21:01:03 #startmeeting networking 21:01:04 Meeting started Mon Apr 27 21:01:03 2015 UTC and is due to finish in 60 minutes. The chair is mestery. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:01:06 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:01:09 The meeting name has been set to 'networking' 21:01:10 Hi 21:01:15 jello! 21:01:18 #link https://wiki.openstack.org/wiki/Network/Meetings Agenda 21:01:40 By announcing the agenda is lean today and I expect a short meeting, I'll likely doom us all to a full hour, but why not try. ;) 21:01:43 #topic Announcements 21:01:43 hi 21:01:51 #info Kilo-RC3 will release later today or tomorrow with 3 changes 21:01:57 The 3 changes are: 21:02:06 https://bugs.launchpad.net/neutron/+bug/1447883 21:02:06 Launchpad bug 1447883 in neutron kilo "Restrict netmask of CIDR to avoid DHCP resync is not enough" [Critical,Fix committed] 21:02:09 https://bugs.launchpad.net/neutron/+bug/1441107 21:02:10 Launchpad bug 1441107 in neutron kilo "Missing entry point for haproxy namespace driver" [High,Fix committed] 21:02:13 https://bugs.launchpad.net/neutron/+bug/1448813 21:02:13 Launchpad bug 1448813 in neutron kilo "radvd running as neutron user in Kilo, attached network dead" [High,In progress] - Assigned to Henry Gessau (gessau) 21:02:26 Waiting on 1448813 to merge in master so we can get it in stable/kilo, and then we'll cut RC3 21:02:38 Thanks to all the testing, in particular 1448813 was found by someone testing RC1! 21:03:05 Once we release RC3, we'll be done with Kilo. 21:03:13 A few more announcements 21:03:22 #info Reminder: neutron mid-cycles 21:03:24 #link http://lists.openstack.org/pipermail/openstack-dev/2015-April/060713.html 21:03:42 I've made some changes to the neutron-drivers team. 21:03:49 mestery: I added an item to the agenda of the sprint 21:03:49 #link Salvatore and Mark have been swapped out with Maru and Doug on the neutron-drivers team. 21:03:55 Sukhdev: Thank you! 21:04:03 Yes, please add items to the coding sprints as you see fit. 21:04:15 I've found the last two to be very useful for getting thigns done, and I've heard that feedback from others. 21:04:32 OK, I'll wait 60 seconds now in case anyone has any other announcements. 21:04:35 Before moving on. 21:04:57 mestery: The item I added has to do with Task Flow 21:05:09 In case others have interest and expertise in this area 21:05:11 Sukhdev: For ML2, right? 21:05:17 Yes 21:05:23 ML2 Sync and error handling 21:05:41 OK, lets move on 21:05:45 #topic Bugs 21:05:50 * mestery looks for enikanorov and/or enikanorov2 21:06:44 In lieu of enikanorov and/or enikanorov2, does anyone have any bugs they'd like to bring up for the broader team? 21:06:50 I realize we're in the "quiet" period of the cycle. 21:07:02 I know some of you relish hte peace and quiet :) 21:07:30 mestery: there's a storm brewing 21:07:37 salv-orlando: when isn't there 21:07:42 * marun goes to batten down the hatches 21:07:47 * mestery mixes a drink 21:07:49 :) 21:07:55 OK, lets move on! Maybe this will be a short meeting! 21:08:05 #topic Docs 21:08:07 we should talk about naming something. 21:08:12 Hello! 21:08:14 * mestery was hoping emagana would be here to give us an update on the Document sprint 21:08:17 you're here emagana! Great! 21:08:24 dougwig: I like quiet 21:08:38 Yes! Well, the good news is that we have done a terrific progress on the Networking Guide! 21:08:57 All seven scenarios have content! 21:09:03 most of them already merged! 21:09:04 Nice! 21:09:22 the bad news is that we did not have NEW folks helping during the docs day! 21:09:49 :( 21:09:50 emagana: end of cycle is a hard time for new people 21:09:55 a recrutiment failure 21:09:56 we should discuss during the summit how we "ensure" participation on the documentation part!! 21:10:08 emagana: perhaps a doc sprint in milestone-1? 21:10:09 emagana: ++! 21:10:21 Make atc codes based on docs alone :) 21:10:22 mestery: sorry, no update from me this weel 21:10:25 *week 21:10:26 yes, mestery and salv-orlando our post review indicated that we did not chose the best time for this! 21:10:32 enikanorov2: No worries :) 21:10:48 emagana: anteaya's idea about an earlier sprint may make sense 21:10:48 emagana: at least you got the practice of running it 21:10:52 anteaya: ++ 21:11:01 emagana: people are not joining because you're already doing 10 people's worth of work in the area 21:11:16 Absolutely, next time will be better anteaya ;-) 21:11:25 large groups need regular patterns 21:11:34 so whatever pattern you pick do that 21:11:41 either once per milestone 21:11:50 or milestone-1 and 2 or whatever 21:11:54 salv-orlando: Thanks... but it is mostly work done by the Docs Guide! 