22:01:12 <armax> #startmeeting neutron_drivers 22:01:13 <openstack> Meeting started Thu Oct 20 22:01:12 2016 UTC and is due to finish in 60 minutes. The chair is armax. Information about MeetBot at http://wiki.debian.org/MeetBot. 22:01:14 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 22:01:15 <ihrachys> kevinbenton: next one 22:01:16 <openstack> The meeting name has been set to 'neutron_drivers' 22:01:16 <armax> kevinbenton: next week's 22:01:19 <kevinbenton> ah 22:01:23 <armax> that’s ok 22:01:44 <armax> if your mindset is already at the summit we can have a short sync 22:02:45 <armax> right now the bulky of my energy is going into reviewing/revising the stadium 22:03:04 <armax> the list of open current assessments is available at the following link 22:03:05 <armax> #link https://review.openstack.org/#/q/status:open+project:openstack/neutron-specs+branch:master+topic:stadium-implosion 22:03:10 <njohnston> o/ 22:03:35 <ihrachys> armax: what's the plan for merging them? 22:03:38 <armax> I also started the shutdown procedure for neutron-pd-driver 22:03:40 <armax> https://review.openstack.org/#/q/status:open+topic:retire-neutron-pd-driver 22:03:43 <armax> #link https://review.openstack.org/#/q/status:open+topic:retire-neutron-pd-driver 22:03:50 <armax> ihrachys: I am glad you ask 22:04:21 <armax> my current line of thinking is the following 22:05:35 <armax> I was going to summarize the assessments in yet another (short) document 22:05:47 <armax> a side-by-side comparison if you will 22:06:01 <armax> that side 22:06:15 <armax> I think the assessments so far have unvelied a number of gaps across the board 22:06:23 <armax> while some gaps are more grave than others 22:07:36 <armax> I think it would be reasonable to decide collectively to let some projects go out of the stadium only if a subset of the criteria have no chance of completion by the first milestone of the current cycle 22:07:52 <armax> which is the week of Nov 14 22:08:55 <armax> tasks must not finish by the deadline but they must be started and show meaningful progress by then 22:09:35 <njohnston> currently the answers to criteria are expressed as a binary true/false; will they change to show items in progress? 22:10:06 <armax> the links attached to each criterion section should help depict a more complete picture 22:10:23 <armax> njohnston: and I’ll plan to refresh the assessment patches on an ongoing basis 22:10:26 <armax> until the deadline 22:11:15 <armax> but I’ll ask people to kindly to avoid nit-picking 22:11:46 <armax> because refreshing 10+ assessment is a pain I am willing to endure only if there’s something grossly captured 22:12:22 <armax> as for the subset of must haves 22:12:25 <armax> I intend to include 22:12:30 <armax> OSC bindings 22:12:58 <armax> so L1 22:13:02 <armax> S1 22:13:32 <armax> all of the R? requirements 22:14:21 <armax> C1, C2, C3 (db migration validation where it applies), a gating and stable C5 22:14:34 <armax> and C8 22:15:05 <armax> D1, D2, D3 22:15:39 <armax> and naturally all of the N? criteria 22:16:52 <armax> C5 is key to ensuring that moving APIs into neutron-lib does smoothly and can be validated, for those projects that expose APIs 22:17:19 <armax> the patch that shows how that takes place is https://review.openstack.org/#/c/353131/ 22:17:53 <armax> I will assist the various projects in the review process, and I’ll proceed over the course of the cycle to move the neutron in-tree APIs over too 22:18:30 <armax> once that’s done, the stadium evolution effort can be considered officially complete 22:19:01 <njohnston> So just to be clear, stadium projects won't contain their own api definitions anymore, they will live in neutron-lib? That is not something I realized before now. 22:19:17 <armax> njohnston: yes, and wow! 22:19:35 <armax> njohnston: that’s been something I advertsized from day 0 22:19:54 <kevinbenton> not very well :) 22:19:57 <armax> being part of the stadium for a project means giving up their control over the API 22:20:17 <armax> kevinbenton: can’t help people if they skim documents while reading :) 22:20:39 <ihrachys> people were too busy to struggle for their place inside 22:20:46 <armax> http://specs.openstack.org/openstack/neutron-specs/specs/newton/neutron-stadium.html#when-is-a-project-part-of-the-stadium 22:21:12 <armax> http://specs.openstack.org/openstack/neutron-specs/specs/newton/neutron-stadium.html#how-do-we-build-a-networking-community 22:21:41 <armax> though we settled with neutron-lib rather than creating yet another project repo 22:23:24 <armax> comments? Thoughts? Virtual rotten tomatos? 22:23:28 <armax> tomatoes? 