14:01:11 #startmeeting networking 14:01:12 Meeting started Tue Feb 2 14:01:11 2016 UTC and is due to finish in 60 minutes. The chair is ihrachys. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:01:14 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:01:16 The meeting name has been set to 'networking' 14:01:23 o/ 14:01:23 o/ 14:01:25 hi all, thanks for joining in this early hour (for some of you) 14:01:37 as you now probably know, we are back to doing bi-weekly alternating meetings 14:01:45 #link https://review.openstack.org/#/c/272494/ 14:01:54 and today is the first Tue meeting in the new year, yay! 14:02:08 aloha 14:02:11 (thanks everyone to make it revived) 14:02:18 hi 14:02:25 Armando leans toward delegation [or maybe just some more sleep] :), so today I will run this meeting for you. 14:02:31 Hello 14:02:39 and probably some more in the future. We’ll see. 14:02:49 ok... so... 14:02:52 #link https://wiki.openstack.org/wiki/Network/Meetings Agenda 14:02:59 ihrachys: thanks for reviving!:) 14:03:05 +1 14:03:06 today we’ll try to keep the meeting focused, as usual 14:03:26 the main goal for today is to follow up on milestone3 deliverables that we have not managed to cover the previous week 14:03:40 I hope other sections won’t take much time. But they always say that. ;) 14:03:47 but first, some announcements 14:03:52 #topic Announcements 14:04:01 I remind everyone that we have a neutron mid-cycle in the cold and unforgiving Minnesota end of Feb 14:04:13 #link http://lists.openstack.org/pipermail/openstack-dev/2016-January/084672.html 14:04:23 brrrr 14:04:32 right. I won't go. :P 14:04:37 so get some warm clothes and get ready ;) 14:04:49 :) 14:05:08 also, we have a code sprint for rolling upgrades, specifically for getting versioned objects for neutron core resources, in Brno mid March 14:05:12 #link http://lists.openstack.org/pipermail/openstack-dev/2016-January/085328.html 14:05:27 I will update the sprint etherpad with some travel tips later today 14:05:30 how warmer is Brno?! :) 14:05:37 thanks ihrachys 14:05:41 that one: 14:05:41 #link https://etherpad.openstack.org/p/code-sprint-neutron-objects-brno 14:05:42 compared to Minnesota is like hot 14:05:58 mhickey_: it's 12 degrees Celsius right now here, pretty fine for start of Feb :) 14:06:08 note: it's PLUS ;) 14:06:13 ihrachys: sure is! :) 14:06:14 :) 14:06:21 it's pretty warm everywhere in Europe this year 14:06:23 warmer than tokyo 14:06:36 hi, sorry for being late :( 14:06:52 in March, sometimes you can walk in shorts. but no guarantees attached ;) 14:06:57 akamyshnikova: hi! :) 14:07:03 final announcement is: call for papers for Austin is now closed as of yesterday 14:07:13 next stop: voting for talks and actually delivering them for those who will get through the filter. :) 14:07:16 ihrachys: closing today 14:07:24 it was extended 1day 14:07:31 ihrachys its open till today 14:07:36 mhickey_: oh? I missed the extension. 14:07:42 amotoki: yes, thanks 14:07:51 ihrachys: there was an issue yesterday 14:07:52 ihrachys: some issues with the submission website apparently 14:08:15 late o/ (sorry) 14:08:31 ok, then we are still good to do final update. 14:08:42 ok. it's all from me. no more announcements anyone? 14:09:05 FYI: The new deadline to submit a talk is February 2 at 11:59pm PST (February 3 at 7:59 UTC) 14:09:06 ok, let’s move on 14:09:12 mhickey_: thanks 14:09:13 #topic Bugs 14:09:19 hi 14:09:24 mlaville, the stage is yours. anything worth team attention from the previous week? 14:10:01 sorry, mlavalle 14:10:04 it was a relatively quiet week.... Many bugs as usual, but not very many of importance 14:10:19 worth highlighting: https://bugs.launchpad.net/neutron/+bug/1540411 14:10:21 Launchpad bug 1540411 in neutron "kilo: ValueError: git history requires a target version of pbr.version.SemanticVersion(2015.1.4), but target version is pbr.version.SemanticVersion(2015.1.3)" [Critical,In progress] - Assigned to Dave Walker (davewalker) 14:10:46 we have this critical one. It affects many projects, it is well known an a solution is being worked on 14:10:49 I think we struggle to merge https://review.openstack.org/271396 now 14:11:07 because there are some issues with linuxbridge job in both stable branches lately 14:11:16 I need to look at specific causes 14:11:29 njohnston_: I see you are the bug deputy for this week. thanks! 