16:00:09 #startmeeting Ansible Network Working Group 16:00:09 Meeting started Wed Oct 16 16:00:09 2019 UTC. 16:00:09 This meeting is logged and archived in a public location. 16:00:09 The chair is Qalthos. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:09 Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:00:09 The meeting name has been set to 'ansible_network_working_group' 16:00:38 #chair ganeshrn ikhan justjais dmellado 16:00:38 Current chairs: Qalthos dmellado ganeshrn ikhan justjais 16:00:49 #topic Core Updates 16:01:52 #info It looks like 2.9 has slipped a few weeks. 16:02:28 #info The rooadmap hasn't updated, so here are the dates from the 2.9.0rc3 announce email 16:03:04 #info 2019-10-10 Release Candidate 3 16:03:15 #info 2019-10-17 Release Candidate 4 (if needed) 16:03:26 #info 2019-10-24 Release Candidate 5 (if needed) 16:03:36 #info 2019-10-31 Final Release 16:04:08 Oh, hah, it's updated in devel, but not latest 16:04:11 #link https://docs.ansible.com/ansible/devel/roadmap/ROADMAP_2_9.html 16:04:53 In any case, that means we're still working on getting 2.9 out the door. 16:05:31 #info As for the networking team, we have published a tentative roadmap for 2.10 16:05:36 #link https://github.com/ansible/community/wiki/Network:-2.10-Roadmap 16:07:13 That's all the things I can think of at the moment 16:07:28 #topic Many EXOS resource modules 16:07:48 ujwalkomarla: I've reviewed most of these, and they look fine 16:07:59 Thanks. 16:08:09 One PR with the failing test? 16:08:16 https://github.com/ansible/ansible/pull/61914 16:08:38 My main comments are nitpicks... merging multiple `assert` calls into one, using `bool()` instead of `True if var else False` 16:08:57 Yeah... I'm not entirely sure wht you need there. 16:09:19 Yes, saw the comments, do you want them fixed for this PR? 16:09:24 It has been a while since all that was put in place, so I'll have to refamiliarize myself 16:09:54 ujwalkomarla: You can fix them before I merge, or just keep it in mind for later, they really are nitpicks 16:09:54 I see 16:10:54 I will see if Jaya, the author, can get to it before that. 16:11:36 Just let me know if you want to touch them up before I merge. I imagine there will be conflicts on the facts base class between each merge, so that would be a good time to make changes if you want 16:12:16 Yes, that is right. 16:12:29 Will do so. 16:13:11 Once 61914 finishes tests, I'll take a look and see if I can't figure out what it needs. If I can, I'll probably just commit the fix directly to the PR, so be aware for that 16:13:51 Anything else we need to talk about? 16:13:52 Regarding the PR with failing test case, the other NOS with single action file, the way they were fixed seemed to be by ignoring the particular file. 16:15:46 I will try to find the ignore file and comment on the PR, if required. 16:16:58 Yeah, the way they work is something of a kludge. My hope is that we can get away from the action plugin entirely, for which exos might be a nice test case as it has no historical local connection baggage 16:17:20 ujwalkomarla: If you get it working before I do, then so much the better 16:19:09 Sure 16:20:02 I added this PR couple of minutes before the meeting - https://github.com/ansible/ansible/pull/63574. I'm mapping all the different EXOS network OSes to their facts module. 16:21:08 Yeah, I just saw that 16:21:52 Great. Wanted to make sure :) 16:22:44 That is all the items I had. 16:22:51 ujwalkomarla: Thanks 16:23:00 #topic Open Floor 16:23:35 Hi, I have a question. 16:23:36 I saw some resource modules using the new architecture. 16:25:31 xuxiaowei0512: What's the question? 16:26:26 I saw that network resource modules are different with older 16:28:15 The code has changed structure. 16:28:43 Provider is unnecessary. 16:30:29 Yes, we are trying to move forward with a new structure that lets us drop some of the historical baggage of the older modules and encourages code reuse 16:31:33 The historical baggage will be discarded ? 16:31:39 We're not mandating that anyone else follow us on this, though it is encouraging seeing that some (like ujwalkomarla with exos) have done so 16:33:03 I see. 16:33:05 For core-supported platforms, the replaced modules have been marked as deprecated, and will be removed after four releases. 16:33:31 For other platforms, it is up to the maintainer if they want to continue to support local and provider 16:33:53 I see. 16:34:12 We're not going to intentionally break `connection: local` on other modules 16:37:54 (I should say not never, but not without a lot of warning) 16:38:55 Anyone else have anything to bring up? 16:38:59 Qalthos I got it,thank you. 16:41:03 Qalthos: I commented on the PR on how to fix the test failure. 16:41:12 https://github.com/ansible/ansible/pull/61914 16:41:34 Rather than fix, a hack. 16:43:26 ujwalkomarla: It's the same hack that is applied to all the other action plugins, so I think it's fine 16:44:16 Ok, I'll go ahead and commit the change 16:44:28 +1. That said, I'd still like to see if there is a way to avoid the action plugin in the first place. 16:45:23 All it is really doing is restricting the connection plugin type... I feel like there should be a better way of handling that 16:45:57 Agree. Any ideas on that? Do you want me to try something? 16:46:35 I'll work on that and accept this in the meantime. If I come up with a solution I think works, I'll open a PR and let you review it 16:47:24 Okay. Sounds good :) 16:47:45 Can you re-run the CI on this https://github.com/ansible/ansible/pull/61914? 16:48:01 Unrelated failures as I see. 16:48:49 If you append the cahnge you need to make to the PR, it will start a new test 16:49:15 My bad, this PR https://github.com/ansible/ansible/pull/63574 16:49:42 ujwalkomarla: kicked 16:49:51 Thanks 16:50:22 Anyone else have anything they want to talk about in the last ten minutes? 16:54:00 Thanks for coming, everyone! 16:54:03 #endmeeting