14:01:09 <boris-42> #startmeeting Rally
14:01:10 <openstack> Meeting started Mon Jan 25 14:01:09 2016 UTC and is due to finish in 60 minutes.  The chair is boris-42. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:01:11 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:01:13 <openstack> The meeting name has been set to 'rally'
14:01:36 <andreykurilin> hi hi
14:01:42 <ikhudoshyn> o/
14:01:47 <andreykurilin> \o
14:01:48 <redixin> sup
14:01:52 <boris-42> hi hi
14:02:19 <andreykurilin> do we have agenda for meeting today? :)
14:02:22 <yfried_> bogdando: hi
14:02:37 <bogdando> yfried_, hi
14:03:41 <boris-42> bogdando: hi
14:03:48 <boris-42> yfried_: hi hi
14:04:25 <boris-42> #topic missing agenda
14:04:28 <boris-42> #link https://wiki.openstack.org/wiki/Meetings/Rally
14:04:36 <boris-42> Okay seems like we don't have agenda for today
14:04:59 <boris-42> #topic Broken gates and pain *
14:05:56 <boris-42> Okay I just would like to say that whole previous week we were fighting to make everything work back
14:05:59 <boris-42> results are next
14:06:41 <boris-42> 1) We refactored rally verify command tempest installation part (so we don't use tempests/tools which is going to be removed and actually doens't work well)
14:07:21 <boris-42> 2) We upgrade pip to the version <8 in rally instal script
14:07:40 <boris-42> as well amaretsky improved our dsvm-cli job to show how long different tests takes
14:07:55 <boris-42> so overall we didn't do much
14:08:09 <boris-42> #topic free discussion
14:08:25 <boris-42> yfried_: bogdando redixin andreykurilin ikhudoshyn andreykurilin does anybody have something to share?
14:08:30 <redixin> yep
14:08:40 <redixin> I made a patch WIP
14:08:50 <redixin> https://review.openstack.org/#/c/272101/
14:09:07 <redixin> Is just a helper class for heat client
14:09:40 <redixin> boris-42: I have a doubt where this file should be
14:09:58 <boris-42> redixin: please do not add any more wrappers=(
14:10:05 <boris-42> redixin: wrappers are really bad approach
14:10:20 <boris-42> redixin: each line that we are adding to wrappers/* module should be replaced in future
14:10:43 <boris-42> redixin: we really need to finish spec
14:11:33 <boris-42> redixin: https://review.openstack.org/#/c/172831/
14:12:03 <boris-42> redixin: and at least implement new functionality in new way
14:13:04 <redixin> boris-42: so what can I do?
14:13:15 <boris-42> redixin: why do you need that helper at all?
14:13:50 <boris-42> redixin: and yes you can review a bit code and specs of others ><
14:14:15 <boris-42> redixin: http://stackalytics.com/report/contribution/rally/30 7 review / 30 days is not serious =(
14:14:18 <redixin> I need this helper to avoid code duplication
14:14:42 <boris-42> redixin: so we need to work together with yfried_ to finish that spec
14:14:52 <boris-42> redixin: after that we will implement your code in new style
14:14:57 <boris-42> redixin: this is what we need to do
14:15:14 <redixin> so I cant continue before this spec is done?
14:15:29 <redixin> or should I remove that class and just copy-paste code?
14:15:49 <yfried_> redixin: boris-42: have you seen shade?
14:16:03 <yfried_> redixin: it does a lot of what the wrappers do
14:16:34 <redixin> yfried_: hever heard
14:16:34 <boris-42> yfried_: repo?
14:17:07 <yfried_> boris-42: http://docs.openstack.org/infra/shade/
14:17:26 <boris-42> yfried_: ah that one
14:17:31 <boris-42> yfried_: no we won't use that one
14:17:38 <boris-42> yfried_: it's just hardcoded bunch of commonly used methods
14:17:42 <redixin> boris-42: yfried_: btw it is not actually wrapper. is just helper class which remember the state and allows to update stack in one line of code
14:17:45 <boris-42> yfried_: it won't work for us
14:17:56 <boris-42> redixin: we need to finish that spec first
14:18:12 <boris-42> redixin: that spec is all about how to deal with the situation that you faced
14:19:59 <boris-42> redixin: so please help with reviews of that proposal
14:20:20 <boris-42> redixin: and let's finish it asap
14:20:44 <redixin> boris-42: so, in few words, where should be such helper classes?
14:20:54 <boris-42> redixin: did you read spec?
14:21:26 <redixin> no
14:21:39 <redixin> I can see only 300 lines of samples and ascii trees
14:23:31 <boris-42> redixin: there is code as well there and description
14:25:09 <boris-42> redixin: so please take a look at spec and put your commens in all unclear place
14:25:17 <boris-42> redixin: I will find some time today to re-review it
14:25:39 <boris-42> ikhudoshyn: andreykurilin ^ please take a look as well
14:25:49 <andreykurilin> sure
14:25:51 <ikhudoshyn> will do
14:27:41 <ikhudoshyn> if we're done with the topic I'd like to discuss some other things
14:28:52 <boris-42> redixin: so do you have something else to discuss?
14:29:04 <boris-42> redixin: how work on workload framework is going?
14:29:50 <redixin> boris-42: almost done
14:30:16 <redixin> boris-42: but wait. what do you mean "workload framework"?
14:30:41 <redixin> boris-42: it is not actually framework, it is just scenario
14:30:53 <boris-42> redixin: ya but it should allow to run different stuff
14:31:14 <boris-42> redixin: so it's some kind of framewrok for doing different kinds of workloads
14:31:17 <redixin> yes, this scenario can run anything
14:31:47 <boris-42> redixin: btw why you don't propose a patch?
