17:00:36 #startmeeting murano 17:00:37 Meeting started Tue Nov 18 17:00:36 2014 UTC and is due to finish in 60 minutes. The chair is serg_melikyan. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:38 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:00:40 The meeting name has been set to 'murano' 17:00:48 #topic agenda 17:00:59 1. Action Items Review 17:01:14 2. Review Roadmap for K 17:01:19 hi! 17:01:22 o/ 17:01:24 3. Update blueprint for K1 17:01:34 4. Open Discussion 17:01:46 #topic Action Items Review 17:01:48 Hi guys! 17:02:21 stan_lagun: you've joined us today? 17:02:28 hi! 17:02:42 serg_melikyan: sure :) 17:02:45 Hi! 17:02:51 janalex: hi! 17:02:51 o/ 17:03:10 Let's start with action items from previous meeting :) 17:03:22 yes! 17:03:34 meeting minutes: http://eavesdrop.openstack.org/meetings/murano/2014/murano.2014-11-11-17.00.html 17:03:44 1. ruhe will create murano-specs 17:03:52 #link http://git.openstack.org/cgit/stackforge/murano-specs 17:04:02 thank you! 17:04:16 Now we able to start working on specifications for K :) 17:04:22 repo is already there. but it's empty atm. patch with initial project setup is on review: 17:04:23 #link https://review.openstack.org/#/c/134907/ 17:04:48 and what about the template for specs? 17:04:50 i've asked stan_lagun to update spec template. i agree with his comment about general versioning impact 17:05:05 2. Stan Lagun will take a look at what exactly need to be fixed to resolve " Re-surrect support for configuration changes in deployed environment" 17:05:13 stan_lagun: had a chance to take a look? 17:05:23 ruhe: I'm going to update it immediately after this meeting 17:05:27 katyafervent2: tempalte is here on review https://review.openstack.org/#/c/134907/ 17:05:44 serg_melikyan: yes 17:06:01 ok 17:06:12 Just finished investigation. There are 2 bugs in dashboard that prevent redeployment of failed environments 17:06:29 links? 17:06:37 Environments that were deployed successfully can be redeployed 17:06:57 no tickets yet. Just finished researching 17:07:24 Ok, I assume that you will create appropriate tickets for them? 17:07:57 #info we have two bugs that prevent proper re-deployment of environments 17:07:58 yes. Maybe they already exist. Probably need to talk to Kate 17:08:23 #action stan_lagun will insure that tickets about issues with re-deployment will be created 17:08:27 stan_lagun, ping me anytime 17:08:52 3. stan_lagun, will think about implementation of one-time properties 17:08:59 katyafervent2: thanks! 17:09:22 I think that we need to create a document with all improvements in our DSL that we need to do in K 17:09:53 serg_melikyan: create a spec in murano-specs maybe? 17:09:55 and file appropriate blueprints for each of them 17:10:07 I have some ideas on how this should be supported on engine's side but I don't see how this can be implemented with current API 17:10:21 ruhe: I think one spec will be not enough 17:10:37 because with current API dashboard has no access to class metadata 17:12:00 Ok, but we plan fix API 17:12:20 stan_lagun: what do you think about document with list of all changes in dsl? 17:12:49 I think we need to have BPs+specs for each change 17:12:49 can you start working on that document? I believe you already have few thoughts about that 17:12:54 stan_lagun: =1 17:12:57 +1 17:13:17 I'm going to submit many of them during thins week 17:13:25 *this week 17:14:02 but the problem with one-time properties is not one of them 17:14:10 stan_lagun: let's anyway create some etherpad with list 17:14:11 because we need to design new API first 17:14:35 sure. I'll summarize them in one place for next meeting 17:14:51 stan_lagun: why? we can start with specification, and describe what needs to be changes in API 17:15:20 specification is a designed solution. I don't have one at the moment 17:15:43 I think there is no reason why we can't create spec for one-time properties before