16:01:24 #startmeeting #openstack-salt 16:01:25 Meeting started Tue Feb 2 16:01:24 2016 UTC and is due to finish in 60 minutes. The chair is jasondotstar. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:01:27 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:01:30 The meeting name has been set to '_openstack_salt' 16:01:37 #topic Roll Call 16:01:39 o/ 16:01:49 o/ 16:02:13 hi cznewt :-) 16:02:15 GM 16:02:42 another 60secs 16:02:45 hello salters 16:02:51 to see who else joins us :-) 16:03:00 salters unite! 16:03:34 ok.... 16:03:44 #topic Introduction 16:03:59 this is the second irc meeting for the openstack-salt team 16:04:10 If you're interested in contributing to the discussion, please join #openstack-salt 16:04:24 Meetings are Weekly on Tuesdays at 1600UTC <-now 16:04:41 #link http://eavesdrop.openstack.org/#OpenStack_Salt_Team_Meeting 16:04:50 #link https://wiki.openstack.org/wiki/Meetings/openstack-salt 16:05:11 #topic Review past action items 16:05:32 first order of business: 16:05:34 cznewt to finish out the chapter on local dev environment (i.e vagrant) 16:05:57 yes, the documentation is now up-to-date, how to provision vagrant 16:05:58 so I think this is completed 16:06:10 thanks cznewt - I'll try this out 16:06:24 single node deployment, meaning 1 config, 1 control 1 compute node 16:06:34 do you have that link handy to post here? 16:07:12 * jasondotstar looks 16:07:32 #link http://openstack-salt.tcpcloud.eu/develop/quickstart-vagrant.html 16:07:42 any other comments on this? 16:07:42 Until the official documentation is built use the preceding link 16:07:54 thanks, jasondotstar 16:07:57 +1 16:08:18 this is ok, well discuss heat in next topic 16:08:25 cool.. 16:08:34 next: cznewt to report on the completion of the core set of salt formulas 16:09:15 the core formulas have been reviewed and all required parts are there 16:09:17 any news here? 16:09:28 we've added tests 16:09:35 nice 16:09:50 for the code style and metadata validity 16:10:03 +1 16:10:08 we've cought several errors by this mechanism already 16:10:20 one question about the last topic 16:10:24 i just thought about 16:10:30 for the AIO deployment 16:10:50 would it be beneficial to have a aio script to run all the tasks required? 16:11:00 the puppet guys do something similar 16:11:12 ah... 16:11:18 the question is the other needed formulas are to be included in the openstack as well 16:11:25 i see you have something there already for booting up the salt formulas... 16:11:27 You mean something like Salt orchestrate? Or Fabric in eg. opencontrail case? 16:11:32 salt master i mean 16:11:54 yes there's a script to bootstrap salt-master from packages or git 16:12:22 and for AIO deploy it's just matter of highstating the nodes 16:12:29 gotcha. 16:12:41 jasondotstar: basically this - https://github.com/tcpcloud/heat-templates/blob/master/templates/salt_single_public.hot#L120 16:13:09 yes 16:13:15 i haven't looked at the details of this script but 16:13:27 I've cleaned that up a little but basically that's it 16:13:35 there's a modular way that the puppet guys do it 16:13:42 you can add remove modules at will. 16:14:01 im wondering if that's implemented here for the salt states... 16:14:11 forgive me if it's there, and i just haven't seen it yet. 16:14:22 you mean librarian? 16:14:34 well 16:14:39 dependencies among modules? 16:14:44 versions and so on 16:14:50 yes 16:15:04 they've got the Puppetfile 16:15:10 with the list of modules in it 16:15:14 we had to do this for the testing purpose 16:15:21 tagged with version numbers and such. 16:15:23 ok 16:15:27 jasondotstar: there are nothing like puppetfiles in salt, but we have done this for testing dependencies - https://github.com/openstack/salt-formula-heat/blob/master/metadata.yml 16:15:34 ok 16:15:35 gotcha. 16:15:39 define the dependencies, but it is not official 16:15:55 jasondotstar: anyway we prefer deployment using debian packages which can handle these things 16:15:56 i see. 16:16:06 makes sense 16:16:07 ok 16:16:17 but we could propose something like that to the salt community 16:16:37 if there's demand 16:17:06 #agreed Perhaps proposing a dependency tool to the salt community is a good idea based on demand 16:17:11 also that's another topic to adress: how to contribute to salt community as well 16:17:22 cznewt: go for it 16:17:42 #topic Contributing to the Salt communit 16:17:46 #topic Contributing to the Salt community 16:18:01 i've yet to do that.... thoughts? 16:18:25 the testing mechanisms for salt formulas 16:18:37 the formula dependencies 16:19:13 these are the things that might be valuable to salt community as well 16:19:31 we should discuss with salt community some standarized way to do this and possibly help with implementation 16:19:44 maybe the same for debian packaging of formulas which we have already done 16:19:53 +1 16:20:28 the salt-linters out in the wild are outdated and need to be cleaned. 16:21:14 so upstream from us might be work in the salt project itself 16:21:45 and where it makes sense we should push those items up there. 16:21:46 yes i think so 16:21:49 #agreed work with salt community - agree on standard for versioning and satisfying formula dependencies (aka Puppetfile) and for packaging 16:22:10 very well. 16:22:17 any other thoughts on this? 16:22:48 i think this is aggreed 16:22:49 if not, we'll move on. 16:23:02 cznewt: sounds good. 16:23:18 #topic Today's Agenda 16:23:46 ok, first: setup core/release teams: 16:24:03 cznewt, genuing, pupapaik, Jay_Clark, majklkcz anyone else? 16:24:12 genunix 16:24:16 +1 16:24:40 +1 16:24:47 now nothing block the team creation? 