14:00:38 <mlavalle> #startmeeting networking 14:00:40 <openstack> Meeting started Tue Aug 27 14:00:38 2019 UTC and is due to finish in 60 minutes. The chair is mlavalle. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:41 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:43 <openstack> The meeting name has been set to 'networking' 14:01:21 <rubasov> hi 14:01:26 <amotoki> hi 14:01:33 <haleyb> hi 14:01:47 <ralonsoh> hi 14:01:47 <njohnston> o/ 14:01:53 <bcafarel> \o 14:01:58 <mlavalle> ok, let's get going 14:02:11 <mlavalle> #topic Announcements 14:02:36 <lajoskatona> o/ 14:03:03 <mlavalle> Our T-3 milestone will be the week of September 9th 14:03:43 <mlavalle> Please remember that we are in the PTL candidates self nomination week: https://releases.openstack.org/train/schedule.html#u-ptl-nominations 14:04:14 <mlavalle> if you want to run for PTL, please don't miss the deadline 14:05:18 <mlavalle> which is Sep 03, 2019 23:45 UTC 14:05:26 <mlavalle> that is next week 14:05:44 <mlavalle> #link https://governance.openstack.org/election/ 14:06:51 <mlavalle> Don't forget to propose topics for the PTG in Shanghai. Here's the etherpad: https://etherpad.openstack.org/p/Shanghai-Neutron-Planning 14:07:26 <mlavalle> many of you registered "tentative" in the etherpad 14:08:18 <mlavalle> please update it as soon as you get a yes or no from your sponsors 14:08:32 <mlavalle> any other announcements from the team? 14:09:03 <amotoki> the next week is "Final release for non-client libraries" 14:09:21 <amotoki> we need to release neutron-lib next week if needed. 14:09:38 <amotoki> this is just a reminder. 14:09:56 <mlavalle> ok, let's get ready for that 14:10:01 <mlavalle> thanks for the reminder 14:10:27 <mlavalle> ok, let's move on 14:10:58 <mlavalle> #topic Blueprints 14:11:16 <mlavalle> These are the blueprints we are targeting for T-3: https://launchpad.net/neutron/+milestone/train-3 14:11:47 <mlavalle> any updates on any of those blueprints? 14:12:57 <rubasov> the extraroute change's neutron-lib dependency was merged and released, thank you all 14:13:31 <rubasov> I have a random question I'd like to raise at the end of the meeting 14:13:47 <mlavalle> yeah I saw that 14:13:51 <rubasov> I did write it up here: https://wiki.openstack.org/wiki/Network/Meetings#On_Demand_Agenda 14:14:05 <mlavalle> is it related to the extraroute work? 14:14:10 <rubasov> yes 14:14:19 <mlavalle> ok 14:14:39 <mlavalle> njohnston: any updates on custom ethertypes? 14:15:20 <njohnston> I've been working away at it in the background and I have figured out I need to refactor it into multiple smaller changes; going to start pushing those this week 14:16:01 <mlavalle> thanks for the hard work! 14:17:15 <mlavalle> #topic Community goals 14:17:27 <mlavalle> any updates on community goals? 14:17:42 <amotoki> I have an update on the PDF goal. 14:18:21 <amotoki> there are a lot of progress since last week. I worked with the docs/infra team to support the common job configuration and it has landed. 14:18:29 <amotoki> now we can work on individual projects. 14:18:46 <amotoki> I started to send patches to neutron stadium projects as you may notice. 14:19:06 <amotoki> we can find review links via https://review.opendev.org/#/q/topic:build-pdf-docs+(status:open+OR+status:merged) 14:19:10 <mlavalle> Thanks! 14:19:58 <amotoki> the effort is tracked by https://storyboard.openstack.org/#!/story/2006099 and I will send more patches for projects uncovered yet. 14:20:22 <amotoki> I don't see a big problem so far and time will complete the goal :) 14:20:24 <njohnston> FOr python 3 in the stadium - https://etherpad.openstack.org/p/neutron_stadium_python3_status - lajoskatona can you update us for networking-odl? amotoki has done some work to finish up networking-bagpipe. I haven't seen yamamoto online to ask him about networking-midonet. 14:20:57 <lajoskatona> njohnston: sure 14:21:12 <bcafarel> amotoki: great progress on that PDF goal! 14:21:13 <amotoki> re: networking-bagpipe, I updated the etherpad to clarify the remaining topics. 