17:00:01 #startmeeting murano 17:00:02 Meeting started Tue Jul 29 17:00:01 2014 UTC and is due to finish in 60 minutes. The chair is serg_melikyan. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:03 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:00:05 The meeting name has been set to 'murano' 17:00:16 o/ 17:00:25 Hi :) 17:00:37 hi 17:01:19 Agenda for today's meeting is available here: https://wiki.openstack.org/wiki/Meetings/MuranoAgenda 17:01:42 Hi 17:01:51 Hi 17:02:07 Let's start as usual with AI review :) 17:02:15 #topic Action Items Review 17:02:25 hi 17:02:34 hi 17:02:42 katyafervent, tsufiev first action item is on you :) 17:02:49 oops 17:02:56 "update etherpad with results of our investigations about unit and integration tests in Horizon/Muranodashboard" 17:03:04 Any news since last week? 17:03:20 hi 17:03:32 * tsufiev is ashamed 17:04:02 I assume that there are no news about this AI? 17:04:11 :) 17:04:33 serg_melikyan, I think there should be one person responsible :) 17:05:03 katyafervent: had you time to update etherpad with results of investigations? 17:06:28 No, I don't know what else to investigate 17:06:48 katyafervent: so all results are already published? 17:06:48 i think its alrwady in progress 17:07:21 katyafervent is moving dashboard tests to work inside devstack 17:07:26 I'm searching for etherpad 17:07:39 katyafervent: thx! 17:07:53 Then let's to move to another AI 17:08:19 sjmc7: "Get more details on Apps Licencing from HP side" 17:08:34 what ryan already did is fine for this for now 17:08:46 I think patch from drupalmonkey is already merged :) 17:08:49 yeah 17:08:52 yep 17:08:55 :) 17:08:56 we'll revisit if and when proper licensing comes up 17:09:05 sjmc7: ok, thx 17:09:32 Looks like this is all on AI Review topic 17:09:44 Did I miss somethings? 17:10:25 #topic Announcements 17:10:47 #info murano-apps project is created 17:11:02 https://etherpad.openstack.org/p/horizon_unit_tests_analysis here it is, but there is no any update 17:11:09 Cool. Does it have gerrit? 17:11:15 gokrokve: sure 17:11:44 This repository is intended as storage for applications that may serve as example for community 17:12:56 yes, i wouldnt like to see our complex examples. theyre too scary for newcomers 17:13:23 I agree with ruhe that we need to start filling-in this repo with examples based on SoftwareConfiguration, with Hot Packages and so on. 17:13:32 https://github.com/murano-project/murano-app-incubator/tree/master/io.murano.apps.java.HelloWorldCluster 17:13:36 How about that? 17:13:43 Just added this night. 17:14:01 Can use both single instance Tomcat and Tomcat Cluster with LB 17:14:49 New Tomcats are added too 17:14:52 gokrokve: cool! I think this one will eventually end-up in murano-apps :) But you need to make change to Gerrit 17:15:02 gokrokve thats an example of a complex package 17:15:10 SoftwareCongiguration doesn't make applications simpler by itself. It just a replacement for MuranoAgent which is rarely a reason for complexity 17:15:36 stan_lagun: agree, but SoftwareConfiguration has much more demand for now 17:15:49 I don't say not to use it 17:16:23 I will write-up a little email to mailing-list tomorrow about this repository and our plans 17:16:27 As I heard there is an issue with SoftwareConfigs 17:16:36 they are executed on each stack update 17:16:48 #action sergmelikyan need to write-up a little email to mailing-list tomorrow about this repository and our plans 17:16:55 But having lots of simple applications won't help anyone as most real life applications are complex and the person who tries to write such application will have no proper example 17:17:27 Agree. That is why we have examples with inheritance and multiple selections 17:17:30 then we can add a proper, well documented example 17:17:34 i agree that a mix of simple to complex would be ideal 17:17:38 but we need short introduction ones 17:17:52 +1 17:17:54 that teach concepts 17:17:56 lets rate them 17:18:04 I think the best candidates are databases 17:18:13 as for books. Intended reader level 17:18:18 MariaDB, MySQL, PostreSQL, Mongo 17:18:18 eventually this repository will be filled with all sorts of applications, but we need to always mind that application in murano-apps are public examples about how to create application in particular case. Starting from apps focused on SoftwareConfig is logical 17:18:48 sergmelikyan, +1 17:18:49 Sure. DBs are godd candidates as they usually require less steps for installation 17:19:16 agreed, teaching concepts is crucial. i think that should be the primary focus vs # of steps 17:19:28 if gokrokve is right about configs being reaplied on each update this would mean they are still useless in Heat 17:19:28 Looks like we are on the same wave here :) 17:19:47 Let's move to the next topic 17:19:59 #topic Plans for Juno 17:20:22 * serg_melikyan meen juno-3 17:21:14 part of blueprints that were approved for J2 but not finished were moved to j3 17:21:46 let's go through blueprints mentioned in agenda 17:21:59 1)https://blueprints.launchpad.net/murano/+spec/disable-murano-agent 17:22:13 there's a patch in review 17:22:26 sjmc7: thank you! 17:23:07 And since we agreed on implementation this BP may be approved, right? 17:23:19 i think it got agreed upon previously 17:23:19 i guess we can mark this as approved 17:23:21 any objections? 17:23:43 sjmc7: yeah, but BP in LP was not marked so, I don't know why 17:23:52 done 17:24:15 2) https://blueprints.launchpad.net/murano/+spec/engine-test-based-on-murano-pythonclient 17:25:16 Why need to disable Agent? You can just use images without Agent 17:25:28 I think this is really useful thing, but which implementation scenarios we are going to choose? 