20:00:05 #startmeeting Windows Working Group 20:00:05 Meeting started Tue Aug 1 20:00:05 2017 UTC. The chair is jborean93. Information about MeetBot at http://wiki.debian.org/MeetBot. 20:00:05 Useful Commands: #action #agreed #halp #info #idea #link #topic. 20:00:05 The meeting name has been set to 'windows_working_group' 20:00:20 Howdy all 20:01:32 yo 20:01:38 #chair nitzmahone 20:01:38 Current chairs: jborean93 nitzmahone 20:02:09 heya 20:02:19 #chair jhawkesworth_ 20:02:19 Current chairs: jborean93 jhawkesworth_ nitzmahone 20:03:11 #topic https://github.com/ansible/community/issues/195#issuecomment-314705308 new modules 20:03:25 anything anyone wants to add regarding the new modules for 2.4 20:03:37 there is a pretty substantial list which is great to see 20:04:10 Yep- should have everything that's ready merged for 2.4, but probably won't be much movement there until after core freeze (at least by me) 20:04:18 I've looked at the win_xml very briefly and ok with it. I don't have really anything substantial to test it with but feel like we should just merge and se where it takes up 20:04:25 WFM 20:04:25 cc karstensrage 20:04:49 win_toast basically needs to run async. seems weird though so not really sure what I'm going to do with that 20:04:53 community + preview, so bases/asses more or less covered ;) 20:05:33 jhawkesworth_: yeah, seems like it needs a shared host process or something to keep the notifications alive... Maybe just nuke the PR for now? 20:05:38 would be nice to see how win_xml gets on with tests from xml module 20:05:56 are there currently tests for the xml module? 20:06:29 If I recall there are loads of tests for xml module - lots of xml files about beer IIRC 20:06:38 win_xml has some basic smoke tests 20:06:51 (to be clear I mean the python xml module has lots of tests) 20:07:03 yep, I've seen only a few for win_xml in the PR 20:07:13 Will see if we can use the same tests for win_xml 20:07:33 #action jborean93 to follow up karstensrage regarding win_xml tests 20:07:44 cool thanks 20:07:50 anything else for new modules 20:08:11 nawp 20:08:13 I've got a few but they aren't critical and an wait until people have more time to test 20:08:14 not from me 20:08:21 #topic https://github.com/ansible/community/issues/195#issuecomment-318914470 WSL support 20:08:31 I know dag isn't online but he added it as a topic 20:08:47 yeah I don't think much changes really. 20:08:55 I personally don't think we should support it as MS don't support it for non development purposes 20:09:02 I'm still not going to set forks to 10000 and run my datacenter off it 20:09:21 yup, its still a tool for developers 20:09:22 I'm also under the assumption Server 2016 doesn't even run it yet, just the preview 20:09:25 I'm not going to say it's *impossible* to test in CI, but it's probably not worth the trouble. Any issues we have there are likely to be scalability issues, which we won't be hitting during CI anyway 20:09:32 Nope, no servers will run it 20:09:55 if its not going on servers its not suitable for production workloads. ergo its still a dev tool 20:10:00 + . the only issue I've come across is windows line endings when trying to source ansible pulled down using windows git 20:10:05 I did some basic testing with like 20 forks over ssh with no problems at all (even on my junky old Core 2 Duo from 2007) 20:10:31 cool sounds like we have a consensus 20:10:38 I have used like 30 forks but don't have much inventory in test datacenters 20:10:42 I think there might be some git flags we could use to force things to *nix line endings, but my gitfu is weak in that area (never used it on Windows much) 20:10:54 there si for Windows and I ended up doing that 20:11:06 just wans't fun when the inventory just stopped working for no reason 20:11:11 heh 20:11:14 plus vault fails with \r\n 20:11:14 I never used git for windows either 20:11:24 arguably we'll need to figure that out anyway, esp if we ever end up doing native Windows controller 20:11:34 * jborean93 fumes with Windows line endings 20:11:37 I think I saw a PR re \r\n and vault earlier 20:11:46 alkins I think 20:11:49 I'm debating putting it on the 2.5 roadmap as an exploratory/experimental item 20:12:05 Yeah,I was just reviewing that PR today 20:12:29 At least they're not the bastard old Mac \r 20:12:40 native win controller would be awesome - help me with adoption by devs if nothing else 20:12:45 Exactly 20:13:08 #topic https://github.com/ansible/community/issues/195#issuecomment-319161331 member/membership 20:13:20 It'd depend on us bulletproofing threaded workers, and I honestly don't know how well it'd perform at scale on Windows without PyPy 20:13:38 plus there is the Kerberos ticket management issue 20:14:02 Yeah, though I have a couple ideas there 20:14:20 WRT renaming win_group_member- anyone else have an opinion? 20:14:26 for your member/membership I'm pretty ambivalent 20:14:36 if we already have a standard would be good to just align with it 20:14:40 both make sense to me 20:14:42 I like consistent names 20:15:04 I also don't mind typing whole words so happy with member -> membership 20:15:16 those 4 extra chars though :) 20:15:24 might get RSI 20:15:26 :-) 20:15:29 :P 20:15:41 cool, happy to make that change nitzmahone? 20:15:57 good names are worth the typing 20:16:00 OK- it's definitely an OCD thing, but I like having the module name map to the resource it's managing 20:16:26 jborean93: were you volunteering, or asking me if I'm happy to make it (yes) 20:16:31 #action nitzmahone to rename win_group_member to win_group_membership 20:16:35 done :) 20:16:38 #topic open floor 20:16:38 +1 20:16:45 Anything else to talk about 20:16:56 nothing from me oday 20:16:59 *today 20:17:15 I will be a bit offline till middle of the month (family holiday) 20:17:23 Nice- have fun! 20:17:29 only thing to add is I'm making changes to win_copy so hopefully people haven't been working on it 20:17:36 Going anywhere cool? 20:17:37 cool hope it is enjoyable for you 20:18:07 thanks. Northern France. Hopefully slightly warmer than soggy UK! 20:18:46 might get to see the sun in summer at least 20:18:56 Sweet... 20:18:59 saw your note about win_copy so wil leave that alone for now 20:19:20 Perfect timing, my lunch just arrived... See ya'll later! 20:19:40 cool, fast meeting is a good meeting 20:19:46 I'll let you get on. Have a good one 20:20:02 #endmeeting