22:23:35 <njohnston> Do api tests also move into neutron-lib as well, then? 22:23:42 <armax> njohnston: no 22:23:57 <njohnston> ok 22:24:04 <armax> for a reason 22:24:24 <njohnston> that makes sense to me, but I just wanted to be sure :-) 22:24:24 <armax> if we did that, we can accidentally merge nn bw compatible changes 22:24:40 <armax> and we want to make sure that doesn’t happen 22:25:29 <njohnston> indeed 22:25:51 <armax> so we have 3 weeks, summit one included 22:26:12 <armax> to make sure that the must have’s are well underway 22:26:19 <armax> for those projects that don’t ACK 22:26:25 <armax> they are being let go 22:26:56 <dasm> question about api part in neutron-lib, but i'd like to hear confirmation/denial. subprojects are allowed to move their apis, during release cycle, without any objections? to prepare for future inclusion? 22:27:26 <armax> dasm: at this point we’ll have a quick pass/amnesty 22:27:54 <armax> dasm: but of those that have APIs I can only see midonet, bgpvpn, and sfc 22:28:02 <armax> because I think l2gw is on track for removal 22:28:06 <armax> as well as vpnaas 22:28:12 <dasm> ok, ack 22:28:14 <ihrachys> armax: fwaas? 22:28:16 <armax> and a few others that have no APIs 22:28:47 <armax> ihrachys: fwaas has 3 more weeks to tick all the must haves 22:28:54 <armax> if they do, great 22:28:57 <armax> if they don’t 22:29:07 <armax> well, you know the answer 22:29:36 <armax> ihrachys: or are you referring to something else in particular? 22:30:09 <ihrachys> merely saying that you haven't mentioned fwaas while it may be in ;) 22:30:19 <njohnston> :-) 22:30:54 <armax> I didn’t mention it because the API was agreed and merged in neutron-specs 22:31:12 <armax> with this effort we would have shrank the stadium to 22:31:37 <armax> ~6 projects neutron/neutron-lib/neutron-specs excluded 22:32:04 <armax> and that’s something a single PTL with his valiant knights can manage well 22:32:31 <armax> and that can give us some headroom for growth 22:32:32 <mfranc213> or her :) 22:32:56 <armax> mfranc213: yes, indeed 22:34:33 <armax> I would love for the individual assessment to be done and merged by next week’s end 22:34:57 <armax> or in the early days of the week after the summit 22:35:25 <ihrachys> do you get enough reviews to proceed? 22:35:41 <armax> I did get feedback from the people I expected 22:35:49 <armax> for the feedback I didn’t get 22:35:50 <armax> well 22:35:54 <armax> that’s feedback too ;) 22:36:12 <ihrachys> especially if it's lack of feedback from project owners I guess 22:36:19 <armax> correct 22:36:33 <armax> mostly were pretty responsive 22:36:49 <armax> one project which I fear might be on the way to its demise right off its infancy 22:36:53 <armax> is dynamic-routing 22:37:24 <armax> but we can cross that bridge when we get to it 22:38:00 <armax> I am unclear at this point who the core go-to people are 22:38:03 <amuller> armax: do you think we should organize a working session at the summit and do live reviews of your patches? 22:38:17 <armax> amuller: there’s a stadium session I reserved 22:38:18 <ihrachys> armax: vikram? 22:38:29 <amuller> ack 22:38:34 <armax> ihrachys: yes, vikram but again it can’t be a one man gig 22:38:42 <ihrachys> he seems like the only person showing up 22:38:44 <armax> amuller: though we have the Friday too 22:38:51 <amuller> yeah I was thinking about Friday 22:38:53 <armax> which I personally going to be available all day 22:39:19 <armax> I have been in Barcelona like 10 times, I don’t need any sightseeing 22:39:23 <armax> :) 22:39:38 <armax> though I wouldn’t mind doing the reviews at a caffee in plaza major 22:39:56 <armax> which actually is in Madrid, so never mind ;) 22:41:08 <armax> comment/questions? 22:41:25 <kevinbenton> armax: perhaps you do need some sight seeing 22:42:18 <armax> kevinbenton: we’ll see 22:42:36 <armax> ok folks 22:42:45 <armax> if there’s nothing else 22:43:07 <dasanind_> There are 2 rfe's that need inputs 22:43:11 <armax> I’ll leave you to it, and I wish you a safe and uneventful journey for those of you who go to the summit 22:43:13 <dasanind_> https://bugs.launchpad.net/neutron/+bug/1633280 22:43:15 <openstack> Launchpad bug 1633280 in neutron "[RFE]need a way to disable anti-spoofing rules and yet keep security groups" [Undecided,Opinion] - Assigned to Rui Zang (rui-zang) 22:43:23 <dasanind_> https://bugs.launchpad.net/neutron/+bug/1634798 22:43:24 <openstack> Launchpad bug 1634798 in neutron "[RFE] Qos DSCP to vlan priority mapping" [Undecided,New] 22:43:25 <armax> dasanind_: RFEs are low priority right now 22:43:40 <ihrachys> dasanind_: note Ocata is low on features 22:43:41 <armax> we’ll get to them when we get to them 22:44:27 <dasanind_> armax: makes sense. Should they be marked as wishlist or leave them undecided 22:45:04 <armax> dasanind_: RFEs are by definition wishlist 22:45:22 <armax> if they are not confirmed leave them alone until a member of the drivers team get a pass at them 22:45:32 <dasanind_> armax: thank you 22:46:20 <armax> ok 22:46:23 <armax> good bye! 22:46:27 <armax> #endmeeting