08:04:18 <yasufum> #startmeeting tacker
08:04:18 <opendevmeet> Meeting started Tue Jun 29 08:04:18 2021 UTC and is due to finish in 60 minutes.  The chair is yasufum. Information about MeetBot at http://wiki.debian.org/MeetBot.
08:04:18 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
08:04:18 <opendevmeet> The meeting name has been set to 'tacker'
08:05:45 <yasufum> We have two topics today from h_asahina and masaki-ueno.
08:05:52 <yasufum> #link https://etherpad.opendev.org/p/tacker-meeting
08:06:23 <yasufum> h_asahina: Can you start from your topic?
08:06:47 <h_asahina> Sure
08:07:08 <h_asahina> I pushed a patch today. https://review.opendev.org/c/openstack/tacker/+/798409
08:07:55 <h_asahina> This patch will reorganize the User Guide in the official document according to the revise document plan proposed previsouly.
08:08:24 <h_asahina> Since it makes a big change, I'd like to share you all in advance.
08:08:33 <h_asahina> That's all from my side.
08:10:50 <yasufum> thanks, I agree with your update.
08:11:14 <h_asahina> thank you
08:12:26 <yasufum> Go to the next topic if no more comments.
08:13:17 <yasufum> masaki-ueno: please share your topic.
08:13:21 <masaki-ueno> Sure.
08:14:08 <masaki-ueno> We've uploaded and discussed on two specs in the previous vPTG,
08:14:34 <masaki-ueno> one is " Alert server for Prometheus with K8s cluster VNF" https://review.opendev.org/c/openstack/tacker-specs/+/786573
08:14:57 <masaki-ueno> and another is Support Multi-tenant for Kubernetes VIM https://review.opendev.org/c/openstack/tacker-specs/+/786580
08:16:46 <masaki-ueno> In those specs, there are some remaining `TBD` part, as Manpreet-san pointed out on Gerrit comments.
08:18:14 <masaki-ueno> Through our discussion after the previous vPTG, we found that we need more precise consideration to determine API and Data model in Tacker implementation to fill in those `TBD` parts..
08:18:54 <masaki-ueno> So we'd like to postpone merging those two specs, and we'd like to re-discuss in Tacker Y vPTG.
08:19:41 <masaki-ueno> That's all from my side, thank you.
08:22:18 <yasufum> I’ve understand current discussion on these topics.
08:25:06 <ueha> Should it change to "abandoned" once? or should it change to WIP and move it when the Y release folder is created?
08:26:05 <yasufum> No need to do so, but it’s helpful for reviewers to add a comment for describing current your decision.
08:27:14 <masaki-ueno> Sure, I'll post the discussion above and decision on Gerrit comment later.
08:27:30 <yasufum> It’s simply listed on the specs of next release.
08:27:35 <yasufum> thanks
08:28:18 <masaki-ueno> Thank you :)
08:31:18 <yasufum> Is there any comment?
08:32:19 <yasufum> seems nothing.
08:32:25 <takahashi-tsc> Not from my side, they seems not core features for k8s, so postpone is OK.
08:37:39 <yasufum> Do you have any other topic?
08:37:43 <manpreetk> hi
08:38:10 <manpreetk> sorry I added topic quiet late. Its regarding SQL migration
08:38:47 <manpreetk> L29
08:40:24 <manpreetk> One of the problem in that proposed patch is that it drops "sqlalchemy_filters" and adds new method which just apply filters.
08:40:24 <manpreetk> Following are few possibilities to look into this issue
08:40:24 <manpreetk> 1. Align our method with the method definition provided by "sqlalchemy_filters". I have less/no experience in SqlAlchemy, so probably this solution will take time.
08:40:24 <manpreetk> 2. Ask the SqlAlchemy community to fix the reported bug https://github.com/juliotrigo/sqlalchemy-filters/issues/61
08:40:25 <manpreetk> 3. Temporarily switch to the older version of SQLAlchemy for eg. 1.3.23, until the current release 1.4 issues are resolve. This might not be a good idea, and not sure whether it's doable.
08:40:45 <manpreetk> Thanks, thats all from my side.
08:42:38 <yasufum> Do you mean your current fix is not enough?
08:42:55 <yasufum> I mean about your comment on etherpad “various other tasks”.
08:44:25 <manpreetk> Yes, what I understood from method imported from sqlalchemy_filters definition does auto join query, then apply filters
08:49:11 <yasufum> Umm...
08:51:03 <yasufum> I prefer to replace sqlalchemy_filter with another code in tacker local repo working similar as you suggested.
08:51:53 <yasufum> I remember that some other projects using similar way, right?
08:52:32 <yasufum> Is it not helpful for us actually?
08:54:31 <yasufum> the project is aodh correctly.
08:55:05 <manpreetk> In aodh, customize method was just applying filter nothing more than that. But in tacker we are using ready-made function, and i would like to know why are we using it as I couldn't find any comment regarding the same.
08:57:23 <manpreetk> https://review.opendev.org/c/openstack/tacker/+/716249 this was review ID when this method was introduce.
09:01:30 <yasufum> I agree your option 3 is not good.
09:01:49 <manpreetk> yes that doesnt seem doable also
09:02:52 <yasufum> I think we had better to try to replace it while waiting for update of sqlalchemy_filters.
09:04:35 <manpreetk> ok
09:05:29 <yasufum> I also try to do that.
09:06:07 <manpreetk> Thanks
09:06:19 <ueha> I'm also trying in https://review.opendev.org/c/openstack/tacker/+/797809
09:06:37 <yasufum> manpreet: By the way, I have make some changes for your patch, and uploaded on your PS wrongly. Please ignore if it does not work.
09:06:47 <manpreetk> Ok
09:07:05 <manpreetk> ueha: thanks for help
09:07:48 <ueha> I have removed one error, so I would like to give you feedback. https://review.opendev.org/c/openstack/tacker/+/797809/1/tacker/objects/vnf_lcm_op_occs.py
09:09:04 <yasufum> thsx
09:09:06 <ueha> I will try modifying apply_filter too. thanks
09:10:17 <manpreetk> Thanks will update patch as per suggestion
09:11:05 <ueha> :)
09:12:12 <yasufum> OK
09:12:41 <yasufum> If no other comments or topics, lets close this meeting.
09:13:38 <ueha> Nothing from my side.
09:15:04 <yasufum> good
09:15:14 <yasufum> Thank you for joining.
09:15:15 <yasufum> Bye
09:15:22 <ueha> thanks, bye
09:15:24 <takahashi-tsc> thanks bye
09:15:27 <manpreetk> thanks, bye
09:15:34 <masaki-ueno> bye
09:15:41 <yasufum> #endmeeting