14:32:25 <redixin> because almost done != done
14:32:44 <redixin> heat helper class is first part of it
14:34:12 <boris-42> redixin: so then we have a blocker that spec
14:34:16 <redixin> boris-42: also I cant finish this patch because of spec "Refactoring scenario utils@ is not finished yet
14:35:18 <boris-42> redixin: so where you use that code? and why you can't for now reuse the scenario/heat/utils?
14:35:58 <redixin> because I need to save the state of stack to update it later
14:36:11 <redixin> scenario/heat/utils cant help with this
14:37:11 <boris-42> redixin: but as far as I know scenario heat utils will return stack instance
14:37:31 <boris-42> redixin: so you can operate with stack instance on upper level
14:38:58 <boris-42> redixin: https://github.com/openstack/rally/blob/master/rally/plugins/openstack/scenarios/heat/utils.py#L155
14:39:04 <boris-42> redixin: really not sure why you need wrapper
14:40:41 <boris-42> redixin: are you around?
14:41:17 <redixin> yes
14:41:33 <boris-42> redixin: so ?
14:41:51 <boris-42> redixin: if you need some time to think let's dicuss this after the meeting in rally chat
14:42:22 <redixin> I made this helper method to simplify things
14:42:34 <redixin> thats why people are making helpers
14:44:05 <boris-42> redixin: we will move this to rally chat
14:44:13 <boris-42> ikhudoshyn: okay what you have to discuss
14:44:19 <ikhudoshyn> at last..
14:44:24 <ikhudoshyn> i got some
14:44:51 <ikhudoshyn> 1. our rally-ci has issues, it just does not start at least for my patch
14:45:04 <ikhudoshyn> i'd like to ask redixin to help
14:45:23 <boris-42> ikhudoshyn: you mean it doesn't test propsed patches?
14:45:30 <ikhudoshyn> redixin, could u pls either look yourself or tell me wehrer to look at
14:45:44 <ikhudoshyn> boris-42, yes
14:45:50 <boris-42> ikhudoshyn: is it related to all jobs in rally-ci? or just newly added?
14:46:21 <ikhudoshyn> i only tried my own patch -- I submitted new patchset and it didn't start
14:46:36 <ikhudoshyn> i asked 'rally-ci recheck' ant it wont start anyway
14:47:24 <ikhudoshyn> redixin ?
14:47:41 <redixin> ikhudoshyn: it should work now
14:47:50 <ikhudoshyn> redixin, tnx
14:48:00 <ikhudoshyn> so the next one...
14:48:10 <ikhudoshyn> it's db migration patch
14:48:25 <boris-42> ikhudoshyn: ok
14:48:32 <ikhudoshyn> boris-42, u suggested to add 'rally-manage db list_revisions'
14:48:39 <boris-42> ikhudoshyn: yep
14:48:57 <ikhudoshyn> i think 'rally-manage db history' would be better
14:49:12 <ikhudoshyn> so we keep single-word commands
14:49:21 <ikhudoshyn> and it is aligned with alembic
14:49:45 <ikhudoshyn> And.. I'd like to add it in separate patch
14:50:00 <ikhudoshyn> So that I could finish README now and submit it
14:50:31 <ikhudoshyn> boris-42, ?
14:51:31 <boris-42> ikhudoshyn: okay
14:51:35 <boris-42> ikhudoshyn: I am ok with history
14:51:48 <ikhudoshyn> good
14:51:57 <boris-42> ikhudoshyn: btw we can add a special variable in migrations
14:51:58 <boris-42> ikhudoshyn: tag
14:52:04 <boris-42> ikhudoshyn: that will point to current tag
14:52:27 <ikhudoshyn> boris-42, dont follow u
14:52:46 <boris-42> ikhudoshyn: you should display table
14:52:58 <boris-42> ikhudoshyn: revision_uuid | tag_when_it_was_added
14:53:10 <ikhudoshyn> u mean git tag?
14:53:20 <boris-42> ikhudoshyn: yes current release tag
14:54:17 <boris-42> ikhudoshyn: othewise it hisotry of revisions will be useless for end users
14:54:23 <ikhudoshyn> hm let discuss it separately. Cos currently history is built by alembic from existing migration script files
14:54:35 <boris-42> ikhudoshyn: we can just do this later
14:54:43 <ikhudoshyn> ok
14:54:43 <boris-42> ikhudoshyn: not in first patch
14:54:56 <boris-42> okay antyhing else to discuss?
14:55:00 <boris-42> seems like we are running out of time
14:55:01 <ikhudoshyn> what about history itself?
14:55:11 <ikhudoshyn> now or in separate patch?
14:55:16 <boris-42> ikhudoshyn: separated patch
14:55:26 <ikhudoshyn> ok and the last from me
14:55:38 <ikhudoshyn> this boot_runcommand_delete test
14:55:48 <boris-42> ikhudoshyn: so what is with it?
14:55:48 <ikhudoshyn> there seems to be a race with LOG
14:56:00 <boris-42> ikhudoshyn: could you fix it?
14:56:12 <ikhudoshyn> I guess I have too
14:56:20 <ikhudoshyn> cos it really pissed me off
14:56:39 <ikhudoshyn> it's just.. if anybody have an idea -- ur welcome
14:56:54 <ikhudoshyn> EOM
14:56:56 <boris-42> ikhudoshyn: ok I don't face it
14:57:02 <boris-42> okay anybody else?
14:57:42 <boris-42> okay we have to end meeting see you next week
14:57:45 <boris-42> #endmeeting