17:03:07 #startmeeting CongressTeamMeeting 17:03:08 Meeting started Tue Jun 24 17:03:07 2014 UTC and is due to finish in 60 minutes. The chair is pballand. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:03:09 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:03:11 The meeting name has been set to 'congressteammeeting' 17:03:47 Hi all 17:03:53 Hi All 17:03:58 Hi 17:04:12 Hi 17:04:30 hi guys 17:04:46 last week, we were looking at the use case document, and voting 17:05:17 there has been some progress, but I think there’s still some work to do 17:05:24 Yup 17:05:45 it looks like the majority of use cases require some sort of action, which is good, but will take a bit longer to get to 17:06:04 We need to get a few blueprints targeted for Juno-2 and 3 17:06:05 Hi Peter, I have added some policies for the security use cases 17:06:41 it is also helpful to have the data soruces listed 17:07:22 my preferece are the cases that have multiple data sources, since they leverage a unique capability of congress 17:07:28 I see we have data sources/policy/actions listed for almost all the use cases—this is super helpful! 17:07:37 Yeah, I have some data sources too 17:07:57 I'll need to review again 17:08:31 if anyone has yet to review or vote, please do so ASAP, so we can push as blueprints 17:08:42 When do we want to assign blueprints? 17:09:08 That we give us a date to work backward from 17:09:26 some data sources are listed as generic : RDBMS or some name which i dont understand 17:09:28 When we need to finish up specs and the use cases 17:09:30 I think we should target this week - does that sound reasonable? 17:09:32 There are 2 things here: writing up the blueprint and doing the work required for the blueprint. 17:09:35 it will be great to elaborate 17:09:54 I *think* that the work required to implement the blueprint in most cases is writing a datasource driver or two. 17:10:22 Except for adding the ability to execute actions. Which I’m guessing I’ll end up doing. 17:10:39 sarob: I think you had offered to do the blueprint conversion from the doc - is that something you can do this week? 17:10:50 Def 17:11:04 Just when to pull the trigger 17:11:18 so let’s put the deadline for use cases and voting at end of day tomorrow (Wed) 17:11:26 yup 17:11:28 Sounds good 17:11:35 …at which point we deprecate the google doc 17:11:56 #action Final use case submission and voting deadline is Wednesday 17:12:00 So blueprints on diff features of the use cases 17:12:09 #action sarob to convert top use cases to blueprints 17:12:25 other tracking items… 17:12:27 Like per data source, project action 17:13:05 sarob: are we talking about 1 blueprint per use case? Or are we talking 1 blueprint per datasource driver? 17:13:33 Blueprint per specific features 17:13:56 we use blueprint like storyboard cards 17:14:14 Specific actionable dev 17:14:16 it seems like a blueprint per datasource is appropriate, then the policy use cases should be fairly straightforward 17:14:30 As long as we don’t lose the info currently in the gdoc, I’m happy. It’s useful to see all of that info all in 1 place. 17:14:43 If it's too large of effort then difficult to gauge progress 17:15:11 each datasource is pretty well self-contained 17:15:12 Thinichs I'll move it to the wiki 17:15:41 sarob: why move it? 17:15:46 We want the info to be public and debatable 17:16:05 the google doc is public - am I missing something? 17:16:07 Wiki is more standard 17:16:23 But people can’t leave comments (as easily) on the wiki. 17:16:24 Not that big of a deal 17:16:37 Just not how the rest of the teams works 17:17:03 we’re agreed on the blueprints, lets start there, and see how it goes 17:17:13 We are still planning on gdoc to spec to blueprint. Right? 17:17:24 sarob: Looks like it 17:17:49 I assumed you guys were using shorthand 17:18:01 Just wanted to double heck 17:18:05 Check 17:18:32 sounds right 17:18:39 Coolo 17:18:49 great, lets move on 17:19:10 we seem to be back on track with code reviews, thanks to everyone who jumped in 17:19:36 on the code front, we are pushing toward getting an alpha release ready... 17:19:49 thinrichs: how’s the policy/dse integration going? 17:19:58 Making good progress. 17:20:15 Late last week I hooked up a live Neutron to the system and evaluated policy over it. 17:20:21 sweet! 17:20:32 thinrichs: thats awesome 17:20:32 This week I’m adding hooks for pballand’s API work. 17:20:49 rajdeep’s driver’s worked without a hitch! 