03:08:58 <Yumeng> #startmeeting openstack-cyborg
03:08:59 <songwenping__> yeah
03:08:59 <openstack> Meeting started Thu Sep  3 03:08:58 2020 UTC and is due to finish in 60 minutes.  The chair is Yumeng. Information about MeetBot at http://wiki.debian.org/MeetBot.
03:09:00 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
03:09:02 <openstack> The meeting name has been set to 'openstack_cyborg'
03:09:17 <Yumeng> #topic Roll call
03:09:22 <Yumeng> #info Yumeng
03:09:26 <xinranwang__> #info xinranwang__
03:09:42 <songwenping__> #info songwenping__
03:09:47 <brinzhang> #info brinzhang
03:09:48 <Yumeng> agenda for today https://wiki.openstack.org/wiki/Meetings/CyborgTeamMeeting#Agenda
03:10:49 <Yumeng> I didn't have enough time to put all patches that needs review. pls just bring up yours if any.
03:12:43 <brinzhang> https://review.opendev.org/#/q/project:openstack/cyborg-specs+status:open
03:13:20 <brinzhang> add implemented directory to the cyborg-specs project, to record the implementd specs
03:14:13 <brinzhang> add judgament the specs we are doing but not completed in each release, such as policy refresh feature
03:15:07 <brinzhang> Adding nova-cyborg-interaction spec, copied from nova-specs, I think we should leave it in cyborg and let others know
03:15:44 <Yumeng> brinzhang: I see gmann re-propose policy refresh feature in nova-spec, because the spec is really "re-proposed"
03:16:23 <brinzhang> others is check the requirement of doc8, the tox -e doc8, cannot be run, so I removed and add to the [docs] model
03:17:10 <brinzhang> Yumeng: yes, like nova, if the spec isnot completed in Ussuri, if you want to continue work in Victoria, you should re-proposed it
03:17:12 <Yumeng> but I didn't see the value why we need re-propose it in cyborg. we didn't do any change in Victoria release compared to that of U
03:17:35 <brinzhang> it's a process
03:17:47 <brinzhang> I think so
03:18:48 <brinzhang> if not do this, one people read the spec, which is dont know when we completed it, and when we proposed it, I dont think it's a good way to show developers
03:19:27 <brinzhang> I hope we can abide by such rules so that everything can be tracked
03:19:34 <xinranwang__> IMHO, once we need to improve the spec/implementation, or continue work on this spec, we should repropose it.
03:20:04 <brinzhang> xinranwang__: yes
03:21:29 <Yumeng> although I was not quite clear how nova doc track this kind of process. can we do it more simple?  like we just record the spec when approved and implemented?
03:22:16 <brinzhang> Indeed, if the spec were not re-proposed, we can say it wont be continue work in this release, that we can select dont to review the codes
03:23:19 <brinzhang> Yuemng: no, we also need to record it's development process
03:23:29 <openstackgerrit> wangzhiguang proposed openstack/cyborg-tempest-plugin master: add delete multiple device profile by names api test  https://review.opendev.org/749622
03:24:01 <Yumeng> code review is according to our PTG planning and release goals. spec is just part of this goal
03:25:14 <Yumeng> I think this is not a big deal. once it is decided to do this way, we might need to cotinue in the future.
03:25:23 <xinranwang__> IMO, if there is no changes on the previous spec, either on implementation, we may not need to repropose it. We just re propose it if we need improve the spec or implementation.
03:25:47 <brinzhang> Because the impemented directory has been added, I think we need to follow this rule. The content discussed on PTG is a milestone, what we will do.
03:26:35 <brinzhang> if we dont re-proposed the SPEC in a new relese cycle, may we cannot track it.
03:27:32 <brinzhang> We can know what we do now, because there are little things todo, if there are more feature need to do in the future, how we track them?
03:28:45 <Yumeng> I don't know why you think we cannot track feauters.    we have gerrit, storyboard and specs
03:29:48 <Yumeng> also the wiki release schedule, PTG etherpad
03:31:19 <songwenping__> track them in doc is better. i think we should unified manage these features.
03:31:49 <Yumeng> I agree that we can add impemented directory, and if we don't have anything new to refresh the spec, we don't need one more copy of it.
03:32:06 <brinzhang> The spec specification is for the general public. If someone wants to know which features we have made in which version, specs will be a good reference.