14:21:32 <njohnston> amotoki: Thanks for your work, I just gave a +2 to both patches you point out 14:21:41 <lajoskatona> the most important patches for networking-odl are on the master, and I have few other (perhaps 4 with elod's patch) to make te old job efinitions zuulv3 compatible 14:22:57 <lajoskatona> njohnston: so as I have written on the pad, the must be done part is ready, but I try again to find somebody from qa team as well for zuul related review to push the other patches. 14:24:44 <njohnston> lajoskatona: let me know if you need any assistance 14:25:02 <amotoki> lajoskatona: which patches need to be reviewed by the qa team? 14:25:12 <amotoki> on zuulv3 migration? 14:25:40 <lajoskatona> amotoki: these: https://review.opendev.org/#/q/status:open+project:openstack/networking-odl+branch:master+topic:use-latest-odl 14:26:08 <mlavalle> I can help with those 14:26:16 <mlavalle> I'll take a look later today 14:26:23 <mlavalle> or early tomorrow morning 14:26:32 <lajoskatona> amotoki: it is a series, and the top one, which is for tempest is under investigation from ODL team, as the job ends with timeout 14:26:53 <amotoki> lajoskatona: thanks 14:27:08 <lajoskatona> mlavalle: thanks, I got some promise from gmann as well 14:27:17 <mlavalle> so the first two are good to go 14:27:19 <mlavalle> ? 14:27:46 <lajoskatona> njohnston: thanks, I will check the patches, if I need ot rebase or similar 14:28:34 <mlavalle> ok, let's move on 14:28:40 <amotoki> networking-bagpipe fullstack job is failing by some neutron change even after https://review.opendev.org/#/c/677711/ lands. it is nice if someone can take a look (as I will be travel since tomorrow this week) 14:28:43 <lajoskatona> mlavalle: yes 14:28:49 <mlavalle> + 14:29:45 <mlavalle> ok, let' move on 14:29:49 <mlavalle> #topic Bugs 14:31:05 <mlavalle> last week our deputy was Swami. Did anybody see a report from him in the ML? 14:31:47 <amotoki> I see him message in #openstack-neutron channel. Looking for it. 14:31:49 <mlavalle> he confirmed last week by email that he was going to take care of his duty week 14:32:03 <amotoki> Found! https://docs.google.com/spreadsheets/d/1JjcSHL6hFYsAvlrcgK7_Eje1S4vvJdvm363_Mmy557I/edit?usp=sharing 14:32:11 <bcafarel> https://docs.google.com/spreadsheets/d/1JjcSHL6hFYsAvlrcgK7_Eje1S4vvJdvm363_Mmy557I/edit?usp=sharing 14:32:22 <bcafarel> he, a second too late :) 14:34:10 <mlavalle> Thanks 14:35:29 <mlavalle> so everything high has been fixed or has an owner 14:35:59 <haleyb> or is incomplete 14:36:31 * haleyb just marked 1840952 incomplete 14:36:38 <mlavalle> yes, the undecided ones are incomplete 14:37:00 <mlavalle> thanks for the undecided tag an that last one haleyb 14:37:14 <mlavalle> I mean, incomplete 14:37:18 <amotoki> bug 1841067 is related to nova interaction. do we have any tag for nova interaction? 14:37:20 <openstack> bug 1841067 in neutron "SR-IOV agent depends on mac addresses for getting bound ports" [Medium,Confirmed] https://launchpad.net/bugs/1841067 14:37:30 <haleyb> np, i forgot to change when i looked yesterday 14:38:33 <mlavalle> no, but maybe we should create one 14:39:59 <mlavalle> in any case, I'll look at that bug 14:40:22 <mlavalle> any other bugs we should discuss today? 14:41:35 <mlavalle> ok, for this week our deputy is lajoskatona 14:41:49 <lajoskatona> mlavalle: my eyes on lunchpad 14:42:04 <mlavalle> and next week it's njohnston's turn 14:42:07 <lajoskatona> launchpad--^ 14:42:17 <njohnston> aye aye captain 14:42:26 <mlavalle> cool 14:42:44 <mlavalle> #topic CLI / SDK 14:43:08 <mlavalle> amotoki: anything else to say about this in preparation for end of cycle? 14:43:38 <amotoki> nothing special except extra route support patch. 14:43:58 <mlavalle> Thanks! 14:44:07 <mlavalle> #topic nuetron-lib 14:44:17 <mlavalle> #undo 14:44:18 <openstack> Removing item from minutes: #topic nuetron-lib 14:44:29 <mlavalle> #topic neutron-lib 14:44:42 <mlavalle> boden: any updates? 14:44:46 <boden> hi 14:45:00 <mlavalle> nice to see you back :-) 14:45:13 <mlavalle> the new Idahoan! 