16:25:00 that was my next question 16:25:10 we need to go back to -infra and make sure you are added, cznewt 16:25:14 if we haven't already 16:25:47 yes, I'll after the meeting 16:26:02 good deal. 16:26:19 next: development guideline 16:26:45 proposal - repositories imported under github.com/openstack 16:26:57 the repositories are under the openstack nameespace now 16:27:28 cznewt: ok. how do we keep the github project(s) and the openstack project(s) in sync? 16:27:47 and in the openstack git, is the infromation here correct? 16:27:51 #link http://openstack-salt.tcpcloud.eu/develop/extending-contribute.html 16:28:17 github - the tcpcloud namespace - the repos will be removed 16:28:24 jasondotstar: our repos in tcpcloud namespace will be removed so there's no need for sync 16:28:41 +1 16:28:52 the question is about building packages 16:28:56 as they are being added to the openstack 16:29:07 it's currently done in our Jenkins and uploaded to our PPA 16:29:18 but I would like to do this in openstack way 16:29:19 that has to be moved over 16:29:20 right 16:29:32 so these packages may reach ubuntu cloud and debian once 16:29:49 but I don't know how it works for openstack packages so this is probably action item to me 16:30:19 also maybe do some cleanup on packaging, specify dependencies for some formulas, etc. 16:30:39 #action genunix to research how to openstack packaging works 16:31:54 The integration setup of the launchpad needs to be checkedd 16:32:45 agreed 16:32:47 I think the openstack access is not well configured there 16:33:00 I can take a look at that one 16:33:24 #action jasondotstar to look at the integration of launchpad with our repos 16:33:52 jasondotstar: can we go over the complete process of contributing together? I'd like to evangelise further, but i want to certain that i do it right 16:34:02 of course 16:34:26 the contributor guidelines 16:34:59 let's take a quick at that 16:35:05 see if there is any feedback 16:35:34 #link http://openstack-salt.tcpcloud.eu/develop/extending-contribute.html 16:35:38 (again) 16:36:56 yes, I'll do it after the meeting to see how it works from both sides 16:37:31 I'd like to 16:37:51 +1 16:38:08 get the heat stack for development of salt stack tested 16:38:14 and ready 16:38:28 +1 16:38:44 we provide labs for openstack-salt creating to the customers so we can do it to the community as well 16:39:18 jasondotstar: there'a account for you ready already and it will be open for openstack-salt team members 16:39:28 nice 16:40:12 yes, the full scale HA deployment lab takes about 20 cores and 60gb of ram 16:40:26 it's out of scope of vagrant testing :) 16:40:31 hehehe 16:40:34 i think so 16:40:43 unless you've got an entire LAB 16:40:50 however, that might be a point to discuss 16:41:03 perhaps we can get talk to -infra about a dev environment 16:41:10 or have someone sponsor some resources 16:41:24 so that we can fully test HA 16:41:42 yes, this is for further discussion 16:42:12 it would be nice to get more sponsors 16:42:14 +1 16:42:21 #agreed Further research is needed on how we acquire resources for HA testing 16:42:51 I can ask around 16:43:15 my company *might* have something we can leverage, but don't quote me on that :-) 16:43:47 I've known rackspace to sponsor some lab space IIRC 16:43:48 Super, the openstack environments are preferrable 16:43:57 +1 16:44:25 #action jasondotstar to look into dev environment sponsorship 16:45:00 other items on collaborative efforts.....? 16:45:01 #action newt test the aio/ha openstack-salt heat setup 16:45:27 just the salt community involvement 16:45:33 +1 16:45:48 it has been already put into action, I think 16:45:53 good deal 16:46:15 we touched on this, but are we good on building formula packages under openstack? 16:46:31 RHEL needs love 16:46:34 yes? 16:47:11 we are good for debian, and the packages are not very complicated, just copying around 16:47:55 so i think we can provide rhel packages as well in short time 16:48:02 cool. 16:48:47 oh.... the contributor license 16:49:15 Apache License, Version 2.0 16:49:31 is that the license we're using? 16:50:15 are there comments about that? 16:50:37 the formulas use the apache2 all of them 16:50:55 and what about the contributor's license, which one applies 16:51:47 #link https://review.openstack.org/static/cla.html 16:52:23 yes, thank you 16:52:31 i think mainly this means that the code you contribute is originated by you 16:52:40 once you offer it 16:52:48 it goes under the apache license at that point 16:52:51 IIUC 16:53:26 i can't take closed-source code from IBM, and offer it as a patch 16:53:28 for example. 16:53:48 yes i get it even better now :) 16:53:59 so both kinda come into play 16:55:18 "I offer MY (and I mean MY code - as stated in the Contributor license) code to this project, which is then protected by the Apache License" 16:55:42 something to that affect 16:56:02 ok we've got 5 mins left 16:56:06 #topic Open Discussion 16:57:00 no additional comments from me. we've got a few things to look int. 16:57:04 well i'm out of topics, I'll be here to help you with the install 16:57:05 s/int./into. 16:57:19 which ones? 16:57:30 just the action items 16:57:50 im saying nothing else besides what we've already talked about :-) 16:58:03 oh... summit 16:58:09 you guys going? 16:58:38 yes, most probably 16:58:46 yep, I'll be there 16:58:57 we can talk more about that later 16:59:09 yes, looking forward to summit :) 16:59:22 cool. if nothing else, i think we can close 16:59:23 let's continue on #openstack-salt 16:59:27 thx guys 16:59:38 thx all 16:59:39 until next week 17:00:00 #endmeeting