21:12:11 However, we STILL need you! 21:12:32 #info Future doc sprints to be done earlier in the cycle. 21:12:34 we need people with deployment experience that validate what we are saying on the current docs 21:12:35 hopefully the proposal to produce dev docs as we go will ease the way for user docs 21:12:37 Thanks for your continued work here emagana. 21:12:46 marun: That's part of the plan. 21:13:28 ok.. I will update wiki with existing open reviews but again, even what has been merged should eb reviewed! 21:13:50 #info Reviews of content still encouraged. 21:13:52 nothing more from me! 21:13:56 Thanks emagana! 21:14:10 #topic Liberty Design Summit 21:14:16 thanks emagana! 21:14:21 I hope to have a first cut of our summit schedule done by Friday 21:14:31 The etherpad no longer works for me in Firefox it's so full of stuff 21:14:32 #link https://etherpad.openstack.org/p/liberty-neutron-summit-topics 21:14:51 If you reject my session can I appeal to somebody? 21:14:59 mestery: try chromium 21:15:03 like the "board for openstack complains and various queries" 21:15:06 salv-orlando: Yes, me. :P 21:15:18 anteaya: :) 21:15:27 mestery: we should totally have a complaints department 21:15:38 I think most people know how these design summit's work, and how they are supposed to work. 21:15:46 I'm leaning to run this one like the latter rather than the former. 21:15:50 mestery: I have been having etherpad grief with firefox 21:15:55 salv-orlando: If we do, I nominate armax to lead it. 21:16:20 So, stay tuned for our schedule. 21:16:29 I will operate the arguments table right beside armax's table 21:16:31 mestery: pretty sure armax will be in jail for GBH in a few weeks 21:16:47 lol 21:17:10 arm wrestle for spec acceptance at the neutron pod 21:17:16 I am not sure if people are serious but I am starting to worry 21:17:21 salv-orlando: As long as his review stats remain high, to me it doesn't matter from where armax works :) 21:17:26 anteaya: trying clearing your cache. I too had recent issues with etherpad. 21:17:27 gbh? 21:17:30 sc68cal: ha ha ha 21:17:53 cbouch: thanks 21:18:03 OK lets move on 21:18:07 I have one more topic before Open Discussion today 21:18:17 #topic New Feature Request Process for Liberty 21:18:20 #link https://review.openstack.org/#/c/177342/ 21:18:33 That review proposes a new way to get features into neutron rather than neutron-specs 21:18:35 dougwig: http://en.wikipedia.org/wiki/Grievous_bodily_harm (sorry I forgot some things apply only to one side of the atlantic_ 21:18:37 I encourage people to review it 21:19:04 mestery isn't happy until gerrit is on fire. 21:19:09 Thanks to those who have already reviewed (marun, blogan, dougwig, rossella_s, sc68cal, etc.) 21:19:25 amuller as well 21:19:45 The process we have in place wasn't working in many regards, thus I'm changing it so we get out of the way 21:19:46 mestery: I read it but not voted, does that count? 21:19:46 No deadlines 21:19:48 Streamlined 21:19:49 :) 21:19:55 mestery: I'll ping you next time gerrit is stuck on threads it is exciting 21:19:56 armax: Now you're just teasing me ;) 21:20:24 * marun stands by to put the specs repo to the torch 21:20:39 mestery: I am ok with slimming the process. The only open question is what happens to the public face of neutron-specs which is published on specs.openstack.org 21:20:43 marun: I get to submit that gerrit review 21:20:53 * kevinbenton 21:20:57 * marun stands by to hand mestery the torch 21:21:05 I am ok with just saying we won't publish specs anymore btw 21:21:07 salv-orlando: Yes, I saw your comment. I think it will remain for those specs pre-Liberty. The ones approved for Liberty I will migrate just to LP. 21:21:12 i almost want to -2, just to kill their fun. 21:21:13 salv-orlando: Excellent 21:21:27 mestery: Should we move them over into devref? 21:21:34 salv-orlando: the design part of specs will be in-tree and published with devref 21:21:36 mestery: at some point in the future? 