00:01:04 <nitzmahone_> #startmeeting Ansible Azure Working Group 00:01:04 <zodbot> Meeting started Thu May 17 00:01:04 2018 UTC. 00:01:04 <zodbot> This meeting is logged and archived in a public location. 00:01:04 <zodbot> The chair is nitzmahone_. Information about MeetBot at http://wiki.debian.org/MeetBot. 00:01:04 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic. 00:01:04 <zodbot> The meeting name has been set to 'ansible_azure_working_group' 00:01:27 <zikalino> hello, anybody here? 00:01:35 <nitzmahone_> hey! 00:01:49 <zikalino> Hi Matt! 00:01:50 <nitzmahone_> #chair zikalino 00:01:50 <zodbot> Current chairs: nitzmahone_ zikalino 00:01:57 <nitzmahone_> #chair jborean93 00:01:57 <zodbot> Current chairs: jborean93 nitzmahone_ zikalino 00:02:05 <zikalino> just saw you merged AKS, thanks! :-) 00:02:14 <nitzmahone_> One down, many to go :D 00:02:19 <jborean93> hi 00:02:25 <zikalino> yeah 00:02:51 <zikalino> i think today just me here, entire team went for a trip 00:03:01 <nitzmahone_> Must be nice ;) 00:03:16 <nitzmahone_> I just got back from Red Hat Summit- I feel like I need a few days off to recover 00:03:21 <zikalino> hehe, working from home 00:03:55 <nitzmahone_> OK, so I've been looking at the app gateway and webapp modules 00:04:30 <zikalino> yes, pls add any comments and i will just update anything needed today. 00:04:35 <nitzmahone_> I haven't had to use those resources yet, so it takes me awhile to look at them and see what makes sense. I think app gateway is pretty close. 00:04:47 <zikalino> these are pretty complex 00:05:19 <zikalino> well, app gateway is just an aggregation of many different things. so it's large but not super complex logically 00:05:53 <nitzmahone_> So I see a lot of code in app gateway dealing with munging args- I think we have snake_to_camel and camel_to_snake helpers that you could use for that instead of hardcoding all those parameter conversions. 00:05:59 <nitzmahone_> That's code no human should have to write 00:06:17 <nitzmahone_> (I know we have those helpers, but I'm trying to remember if they're in a module_util somewhere or just inside a module) 00:06:33 <zikalino> yes, i will update that, just some of the params are not convertible by snake_to_camel 00:06:40 <zikalino> but i will update all that is possible 00:06:43 <jborean93> they were recently moved out of the ec2 module_util 00:07:01 <zikalino> ok, i will use these 00:07:32 <nitzmahone_> @jborean93 where did they go? 00:07:46 * nitzmahone_ is currently on a 2.5 branch for obvious reasons ;) 00:07:49 * jborean93 looks 00:08:07 <jborean93> https://github.com/ansible/ansible/blob/devel/lib/ansible/module_utils/common/dict_transformations.py 00:10:44 <zikalino> ok, so pls add any comments, i will update them asap. when is the freeze time exactly for the modules? 00:11:15 <nitzmahone_> The UI on the web app one needs some love- the huge wad of effectively mutually exclusive top-level args and the `linux_fx_version` thing need to be rethought IMO 00:11:25 <nitzmahone_> Next Wednesday 00:11:35 <nitzmahone_> Yeah, I'll fill in some more comments 00:11:37 <zikalino> ah, so moved? 00:11:52 <zikalino> i thought it was on 17th 00:12:04 <yuwei> me too 00:12:28 <jborean93> core modules are this week 00:12:31 <nitzmahone_> Today's freeze is just for core stuff, everything else is May 25 (actually next Friday): https://docs.ansible.com/ansible/devel/roadmap/ROADMAP_2_6.html 00:13:17 <nitzmahone_> s/Today's/Friday's/ 00:13:28 <jborean93> timezones definitely make things confusing... 