18:02:35 #startmeeting networking_policy 18:02:36 Meeting started Thu Nov 19 18:02:35 2015 UTC and is due to finish in 60 minutes. The chair is SumitNaiksatam. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:02:37 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:02:39 The meeting name has been set to 'networking_policy' 18:02:59 #info agenda https://wiki.openstack.org/wiki/Meetings/GroupBasedPolicy#Nov_19th.2C_12th_2015 18:04:04 not everybody seems to have adjusted to the changed time here i believe :-) 18:04:10 #topic Bugs 18:04:15 there are no critical bugs 18:04:27 but i wanted to bring up a couple and their fixes 18:04:56 there was an issue with not being able to clear/unset the protocol field from the CLI 18:05:36 while investigating that it led me to a few other issues including one because of which the setting/unsetting of boolean attrs was not working at all 18:05:55 the fixes are in this #link https://review.openstack.org/246081 18:06:17 also, in the past we discussed adding support for protocol numbers 18:06:36 magesh has a patch here: #link https://review.openstack.org/246895 18:07:00 i believe the above has been done in a backward compatible way 18:07:30 and is also a critical requirement for supporting VPN service 18:07:46 please help review and merge at the earliest 18:08:18 other than that there was this ODL sync up patch posted by yi #link https://review.openstack.org/230760 18:08:41 i am looking for a second core brave enough to +2 this :-) 18:09:02 are there any other critical issues 18:09:03 ? 18:09:47 also, the NAT pool support was added to the UI 18:10:02 with this all the workflows should be now possible through the UI 18:10:08 please validate if you get a chance 18:10:41 i don’t have an update on the functional/integation tests 18:10:50 #topic Packaging 18:11:19 rkukura: over to you if there are any updates 18:12:07 nothing new from me - any ETA for updated stable/juno and stable/kilo releases, or liberty support? 18:12:48 rkukura: yeah, i think for stable we had decided past monday 18:13:00 but i still bug fixes being ported 18:13:29 hence i did not post a new stable release 18:13:38 so at this point i am thinking next week 18:13:53 SumitNaiksatam: OK, thanks 18:14:04 regarding the liberty branch, the issue is that we have dependency on vendor packages 18:14:14 and those need to be updated as well :-( 18:14:42 the dependencies are in the test-requirements to be able to run the UTs 18:14:43 SumitNaiksatam: Are those dependencies needed for UTs and CI? 18:14:52 rkukura: yeah UT, not CI 18:15:26 so i had it on my to do list to take a stab at those as well, but have been distracted with a few other things 18:16:32 ok moving on 18:16:58 first, a couple of items which were on the agenda last week, but we ran out of time 18:17:07 #topic Long pending patches 18:17:34 the following patches from ivar are long pending review: 18:17:36 #link https://review.openstack.org/179327 18:17:49 #link https://review.openstack.org/192424 18:18:15 we need to make a call on those before we close liberty branch 18:19:02 rkukura: on the patch which refactors the required attributes, you had some comments about keeping it in sync with the CLI 18:19:13 rkukura: can you take a look again and see if we can proceed? 18:19:32 OK 18:19:43 These both will need rebasing as well. 18:19:48 rkukura: okay 18:20:04 yes, lets follow up with ivar 18:20:34 ivar-lazzaro: perfect timing 18:22:08 #topic Specs 18:22:19 we have a couple of pending specs 18:22:25 songole: hi 18:22:34 Hi SumitNaiksatam 18:22:39 songole: do you know if hemanth is around? 18:22:56 Hemanth is travelling 18:23:09 i wanted to check on #link https://review.openstack.org/239743 (Network Service Framework) spec 18:23:11 songole: ok 18:23:13 hi 18:23:45 songole: is it ready for review, or more work needs to be done? 18:23:54 if its the latter we should mark it as WIP 18:24:20 ok. I will ping hemanth. 18:24:39 cores/reviewers should take note of the spec anyway 18:25:06 ivar-lazzaro: there were a few review comments on #link https://review.openstack.org/242781 18:25:58 lookinbg 18:26:01 there was also this spec posted by magesh a long time back #link https://review.openstack.org/#/c/189629/ 18:26:32 i am not sure if this is still required 18:26:58 SumitNaiksatam: the cluster_id spec may need some rework 18:27:03 songole: i have put a comment in the spec, could you check this with magesh as well? 18:27:05 SumitNaiksatam: as the API itself 18:27:13 ivar-lazzaro: ah okay, sure 18:27:20 SumitNaiksatam: will do today. 18:27:24 songole: thanks 18:27:32 SumitNaiksatam: we are aware of some limitations that we may want to fix before it's too late (we release) 18:28:32 igordcard: thanks for posting the questions on the other specs, i abandoned them since there was not activity 18:28:42 we can revive them per interest and need 18:29:33 SumitNaiksatam, yeah, housekeeping 18:29:43 igordcard: :-) 18:30:00 and on that note, we can segue to your anticipated spec 18:30:09 #topic QoS policy support 18:30:26 i believe there were some follow up discussions after the IRC meeting last week 18:30:36 igordcard: any update on that front? 18:31:23 not yet, but as I talked a bit with ivar after the other meeting, it might be only for all traffic for now 18:31:53 igordcard: okay 18:32:16 with development regarding the neutron flow classifier that might be something to integrate with QoS, and then both with GBP 18:32:59 I am drafting an initial spec for QoS in GBP, between groups only, for now 18:33:11 it's all 18:33:26 igordcard: i think the GBP classifier is high level enough that is can be mapped to whatever flow classifier gets defined anywhere else 18:33:36 igordcard: thanks, looking forward to it 18:34:00 SumitNaiksatam, yes that would be good 18:34:00 #topic Open Discussion 18:34:45 other house keeping items, no weekly IRC meeting next week on account of Thanksgiving here in the US 18:35:18 we might also be take a break from the meetings in the last two weeks of Dec since people will be on vacation/travelling 18:36:06 let me know if anyone would like to volunteer with documenting the API in-tree ;-) 18:36:54 even if you are able to document a couple of resources, its good, i imagine its mostly going to be mechanical work 18:37:05 anything else we missed today?’ 18:37:34 SumitNaiksatam: do we have auto generated documentation? 18:38:04 SumitNaiksatam: how do we document APIs in tree? 18:38:07 ivar-lazzaro: unfortunately i am not aware of an easy way to do that for the API 18:38:17 ivar-lazzaro: but if you are, we can definitely follow that 18:38:43 SumitNaiksatam: no idea myself, was just wondering what the workflow for API documentation would be 18:38:44 i wanted to follow the lead of what other projects are doing in openstack 18:39:14 ivar-lazzaro: so in that context, we look at how some of the existing projects do it 18:39:46 and basically capture the request/response body for CRUD with resources 18:40:05 ok 18:40:07 either using verbose CLI or postman or something similar 18:40:43 i think the openstack documents uses WADL, but it was pretty painful to deal with it if you are new 18:40:57 so we can probably just do some rst formatting and get going 18:41:32 but i will post a skeleton for review, and perhaps that might make this more concrete 18:41:51 anything else anyone want to discuss today? 18:42:56 alrighty, i guess shorter meeting today 18:43:07 happy thanksgiving to everyone in advance! ;-) 18:43:11 thanks all! 18:43:32 thanks SumitNaiksatam! 18:43:36 bye! Happy thanksgiving! 18:43:39 bye 18:43:41 bye 18:43:44 #endmeeting