21:21:39 sc68cal: Not in their current forms, but in some form, yes. 21:22:22 mestery: one thing to note, if we stop using specs, that means we need to move the API docs too, since docs team have been moving things out of api-site 21:22:24 sc68cal: they are dead 21:22:28 specs repo is a graveyard 21:23:02 sc68cal: Let me follow up with you on that. 21:23:04 we need docs, but ones that are current and maintained with what's in-tree 21:23:29 https://bugs.launchpad.net/neutron/+bug/1438707 21:23:29 Launchpad bug 1438707 in neutron "Security Group API is undocumented in neutron-specs" [Undecided,Incomplete] 21:23:42 there's specs, and API Reference 21:23:59 sc68cal: something fishy happened there with API docs which was moved to WADLs 21:24:03 they're just not rendered anymore 21:24:14 salv-orlando: I think that's the problem. 21:24:18 salv-orlando: ah 21:24:42 sc68cal: we are not moving things out of api-site, what makes you think that? 21:25:28 annegent_: I think because I saw the extra_route API extension get deleted in api-site, but then rendered over in specs.openstack.org 21:25:38 * sc68cal digs up mailing list post 21:25:52 Why are API docs rendered to specs.openstack.org? 21:26:30 I see them at the bottom there when I got to specs.openstack.org. 21:26:35 mestery: the narrative went there 21:26:41 annegent_: Ack 21:26:48 http://lists.openstack.org/pipermail/openstack-docs/2015-March/006156.html 21:27:41 Extra routes (and all extensions) are here: http://developer.openstack.org/api-ref-networking-v2-ext.html 21:27:48 OK, I encourage folks to review the new Feature Request process 21:27:53 In the review I linked above. 21:28:05 I hope to have a more streamlined process for adding features. 21:28:10 #topic Open Discussion 21:28:10 adn get to 100 comments 21:28:12 mestery: API reference is in api-site 21:28:15 blogan: lol 21:28:34 Also, I am working on a spec for API Reference info in Liberty and I'd love input: https://review.openstack.org/177934 21:28:50 #link https://review.openstack.org/177934 21:28:53 HenryG: I think someone re-added it back after I raised it on the ml 21:29:09 sc68cal: yes, me :) 21:29:16 HenryG: :) 21:29:55 I will review it annegent_ 21:30:07 I added myself as well annegent_, I'll take a peek, thanks! 21:30:40 not meeting item per se, but i was trying to add documentation for extension stuff, don't see any other extension api's listed on http://developer.openstack.org/api-ref-networking-v2-ext.html, so not sure where it should go, any one any ideas? 21:31:01 * anteaya pictures anne standing in a doorway partly entering a room but not quite 21:31:09 pritesh: Very relevant I think to the discussion HenryG and sc68cal just had :) 21:31:41 is this were I start the argument that extensions should not be considered anymore neutron APIs? 21:31:53 * mestery rushes to close the meeting 21:31:53 mestery: yep sorry missed that. 21:31:56 are we solving that in liberty? 21:31:57 ahahahahaha 21:32:00 v3? 21:32:03 pecan! 21:32:03 that was an evil laugh 21:32:09 * mestery looks at kevinbenton 21:32:10 salv-orlando: maybe we can just rename extensions to "api-maintainers" 21:32:10 I believe pritesh is talking about documenting new extensions for Kilo. 21:32:13 how does pecan solve the extensions issue? 21:32:14 * mestery and also blogan 21:32:23 * blogan hides 21:32:24 kevinbenton: I jest, why so serious? 21:32:36 OK, I think we'll wrap things up. 21:32:42 * kevinbenton walked into a trap 21:32:42 seriously, are we talking about this at summit? 21:32:47 dougwig: whatever as long as API mainteinance is given to the "bike shedding" sub team and I lead it 21:32:52 if that's not near the top of the list... 21:32:55 marun: yes, in the nuts and legumes session 21:32:55 marun: No, it's on salv-orlando gerrit review 21:33:04 oy 21:33:08 * marun goes to read 21:33:11 marun: salv-orlando doesn't want a summit session for this unless I'm mistaken 21:33:19 pritesh: can you get in touch with annegent_ for the docs? 21:33:20 I think we can resolve it on gerrit iwthout the need for a session as well 21:33:20 salv-orlando as the bike-shedding-as-a-service PTL 21:33:28 HenryG: sure will do 21:33:39 mestery: it depends on whether you want to have a thin hope of achieving anything 21:33:48 +1 for salv-orlando 21:33:57 salv-orlando: I think we know the answer then 21:34:07 mestery: right. 21:34:09 OK, we'll do one more meeting pre-Summit next week 21:34:15 hey is that the end of the meeting I see? 21:34:25 Then take a few weeks off exhausting ourselves by arguing in person in Vanvouver. 21:34:32 Thanks folks! 21:34:33 #endmeeting