17:20:50 Double sweet 17:20:54 On the real system. 17:21:47 And I’m being pretty good about expanding our test suite so we should know when we break things later on. 17:21:59 as thinrichs mentioned - I pushed the API framework to review, and am working with thinrichs to integrate the calls with the policy enginge (which is the source of truth for much of the data exposed) 17:22:21 I’ve not been as good at updating the test suite, but am working on it :) 17:22:44 kudva: are you around? 17:22:51 pballand: yes 17:23:19 how are things going with the policy runtime work? 17:23:51 pballand: Good, testing the integration of builtin with runtime. Making progress. 17:24:01 pballand: hope to have something to report next week 17:24:25 kudva: let me know if you need any more pointers. 17:24:40 kudva found that we were implicitly assuming Python 2.7 or higher 17:25:10 thinrichs: Yes, I will update the requirements document. Had some issues with 2.6, switched to 2.7 17:25:23 thinrichs: now it works great with 2.7 17:25:34 it would be helpful to document all of these dependencies, any takers? 17:25:35 pballand: also getting a tutorial from rajdeep on data source drivers for future ceilometer integration (seems like that will be important to learn) 17:25:52 pballand: I can update the python one. 17:26:05 kudva: sounds great 17:26:22 pballand: I meant data source drivers in general (possibly ceilometer also, but we can decide that later) 17:26:56 kudva: got it - any blockers? 17:27:09 pballand: none so far, moving forward 17:27:27 pballand: rajdeep agreed to take me through the code, so :) 17:27:40 kudva: good to hear :) 17:27:45 cloudtoa_: any progress on oslo.messaging? 17:28:41 sorry guys for being late; stuck in another meeting 17:28:56 I guess cloudtoa_ stepped out 17:29:52 banix: glad you could join 17:30:12 banix: we were just finishing up on updates - do you have anything to share? 17:30:49 pballand: Just got started trying out Congress! 17:31:29 Have been looking ta the code and will be spending more time wrt testing; got some pointers from Tim so will have more to share next week 17:31:44 Banix get your use case input in today 17:31:57 sarob: will do 17:32:09 :) 17:32:50 some people have been asking about status/work items, so I thought it would be worthwhile to quickly summarize where I think we are… 17:34:05 current “sprint” has been toward getting an ‘alpha’ ready, which we’ve defined as getting the code to a point where people can easily download and run, executing some monitoring use cases 17:34:43 the alpha is currently blocking on a integration between policy and API, which thinrichs and I are pushing through 17:35:38 after that, the next goals we’re working toward are additional data sources, and the enforcement use cases that we’ve been discussing for the past few weeks 17:35:50 when you say policy you mean the policy engine? the runtime? 17:36:21 banix: yes, I use those terms interchangably 17:36:40 pballand: thx. 17:37:12 other work including built-ins by kudva also will be very useful as we add additional features 17:38:17 there is a long feature roadmap, but I wanted to share what I see as the current status 17:38:31 does anyone have other high-priority items they are tracking? 17:38:49 is the API piece in for review or not yet? 17:39:17 in stackforge i meant 17:39:37 banix: the framework is in for review (you can actually make requests against it), but it isn’t wired to any actual data 17:40:06 Link: https://review.openstack.org/#/c/101672/ 17:40:33 #link: https://review.openstack.org/#/c/101672/ 17:40:57 sorry… lwas looking for the same link 17:41:17 that’s all I had on the agenda for this week 17:41:20 #topic Open Discussion 17:42:21 anyone have something they wanted to discuss? 17:42:55 Sorry I missed out most of the discussion since I am sort of multitasking. Did we discuss a timeline on the alpha? 17:42:56 ok, I think that’s a wrap for this week 17:43:33 skn_: the alpha will be ready as soon as we get the final api/policy integration pushed - we are hoping for the end of this week 17:44:10 Awesome, once we have something, we should create a plan for the use case implementation for a demo for Juno 17:45:01 Thx guys 17:45:06 skn_: absolutely, let’s talk about in more detail next week 17:45:11 Yup 17:45:14 thanks everyone for joining, we’ll see you next week 17:45:19 skn_: good 17:45:24 #endmeeting