16:01:54 <rakhmerov> #startmeeting Mistral
16:01:56 <openstack> Meeting started Mon May 23 16:01:54 2016 UTC and is due to finish in 60 minutes.  The chair is rakhmerov. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:01:57 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:02:00 <openstack> The meeting name has been set to 'mistral'
16:02:12 <rakhmerov> hi
16:02:20 <rbrady> hey!
16:02:30 <rakhmerov> rbrady: how are you?
16:03:08 <mgershen> o/
16:03:23 <ddeja> Hi
16:03:42 <rbrady> rakhmerov: doing well
16:04:45 <rakhmerov> ddeja: hi
16:04:48 <rakhmerov> mgershen: hi )
16:04:53 <hparekh_> hi'
16:05:02 <rakhmerov> hi hi )
16:05:08 <rakhmerov> #topic Review Action items
16:05:15 <d0ugal> o/
16:05:56 <rakhmerov> 1. rakhmerov: check if PyPi allows to rename releases
16:06:17 <rakhmerov> yes, it does but I still didn't do it
16:06:29 <rakhmerov> was doing something else
16:06:55 <rakhmerov> #topic Current status (progress, issues, roadblocks, further plans)
16:07:04 <rakhmerov> let's quickly tell our updates, if any
16:08:01 <rakhmerov> my status: doing a huge refactoring in Mistral engine, it's challenging but 100% required, https://review.openstack.org/#/c/317879/
16:08:13 <rakhmerov> most of the tests already started passing
16:08:13 <hparekh_> working on mistral-fail-transition-message.
16:08:25 <mgershen> I still didn't get the chance to complete the test for validate action cli call
16:08:47 <ddeja> My status: Not many work unfortunately, I still didn't recover fully and I didn't come back to office. But I have take a look into old patches implementing own RPC layer. I'm starting to working on rebasing them from Wednesday
16:08:54 <rakhmerov> mgershen: I suggest you add a test in a separate patch, IMO it's ok in case of functional tests
16:09:23 <rakhmerov> ddeja: ok, sure. Take care of yourself and get better
16:09:44 <mgershen> OK, I'll push later today or tomorrow.
16:09:56 <rakhmerov> ddeja: once you're back, find me in IRC and we'll discuss your next steps
16:10:02 <rbrady> I worked on the bug 1581649 for a while.  I eventually reverted TripleO to use eventlet again for now and will come back to this bug once the other integration tasks are done
16:10:04 <openstack> bug 1581649 in Mistral "Action Execution Response Timeout" [High,New] https://launchpad.net/bugs/1581649 - Assigned to Ryan Brady (rbrady)
16:10:06 <ddeja> rakhmerov: Thanks, I'm ok now and starting regular work from Wednesday, so I hope to push this case much faster :)
16:10:18 <d0ugal> rakhmerov: Is there a description of the refactoring goals/plans?
16:10:55 <rakhmerov> d0ugal: there's a bunch of blueprints that together explain what needs to be fixed
16:11:20 <d0ugal> rakhmerov: k, I'll take a look. It would be good to reference them in the commit message
16:11:35 <mgershen> ddeja: feel better
16:11:56 <rakhmerov> d0ugal: yes, sure. It's just not complete yet, but you're right. I'll insert all needed info
16:12:03 <rakhmerov> for example, https://blueprints.launchpad.net/mistral/+spec/mistral-engine-error-handling
16:12:10 <d0ugal> rakhmerov: thanks :)
16:12:22 <rakhmerov> once I started it I realized that a lot of things need to be improved
16:13:08 <rakhmerov> #action rakhmerov: add all necessary info (links to BPs and bugs) to engine refactoring patch
16:14:15 <rakhmerov> one more thing, kong also sent a patch to enforce upper constraints, now we're trying to backport it into stable/mitaka
16:14:38 <rakhmerov> because now this branch is stuck due to new version of oslo.messaging
16:15:02 <rakhmerov> #topic Discuss what should be finalized in Newton-1
16:16:00 <rakhmerov> here's what's assigned to Newton-1: https://launchpad.net/mistral/+milestone/newton-1
16:16:23 <rakhmerov> https://blueprints.launchpad.net/mistral/+spec/mistral-alternative-rpc
16:16:40 <rakhmerov> ddeja: I'll move it to Newton-2 for now
16:16:52 <rakhmerov> and reassign it to you
16:16:53 <rakhmerov> ok?
