15:01:28 <rakhmerov> #startmeeting Mistral 15:01:30 <openstack> Meeting started Mon Apr 3 15:01:28 2017 UTC and is due to finish in 60 minutes. The chair is rakhmerov. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:01:31 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:01:34 <openstack> The meeting name has been set to 'mistral' 15:01:37 <d0ugal> Hey all! 15:01:39 <mgershen> o/ 15:01:42 <sharatss> hi 15:01:56 <akuznetsova> hi there 15:02:12 <rakhmerov> hi 15:02:16 <rbrady> o/ 15:02:53 <rakhmerov> hi all 15:02:54 <rakhmerov> 1 sec 15:03:47 <rakhmerov> #topic Review Action items 15:03:50 <rakhmerov> 1. d0ugal to check for doc liaison tasks 15:03:59 <d0ugal> oops, I have not done this. 15:04:05 <rakhmerov> :) 15:04:09 <rakhmerov> #action d0ugal to check for doc liaison tasks 15:04:17 <toure> hello... 15:04:32 <rakhmerov> toure: hey 15:04:34 <rakhmerov> 2. rakhmerov: ask Michal to review the spec about mistral-extra 15:04:35 <rakhmerov> done 15:04:50 <rakhmerov> 3. mgershen/team to talk with Renat about bug https://bugs.launchpad.net/mistral/+bug/1672379 15:04:50 <openstack> Launchpad bug 1672379 in Mistral "default task on error not running when task fail to evaluate input" [Undecided,Invalid] 15:05:03 <d0ugal> rakhmerov: I think maybe that wasn't needed in the end 15:05:13 <rakhmerov> yes 15:05:28 <rakhmerov> we've kind of touched that, it's not an easy thing to solve now 15:05:48 <rakhmerov> 4. rbrady start sensitive data spec 15:06:02 <rakhmerov> it was done 15:06:08 <mgershen> about 3, should I open a blueprint? 15:06:12 <toure> hey rakhmerov once I finish fighting with gerrit I have the client code ready for review 15:06:14 <rakhmerov> it even has one +2 15:06:15 <toure> :) 15:06:28 <rakhmerov> toure: for error analysis? 15:06:32 <toure> yup 15:06:36 <rakhmerov> awesome ) 15:06:41 <toure> merge conflicts 15:06:43 <d0ugal> #link https://review.openstack.org/450853 15:06:43 <toure> :( 15:06:50 <rakhmerov> mgershen: yes, please 15:06:57 <d0ugal> ^ that's the sensitive data spec 15:07:06 <rakhmerov> yes 15:07:30 <rakhmerov> ok 15:07:42 <mgershen> rakhmerov: do you want to add an action to make sure I won't forget? 15:08:16 <rakhmerov> #action mgershen: create a BP about "on-error" semantics 15:08:25 <rakhmerov> #topic Current status (quick info from team members) 15:09:43 <rakhmerov> my status: reviewing, working on advanced publishing (actually got stuck with lang changes unexpectedly, will probably have to do some serious refactoring first), investigating an issue introduced by new SQLAlchemy 15:09:53 <rakhmerov> the last one was finally fixed today 15:10:14 <rakhmerov> so please provide a short update if you have anything 15:10:39 <d0ugal> I'm now working on mistral-extra, and starting the ground work for moving the OpenStack actions over. The spec merged today (thanks for reviews!) 15:10:54 <mgershen> I will have almost no time in the next few weeks to work on mistral, but if something needs reviewing try to ping me. 15:10:55 <rakhmerov> that is great 15:11:11 <rakhmerov> mgershen: sure, understandable 15:11:53 <rakhmerov> just FYI: akuznetsova is now helping us with some tasks again (which is awesome) 15:12:10 <rakhmerov> akuznetsova: could you please tell a little bit about your activities? 15:12:39 <akuznetsova> yes, for the old times' sake 15:12:43 <rakhmerov> :) 15:12:52 <akuznetsova> I've published fixes for two bugs 15:13:14 * d0ugal needs to catch up with reviews. 15:13:23 <rakhmerov> d0ugal: yes! :) 15:13:25 <akuznetsova> and looked through the list of bugs in mistralclient, there were a lot of old bugs which I closed 15:13:39 <d0ugal> bug triage ++ 15:14:08 <rakhmerov> d0ugal: yeah, I was actually going to do that today at the meeting ) 15:14:29 <rakhmerov> but akuznetsova volunteered to do some pre-filtering 15:14:43 <d0ugal> great 15:14:55 <rakhmerov> akuznetsova: thanks a lot! 15:14:56 <mgershen> thank you akuznetsova :) 15:15:11 <apetrich> I'm working on moving some imports from mistral to mistral-lib mostly in tripleo-common (sorry for for the tripleo centric status) 15:15:14 <akuznetsova> rakhmerov, mgershen you are welcome 15:15:57 <d0ugal> apetrich: it is extremely valuable for Mistral to show this stuff working, it is all theoretical so far :) 15:16:05 <rakhmerov> mgershen, d0ugal: we got a bunch of patches passed through check queue today after we rebased them after merging an SQLA issue so please try to find some time to review them 15:16:34 <d0ugal> sure 15:16:52 <mgershen> rakhmerov: do you have links? 