18:01:54 <SumitNaiksatam> #startmeeting networking_policy 18:01:55 <openstack> Meeting started Thu Jun 16 18:01:54 2016 UTC and is due to finish in 60 minutes. The chair is SumitNaiksatam. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:01:56 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:01:59 <openstack> The meeting name has been set to 'networking_policy' 18:02:18 <SumitNaiksatam> #info agenda https://wiki.openstack.org/wiki/Meetings/GroupBasedPolicy#June_16th.2C_2nd.2C_May_26th.2C_2016 18:02:39 <SumitNaiksatam> same agenda from past weeks 18:03:15 <SumitNaiksatam> nothing new on the bugs, documentation, testing front 18:03:22 <SumitNaiksatam> rkukura: anything to discuss on the packaging front? 18:03:28 <rkukura> nope 18:03:32 <SumitNaiksatam> okay 18:04:02 <SumitNaiksatam> to be accurate, there were bugs which got fixed during the last week, one was critical regarding DB migration breaking on Suse 18:04:18 <SumitNaiksatam> tbachman: hi 18:04:24 <tbachman> SumitNaiksatam: hi! 18:04:57 <SumitNaiksatam> there were a couple of other bugs and fixes in review: 18:05:03 <SumitNaiksatam> #link https://review.openstack.org/328821 18:05:18 <SumitNaiksatam> in case if anyone is against using the admin context here ^^^ 18:05:24 <SumitNaiksatam> and second one: 18:05:35 <SumitNaiksatam> #link https://review.openstack.org/328818 18:05:50 <SumitNaiksatam> without the above fix, GBP does not work with Newton 18:06:39 <SumitNaiksatam> but its debatable whether we should merge the above patch as a fix first, an do the real fix as follow up 18:06:55 <SumitNaiksatam> i believe the author of the patch has agreed to investigate, so thats good 18:07:32 <SumitNaiksatam> igordcard: anything new at your end that you need to discuss on QoS patches (I think you planned to post a follow patchset) 18:08:31 <igordcard> SumitNaiksatam: no, not yet :( 18:08:39 <SumitNaiksatam> igordcard: okay, np 18:08:52 <igordcard> SumitNaiksatam: probably not next week as well 18:08:58 <SumitNaiksatam> igordcard: okay 18:09:07 <SumitNaiksatam> #topic NFP impl patches 18:09:10 <SumitNaiksatam> hemanthravi: over to you 18:09:36 <hemanthravi> don't have an update, need reviews on the nfp patches 18:09:49 <hemanthravi> listed at https://wiki.openstack.org/wiki/GroupBasedPolicy/GerritQueries/NFP 18:10:20 <SumitNaiksatam> hemanthravi: you have been planning to update the wiki page? 18:11:07 <hemanthravi> yes, getting delayed due to other tasks. will update it today 18:11:18 <SumitNaiksatam> hemanthravi: thanks 18:11:24 <SumitNaiksatam> hemanthravi: i had a comment on: https://review.openstack.org/#/c/298385 18:12:49 <SumitNaiksatam> oh, just remembered, i have the status attributes patch for review: #link https://review.openstack.org/289530 18:13:09 <SumitNaiksatam> this is useful for GBP, and also for NFP 18:13:12 <hemanthravi> regarding the service profiles 18:13:23 <SumitNaiksatam> so requesting you all to review and help make progress 18:13:30 <hemanthravi> will address that 18:13:30 <SumitNaiksatam> hemanthravi: yes, go ahead 18:13:37 <SumitNaiksatam> hemanthravi: ok thanks 18:14:24 <SumitNaiksatam> #topic Open Discussion 18:14:31 <SumitNaiksatam> anything else for today? 18:14:50 <SumitNaiksatam> songole: hi, just in time before the meeting ends :-P 18:14:55 <tbachman> lol 18:15:05 <SumitNaiksatam> songole: anything you wanted to bring up before we close for today? 