changing API 17:15:44 we can discuss document creation on irc 17:15:57 and changes in API maybe part of the spec 17:16:00 Also I'm not sure about exact requirements for new API though I have some requests 17:16:04 * in 17:16:40 specification is completely designed solution only when it's merged + implemented 17:16:57 we anyway can at least describe what is this about 17:17:16 I have no solution for one-time properties with current API. And there is no point to design small feature that is dependent on whole new API that is not even discussed yet 17:17:29 #action stan_lagun, will create etherpad with list of changes planned in DSL for K 17:17:42 The problem is that i have no solution until we have new API 17:17:54 Ok 17:18:03 Let's move on :) 17:19:15 4. stan_lagun will move his document about versioning to the spec once repo is created 17:19:35 I think we can skip this one - as stan_lagun mentioned before it is not done yet 17:19:44 not done because we haven't merged initial commit for murano-spec 17:19:57 #action stan_lagun will move his document about versioning to the spec repo 17:20:29 5. Serg to send invite to weekly meeting to the Radek and other guys 17:20:43 I've invited guys from HP and Telefonica to our weekly meetinh 17:20:50 *meeting 17:21:00 <_avig_> we are here :-) 17:21:03 hi 17:21:09 Hi, this is Pablo from Telefonica 17:21:22 My colleague Henar couldn't join 17:21:28 Hi guys :) Nice to see you here :) 17:21:31 but next time she will!! 17:21:52 _avig_: sorry about last meeting, I forgot that we had DST changes 17:22:15 Pablo_: ok :) 17:22:47 Let's than move to the next item 17:23:23 #topic Review Roadmap for K 17:23:57 I've composed Roadmap for K cycle - https://wiki.openstack.org/wiki/Murano/Roadmap 17:24:11 We can add new items there, though 17:24:55 I mean add new items later, when we will have better understanding, or time to implement something that we didn't schedule from the begining 17:25:11 Any Blueprints that I've missed to add? 17:25:51 Policy driven fulfillment 17:26:01 <_avig__> where is the policy-guided fulfillment item? 17:26:07 HP is working on the spec proposal for that 17:26:17 <_avig__> i'm from HP :-) 17:26:21 Should we add it to the list now or after we have a blueprint link? 17:26:25 janalex: it not there only because we don't have BP for that right now 17:26:36 We will add that once we will create BP :) 17:26:53 <_avig__> i see, i guess we'll discuss it in tomorrow's meeting, right Serg? 17:27:00 I could'nt figure out how to describe that in BP. 17:27:05 _avig__: right 17:27:11 I guess it should me several BPs/specs 17:27:12 <_avig__> great 17:27:12 ok, let's do an action item to add BP and an item to the roadmap 17:27:17 and right after that I will update roadmap 17:27:28 jfyi: this wiki page is not something final. it is updated during the development cycle 17:27:28 sounds good 17:27:54 it just let's people outside of the project to understand what's going on 17:27:59 #action Serg, discuss policy guided fulfillment and create all necessary BP 17:28:00 *lets 17:30:15 Ok, anything else missing in Roadmap? 17:30:48 a lot of MuranoPL/engine thing but we will fix it soon :) 17:30:55 <_avig__> i see there's an item on the list w/o a blueprint. what does it mean? 17:31:19 serg_melikyan: i'd suggest to add a link to official OpenStack Kilo release schedule 17:31:22 this is just a bug 17:31:33 ruhe: thank you, will do 17:32:01 _avig__: it's a bug. stan_lagun, serg_melikyan can we add a link to that bug? 17:32:09 _avig__: there was action item about that item, but Stan didn't yet created necessary tickets/bp 17:32:11 I think we shouldn't mention bugs in roadmap 17:32:19 we're speaking about "Resurrect support for configuration changes in deployed environment" 17:32:39 stan_lagun: we didn't know is it only a bug ) 17:33:04 You was investigating that - Action Item #3 for today ) 17:33:15 now we know 17:33:33 _avig__: