ansible_windows_working_group
LOGS
20:00:00 <jborean93> #startmeeting Ansible Windows Working Group
20:00:00 <zodbot> Meeting started Tue Aug 25 20:00:00 2020 UTC.
20:00:00 <zodbot> This meeting is logged and archived in a public location.
20:00:00 <zodbot> The chair is jborean93. Information about MeetBot at http://wiki.debian.org/MeetBot.
20:00:00 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
20:00:00 <zodbot> The meeting name has been set to 'ansible_windows_working_group'
20:00:07 <jborean93> boo-yah
20:00:09 <nitzmahone> NOICE
20:00:20 <jborean93> #chair nitzmahone
20:00:20 <zodbot> Current chairs: jborean93 nitzmahone
20:00:23 <nitzmahone> jborean93 got skilz
20:00:40 * jhawkesworth lurking
20:00:56 <jborean93> howdy
20:01:36 <jhawkesworth> Clash with kids bedtime so only half looking tonight
20:01:41 <jborean93> no worries
20:02:09 <jborean93> I've got nothing special to add myself
20:02:41 <nitzmahone> Ditto. Having interesting convo with an internal Red Hat team about ways to do platform-agnostic modules
20:03:14 <nitzmahone> (basically "ew, do we really have to write PS to work on Windows?")
20:03:37 <jborean93> well we would have to add a lot of import guards to basic.py to get that even loadable on Windows
20:04:04 <jborean93> it would be doable but I would be reluctant to look into that without making sure we start testing that
20:04:50 <nitzmahone> Yeah, if they wanted to do something like pyoxidizer, I was going to suggest they just skip AnsibleModule altogether and treat it like a binary module on Windows
20:05:13 <nitzmahone> So same codebase under the covers, but skip most/all the module_utils magic
20:05:29 <nitzmahone> (or at least skip it on Windows)
20:05:37 <jborean93> with binary modules you run into the problem of making sure it is compiled for that platform
20:06:00 <nitzmahone> Yeah, but they're only doing Linux + Windows IIRC, so should be fine there
20:06:16 <nitzmahone> (just ship the compiled bits for Windows and explicitly load them from their custom action)
20:07:02 <nitzmahone> It'd be "fun" to resurrect the C# module stuff for POSIX, but I don't think it's hot enough to warrant an official project right now
20:07:32 <nitzmahone> (might be better as a side project with ansible_platform and pluggable exec subsystems)
20:07:38 <jborean93> yea as you said in that email, I wouldn't be interested until we have more of the exec subsystem in place
20:08:36 <jborean93> the prototype I had before was really just a "plugin" to the Windows exec scripts, would be nice to have that as an official plugin in Ansible that isn't Windows specific
20:08:52 <nitzmahone> I think someone had crusty pwsh modules (as binary/script via shebangs) working on Linux, so that might be an option as well
20:09:04 <jborean93> it works today, you just have no module utils
20:09:28 <nitzmahone> So that might be an option as well
20:09:31 <jborean93> plus IIRC you needed to set `TERM=basic` to avoid pwsh/.net adding some ANSI sequences to the start of the output
20:09:42 <nitzmahone> ick
20:09:47 <jborean93> no idea if that's still a problem
20:10:03 <nitzmahone> Probably- I definitely remember tripping over that issue
20:10:08 <jborean93> https://github.com/ansible/ansible/issues/48881
20:10:34 <jborean93> I had an example of using pwsh with shell, script, and a module there
20:11:42 <nitzmahone> Well anyway, nothing official from me today- just a little Windows tidbit that I thought was interesting from this morning
20:13:23 <nitzmahone> #topic open floor
20:13:39 <nitzmahone> (if it wasn't already obvious ;) )
20:15:15 * nitzmahone eyes kitchen hungrily
20:15:29 <nitzmahone> closing in 2min if nothing new
20:15:42 <jhawkesworth> Nothing from me, let nitzmahone get fed!
20:15:47 <nitzmahone> :D
20:16:03 <jborean93> I'm good as well
20:16:23 <nitzmahone> OK, til next week- thanks all!
20:16:25 <nitzmahone> #endmeeting