16:00:07 #startmeeting Mistral 16:00:07 Meeting started Mon Nov 17 16:00:07 2014 UTC and is due to finish in 60 minutes. The chair is rakhmerov. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:08 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:00:09 hi 16:00:11 The meeting name has been set to 'mistral' 16:00:19 hi ! 16:01:02 let's wait for others 16:01:30 Hi 16:02:47 ok, +1 :) 16:03:01 I wonder if SS folks are going to join 16:03:33 seems like we didn't see any AI from the previous time 16:03:50 so I'm going to skip our usual "Review Action Items" topic 16:03:56 #topic Review Action Items 16:04:08 #info No previous items. 16:05:01 btw, the agenda is here: https://wiki.openstack.org/wiki/Meetings/MistralAgenda 16:05:04 as usually 16:05:13 #topic Paris Summit results 16:07:20 hi there! 16:07:36 Hi, Timur 16:08:13 hey Timur 16:08:55 ok, first of all we had a quick vBrownBag talk: https://www.youtube.com/watch?v=VsGTww7fiLY 16:09:32 we also had a design session with the notes captured here: https://etherpad.openstack.org/p/paris-2014-design-summit-mistral 16:09:51 you can take a look at them 16:09:58 once you have time 16:10:19 so as for results: 16:10:57 * We gathered a good feedback from StackStorm 16:11:15 they started using it weeks ago 16:12:01 all this feedback is not sorted out yet, I'm planning to create BPs for each major item tomorrow 16:12:20 #action rakhmerov, create all needed BPs based on Paris Summit results 16:13:41 * We also got feedback from a number of people outside of StackStorm (just a few names: Angus, Zane, Roshan Agrawal and Adrian from Rackspace) 16:13:57 which mostly was very positive 16:14:50 * There were a number of conversations with folks from HP who want to use Congress, Murano and Mistral together 16:15:20 this week we have a meeting with them to discuss our next steps 16:16:02 * Also we discussed Timur's suggestion about Mistral workbook builder 16:17:07 rakhmerov, thank you for pulling me off during session :) 16:17:27 :)) 16:17:39 that's fine 16:17:41 np 16:17:50 I've got to learn how to improvise )) 16:18:11 you're totally fine 16:18:59 over time you'll be more comfortable at things like that 16:19:15 so here are the most important areas that we'll need to focus on (IMO) in the next cycle: 16:20:01 1. Race-conditions and critical bugs (especially that race condition in engine we discussed last time) 16:20:43 2. Complete 0.2 (fully implemented tested for-each, resume and join) 16:21:15 3. High-load testing & benchmarking using Rally (Anastasia started preparing for this) 16:21:31 4. UI 16:21:52 and one ongoing activity: promoting Mistral 16:21:53 Speaking of Workbook Builder, I have an update as well 16:22:15 ...basically we already started this but need to do much more 16:22:23 sure Timur, go ahead 16:23:05 I was going to ask you about all that effort because I left Paris without talking to you at the end of the summit 16:23:34 well, building UI for Heat is hot topic :) 16:24:05 and is a task we definitely should take part in 16:24:58 ok 16:25:01 so? 16:25:06 thus in order to complete most part of Merlin/Mistral work in Kilo (and do a decent work on mocking up/implementing Heat UI) I've decided that another member of Horizon team, Paul Karikh will join me in working on Merlin 16:25:06 will we? :) 16:25:30 ooh 16:25:31 yes, we will ) 16:25:33 good 16:25:51 is he from Mirantis as well? 16:26:03 yes, we hired him several months ago 16:26:12 ok 16:26:19 you may have seen while you were in Moscow, he's sitting in front of me :) 16:26:24 *seen him 16:26:28 so is he going to work on Merlin full time? 16:26:40 ooh, yes 16:26:45 I have seen him :) 16:27:04 I want him to help with implementing the new UI that Bogdan will design 16:27:16 ok 16:27:42 so what do you say about Merlin in general? 16:27:59 did you see that people were interested in it? 16:28:05 yep 16:28:10 ok, cool 16:28:27 the good news is that some UX guys are willing to help/provide feedback 16:28:47 I've talked with Michael Hinnant (HP UX team, working on OpenStack) at Summit 16:28:52 ooh, really! 16:28:55 that's great! 