13:00:02 <alex_xu> #startmeeting nova api 13:00:03 <openstack> Meeting started Wed Oct 11 13:00:02 2017 UTC and is due to finish in 60 minutes. The chair is alex_xu. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:00:04 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:00:07 <openstack> The meeting name has been set to 'nova_api' 13:00:14 <alex_xu> who is here today? 13:00:17 <stephenfin> o/ 13:00:24 <takashin> o/ 13:01:34 <gmann> o/ 13:01:44 <alex_xu> ok, let us start the meeting 13:01:50 <alex_xu> #topic spec 13:01:59 <alex_xu> #link https://review.openstack.org/#/c/508101/ 13:02:10 <alex_xu> gmann: I just reviewed today 13:02:24 <gmann> alex_xu: thanks for review 13:02:47 <gmann> so should we keep admin only policies as it is? 13:02:47 <alex_xu> I think just want to clear why we want to remove those policies 13:02:55 <alex_xu> gmann: I think so 13:03:19 <gmann> alex_xu: many of them were added while API extensions were added 13:03:42 <gmann> as there is no extensions so those policy does not make sense 13:03:44 <alex_xu> gmann: yea, so some of them is really not very useful I think? 13:04:00 <gmann> yea 13:04:23 <gmann> i think we progress on this spec to eliminate those set by set 13:04:24 <alex_xu> ok, probably we should clear that in the spec 13:04:30 <gmann> sure 13:04:54 <gmann> like who has default admin only it make sense to keep it 13:05:01 <alex_xu> yea 13:05:13 <gmann> and other policy we want deprecation cycle right? 13:05:22 <alex_xu> we also should clear why we want to keep the policy for those admin only attributes 13:05:39 <alex_xu> gmann: I guess so, but good to check with sdague or johnthetubaguy 13:05:41 <gmann> there are 2 things 13:06:08 <gmann> 1. policy has default policies to admin only 13:06:28 <gmann> 2. policies control the admin only attributes 13:06:56 <gmann> sdague: johnthetubaguy feedback will be more helpful 13:07:28 <gmann> let me update spec with your comment by keeping admin related policies and non admin related 13:07:38 <gmann> and more clear reason to remove those 13:07:55 <alex_xu> 1. we remove some of policies since they are useless 13:08:04 <gmann> but we are agree that we need deprecation even for not needed policies ? 13:08:16 <alex_xu> 2. we change some of attributes from the admin-only to the normal user, like user_data 13:08:39 <gmann> alex_xu: 1 is with deprecation cycle ? 13:09:01 <alex_xu> gmann: I think both for 1 and 2 13:09:17 <gmann> humm 13:09:19 <alex_xu> gmann: have we talk about the deprecation in the PTG? 13:09:57 <gmann> we did not get clear on that as policies list was not ready 13:10:18 <gmann> idea was to discuss this on spec when we have list of policies to review 13:10:41 <alex_xu> ok 13:11:19 <gmann> let's get feedback on spec. as of now deprecation is something look required 13:11:27 <alex_xu> gmann: let us clear about the change #1 and #2 in the spec, and then ask wider suggestion like from sdague and johnthetubaguy 13:11:38 <gmann> alex_xu: sure. +1 13:12:07 <gmann> i will update spec by tomorrow 13:12:11 <alex_xu> gmann: thanks 13:12:16 <gmann> alex_xu: one thing on this 13:12:31 <alex_xu> gmann: yea, please 13:12:43 <gmann> as on LP BP we have 4 items for api extensions and 4th one policies is this spec for 13:12:59 <gmann> i do not want to hold other item on this. 13:13:06 <alex_xu> me too 13:13:11 <gmann> but without spec approve BP will not be approved 13:13:26 <gmann> and without BP approved code merge might face issue 13:13:31 <alex_xu> I remember I ask this on the PTG, but I remember sdague said put them all in one is ok, maybe check with sdague again 13:13:44 <gmann> ok. 13:14:05 <alex_xu> I feel the api-extension remove and the policy remove are separate thing 13:14:18 <gmann> or may be we can have 2 BP one like specless and other policy one with sopec 13:14:24 <alex_xu> yea 13:15:04 <gmann> alex_xu: yea actually i encountered this during extension merge and it becomes easy if we get policy things also agreed 13:15:20 <gmann> but policy agreement can take time 13:15:27 <alex_xu> yea 13:15:50 <gmann> let me separate those on LP. 13:15:52 <alex_xu> need a little time to review the whole list of policies 13:16:01 <gmann> yea 13:16:12 <alex_xu> and each attributes 13:16:46 <alex_xu> #link https://review.openstack.org/#/q/project:openstack/nova-specs+status:open+message:%22apiimpact%22 13:17:14 <alex_xu> I don't have any specific more, looks like there are a lot of spec have apiimpact flag, and I missed the spec review sprint 13:18:23 <gmann> alex_xu: in last meeting we discuss about abort cold migration API return code 13:18:30 <gmann> takashin spec 13:18:40 <takashin> #link Abort Cold Migration https://review.openstack.org/#/c/334732/ 13:18:42 <gmann> can you check now if it is ok 13:18:56 <gmann> it looks ok for me now 13:19:01 <takashin> alex_xu: Would you review it? 13:20:01 <alex_xu> takashin: is this already approved in the last release? 13:20:13 <takashin> alex_xu: No 13:20:51 <takashin> Before I submitted it for Ocata. 13:21:04 <takashin> But It has not been approved yet. 13:21:45 <alex_xu> ok, I don't have any problem I think, I probably just need to deep into the detail check the all the API change are good 13:22:24 <takashin> alex_xu: Thanks in advance. 13:22:33 <alex_xu> takashin: np 13:23:36 <alex_xu> #topic open 13:23:43 <alex_xu> anything more people want to bring up? 13:24:21 <gmann> nothing from my side 13:25:00 <stephenfin> nothing here either 13:25:14 <alex_xu> ok, let us close the meeting, thanks all! 13:25:17 <alex_xu> #endmeeting