00:13:38 <nitzmahone_> Heh, I have no excuse 00:14:09 <jborean93> usually I just lop on an extra day, you west coaster are pretty much 1 day behind for me 00:14:15 <nitzmahone_> Oh also, @zikalino you *do* know you can run sanity tests locally very fast, right? I see all those repeated pushes for sanity things 00:14:31 <nitzmahone_> That's gotta be maddening 00:14:42 <nitzmahone_> (well, in fact all tests) 00:15:00 <zikalino> i work on a few things at once, so sometimes, i just don't run locally, but i will try to change that.... 00:15:56 <zikalino> jborean93 how about this pr? https://github.com/ansible/ansible/pull/39722 00:16:03 <nitzmahone_> You're not overlapping our "busy" time for CI anyway, but if you were, you'd be waiting in line for 30-240 minutes for each one of those runs, easily 00:16:29 <nitzmahone_> (but you know how to use `ansible-test sanity`, right?) 00:16:35 <zikalino> yes 00:16:37 <jborean93> zikalino: I'm just trying to catch up on some old Windows PRs, hoping to have that done by midday and will get back onto Azure 00:17:24 <zikalino> ok 00:17:28 <zikalino> nitzmahone_ this is the change we discussed in redmond about facts modules 00:18:07 <zikalino> ok, so i have one more 00:18:15 <nitzmahone_> @zikalino Matt Clay also wanted me to ask if you've got any update on the test/module stability fixes- we're still running with a lot of module tests disabled 00:18:25 <zikalino> azure_rm_resource / azure_rm_resource_facts 00:18:26 <zikalino> https://github.com/ansible/ansible/pull/39274 00:18:35 <zikalino> yes, i know :-( 00:18:56 <nitzmahone_> Yeah, that output is much nicer than the raw SDK output (and theoretically usable for round-trippy kinds of things, though I'm still not convinced how workable that is generally in real-world scenarios) 00:18:57 <zikalino> we have been too busy recently. but i think this week and next week i want to focus on these issues 00:19:33 <zikalino> well, i would like to have this kind of output for all the modules eventually 00:19:43 <nitzmahone_> agreed 00:20:47 <nitzmahone_> I need to do a full review on 39274 still, but the big outstanding question there is around idempotence. 00:21:17 <nitzmahone_> I guess the typical usage pattern would be to use resource_facts first to determine if it exists, then run the resource module if not... 00:21:50 <zikalino> yes, and then idempotence works well for most of the resources. 00:21:52 <nitzmahone_> It seems like it wouldn't be able to support updates generally, but at least allow playbook re-runnabliilty 00:22:10 <zikalino> actually i still want to update idempotence a bit and make it better 00:22:30 <zikalino> (to support more general updates) 00:22:36 <nitzmahone_> So yeah, I'll try to get a full review in on that one, but at a glance it looks fine for 2.6 00:22:56 <zikalino> yes, i just want to have simple thing right now.... 00:23:15 <nitzmahone_> yep, better than nothing for those things that aren't covered by other modules right now 00:23:29 <nitzmahone_> (and better than shell: az ... IMO) 00:23:37 <zikalino> yes :-) 00:23:53 <zikalino> ok, so i think from my side that's all, i will be waiting for comments. 00:24:07 <zikalino> yuwei i think you have a few things too? 00:28:47 <nitzmahone_> Welp... 00:29:06 <yuwei> yes 00:29:16 <yuwei> still 2 prs 00:29:21 <yuwei> https://github.com/ansible/ansible/pull/38643 00:29:36 <yuwei> https://github.com/ansible/ansible/pull/37627 00:30:06 <yuwei> We have discussed these two before 00:30:59 <yuwei> Refer resources from other resource group. I have modified the both two pr to support it 00:31:19 <nitzmahone_> Oh nice 00:33:01 <nitzmahone_> I'll review those and get them merged for 2.6 00:33:14 <nitzmahone_> Looks good at a glance though 00:33:46 <nitzmahone_> @yuwei you good if I just add a commit to tweak up docs text? 00:36:01 <nitzmahone_> OK, well if nothing else, we'll wrap it up for today 00:36:09 <zikalino> yes, i think it will be fine. also, i have one pr from Catherine: https://github.com/ansible/ansible/pull/37909 00:38:00 <zikalino> and i think that will be all :-) 00:38:52 <nitzmahone_> Cool- I'll re-review that one with the changes. Thanks all! 00:39:01 <zikalino> thank you! 00:39:14 <zikalino> so have a nice evening and afternoon :-) 00:39:26 <nitzmahone_> #endmeeting