16:29:27 let's discuss that this week, I'd like to know more details about the whole thing 16:29:36 rakhmerov, okay 16:29:55 so guys, let's move on and quickly write our statuses 16:30:04 #topic Current Status (by team members) 16:30:34 Me: implemented musty 16:30:42 Oh, sorry 16:31:33 Last week I fixed some small bugs and worked on blueprint workflow-resume 16:31:36 my status: Last week I was on vacation, the week before I was attending the summit and YAC conference at Yandex. Planning to create all needed BPs tomorrow and start fixing that race condition (it requires some redesign, but not too serious) 16:32:09 nikolaym, did you face any major issues with "resume"? 16:32:54 Not yet, but I think it needs to be tested much more 16:33:01 ok 16:33:02 Implemented multi tenancy CLI integration tests, reduce number of ignored pep8 codes and fix appropriate pep8 issues in python-mistral client repo. started to look at Rally. 16:33:21 ok 16:33:45 I also fixed "pause" feature 16:33:47 #action rakhmerov, akuznetsova: meet and discuss integration test coverage 16:33:56 pause? 16:34:01 what was wrong with it? 16:34:03 Yes 16:34:42 did you send a patch? 16:34:55 for some reason I don't see anything related to 'pause' 16:35:06 Execution was in the pause state but tasks are continuing to execute 16:35:19 ok, it may have been 16:35:31 I included it in resume patch 16:35:43 we didn't really work through it carefully 16:35:48 ok 16:36:07 guys, do you have any major issues you'd like to discuss? 16:36:30 I only have one topic to discuss: Release 0.2 progress 16:36:56 so if you don't have anything else I would quickly go over what's left on it 16:37:57 #topic Release 0.2 16:38:05 so the link to the milestone: https://launchpad.net/mistral/+milestone/0.2 16:38:28 good news that Winson started working on event listeners 16:38:42 we need to help him with details if needed 16:39:11 We need to update bps' statuses 16:39:25 I think they're mostly up to date 16:40:31 ok, so we have 4 BPs I would consider done or mostly done 16:40:35 including 16:40:36 I know that there are a few already implemented bps 16:40:46 data collections (for-each) 16:40:49 which ones? 16:41:27 https://blueprints.launchpad.net/mistral/+spec/mistral-task-executor-affinity 16:41:40 as for for-each and resume I would hold on for now and keep them in "In progress" state until we make sure they're fully tested 16:41:47 I know that Nikolay worked on it 16:41:57 And tested it 16:42:02 yees, it says it's in "Beta available" which is fine IMO 16:42:10 yeah 16:42:30 Recently I updated bps statuses 16:42:31 I consider "Beta available" as "Completed" with the only difference that it's not released yet 16:42:45 maybe I'm wrong though 16:43:09 so we have also 4 BPs that we didn't start working on yet 16:43:46 which are basically: "join", "pause-before" and HA & Scalability 16:44:41 well, in fact, I already started working on "join" but have to postpone it for a couple of days in order to make changes in the engine 16:44:41 This one https://blueprints.launchpad.net/mistral/+spec/mistral-manual-control-checkpoints looks like duplicate 16:44:46 No? 16:44:51 almost 16:45:20 it's basically a more high-level BP that includes "pause-before" and "resume" 16:45:44 Hmm, ok 16:45:44 we created it just to keep the major user meaningful feature under control 16:46:01 but yes, it's not something we have to work on separately 16:46:05 in that sense you're right 16:46:13 it's a duplicate 16:47:03 so overall I think our progress is ok 16:47:07 just one comment 16:47:33 Engine HA & Scalability I think is not something that we need to do just once and forget about it 16:48:13 I think in 0.2 we'll need to do the most important testing and fix the most important bugs that we'll find (I'm sure we will) 16:48:21 and then continue this work 16:48:37 that's basically why we need to use Rally 16:48:47 to automate it 16:48:59 and keep creating more tests 16:49:25 also I'd like to capture a couple of more screencasts about new features 16:49:36 it will take a day or two 16:50:09 I think that's it from my side 16:50:18 something from yours? 16:50:41 Nothing from me 16:51:23 ok 16:51:42 then 16:51:44 thank you 16:51:53 good bye to all 16:52:00 Bye ;) 16:52:07 #endmeeting