14:11:38 we need more volunteers for the upcoming weeks. anyone willing to cover for the week starting Feb 8? Feb 15? 14:11:52 ihrachys: my pleasure! 14:11:54 ihrachys: I can do the week after njohnston_ 14:12:03 mhickey_++ 14:12:10 mhickey_: ok, please put your name in the Agenda page 14:12:14 #link https://wiki.openstack.org/wiki/Network/Meetings 14:12:19 njohnston++ 14:12:19 :) 14:12:23 specifically, https://wiki.openstack.org/wiki/Network/Meetings#Bug_deputy 14:12:27 ihrachys: sure. I am green so may need some help? 14:12:47 mhickey: I'm green too, I'll share what notes I have 14:12:50 mhickey_: that's fine, let's talk during the week on the way forward. :) 14:12:54 mhickey_, we can help, no worries. Thanks for stepping up 14:12:54 mhickey_, I recommend you syncing with the exiting bug deputy by the end of the week :D 14:12:58 njohnston covers the week of Feb 1. mhickey_ will cover Feb 8. 14:13:06 anyone for the week starting Feb 15? 14:13:28 I can take it ihrachys 14:13:36 rossella_s++ 14:13:43 thanks folks, now we are packed for some more weeks! 14:13:46 njohnston, ihrachys, rossella_s, ajo: thanks 14:13:48 mhickey_: There is a nice document in devref that works you through the process. Ping me 14:13:49 thanks to all volunteers. let’s move on to the juicy bits. 14:13:57 #topic Docs 14:14:04 mlavelle: ok, sure, thanks 14:14:06 Sam-I-Am: anything we should be aware of? 14:14:10 hey 14:14:13 hey 14:14:27 the ongoing mtu stuff 14:14:38 not necessarily a docs thing, but its what i was working on for a while 14:15:00 something we should probably solve for mitaka 14:15:06 Sam-I-Am: ok, I guess we proceed in mailing thread for now and then we'll see what's the way out of the maze, right? 14:15:38 Sam-I-Am: depending on the scope of the changes. if it's just enforcing same mtu for the whole data path in compute hosts, it may be fine for Mitaka. 14:15:39 i think we know how to fix it for the built-in drivers, it just a matter of... do we need a spec? who does the work? etc 14:16:01 for the reference, it's around 14:16:03 there was a spec from a couple releases ago that wasnt fully implemented 14:16:04 #link http://lists.openstack.org/pipermail/openstack-dev/2016-January/084241.html 14:16:36 Sam-I-Am: ok cool, anything else? 14:16:54 yeah 14:17:12 for the folks that werent at last weeks meeting, the networking guide is now versioned 14:17:31 meaning people can go back to versions for prior releases of openstack 14:17:52 Sam-I-Am, : very nice 14:17:54 piece of cake 14:18:15 mitaka version of networking guide is found at http://docs.openstack.org/draft/draft-index.html 14:18:25 other than that, currently working on updates for mitaka, which includes debating the best way to add provider network support to the existing scenarios 14:18:35 The draft index is not linked from the top page of docs.o.o 14:18:49 seems like a lot of people want "hybrid" options - boot VMs on provider networks and self-service/private networks 14:19:07 right now the classic scenarios only support booting VMs on private networks 14:19:18 Sam-I-Am, yes, I've listened to that kind of feedback too 14:19:54 the hybrid scenario gets a bit more complex for first-timers... just another set of diagrams for attaching VMs to provider nets 14:19:55 Sam-I-Am: some people don't take on the new tenant-y thing real quick 14:20:14 the other option is making another set of scenarios, but then they just proliferate :/ 14:21:02 Sam-I-Am: I am also working on a chapter for the new external DNS integration functionality. I will push first version for review later this week 14:21:16 my thinking now is just add on to the existing scenarios, sort of like we do with the install guide... one option built on another or somesuch 14:21:59 mlavalle: cool 14:22:01 ok, I guess we need to move. thanks Sam-I-Am. and now for the most juicy part... 14:22:10 #topic Blueprints 14:22:20 prev week, we started walking thru the BPs targeted for M3. we have not completed the list though. 