16:00:12 #startmeeting Ansible Network Working Group 16:00:12 Meeting started Wed Jun 19 16:00:12 2019 UTC. 16:00:12 This meeting is logged and archived in a public location. 16:00:12 The chair is Qalthos. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:12 Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:00:12 The meeting name has been set to 'ansible_network_working_group' 16:00:48 #chair ganeshrn privateip trishnag dmellado pabelanger 16:00:48 Current chairs: Qalthos dmellado ganeshrn pabelanger privateip trishnag 16:00:57 #topic Core Updates 16:01:32 #info 2.9 freeze dates are up 16:01:35 #link https://docs.ansible.com/ansible/latest/roadmap/ROADMAP_2_9.html 16:02:48 #info The Network project board should be getting cleaned up for 2.9 as well, soon 16:02:58 #link https://github.com/ansible/ansible/projects/10 16:04:03 #link https://github.com/ansible/community/labels/network is where this meeting's agenda lives 16:04:18 Let's get started! 16:04:27 #topic Meraki PRs 16:04:50 mrproper: I made some specific comments about your DOCUMENTATION strings 16:05:07 Yes, I saw those. I will integrate tonight and @ you in the PR when I have resolved them. 16:05:30 There is one other thing I would like to ask about regarding your tests 16:05:35 Of course. 16:06:26 I'm not familiar with /exactly/ how your connections are made, but I don't think you should have to specify `delegate_to: localhost` for every task 16:06:57 I know I could specify the delegate_to as part of a block, correct? 16:07:40 I don't think you should need it at all. What connection are you using? 16:08:18 It's an API based connection. So it does connect from local to the Meraki API (n110.meraki.com as an example). I'd have to look at the exact URL, but it shouldn't matter. 16:08:30 Or are you not specifying a connection type? 16:09:00 Am I able to globally specify a connection type for an integration test and have it apply to all tasks in there? That would help. 16:09:47 I expect this behavior will change when I can implement httpapi, but that's a topic for another day. 16:11:25 does meraki give a VM to download, to test locally? 16:11:36 The way the core tests work is to not have the tests themselves in main.yaml, and have main.yaml import the those tasks with a connection applied 16:11:44 pabelanger: No. It's fully API and web driven management. Picture is like interfacing with AWS in a way. 16:11:56 let me get you some examples 16:13:04 https://github.com/ansible/ansible/blob/devel/test/integration/targets/eos_config/tasks/cli.yaml#L13 further abstracts this to files in a tests folder 16:13:46 But you could just as easily do something like 16:14:40 - include: "local.yaml ansible_connection=local" 16:15:00 Then local.yaml would have the tests for the respective module? 16:15:07 Exactly 16:15:37 That's a good idea. I will update the two new modules you commented on today to take that approach and see how well it works (I expect it will work well). 16:15:57 mrproper: Thanks! 16:16:25 Qalthos: Before we move on, if I have some design questions and "brainstorming" I'd like to do on a future enhancement, is this an appropriate forum for it? 16:17:07 mrproper: We can get in to that 16:17:39 I'll add it to the agenda for next week. I need to do a little more research to make sure I articulate the problem properly. 16:18:23 mrproper: Sounds good. 16:18:27 Thank you. 16:18:33 Anything else you want to cover today? 16:18:58 Not this week. I'm probably going to have a full week next week though. But thanks for your help today. 16:21:43 #topic Open Floor 16:23:14 I mostly wanted to point out that we finally have our network appliances, that on zuul.ansible.com, passing tests on ansible/ansible devel. 16:23:25 so, thanks to all how helps with that effort 16:23:44 and moving forward, should make PRs much easier to review, by provide test results 16:24:34 pabelanger: For my Meraki modules, it's all API driven so I can't provide a VM. But is it possible I can provide an API key we keep secret and it can possibly automate some of them? 16:25:18 mrproper: possible, but there is some concerns with that (like leaking data). But it would be interesting to learn more about it 16:25:35 pabelanger: Lets talk about it on the side. I'd like to investigate if it's an option. 16:25:42 +1 16:34:12 Anyone else have anything to bring up here? 16:35:15 Cloud you please review my PRs? 16:35:21 xuxiaowei0512: Unfortunately I have to once again refer you to ikhan jtanner or gregdek for more information. 16:37:28 ok 16:43:35 Alright then, thanks for coming, everyone 16:43:42 #endmeeting