15:17:04 <rakhmerov> d0ugal, apetrich: this is very very good, the only thing I'm worried about is that isn't it too early? :) 15:17:13 <apetrich> d0ugal, I can see that 15:17:22 <rakhmerov> mgershen: there's too many of them, just look at gerrit pls 15:17:40 <d0ugal> rakhmerov: possibly - but we need to start some time and it'll help direct us what we need to do next. We wont land anything in tripleo yet. 15:17:41 <apetrich> rakhmerov, it is more of proof of concept and identify the early pitfalls 15:17:46 <d0ugal> +1 15:17:47 <rakhmerov> ooh, ok 15:17:55 <rakhmerov> if so then it's fine 15:18:10 <rakhmerov> because otherwise the APIs may not be stable yet etc. 15:18:48 <rakhmerov> then this is good really 15:18:49 <mgershen> rakhmerov: ok, I will. 15:18:52 <rakhmerov> ok 15:19:35 <rakhmerov> mgershen: I just usually look at "Watched Changes" tab and see patches that recently changed on the top 15:21:13 <rakhmerov> ok, is there anything else on that? 15:21:24 <rakhmerov> rbrady: do you have any updates today? 15:22:07 <rbrady> rakhmerov: thrash took care of the spec, I've only done reviews since last week 15:22:26 <rakhmerov> ok, thanks 15:22:27 <rakhmerov> good 15:22:58 <rakhmerov> rbrady: as far as mistral-lib, are you planning to make more changes in the near future? 15:23:18 <rakhmerov> rbrady: what do you think is now left? 15:23:28 <rakhmerov> from what we planned 15:23:32 <rbrady> rahkmerov: I think the next steps are the keystone_utils and the openstack actions 15:23:39 <d0ugal> We need Mistral to depend on msitral-lib and to pass the context 15:23:56 <d0ugal> rbrady: ^ I think we need that first maybe, but it is easy to hack in for testing :) 15:23:58 <rbrady> ^^ and what d0ugal said :) 15:24:28 <d0ugal> I have a patch locally that does it, I could put it up for review but we need mistral-lib in global-requirements first 15:24:43 <d0ugal> I'll post it as a starting point at least. 15:24:56 <rakhmerov> yes 15:24:58 <rakhmerov> agree 15:25:27 <rakhmerov> d0ugal: having it in global-requirements should not be a huge problem, right? 15:25:38 <rakhmerov> afaik 15:25:44 <d0ugal> rakhmerov: no, that should be easy 15:25:52 <d0ugal> I just need to find the process and follow it 15:26:04 <rakhmerov> ok 15:26:32 <d0ugal> and we need to figure out rpm packaging for the tripleo side - just adding the dep will break tripleos CI I think 15:26:44 <d0ugal> but again, that should be quite easy. 15:27:00 <rakhmerov> yep 15:27:26 <rakhmerov> so, we can probably go to an open discussion 15:27:46 <rakhmerov> I don't have any special topics today 15:28:03 <rakhmerov> if you do please bring them up 15:28:12 <rakhmerov> #topic Open Discussion 15:29:04 <rakhmerov> one thing I'd like to share about what I'm doing is that I stumbled on making changes in workflow language in the last few days 15:29:22 <rakhmerov> I found that it became very hard to make such changes 15:29:51 <d0ugal> What makes it so difficult? 15:29:52 <rakhmerov> this Spec class hierarchy became enormously complex 15:29:55 <d0ugal> ah 15:30:12 <d0ugal> The code in mistral.lang.*? 15:30:18 <rakhmerov> d0ugal: too many places to change in order to make even relatively small changes 15:30:21 <rakhmerov> yes 15:30:35 <d0ugal> Interesting, I wonder how we can make it easier 15:30:39 <rakhmerov> and then if you look at how to alter tests it's not easier ) 15:30:53 <rakhmerov> d0ugal: yeah, that's what I'm thinking about now 15:31:04 <rakhmerov> I have some ideas, will try to implement them 15:31:31 <rakhmerov> one more thing related to that is we may want to have a new version of the language some time in the future 15:31:39 <rakhmerov> and I believe it may happen relatively soon 15:32:02 <rakhmerov> because we've been adding more and more feature into the language 