14:22:31 so let’s do it. :) 14:22:36 #link https://launchpad.net/neutron/+milestone/mitaka-3 14:22:52 and the first one is 14:22:53 #link https://blueprints.launchpad.net/neutron/+spec/vm-without-l3-address 14:23:13 ajo: can you update us on ^? 14:23:28 I see some pieces are merged 14:23:35 ihrachys, give me a sec, you got me off guard :) 14:23:45 can we move on, and get back to this one in a minute? :) 14:24:01 ok, let's move. we'll get back to it. 14:24:12 #link https://blueprints.launchpad.net/neutron/+spec/ml2-lb-ratelimit-support 14:24:22 it's on me (and slawek) 14:24:55 I think we are still waiting for fullstack resources for linuxbridge, but apart from it, the qos patch itself is good to go. so hopefully we get it in M. 14:25:05 the next is... 14:25:06 #link https://blueprints.launchpad.net/neutron/+spec/add-tags-to-core-resources 14:25:27 amotoki: ^ any update? 14:25:40 hichihara volunteered for implementations. 14:25:47 Yes. I guess that the spec will be merged soon. 14:25:56 I need to do the final review on -spec. 14:25:57 I have posted PoC patch 14:26:15 do we think we get it in M? 14:26:28 ihrachys: Of cource. 14:26:29 I think we can. 14:26:43 that's cool. 14:27:06 note everyone M3 is sometime start of March, so we have a month to get clear on all deliverables for M. 14:27:23 #link https://blueprints.launchpad.net/neutron/+spec/add-timestamp-attr 14:27:31 we can revisit the status two weeks later 14:27:43 I suspect kevinbenton is not here. 14:27:57 anyone has the status for the bp? 14:28:17 I see some 3 patches in review at https://review.openstack.org/#/q/topic:bp/add-port-timestamp,n,z 14:28:32 week of feb 29 is m3: http://docs.openstack.org/releases/mitaka/schedule.html 14:28:52 anteaya, thanks for the reminder 14:28:57 anteaya: thanks for the correction! so yeah, even less time to push code in. 14:29:17 or review patches, depending on how you look at it 14:29:25 * ajo tightens his seatbelt 14:29:30 ok, I guess we don't have anyone here to update on the add-port-timestamp spec. let's move on. 14:29:38 #link https://blueprints.launchpad.net/neutron/+spec/network-ip-usage-api 14:29:56 that's also on kevinbenton 14:30:23 the patch is under active review 14:30:24 #link https://review.openstack.org/#/c/212955/ 14:30:38 and I see the right folks shaving it :) 14:30:53 ok, next 14:30:55 yaks gotta be shaved 14:31:01 #link https://blueprints.launchpad.net/neutron/+spec/improve-dvr-l3-agent-binding 14:31:13 carl_baldwin: obondarev: what's on that one? ^ 14:31:14 implemented 14:31:28 last patch was merged yesterday 14:31:37 thanks to reviewers 14:31:46 o/ 14:31:47 good to hear that :) 14:32:04 :) 14:32:06 obondarev: what's the https://review.openstack.org/#/c/143169/ then? 14:32:11 yay! 14:32:12 :) 14:32:20 it has the topic referring to the bp 14:32:33 ihrachys: it's by mistake I guess 14:32:46 it was based on dvr refactiring 14:33:11 that's great news actually. thanks to reviewers and obondarev for driving it. 14:33:26 obondarev: that was re-based on top of yours, so maybe picked up the tag 14:33:31 obondarev: could you update the topic of 143169? 14:33:45 yeah, probably. better clear the topic not to confuse. 14:33:46 amotoki: sure 14:33:59 ok cool. we have stuff delivered woot. 14:34:03 #link https://blueprints.launchpad.net/neutron/+spec/address-scopes 14:34:20 I guess carl_baldwin is not here, but salv-orlando is ;) 14:34:28 salv-orlando: are we far for ^ ? 