 14:01:51 <mlavalle> o/ 14:01:58 <rubasov> o/ 14:03:00 <slaweq> #topic Announcements 14:03:06 <lajoskatona> o/ 14:03:26 <slaweq> First announcement is good news actually 14:03:46 <slaweq> We found new maintainer for neutron-vpnaas yedongcan stepped up to help maintaining it 14:03:54 <mlavalle> yayy 14:05:01 <slaweq> so we are still looking only for maintainers for neutron-fwaas: http://lists.openstack.org/pipermail/openstack-discuss/2019-November/010929.html 14:05:53 <slaweq> which is somehow connected to my last announcement: 14:05:55 <slaweq> We are getting close to Ussuri-2, it will be in the week of February 10: https://releases.openstack.org/ussuri/schedule.html 14:06:29 <slaweq> so if we will not have anyone interested in maintaining fwaas before Ussuri-2, we will mark this project as deprecated 14:06:41 <slaweq> that's what we discussed in Shangahi 14:06:57 <mlavalle> yeah, maybe time to send a warning to the ML again 14:07:04 <bcafarel> we'll see if the "last call" mail will trigger a reaction 14:07:18 <slaweq> I actually already sent such message last week 14:07:35 <mlavalle> Yeah, include "LAST CALL" in the subject 14:07:36 <slaweq> but I will sent one or two more before Ussuri-2 14:07:44 <slaweq> mlavalle: sure, I will do that 14:07:55 <amotoki> yeah. in addition, "deprecation" notice in Ussuri release may bring another chance someone notices it... (though deprecation is not a great thing) 14:08:23 <slaweq> amotoki: yes, I agree, that will be "very last call" in fact :) 14:09:42 <slaweq> and that's all announcements from me 14:09:51 <slaweq> anyone has got anything else to share with the team? 14:10:42 <ralonsoh> hi 14:12:01 <bcafarel> on stable branches if you had recent failures this should be fixed now with http://lists.openstack.org/pipermail/openstack-discuss/2020-January/011970.html 14:12:15 <bcafarel> feel free to send rechecks :) (even if that adds to slaweq's stats) 14:12:59 <slaweq> bcafarel: You read my post, thx :) 14:13:13 <bcafarel> :) 14:13:16 <slaweq> and thx for sharing info about this CI issue 14:13:34 <slaweq> actually it's not only for stable branches, for master it should works also if You recheck now 14:15:17 <slaweq> ok, next topic than 14:15:19 <slaweq> #topic Blueprints 14:15:26 <slaweq> Blueprints for Ussuri-2: https://launchpad.net/neutron/+milestone/ussuri-2 14:15:40 <slaweq> I updated status of some of them today 14:15:53 <slaweq> anyone has got any updates about BPs? 14:16:25 <njohnston> I think stateless security groups is coming along well, not that long before I can imagine it merging 14:16:32 <mlavalle> in regards to https://launchpad.net/bugs/1815933, I think this patch is good to go: https://review.opendev.org/#/c/700755 14:16:32 <openstack> Launchpad bug 1815933 in neutron "[RFE] Allow bulk-tagging of resources" [Wishlist,In progress] - Assigned to Miguel Lavalle (minsel) 14:16:45 <njohnston> mlavalle: Excellent! I was going to review that later 14:17:20 <mlavalle> this patch will be followed by other two: api-ref and a tempest api test 14:17:36 <mlavalle> in neutron-tempest-plugin of course 14:17:51 <slaweq> great mlavalle, thx a lot 14:18:24 <mlavalle> after that I will move to work on the policy engine update 14:18:38 <slaweq> njohnston: yes, I was checking stateless security groups patch recently and it was pretty close IMO 14:18:51 <slaweq> mlavalle: thx for working on both :) 14:20:11 <slaweq> I would like to ask about https://blueprints.launchpad.net/neutron/+spec/metadata-over-ipv6 14:20:30 <slaweq> I got today info from rubasov that he don't have time to work on this one 14:20:43 <slaweq> thx rubasov for this info 14:20:52 <slaweq> so I have actually 2 questions: 14:21:01 <slaweq> 1. is there any volunteer to work on that? 14:21:10 <rubasov> sorry for not making progress with that 14:21:48 <slaweq> 2. haleyb: did You get some info about IPv6 address which we should use and if we should "reserve" it with some RFC? 14:22:43 <slaweq> rubasov: no problem :) 14:23:19 <slaweq> I think haleyb is not here now so I will probably ping him later during this week about it 14:23:53 <mlavalle> This RFE is important in China, which is a big market for OepnStack 14:23:54 <haleyb> hi 14:24:14 <slaweq> mlavalle: I think this RFE should be important for everyone in 2020 ;) 14:24:19 <slaweq> hi haleyb :) 14:24:37 <haleyb> slaweq: i had a few suggestions, but in the end they raised more questions than answers 14:24:44 <mlavalle> so, while I don't want to commit now, please leave it in the dashboard and I will try to come to it later 14:24:54 <slaweq> mlavalle: sure 14:25:11 <slaweq> I will also try to help with this if I will have some time 14:25:14 <haleyb> so i think we just talk to the cloud-init people and agree on an address, otherwise it could take forever 14:25:36 <slaweq> haleyb: that's what I was affraid of actually :) 14:25:50 <mlavalle> unless our fearless leader tells me that this RFE has priority over the policy engine update. The I'll start with it after the bulk tagging 14:26:06 <mlavalle> Then* 14:26:13 <slaweq> mlavalle: TBH I think this is bigger priority 14:26:25 <haleyb> i had these three questions i remember, "what about security?", "why are you making this look like IPv4?", "what about using mDNS instead?" 14:26:29 <slaweq> as this would be something "big" which we can share for Ussuri release :) 14:26:38 <haleyb> i let out a sigh after that 14:27:43 <mlavalle> ok in that case, I will explore how to move forward with this RFE. I'll discuss with haleyb his concerns to make sure we can move forward 14:28:00 <slaweq> mlavalle: haleyb: great, thx a lot 14:28:07 <haleyb> mlavalle: ack, let's talk about it offline 14:28:12 <slaweq> please keep me in the loop if You would have something interesting :) 14:28:28 <mlavalle> slaweq: if I drop the policy engine update, you deal with the Keystone lady 14:28:37 <slaweq> mlavalle: ok :) 14:29:41 <slaweq> ok, so lets move on to the next topics than 14:29:43 <slaweq> #topic Community goals 14:29:57 <slaweq> njohnston: any updates on dropping py2 support? 14:30:25 <njohnston> Our part of the drop-py27-support goal is almost done. Remaining changes are: 14:30:25 <njohnston> neutron: https://review.opendev.org/701409 14:30:25 <njohnston> networking-bagpipe: https://review.opendev.org/701356 14:30:27 <njohnston> neutron-fwaas: https://review.opendev.org/688278 14:30:36 <njohnston> The zuulv3 goal is not as close. The following projects don't appear to have zuulv3 conversion changes underway: networking-bgpvpn networking-midonet networking-odl neutron-dynamic-routing neutron-vpnaaa. Everything else is completed. 14:30:37 * mlavalle assigned the metadata RFE to himself 14:31:26 <njohnston> thanks everyone for all the attention and reviews for these changes, this has been going quite smoothly all things considered 14:31:32 <slaweq> njohnston: and what about networking-midonet? What's status of it according to dropping py27 testing? 14:31:54 <njohnston> slaweq: That change merged already https://review.opendev.org/701210 14:32:53 <slaweq> ok, but this basically means that networking-midonet is doing only UT jobs now 14:32:55 <slaweq> :/ 14:33:56 <mlavalle> dang 14:33:56 <njohnston> Yeah, I can't say what is the right or wrong strategy for that project. But they are fully compliant now. 14:34:21 <amotoki> njohnston: the patch in netwokring-midonet has not merged yet. it depends on a grenade patch. 14:35:18 <njohnston> amotoki: Oh gosh, thanks for pointing that out. I'll add https://review.opendev.org/#/c/695094 to my tracking. 14:35:51 <slaweq> thx for update njohnston, so we are quite close to accomplish this goal soon 14:36:24 <slaweq> according to the "Project Specific PTL and Contributor Documentation" goal, I started looking into our docs this week 14:36:38 <slaweq> and I will start writing some guide for that 14:36:49 <njohnston> for dropping py27, yes. I would not say we are as close with the zuulv3 since there are a lot of changes needed that do not exist yet. 14:36:52 <slaweq> than I will send it to mlavalle to review as much more experience ptl :) 14:37:30 <slaweq> njohnston: yes, sure, but community goal is only about dropping py2 support, zuulv3 is another story and it's not part of this goal community wide 14:37:38 <slaweq> it was our "internal goal" IIRC 14:39:01 <amotoki> njohnston: is the zuulv3 stuffs a part of the py27 drop goal? I see another goal proposal on zuulv3 in https://review.opendev.org/#/c/691278/ 14:39:23 <amotoki> of course it is worth tracking them. 14:39:37 <njohnston> amotoki: No they are separate goals, I am just tracking their progress in the same etherpad 14:39:52 <amotoki> njohnston: thanks for clarification. 