if it's only a bug we will remove that - features == blueprints 17:34:49 added policy guided fulfillment there too, 17:35:07 #action Serg, fill all missing BP in Roadmap 17:35:38 #topic Update blueprint for K1 17:36:16 I've assigned part of the blueprints for K1, mostly ones that are already near completion: https://launchpad.net/murano/+milestone/kilo-1 17:36:50 Deadline for K1 is December 18, exactly one month from now 17:36:51 serg_melikyan: i have a question. should we expect all the blueprints scheduled for Kilo-1 (https://launchpad.net/murano/+milestone/kilo-1) to have a spec? 17:37:52 ruhe: I suggest to have specs only for big BPs + write them on request. But not to have this as a mandatory rule 17:38:13 ruhe: I hope so, but I think since we only started with spec-repo part of specs may be in progress 17:38:21 stan_lagun: disagree 17:38:49 It is difficult to write spec for each and every tiny feature 17:39:05 stan_lagun: tiny feature - tiny spec 17:39:21 I don't want to have such high bar for newbies 17:39:48 stan_lagun: that's why I say that we may not have all BP merged at the end of K1 17:39:59 we will help with spec creation 17:41:29 seems like too much bureaucracy for me. I like the idea of specs but don't like being to formal with that 17:42:36 I think we have pretty tentative development schedule for https://launchpad.net/murano/+milestone/kilo-1 already and we don't want any other BPs 17:42:43 if you get 100 formally written specs for each little improvement in UI it would be hard to find important ones 17:43:09 It gives us much more time to design features like policy guided fullfillment and support for puppet and chef 17:43:13 Or new API 17:43:24 i suggest to move discussion about scope of specs outside this meeting 17:43:56 ruhe: +1 17:44:19 but sure, we as a team need to agree on this item 17:46:26 Guys, you agree that we don't need to include other blueprints to K1 17:46:58 does this implies that all MuranoPL BPs will go to K2? 17:47:28 i'd say that some UI-related work could be done in kilo-1 if there is anyone willing to do that work :) 17:48:35 stan_lagun: it depends - I would prefer to spend time on designing heavy features during k1 17:49:06 ruhe: yes ) Unfortunatelly Brian not with us today, I will ping him on this week about UI part 17:49:19 agree. But lets not move all of them to the end of the cycle 17:49:20 also we need to reserve some time for code-freeze and testing of milestone release 17:50:21 serg_melikyan: I haven't seen TOSCA in K roadmap but it is in K1 17:50:35 stan_lagun: it is in Roadmap 17:50:39 stan_lagun: look closer ;) 17:50:51 sorry. Missed it 17:51:17 But as we discussed we are talking about most basic integration that relies on Heat + heat-translator 17:52:39 propper implementation of pluggable translators need some packages API refactoring. I'm not sure how big it is going to be 17:52:56 just be aware 17:53:15 ruhe: agree about code-freeze, let's do as always - one week for CF 17:53:46 I think we already moved to the latest part of our meeting 17:53:52 #topic Open Discussion 17:54:18 stan_lagun: you have in mind some nice wrapping around that? 17:54:52 serg_melikyan: around what? 17:55:04 plugins? 17:55:28 just an update about failing CI jobs. fix for failing gate-murano-devstack-dsvm jobs should be resolved with https://review.openstack.org/#/c/134901/ . it should be merged in a few minutes 17:56:12 murano-ci jobs "gate-murano-integration" are in a good shape at this moment. leave a comment in gerrit "retrigger murano-ci" if you have failing murano-ci tests 17:57:51 There was problem with HOT support: MuranoPL class was generated on the fly but UI form was generated at package upload and I couldn't find an easy way to do otherwise 17:59:15 ruhe: thanks 18:00:07 stan_lagun: ok, let's discuss that later, I don't think we need full fledged plugins but something we definitely need 18:00:11 #endmeeting