16:17:02 <ddeja> rakhmerov: OK
16:17:46 <rakhmerov> honestly, I'm not even sure if we'll get it done in Newton-2 but if we do it would be awesome
16:18:00 <rakhmerov> https://blueprints.launchpad.net/mistral/+spec/mistral-custom-actions-api
16:18:17 <ddeja> rakhmerov: I hope it can be done in Newton-2, a lot of work is already done
16:18:25 <rakhmerov> same here, just one week left in N-1 so need to move it to N-2 too
16:18:31 <rakhmerov> yes
16:18:36 <rakhmerov> hope so
16:19:21 <rakhmerov> rbrady: with Actions API, technically we can start working on it
16:19:42 <rakhmerov> at least writing some minimal spec with goals, approaches etc.
16:20:27 <rakhmerov> like we discussed, I can propose an initial spec before the end of N-2 and then we'll complete it and decide who's going to contribute and how
16:21:19 <rakhmerov> obviously we'll not finish it during a week but I think N-2 looks feasible
16:21:44 <rakhmerov> we discussed this BP internally and decided that it's a high priority thing for us too
16:21:48 <rakhmerov> I mean Nokia
16:22:45 <rakhmerov> #action rakhmerov: propose an initial spec for Custom Actions API
16:23:31 <rakhmerov> https://blueprints.launchpad.net/mistral/+spec/mistral-fail-transition-message
16:23:41 <rakhmerov> hparekh_: I think this one will be done
16:23:56 <rakhmerov> tell me if I'm wrong
16:23:58 <hparekh_> rakhmerov: yrah it will be done in N-1
16:24:05 <rakhmerov> ok, great
16:24:10 <rakhmerov> I just updated its status
16:24:47 <rakhmerov> https://blueprints.launchpad.net/mistral/+spec/mistral-priority-task-scheduler
16:25:18 <rakhmerov> this one is on Winson (StackStorm) but he's now working on a different task so this one will be moved to a later milestone
16:25:29 <rakhmerov> I even think it should be N-3 for now
16:25:59 <rakhmerov> https://blueprints.launchpad.net/mistral/+spec/validate-ad-hoc-action-api
16:26:17 <rakhmerov> I expect that this one will be done in N-1
16:26:30 <rakhmerov> mgershen: seems like there's not too much left on it
16:26:38 <rakhmerov> right?
16:26:52 <rakhmerov> fix a client a little bit and functional tests
16:27:02 <mgershen> I think I can move this to done now
16:27:19 <mgershen> The client is a different project
16:27:33 <rakhmerov> formally yes
16:28:06 <mgershen> What do you think? Should I? The code is In for the API
16:28:09 <rakhmerov> for now, I'd keep it "IN PROGRESS" because we don't pay too much attention to a client project at LP
16:28:46 <rakhmerov> if you move it to done then please make sure there's a BP at a client LP project
16:29:00 <rakhmerov> so that we know how to track it
16:29:01 <mgershen> Ok
16:29:20 <rakhmerov> as far as other BPs, I'm confident that they will be done
16:29:42 <mgershen> If It doesn't have to wait for the test, there is no problem
16:29:42 <rakhmerov> except UI BPs that I first need to discuss with a UI developer (he's not here)
16:30:23 <rakhmerov> as far as bugs
16:31:03 <rakhmerov> it'd be good to fix https://bugs.launchpad.net/mistral/+bug/1581649 and https://bugs.launchpad.net/mistral/+bug/1584731 in N-1
16:31:06 <openstack> Launchpad bug 1581649 in Mistral "Action Execution Response Timeout" [High,New] - Assigned to Ryan Brady (rbrady)
16:31:08 <openstack> Launchpad bug 1584731 in Mistral "Intermitent ActionExecution not found" [Critical,New]
16:31:35 <rakhmerov> we've already done some investigation with rbrady on the first one and know approximately where the problem is
16:32:10 <rakhmerov> seems like it's an issue with eventlet under Apache
16:32:28 <rakhmerov> as far as the second one, any volunteers?