14:41:00 <slaweq> ok, lets move on to the next topic 14:41:01 <njohnston> that is it from me 14:41:05 <slaweq> thx njohnston :) 14:41:07 <slaweq> #topic Bugs 14:41:28 <slaweq> hongbin was on bug deputy. His report is on http://lists.openstack.org/pipermail/openstack-discuss/2020-January/011955.html 14:42:00 <slaweq> there is couple of new bugs related to our ci there 14:42:08 <slaweq> and those needs some volunteers 14:42:21 <slaweq> like e.g. https://bugs.launchpad.net/neutron/+bug/1858645 - neutron-fwaas scenario job is unstable, probably some race condition 14:42:21 <openstack> Launchpad bug 1858645 in neutron "Neutron-fwaas tempest job is not stable" [Critical,Confirmed] 14:42:41 <slaweq> for now neutron-fwaas job in neutron-tempest-plugin is set as non-voting due to this bug 14:42:55 <slaweq> or https://bugs.launchpad.net/neutron/+bug/1858642 - ssh failure in scenario jobs with dvr 14:42:56 <openstack> Launchpad bug 1858642 in neutron "paramiko.ssh_exception.NoValidConnectionsError error cause dvr scenario jobs failing" [High,Confirmed] 14:43:19 <slaweq> and this is (probably) different ssh failure than failures which we had in the past many times 14:43:28 <ralonsoh> I can take a look at the SSH failure in DVR 14:43:35 <slaweq> ralonsoh: thx a lot 14:43:46 <slaweq> and one last thing from me regarding bugs 14:44:09 <slaweq> there is https://bugs.launchpad.net/neutron/+bug/1858419 which is opened to track progress on docs update for large scale deployments 14:44:09 <openstack> Launchpad bug 1858419 in neutron "Docs needed for tunables at large scale" [Medium,Confirmed] - Assigned to Slawek Kaplonski (slaweq) 14:44:33 <slaweq> liuyulong already commented a lot there but if anyone else has got any experience with that, please check it and comment there too 14:45:15 <slaweq> You can also push patches with changes related to this bug directly to gerrit :) 14:45:31 <slaweq> this week our bug deputy is haleyb 14:45:45 <slaweq> and next will be lajoskatona 14:45:47 <haleyb> o/ 14:45:59 <slaweq> any other bugs You want to discuss with the team? 14:46:03 <lajoskatona> slaweq: my eyes on it (from Monday of course) :-) 14:46:13 <slaweq> lajoskatona: thx :) 14:48:43 <slaweq> ok, so I guess there is no other bugs to discuss for today :) 14:48:52 <slaweq> next topic then 14:49:02 <slaweq> #topic Networking OVN and ML2+OVS+DVR Convergence 14:49:42 <slaweq> I don't think lucasgomes or maciejjozefczyk are around 14:49:53 <slaweq> but ralonsoh maybe You have any updates on it? 14:49:55 <slaweq> or haleyb 14:50:03 <ralonsoh> I pinged them 14:50:11 <ralonsoh> current status (so far): 14:50:25 <ralonsoh> merging the FTs and solving some problems with the contexts 14:51:01 <ralonsoh> also the devstack entry points are added, just to be able to test within the neutron repo 14:51:04 <maciejjozefczyk> hello! 14:51:22 <ralonsoh> maciejjozefczyk, can you update the OVN migraiton status? 14:52:03 <maciejjozefczyk> Sure! For now we just merged OVN mechanism driver to Neutron tree 14:52:27 <maciejjozefczyk> In meantime we're starting to run tempest tests, for now all seems to be fine, Lucas is on it. 14:53:13 <maciejjozefczyk> We found one issue with Openflow protocols with using OVS firewall driver, it should be fixed with: https://review.opendev.org/#/c/698715/ 14:53:35 <ralonsoh> (OVS community is aware of this) 14:53:48 <maciejjozefczyk> Ok 14:54:05 <maciejjozefczyk> If above bug will be merged then we would be able to start merging functional tests for OVN 14:54:30 <maciejjozefczyk> I just pushed all of them, we're waiting for CI status 14:55:50 <maciejjozefczyk> In meantime we're starting setup of OVN Provieder driver repository: https://review.opendev.org/#/c/701947/ (its under Neutron umbrella) 14:56:45 <maciejjozefczyk> that's all from my, lucasagomes anything else? 14:57:20 <lucasagomes> maciejjozefczyk: well, I guess. I missed most of the update (I wasn't in the room) 14:57:46 <ralonsoh> +1 for maciejjozefczyk and lucasagomes 14:57:58 <slaweq> thx guys for the update 14:58:01 <lucasagomes> but yeah, tl;dr good progress last week. We got up to the point where we can enable the ovn driver in neutron 14:58:04 <lucasagomes> and tempest are passing 14:58:23 <slaweq> lucasagomes: that's great news :) 14:58:23 <lucasagomes> maciejjozefczyk: is working on the functional tests now that are missing 14:58:28 <lucasagomes> sorry if I repeated anything 14:58:48 <maciejjozefczyk> lucasagomes, yes, I just pushed all of functional tests :) all were passing on my env 14:59:02 <maciejjozefczyk> slaweq, any update on doc topic? 14:59:15 <slaweq> maciejjozefczyk: not too much 14:59:33 <slaweq> I started merging networking-ovn docs into neutron but it's still failing on docs build 14:59:39 <slaweq> so I will continue this during this week 14:59:49 <slaweq> ok, we are almost out of time now 14:59:55 <slaweq> thx for attending the meeting 14:59:59 <bcafarel> o/ 15:00:00 <njohnston> o/ 15:00:00 <slaweq> have a great week 15:00:02 <slaweq> o/ 15:00:04 <slaweq> #endmeeting