14:01:38 #startmeeting glance_artifacts_sub_team 14:01:39 Meeting started Mon Sep 28 14:01:38 2015 UTC and is due to finish in 60 minutes. The chair is nikhil. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:01:40 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:01:43 The meeting name has been set to 'glance_artifacts_sub_team' 14:01:53 Hi folks 14:01:57 o/ 14:02:09 sigmavirus24: ativelkov mfedosin flaper87 rosmaita courtesy meeting reminder 14:02:16 Thanks nikhil 14:02:22 #chair ativelkov 14:02:22 Current chairs: ativelkov nikhil 14:02:38 #topic agenda 14:02:54 There's no official agenda for today and I don't have anything new to add. 14:03:18 ativelkov: if you would like to run the rest of the meeting that would be great or should we start a open discussion? 14:03:32 I can give brief update 14:03:43 Thanks. feel free to use the # topics 14:03:59 #topic Status Update 14:04:11 I have actually two items here 14:05:18 1) We've reached an agreement with the Community Application Catalog team that we are going to make some kind of PoC work for them to demonstrate the potential of grance v3 usage to staore app catalog's data and metadata 14:05:30 to store* 14:05:57 ativelkov: I would like to learn more about their req use cases 14:06:06 what would be a good place to start? 14:06:35 * nikhil shuts up and waits for the status update to complete 14:07:21 nikhil: yup, I'll give the links a bit later 14:07:37 thanks! 14:08:26 2) mfedosin, me and the other folks from the artifacts team have made some kind of task breakdown for the refactorings of v3 API, which is likely to be proposed as artifacts roadmap for Mitaka 14:09:08 cool 14:09:28 We'll put that into the spec when we migrate it to M branch 14:09:38 I think we should come up with a plan as ask for feedback prior to summit 14:09:43 Sure 14:10:31 I still haven't complited the first version of the wiki article, but I am working on that as well 14:11:27 will try to provide something for the early feedback on this week 14:11:35 ativelkov: how about sharing a etherpad for some drafting help with a small group and then propose it to wiki? 14:11:52 nikhil: yup, thats exactly what I plan to do 14:12:19 will try to have that till Thursday meeting 14:12:32 oh nice 14:12:53 That was all for the update part 14:13:05 #topic App Catalog PoC 14:13:06 I have one item that I just realized 14:13:12 oh, sure, please proceed 14:13:18 :) 14:13:30 ok, this will be fast 14:13:44 #topic Liberty-artifacts-work-documentation 14:14:25 So, I had requested a BP for API and one for other foundation work that illustrates what was done in Liberty (that was a lot!) 14:15:02 I think prior to release (late this week or next week) we should update specs so that the work shows up on specs.openstack.org 14:15:28 There are a few other specs that I am working on cleaning up however, these are sort of important as a lot was done 14:15:44 I would need help from you, ativelkov and possible mfedosin on this. 14:15:54 possibly* 14:15:57 * nikhil done 14:16:09 sure, we'll give any needed help 14:16:22 awesome 14:16:28 #topic App Catalog PoC 14:16:32 ativelkov: all yours 14:16:35 Thanks 14:17:05 So, for now the website on http://apps.openstack.org has no backend 14:17:37 :/ 14:17:55 all the data which describes its "assets" (which is technically very similar to the artifacts) are described with simple jason files 14:17:59 json* 14:18:19 this is obviously does not scale well 14:19:06 So, the team of the app catalog is currently investigating the possibilities for the new backend, with appropriate APIs etc 14:19:35 They are gathering the requirements for that backend at the etherpad at https://etherpad.openstack.org/p/app-catalog-v2-backend 14:20:04 excellent 14:20:35 I am gunna ask if some prod mgrs would be interested in adding feedback 14:20:55 I truly believe that the overall idea of the service which catalogs the artifacts metadata and data maps perfectly maps to what we are doing in glance v3 14:21:36 however, we need to demonstrate that, as for now they are not buying the idea and prefer to proceed on their own instead 14:22:06 hmm, that's weird 14:22:28 what are they worried about? 14:22:59 They do not want to run the Glance for this backend 14:23:25 But we definitely need to have more discussions 14:23:37 sure. I am curious why not. 14:23:54 ativelkov: perhaps they're also concerned about the API stability? (since it's supposed to change pursuant to API-WG feedback)? 14:23:57 I don't understand it completely as well 14:24:00 (Also has a spec for the API merged yet?) 14:24:36 sigmavirus24: we were having the discussions with them keeping in mind the planned change to stabilize the API 14:24:38 * sigmavirus24 hasn't been part of the discussion, just imagining things that would make him hesitant to use it 14:25:05 tbh I do understand ... I wouldn't start implementing something that might have big scalability needs and stability need on experimental service/api 14:25:22 The spec is not there yet, as we are told not to push it much, given all the buzz with v1->v2 migrations and defcore things 14:25:34 Specially with all the fuzz around Glance 14:26:07 This is not something anybody implements right now 14:26:15 ah 14:26:31 I see 14:26:32 So, this is just a question on choosing the direction 14:26:39 so it's not like supposed to be implemented during mitaka or so? 14:26:43 aha 14:26:46 or at least I see it like that 14:26:59 Probably we'll need to discuss it more with the app-cat team 14:27:05 +1 on more discussions 14:27:28 see the ML thread called "[openstack-dev] [murano] [app-catalog] versions for murano assets in the catalog" 14:28:00 (it's actually started with murano, but turned-out to discuss glance's usage for the app-catalog needs) 14:28:23 The last message is exactly about "We need closer communications between Murano, Glance, and App-Catalog going forward since each project has valuable things to contribute to the over all big picture." 14:28:41 I hope to come back to this before the summit 14:29:02 ativelkov: I wonder if adding [Glance] tag to it is forbidden as well 14:29:04 ? 14:29:41 nikhil: it is not :) 14:29:52 cool 14:30:33 we are out of time now 14:30:42 Yup, seems so 14:30:47 Thanks everybody 14:30:54 Thanks! 14:30:58 #endmeeting