16:32:29 <rakhmerov> :)
16:32:49 <rakhmerov> I think it's something relatively simple
16:33:23 <rakhmerov> hparekh: do you think you'll have capacity this week to look at it?
16:33:28 <rakhmerov> or early next week
16:34:26 <hparekh> yeah sure I will finish my bp then i can look into it sure
16:34:29 <rbrady> rakhmerov:  I plan on taking another go at it towards the end of this week, but I'm definitely open to help or collaborating
16:34:38 <rakhmerov> hparekh: ok, assigned to you
16:35:07 <hparekh> rakhmerov ok
16:35:24 <rakhmerov> rbrady: yes, thanks. I actually didn't understand something
16:35:31 <rakhmerov> you said that you switched back to Apache
16:35:43 <rakhmerov> but it does not work with Apache
16:35:58 <rakhmerov> I may have misunderstood you
16:36:10 <rbrady> rakhmerov: https://review.openstack.org/#/c/319262/
16:36:36 <rakhmerov> ooh, I see
16:36:42 <rakhmerov> thanks
16:37:20 <rakhmerov> anyway, I'd like to solve this since a lot of people will be stumbling on it
16:37:24 <rbrady> rakhmerov: this should keep tripleo undercloud working for now, but the idea is apache would be a better place to serve it like nova and keystone
16:37:35 <rakhmerov> yes
16:38:22 <rakhmerov> mgershen: Michal, please also look at https://bugs.launchpad.net/mistral/+bug/1581514 once you have 30 mins
16:38:23 <openstack> Launchpad bug 1581514 in Mistral "Misinterpretation of action execution state by ActionExecutionsController" [High,Confirmed] - Assigned to Michal Gershenzon (michal-gershenzon)
16:38:39 <rakhmerov> it's pretty simple and we can fix it sooner than later
16:39:42 <rakhmerov> as far as other bugs, I'll move 2-3 to N-2 as they won't be finished this week
16:39:47 <mgershen> rakhmerov: sure np
16:40:41 <rakhmerov> https://bugs.launchpad.net/mistral/+bug/1484521, https://bugs.launchpad.net/mistral/liberty/+bug/1424461
16:40:44 <openstack> Launchpad bug 1484521 in Mistral "wait-before and retry policies directly call task_handler.run_existing_task() method via RPC" [High,Confirmed] - Assigned to Renat Akhmerov (rakhmerov)
16:40:45 <openstack> Launchpad bug 1424461 in Mistral "Published vars lost for complex parallel join" [High,Confirmed] - Assigned to Renat Akhmerov (rakhmerov)
16:41:46 <rakhmerov> ok, let's move to the next topic
16:41:48 <rakhmerov> #topic Discuss Newton-2 scope
16:42:12 <rakhmerov> https://launchpad.net/mistral/+milestone/newton-2
16:42:38 <rakhmerov> there will be several additions as a legacy of N-1
16:43:20 <rakhmerov> for now, I'd like to discuss 4-5 major things probably that we need to tackle in N-2
16:44:04 <rakhmerov> I'll select BPs based on that and on the next meeting we need to discuss that again in details and assign tasks
16:44:41 <rakhmerov> so, the thing #1 - Custom Actions API
16:44:41 <rakhmerov> https://blueprints.launchpad.net/mistral/+spec/mistral-custom-actions-api
16:45:15 <rakhmerov> it causes a lot of pain already and it needs to be done sooner because it's a serious gap in API
16:45:34 <rakhmerov> no matter whether it's a Python API or not
16:46:06 <rakhmerov> I realized that probably we'll have to move that Python API (as well as probably some other stuff) into a separate subproject
16:46:13 <rakhmerov> like mistral-common
16:46:19 <rakhmerov> or something like this
16:46:26 <rakhmerov> but we'll discuss it in the spec
16:46:38 <d0ugal> I would like to help with this if I can
16:47:15 <rakhmerov> d0ugal: sure, if you can please attend the next meeting
16:47:21 <rbrady> rakhmerov: do you want to start a spec patch or want one of use to do it?