03:34:55 <xinranwang__> Yumeng:  +1, we can have implementation repo
03:36:27 <Yumeng> brinzhang: spec is good, but if nothing new, I don't think we need a duplicated copy of it.  if the feature is still under development, who will see this duplicated one?  I just think most people cares more about the start and the outcome(code).
03:38:14 <brinzhang> As i know, if a user want to know in which release what the project doing, he will see the spec in that release
03:38:46 <Yumeng> ok. we don't have much time left for the meeting. let's continue this topic offline.
03:38:55 <Yumeng> and now let's go to next topic
03:39:25 <Yumeng> # topic PTG
03:40:21 <Yumeng> have you checked this time I reserved? https://ethercalc.openstack.org/7xp2pcbh1ncb
03:40:42 <Yumeng> I wanna make sure it is good for all you guys
03:42:26 <xinranwang__> I'm ok with that
03:44:17 <Yumeng> nice xinranwang__
03:44:39 <Yumeng> Oct 27, 28, 29, 30  6:UTC-8:00UTC are planned for cyborg internal meeting.  Oct 30  13:00-15:00 UTC is planned for external meeting
03:44:54 <songwenping__> There should be no problem for me.
03:45:46 <Yumeng> ok. cool
03:46:15 <Yumeng> what about you s_shogo , brinzhang, chenke?
03:47:34 <Yumeng> ok. let's continue discussion while wait for their feedback
03:47:37 <songwenping__> brinzhang is out of seat.
03:47:56 <Yumeng> #topic microversion
03:48:04 <Yumeng> do we need to add micro-version doc as nova does? https://docs.openstack.org/api-guide/compute/microversions.html
03:48:49 <songwenping__> this maybe hard. as cyborg architecture is not same as nova.
03:49:28 <songwenping__> i'm sorry
03:49:42 <brinzhang> the microversion docs will be completed after songwenping's patch merged
03:49:43 <songwenping__> doc is need.
03:50:02 <brinzhang> https://review.opendev.org/#/c/743722/
03:50:44 <brinzhang> this is the microverion docs, required by sean-k-mooney, we need this, as we need to re-proposed the specs too.
03:50:56 <Yumeng> ok. I didn't see it. cool
03:52:36 <Yumeng> I think cyborg microversion doc can be a simple one. don't have to be as complicated as nova ones.
03:53:15 <Yumeng> just to show what we have and some required guidlines for users
03:53:28 <brinzhang> There are still places that need to be covered.
03:54:12 <Yumeng> ok. BTW, wenping's project_id patch is ready for review https://review.opendev.org/#/c/738427/
03:54:18 <brinzhang> After it's ready, you can review, now we dont need to talk this
03:54:46 <Yumeng> brinzhang: yes, just mention this.
03:54:52 <brinzhang> yeah, will review later
03:55:09 <Yumeng> #topic deployable_name unique: https://review.opendev.org/#/c/748366/3
03:55:14 <songwenping__> Yumeng, brinzhang:thanks
03:56:06 <songwenping__> Yumeng: leave some new comments abount deployable_name, please recheck.
03:56:29 <songwenping__> s/abount/about
03:58:41 <Yumeng> songwenping__: thanks. I also think no need to  replace `:` to `_` in PCI address. I will remove that and keep "PCI address" as it is.
03:58:58 <songwenping__> cool
03:59:32 <xinranwang__> Please check fpga driver improvement as well  lol
03:59:46 <Yumeng> oh yes sure xinran
04:00:12 <Yumeng> ok.the last topic
04:00:13 <Yumeng> # topic review programming API: https://review.opendev.org/#/c/698190/
04:00:34 <Yumeng> pls help to review.
04:01:11 <brinzhang> ack
04:01:41 <Yumeng> ok. that's all from my side
04:02:07 <Yumeng> Do you have any other things to bring up?
04:02:26 <songwenping__> nothing from my side.
04:02:58 <xinranwang__> nothing from my side
04:02:59 <Yumeng> BTW ptg etherpad: https://etherpad.opendev.org/p/cyborg-wallaby-goals  you can addd topics now
04:03:23 <Yumeng> ok. So let's wrap up this meeting and I'll see you all again next week
04:03:30 <Yumeng> #endmeeting