17:25:54 stan_lagun: where you have been when we discussed this BP? :) 17:26:13 yes, you can, and then it fails to deploy without any message 17:26:15 for 1 minute? 17:26:27 stan_lagun: nope, this this was discussed for weeks 17:26:38 this isn't the first discussion of it. it's optional, doesn't require any action to leave it enabled 17:26:39 If you disable agent you are not supposed to use agent-based apps anyway 17:27:18 stan_lagun: this is idea 17:27:25 let's re-discuss this later if need be 17:27:33 sjmc7: +100 17:27:38 its a feature request for deployments where agent is not usable at all 17:27:42 what about second one? 17:27:57 I mean second link that I have shared: https://blueprints.launchpad.net/murano/+spec/engine-test-based-on-murano-pythonclient 17:28:01 if we make murano engine tests inhereted from tempest scenarios, we complicate murano-ci 17:28:47 we will need to clone tempest, and configure it 17:29:12 But eventually we can run some set of tests on OpenStack CI? 17:29:33 (for example without deployment) 17:29:41 or with faked one 17:30:14 all engine tests with deployment 17:30:18 isn't it? 17:30:52 I think I like the idea, but we need to discuss implementation more. 17:30:57 any objections? 17:31:18 i expect our tests to run in a limited devstack env, where murano, dashboard and tempest are installed via devstack and the rest is used from a stable cloud installation 17:31:43 but that's future 17:31:43 + 17:32:24 ruhe_: I think we definitely need this feature, and I have already marked Direction as Approved. But let's discuss this thing a little bit more 17:32:54 3) https://blueprints.launchpad.net/murano/+spec/muano-api-exception-handling 17:33:12 First of all, we need to fix URI for this BP :) 17:33:19 katyafervent: :) 17:33:55 https://blueprints.launchpad.net/murano/+spec/murano-api-exception-handling 17:34:05 sjmc7: I think you had some BP/Bug related to this? 17:34:23 i think it's been gradually improving 17:34:27 https://bugs.launchpad.net/murano/+bug/1328662 17:34:28 Launchpad bug 1328662 in murano "[api] Should return JSON/XML error response bodies" [High,Confirmed] 17:34:31 there are a number of bugs/blueprints, will keep working on it 17:34:38 stan's improvements have helped too 17:34:38 Here it is 17:35:16 But murano-api still have no special exception handing 17:36:11 or similar blueprint is already exist? 17:36:16 i've stated my -1 on teh whiteboard 17:36:33 "[ruhe] +1 for idea, -1 for spec. needs more details - you need to be more explicit, you can even list exceptions you'd like to declare in murano codebase" 17:36:37 (c) 17:36:41 ruhe_: +1 17:37:10 We can assign this issue with j3, and katyafervent would update description 17:37:42 Ok 17:37:52 any objections? BP would be approved after description will be updated 17:37:58 yep, definitely needs more work and unputs 17:38:26 4) https://blueprints.launchpad.net/murano/+spec/support-keystone-v3 17:38:39 i think this was approved already 17:38:46 i'm going to try to work on it in the next couple of weeks 17:39:22 oh.. there was one other that i just remembered 17:39:29 This one is a one of the major features in J3, Direction is Approved but Definition is Discussing. 17:39:31 about more efficiently retrieving packages during deployment 17:39:46 #action katyafervent Extend the description of https://blueprints.launchpad.net/murano/+spec/murano-api-exception-handling 17:40:07 oops, sorry. regarding keystone - i haven't had a chance to look at it yet, but i think it's straightfoward 17:40:15 most of the other openstack services support v3 already 17:40:50 sjmc7: agree with you. I don't think that we need additional discussion on this 17:41:00 any objections? ruhe_ ? 17:41:31 approve it! :) 17:41:35 done :) 17:41:56 sjmc7: yeah, issue with package caching is very important... 17:42:02 * serg_melikyan searching for the link 17:43:05 https://bugs.launchpad.net/murano/+bug/1334352 17:43:08 Launchpad bug 1334352 in murano "package loading makes repeat API calls" [High,Confirmed] 17:43:32 thats actually the biggest one 17:43:35 Assigned for j3 and raised priority 17:45:05 sjmc7: you agree with this? :) 17:45:11 yeah, i think it's pretty important 17:45:21 Ok 17:46:33 all, https://launchpad.net/murano/+milestone/juno-3 - please, take a look, do we miss some important BP that is not listed, but you do like to see it here? 17:46:54 i took a quick look earlier 17:47:03 nothing else caught my eye 17:47:52 Than this is all for plans for Juno 3 :) 17:48:03 #topic Open Discussion 17:49:17 https://blueprints.launchpad.net/murano/+spec/dynamic-ui-specify-no-explicit-name-field 17:49:33 What about this blueprint? will we continue to discuss it? 17:51:10 This one is approved in direction, and I have no objections about implementation details. 17:52:29 serg_melikyan, there were some concerns from sjmc7 IIRC 17:53:02 not particularly. it just seems a bit unclean 17:53:15 i don't understand the issue enough to complain one way or the other 17:54:11 I am not sure that we should do this BP in J3, though 17:54:19 sjmc7, could you please specify how the description should be extended to make it more clean? 17:54:48 sjmc7, I mean, which part - dashboard/dynamic_ui/engine should be clarified? 17:55:03 split it up into paragraphs; exlpain exactly why something fields are going directly from the ui to the engine 17:57:45 sjmc7, have split it into paragraphs. I think stan_lagun could provide more exact info on the lifecycle of object model once it left muranodashboard and entered murano-api 17:57:48 https://github.com/murano-project/murano-app-incubator/blob/master/io.murano.apps.linux.Telnet/UI/telnet.yaml#L13 17:57:54 this part 17:58:43 we are running out of the time, let's move discussion to #murano. 17:59:00 #endmeeting