16:47:29 <d0ugal> rakhmerov: will do
16:47:30 <rbrady> s/use/us/
16:47:49 <rakhmerov> rbrady: if you have time, please go ahead
16:47:58 <rakhmerov> it doesn't matter who starts it
16:47:59 <rbrady> rakhmerov: ack
16:48:15 <rakhmerov> kind of related to this: https://blueprints.launchpad.net/mistral/+spec/mistral-separate-openstack-actions
16:48:25 <rakhmerov> I think it's important too
16:48:36 * d0ugal reads
16:49:07 <d0ugal> oh, I hadn't seen that one. Sounds like a great plan.
16:49:19 <rakhmerov> btw, this all comes from my head based on our summit discussions but if you disagree with something pls disagree out loud )
16:49:56 <rakhmerov> d0ugal: yeah, there's many reasons why it should be done, better sooner
16:50:08 <rakhmerov> assigned it to N-2
16:50:57 <rakhmerov> ok, what else
16:51:37 <rakhmerov> I'd finally like to propose a spec for API v3: https://blueprints.launchpad.net/mistral/+spec/mistral-api-v3
16:52:13 <rakhmerov> it's been going on for a long time, I see that we need to start improving usability
16:52:27 <rakhmerov> speaking of API v3, I don't mean workflow language
16:52:55 <rakhmerov> language is ok, although we're planning a number of small additions to it (backwards compatible)
16:53:11 <rbrady> it would be great to expand that spec.  it's not entirely clear from the lp description
16:53:18 <rakhmerov> #action rakhmerov: Propose a spec for API v3 in N-2
16:53:33 <rakhmerov> rbrady: yeah-yeah, true
16:53:56 <rakhmerov> I already started making some notes on what we need to fix in API v2 but I didn't sort them out yet
16:54:06 <rakhmerov> will try to share soon
16:54:09 <rbrady> ack
16:55:39 <rakhmerov> most likely, we'll also need https://blueprints.launchpad.net/mistral/+spec/mistral-multi-region-support
16:55:41 <rakhmerov> soon
16:55:48 <rakhmerov> will assign it to N-2 too
16:57:51 <rakhmerov> there's also one thing that I'd like to do: https://blueprints.launchpad.net/mistral/+spec/mistral-logo
16:57:51 <rakhmerov> :)
16:58:03 <rakhmerov> but we need a designer here I guess
16:58:14 <rakhmerov> anyway, share your ideas!
16:58:23 <d0ugal> rakhmerov: I know who to ask :)
16:58:26 <rbrady> rakhmerov: I know a designer (the one who made the tripleo logo)
16:58:30 <d0ugal> jrist: ^
16:58:36 <d0ugal> rbrady: ha, snap.
16:58:42 <rbrady> yes, jrist!
16:58:44 <rakhmerov> d0ugal: ooh, that would be awesome
16:58:45 <jrist> haha
16:59:05 <rbrady> rakhmerov: how did Mistral get its name?
16:59:09 <jrist> designer might be a stretch
16:59:21 <d0ugal> I joked to jrist on Friday that he should make a Mistral logo, so I am glad one is wanted :)
16:59:22 <rakhmerov> ok, we're running out of time. I'll propose more BPs and bugs till the next meeting, let's discuss more next time
16:59:34 <d0ugal> +1
16:59:41 <rakhmerov> rbrady: it was an idea of one of my managers at Mirantis :)
17:00:08 <rakhmerov> I don't know how it came up, I guess just because the word itself is pretty beautiful
17:00:27 <d0ugal> Yeah, it's a cool name
17:00:27 <rakhmerov> d0ugal: haha, you really discussed that?? )
17:00:40 <d0ugal> rakhmerov: Yup, because he just done a logo for my python usergroup
17:00:44 <d0ugal> anyway... time up :)
17:00:48 <rakhmerov> hahaha ))
17:00:50 <rakhmerov> awesome
17:00:59 <rakhmerov> yes, thanks for joining guys
17:01:01 <rakhmerov> bye
17:01:05 <rakhmerov> #endmeeting