18:15:21 <songole> Ah 18:15:25 <SumitNaiksatam> songole: if we are ready, we can bring up the UI discussion next week 18:15:27 <songole> No 18:15:33 <SumitNaiksatam> songole: will touch base offline on that 18:15:35 <songole> ok 18:15:39 <SumitNaiksatam> that might be a longer discussion 18:15:41 <ivar-lazzaro> hemanthravi: there are tons of rebase on the NFP patches 18:15:50 <ivar-lazzaro> have you addressed my comments of a while back? 18:16:29 <ivar-lazzaro> it's kind of hard to figure out 18:16:35 <hemanthravi> i think they have been, but will review to make sure 18:17:20 <SumitNaiksatam> so in general, regarding the NFP patches, shall we as a team put a tentative schedule in place to review and attempt merging the patches? 18:17:42 <SumitNaiksatam> if there are critical issues found in the review, we change the schedule 18:17:43 <songole> SumitNaiksatam: that helps 18:18:09 <SumitNaiksatam> but i think in the current state, lots of cycles are being spent in rebases 18:18:18 <SumitNaiksatam> but we are not moving forward 18:18:29 <SumitNaiksatam> and i think reviews are also sporadic 18:18:38 <SumitNaiksatam> so thats probably not helping the authors 18:19:41 <SumitNaiksatam> we dont have to decide here, but i think we are getting to a point where we need to make a call on the NFP patches one way or the other, since they have been in the review queue for a long time 18:20:01 <songole> SumitNaiksatam: can we schedule a hackathon for next week 18:20:19 <SumitNaiksatam> songole: sounds good, let me touch base with offline on this 18:20:26 <SumitNaiksatam> *with you 18:20:31 <songole> ok 18:20:54 <SumitNaiksatam> please let me know if you anyone else in the team has any thoughts on this 18:21:02 <hemanthravi> rebases were done to sync up with the dependent patches, is there a way to avoid this? 18:21:04 <SumitNaiksatam> (or you can speak up here, we have plenty of time :-P) 18:21:27 <SumitNaiksatam> hemanthravi: i dont think anyone is saying that you should not be rebasing 18:21:58 <SumitNaiksatam> hemanthravi: its just that its difficult to keep track of review comments when there are several rebases 18:22:15 <hemanthravi> agree, just making sure 18:22:15 <SumitNaiksatam> hemanthravi: the way to mitigate that is to respond to the review comments after they have been fixed 18:22:44 <SumitNaiksatam> hemanthravi: that way the reviewer knows that a particulay comment has been addressed 18:22:53 <hemanthravi> i'll work on this to make sure they are all addressed 18:23:00 <SumitNaiksatam> hemanthravi: thanks 18:23:42 <SumitNaiksatam> hemanthravi: the wiki page could help in this regard as well 18:23:56 <SumitNaiksatam> anyway 18:24:15 <SumitNaiksatam> songole: so no bugs are at your end that you need discuss? :-P 18:24:17 <hemanthravi> yes, will update the wiki page 18:24:36 <SumitNaiksatam> songole: the neutron service_profiles is a tough one 18:24:52 <SumitNaiksatam> perhaps we can have a quick discussion about it here 18:25:09 <songole> SumitNaiksatam: we addressed the floating ip issue. 18:25:14 <SumitNaiksatam> songole: okay good 18:25:33 <songole> Not sure if I should monkey patch gbp patch.py or do it as part of NFP 18:25:40 <SumitNaiksatam> songole: you want to summarize the resource name conflic issue (service_profile) 18:25:58 <SumitNaiksatam> songole: that would depend on what the fix is 18:26:28 <songole> monkey patching l3_db call. 18:26:35 <SumitNaiksatam> songole: i think patch.py should be the place 18:26:48 <songole> ok 18:27:00 <SumitNaiksatam> that way its easy to find everything in one place 18:27:11 <SumitNaiksatam> songole: you want to summarize the resource name conflic issue (service_profile) 18:27:17 <SumitNaiksatam> *conflict 18:27:23 <songole> ok 18:27:47 <songole> neutron as part of flavors framework is exposing CRUD APIs for its own service-profile resource 18:28:04 <songole> the name directly conflicts with our service-profile resource 18:28:30 <songole> my original idea was to rename GBP resource 18:28:47 <SumitNaiksatam> songole: that comes with the downside that its a backward incompatible change 18:29:14 <SumitNaiksatam> songole: i was hoping that there would be a namspace resolution to this 18:29:26 <ivar-lazzaro> is the flavor framework merged? 18:29:43 <songole> yes, it is merged 18:29:47 <SumitNaiksatam> ivar-lazzaro: yes, its been merged for a while now (i think liberty) 18:30:18 <SumitNaiksatam> so i do have a patch out for review in GBP, which adds loading the flavors plugin 18:30:43 <SumitNaiksatam> this is to try and validate the coexistence of GBP and the neutron flavors framework 18:30:57 <SumitNaiksatam> in that patch i also invoke Neutron flavors CLI 18:31:01 <SumitNaiksatam> all this seems to work 18:31:23 <SumitNaiksatam> the issue is specifically with regards to the ‘service_profile” resources 18:31:39 <songole> Do you excercise neutron service-profile resources? 18:32:14 <SumitNaiksatam> songole: no, per our discussion, i was waiting for you to provide me with the CLI for the workflow ;-) 18:32:57 <songole> Trying to find time for that .. :) 18:33:42 <SumitNaiksatam> this is the patch #link https://review.openstack.org/#/c/296098/ 18:33:46 <SumitNaiksatam> songole: np 18:34:19 <SumitNaiksatam> so anyway, rkukura ivar-lazzaro igordcard, if you have any suggestions on how to handle this resource name conflict, please let us know 18:35:20 <ivar-lazzaro> ok 18:35:23 <rkukura> I think we need to change our resource name, because we are the unofficial project 18:35:37 <SumitNaiksatam> rkukura: but that is backward incompatible 18:35:46 <rkukura> that goes with the territory 18:35:46 <SumitNaiksatam> for current GBP users 18:36:02 <SumitNaiksatam> rkukura: i dont think we can just throw up hands like that 18:36:07 <rkukura> if we had our own server/endpoint, it wouldn’t be an issue 18:36:16 <ivar-lazzaro> Maybe we should keep both service-profile and the new name for a while? 18:36:16 <SumitNaiksatam> rkukura: if we rename, we will have to provide a migration strategy 18:38:19 <rkukura> Which neutron release introduced their service-profile resource? 18:38:33 <SumitNaiksatam> rkukura: i think stable/liberty 18:39:11 <rkukura> so maybe we could make our kilo version support two names as a migration strategy, and only the new name in liberty forward 18:39:56 <SumitNaiksatam> rkukura: okay 18:40:26 <rkukura> that would be cleanest if we think most users are still on kilo 18:40:44 <SumitNaiksatam> rkukura: i agree 18:41:30 <SumitNaiksatam> thats all i had for today 18:41:41 <SumitNaiksatam> anyone have anything else? 18:42:03 <SumitNaiksatam> i think we should have a proper spec out to resolve the name conflict issue 18:42:12 <SumitNaiksatam> so that we can all agree on the path forward 18:42:45 <rkukura> yes, and hopefully fixed before stable/kilo is EOL 18:42:45 <SumitNaiksatam> alrighty, thanks everyone for joining! 18:43:10 <rkukura> thanks SumitNaiksatam! 18:43:11 <tbachman> SumitNaiksatam: thanks! 18:43:16 <SumitNaiksatam> rkukura: yes 18:43:23 <SumitNaiksatam> bye all! 18:43:30 <SumitNaiksatam> #endmeeting