15:32:36 <rakhmerov> they are all backwards compatible and hence we lose some opportunities to make the lang more concise and elegant 15:32:36 <d0ugal> Migrating workflows doesn't sound like fun :) 15:32:46 <rakhmerov> no, wait :) 15:32:50 <rakhmerov> I didn't mean that ) 15:33:00 <d0ugal> haha, I know 15:33:09 <rakhmerov> what I meant is that we'll have to support more than one language version 15:33:18 <rakhmerov> I know, sounds scary 15:33:28 <d0ugal> yup, but even in that case we would want to upgrade older workflows over time. 15:33:37 <rakhmerov> and especially scary for me after I looked again at how it's now implemented :) 15:33:42 <d0ugal> lol 15:34:15 <rakhmerov> d0ugal: yeah, true we'll probably need to provide some aid to upgrade to a newer version 15:34:54 <rakhmerov> anyway, this is something that I've discussed with a number of people already and seems like this idea of the new lang version may not be that theoretical 15:35:11 <rakhmerov> so, that's just one thing I'm involved in now 15:35:28 <d0ugal> cool, it would be interesting to see those ideas. 15:35:31 <rakhmerov> ideally I'd like to improve it somehow 15:35:46 <rakhmerov> and btw, I found one thing that makes WF validation slower, I'll fix it soon 15:36:12 <rakhmerov> yeah, the mechanism is powerful but it became too heavy 15:36:49 <rakhmerov> that is all from my side 15:37:02 <d0ugal> I don't really have anything to discuss this week. 15:37:14 <rakhmerov> guys, next time let's sync on Pike-1 tasks 15:37:35 <d0ugal> sure 15:37:38 <toure> ack 15:37:49 <rakhmerov> please try to find time to check if you're going to meet the scope of Pike-1 15:38:39 <rakhmerov> another topic I'd like to bring up next week is something I discussed with my boss at Nokia recently 15:39:20 <rakhmerov> so essentially I set up 4 goals for 2017 regarding Mistral 15:39:30 <rakhmerov> mostly from our business perspective 15:39:58 <rakhmerov> but anyway, I think these goals reflect my vision of how Mistral should evolve in the next year 15:40:11 <rakhmerov> I'd be interested in discussing these goals with you 15:40:16 <d0ugal> cool 15:40:18 <toure> sounds good 15:40:20 <rakhmerov> so that we all understand the direction 15:40:29 <d0ugal> Sounds like it'll be a busy meeting :) 15:40:33 <rakhmerov> yes 15:41:13 <rakhmerov> my suggestion would be even to declare these goals (if we agree on them) somewhere, in the spec repo or a wiki (seriously outdated now btw) 15:41:26 <d0ugal> a spec? 15:41:33 <rakhmerov> maybe a spec, yes 15:42:11 <rakhmerov> so, just to be even more clear: I believe these goals is the 3-4 main things that we should be focusing on in order to make the project more successful 15:42:22 <rakhmerov> and attract more people to it (both users and contributors) 15:42:42 <rakhmerov> ok, details next time ) 15:42:52 <rakhmerov> I am done for sure 15:42:55 * toure waits for details 15:42:55 <rakhmerov> now 15:42:56 <toure> :) 15:42:56 <d0ugal> look forward to it :) 15:43:04 <d0ugal> toure: haha, rakhmerov is teasing us! 15:43:11 <rakhmerov> :)) 15:43:11 <toure> indeed 15:43:14 <toure> hehe 15:43:24 <rakhmerov> yeah, may be a little :) 15:43:37 <rakhmerov> I've been teasing d0ugal for a looong time with a UI tool :) 15:43:51 <rakhmerov> that is something I'd really love to demonstrate some time soon 15:44:10 <toure> cool 15:44:16 <toure> can't wait to see it 15:44:32 <d0ugal> +1 15:44:36 <rakhmerov> I'm going to bring that up soon with my management in order to find out when and how we're going to announce it 15:44:55 <rakhmerov> ok 15:45:04 <rakhmerov> anything else may be? 15:45:13 <d0ugal> not from me 15:45:16 <rakhmerov> or we can finish for today 15:45:20 <toure> nothing for me 15:45:28 <mgershen> nothing for me 15:45:31 <akuznetsova> nothing from my side 15:46:37 <rakhmerov> ok 15:46:49 <rakhmerov> thank you all for coming and for the interesting discussion 15:46:54 <rakhmerov> see you next week 15:46:59 <d0ugal> thanks all! 15:47:03 <toure> have a good one 15:47:05 <rakhmerov> have a wonderful week guys ) 15:47:06 <akuznetsova> thanks, bye 15:47:11 <rakhmerov> bye 15:47:12 <toure> ta 15:47:16 <rakhmerov> #endmeeting