14:35:33 the whiteboard is too long.... 14:35:38 We are at the stage where the patches are good to go in 14:35:52 and I thought they were already in honestly as they fell off my review dashboard 14:35:57 salv-orlando: do we have a single topic for all of them? as amotoki noted, the whiteboard is a graveyard 14:36:11 https://review.openstack.org/#/q/topic:bp/address-scopes,n,z shows 3 left, we're pretty close 14:36:17 ihrachys: yes, there is ^ 14:36:46 that's an impressive list of patches already merged. I bet you get it in M. :) 14:36:56 thanks for everyone driving it 14:37:13 haleyb: that's 4 actually but besides th3 l3 agent support they're all little things 14:37:36 and on the other hand without the l3 agent support the feature is quite... non-existent ;) 14:37:48 :) 14:37:54 ah yes, one in merge conflict, the first one is the last big one really 14:38:17 gerrit seems to not be cooperating to look further 14:38:38 gerrit-- 14:38:44 it's back 14:39:00 #link https://blueprints.launchpad.net/neutron/+spec/lbaas-l7-rules 14:39:18 I am surprised it's not on dougwig! :) mestery, ^ 14:39:45 proceeding nicely 14:39:50 :) 14:39:54 ihrachys: For that one, I know that sbalukoff was telling me lovely stories abouti t last week 14:40:03 xgerman likely also knows the tale I speak of 14:40:13 that's a huge list of things to merge though: 14:40:14 yeah, we have the code - need to review 14:40:15 #link https://review.openstack.org/#/q/topic:bp/lbaas-l7-rules,n,z 14:40:39 I have to comment one thing on l7-rules. it should be implemented as a separate extension 14:40:51 it is 14:41:04 as an extension of LBaaS 14:41:06 https://review.openstack.org/#/c/148232/42/neutron_lbaas/extensions/l7.py 14:41:11 you mean top-level? 14:41:27 xgerman: let me check later. 14:41:32 ok 14:41:58 ok, next 14:42:06 #link https://blueprints.launchpad.net/neutron/+spec/rbac-qos 14:42:41 hdaniel posted a new lovely metaclass-y patch for db models and objects for RBAC QoS Policy 14:42:56 code looks lovely IMHO :) 14:42:57 but I suspect it's not enough to land it? is it ajo? 14:43:14 #link https://review.openstack.org/#/c/250081/ 14:43:24 ihrachys, I need to do a final pass of review, my initial pass was good 14:43:59 ok, but do we have client and API covered? 14:44:23 ihrachys, I'm unsure if API or clients needed changes, let me check, 14:44:34 note that the client release deadline is earlier than server projects 14:44:39 I think so. otherwise how would it know which objects support RBAC? 14:44:52 amotoki: oh. any specific date? 14:44:54 please make sure to land patches before requirement deadline 14:45:03 no specific date 14:45:23 but from past experience it is a bit earilier than X-3 cut. 14:45:42 ajo: I think that may actually postpone the thing to N. we need to get everything in place in time or we slip forward. 14:45:51 amotoki: thanks for the heads-up 14:46:17 ajo: can you work with hdaniel to get it all set? 14:46:36 ihrachys, , yes, but we even have API tests 14:46:41 so I guess that's all good 14:47:00 and I think the neutron client is independent, it will just check the object type through the api request 14:47:08 I will double check all that with hdaniel btw 14:47:23 ajo: ok we need to check, especially the client. 14:47:35 thanks ajo and hdaniel 14:47:38 next is... 14:47:38 I'm quite sure that's right, but will double check 14:47:42 client support for qos-rbac would be simple, and I have no concern. 14:47:52 #link https://blueprints.launchpad.net/neutron/+spec/vlan-aware-vms 14:48:06 rossella_s: I bet you can update on vlan awareness 14:48:31 who's aware of vlan awareness? O:) 14:48:55 ihrachys, spec was approved 14:49:02 ihrachys, it took quite some time 14:49:19 I suspect it's not M deliverable since only the spec is merged and it sounds like it will require a lot of code changes. 14:49:24 rossella_s: amirite? 