14:45:24 <boden> no updates from me really... I'm just back from a long vacation so just catching up with some patches that were out prior to vaca 14:45:45 <boden> Montanaian actually, or whatever they call us 14:45:58 <mlavalle> ahhh, well, congrats! 14:46:06 <boden> thanks 14:46:35 <mlavalle> did you see amotoki's comment about releasing neutron-lib next week? 14:47:12 <boden> I saw the comment.. we can release whenever folks want; no problems there 14:47:26 <mlavalle> cool 14:47:33 <mlavalle> let's move on 14:47:42 <mlavalle> #topic On demand agenda 14:48:18 <ralonsoh> hi, we talked about this https://bugs.launchpad.net/neutron/+bug/1841067 14:48:19 <openstack> Launchpad bug 1841067 in neutron "SR-IOV agent depends on mac addresses for getting bound ports" [Medium,Confirmed] 14:48:34 <ralonsoh> but we should raise the importance of it 14:48:45 <ralonsoh> that will imply a nova-neutron coordination 14:48:54 <ralonsoh> and a change on the sriov agent 14:49:05 <ralonsoh> does this rfe need a spec?? 14:49:06 <ralonsoh> just asking 14:49:22 <ralonsoh> (btw, is not marked as rfe) 14:49:50 <njohnston> I just wanted to discharge my oslo liaison responsibility and let everyone know that there is an issue in oslo.service with using SIGHUP signals to reload mutable configs; we worked on this a cycle or two ago. Fix in progress: https://review.opendev.org/#/c/641907/ 14:50:39 <njohnston> basically what happens is that the SIGHUP causes a full service restart instead of a graceful config reload 14:51:05 * bcafarel reads 14:51:09 <njohnston> which means some RPC messages can get lost, and all the other sorts of negative effects an unanticipated service restart might generate 14:51:49 <bcafarel> I wonder if it could help for that neutron sighup issue I was looking into at some point (trying to find lp number) 14:52:07 <mlavalle> njohnston: thanks for the heads up 14:52:51 <amotoki> ralonsoh: is the approach clear enough? If it is just an implementation topic, I think we can skip a spec. 14:53:18 <mlavalle> ralonsoh: yeah, to me it looks like border line RFE / bug 14:53:21 <amotoki> RFE sounds fine for bug 1841067. I will add the tag 14:53:22 <openstack> bug 1841067 in neutron "SR-IOV agent depends on mac addresses for getting bound ports" [Medium,Confirmed] https://launchpad.net/bugs/1841067 14:53:33 <ralonsoh> amotoki, if you agree with https://review.opendev.org/#/c/676713 that's ok 14:53:52 <ralonsoh> that was just a heads-up, because this is a change in the architecture of the sriov agent 14:54:18 <mlavalle> do you forsee a massive change? 14:54:57 <ralonsoh> mlavalle, maybe with the way we assign the white-list 14:55:23 <mlavalle> ok, in that case let's look at it as RFE, as suggested by amotoki 14:55:57 <mlavalle> ok looking at the topic raised by rubasov.... 14:56:05 <rubasov> thanks 14:56:20 <rubasov> I'm asking for directions 14:56:30 <mlavalle> I faced something similar when implementing the activate action for multiple port bindings 14:56:30 <rubasov> if extension2 builds on extension1 14:56:52 <rubasov> then should its api-def repeat extension1's api-def parts? 14:57:02 <rubasov> my assumption was it shouldn't 14:57:17 <rubasov> but then I get strange test failures like referred in the wiki page 14:57:47 <mlavalle> it is possible we have a bug in Pecan 14:57:59 <mlavalle> I mean our code based on Pecan 14:58:22 <mlavalle> becuase I also faced some weird stuff with multiple port binding 14:58:22 <rubasov> mlavalle: so you're thinking the api-def is right as is 14:58:32 <mlavalle> yes 14:58:36 <rubasov> okay 14:58:59 <rubasov> I can continue in that direction I just wasn't sure if that's the direction to take 14:59:03 <amotoki> I think we need to check when member actions are registered. 14:59:26 <amotoki> if they are specified when a resource is registered, it might not work. 14:59:45 <rubasov> amotoki: they are, I think 15:00:14 <rubasov> that's why I had to add something to RESOURCE_ATTRIBUTE_MAP too 15:00:38 <mlavalle> ok, top of the hour 15:00:44 <rubasov> thank you guys 15:00:45 <mlavalle> thanks for attending! 15:00:47 <ralonsoh> thanks 15:00:49 <bcafarel> thanks! 15:00:50 <mlavalle> #endmeeting