14:49:39 ihrachys, I think so too 14:49:58 I wonder whether we should start pushing things like that to N without waiting for M3 14:50:06 need to check with armax 14:50:18 ok, back to ajo's bp 14:50:19 #link https://blueprints.launchpad.net/neutron/+spec/vm-without-l3-address 14:50:45 ajo: give us some hope :) 14:51:12 yeah :) 14:51:43 ok, the neutron side looks much better after a rearrangement of the limit-portsec field 14:52:10 It's in merge conflict state, but yalie addressed all my comments so far, I need to check the latest patch 14:52:25 so, the neutron side of things probably will go in M 14:52:36 do we have client? 14:52:38 but, the nova side wasn't accepted for this cycle 14:52:55 yes: https://review.openstack.org/#/c/218388/ 14:53:05 being worked out 14:53:43 I guess that -2 from armax is obsolete? 14:54:28 if server side patch is good enough from secrutiy point of view, I believe he can drop his -2 14:54:39 correct, I agree with amotoki 14:54:53 but we need to consider his point of view 14:55:00 ok, I will leave you to push for M3 then :) good luck. 14:55:07 and... that's about it for bps 14:55:13 #topic Open Discussion 14:55:29 we have 5 mins only, but if someone has anything to bring up, you are welcome 14:55:41 can someone set milestone for prefix delegation bug https://bugs.launchpad.net/neutron/+bug/1505316 14:55:42 Launchpad bug 1505316 in neutron "compute port lose fixed_ips on restart l3-agent if subnet is prefix delegated" [Medium,In progress] - Assigned to venkata anil (anil-venkata) 14:56:04 we have now keystone v3 devstack http://lists.openstack.org/pipermail/openstack-dev/2016-February/085497.html 14:56:15 amotoki: v3 only? 14:56:34 anilvenkata: why do we need a specific milestone? it's not a High prio bug, right? 14:56:45 (at least not in LP) 14:56:54 it is ongoing. i see problems in some other projects (not neutron) 14:57:19 i don't know the full status on this... 14:57:35 ihrachys: prefix delegation is broken, that patch will fix it 14:57:40 FYI: As far as the mid-cycle meetup in Rochester, MN. I live there. Any questions on the location ... can contact me offline. 14:57:54 anilvenkata: ok, let's raise its Importance then to High I guess 14:57:56 jckasper: thank you 14:58:07 ihrachys: Because this now has the 2 dependencies explicitly listed as dependencies, is the DSCP work eligible for lifting the -2? https://review.openstack.org/#/c/251738 14:58:10 IIRC now we have keystone v3 by default 14:58:12 ihrachys: thanks Ihar 14:58:14 jckasper: is it as cold as they say?! :) 14:58:14 jckasper: you can post your name in the etherpad for the event suggesting help there. 14:58:18 I have something I would like to share 14:58:27 njohnston: it is. I will remove it. 14:58:37 ihrachys: Thanks! 14:58:43 anteaya: shoot 14:58:47 thank you 14:59:02 I'm concerned with what I percieve to be the splintering of communication 14:59:03 amotoki : Yes, v3 is by default enabled. 14:59:19 both within this project and with this project's relationships with other projects 14:59:31 I was surprised to see this meeting take place in this time slot 14:59:48 as well as no mention of items covered in last week's nova mid-cycle 14:59:48 anteaya: why? 14:59:53 anteaya: why? we had it before at that same time right? 15:00:03 as well as a new mid-cycle announced for march 15:00:17 you are welcome to disagree with me 15:00:18 and we had a good announcement on -dev list 15:00:26 anteaya: which March midcycle? 15:00:35 but I think neutron is drifting farther and farther away from other projects 15:00:51 ihrachys: how many march mid-cycles were mentioned at this meeting? 15:00:53 anteaya: ok let me bring it up to armax 15:01:03 I have voiced my concerns 15:01:07 thanks for listening 15:01:07 anteaya: that's code sprint, it's a bit different in its scope. 15:01:16 it is not mnid-cycle. it is just code sprint for a specific topic 15:01:17 thanks folks! time to end the meeting 15:01:21 #endmeeting