20:00:00 <jborean93> #startmeeting Ansible Windows Working Group
20:00:00 <zodbot> Meeting started Tue Aug  1 20:00:00 2023 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 https://fedoraproject.org/wiki/Zodbot#Meeting_Functions.
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'
20:00:00 <nitzmahone> I keep toying with the idea of making `ansible-test shell` work with that :D
20:00:06 * nitzmahone lurks
20:00:08 <jborean93> #chair nitzmahone
20:00:08 <zodbot> Current chairs: jborean93 nitzmahone
20:00:29 <jborean93> #topic Agenda https://github.com/ansible/community/issues/682
20:00:35 <jborean93> Nothing on the agenda
20:00:40 <jborean93> #topic Open Floor
20:01:35 <nitzmahone> 🦗
20:02:35 <jborean93> I'm not got anything too special
20:02:54 <jborean93> Still focusing on win_updates and trying to stabilise the various connection/winrm problems that are causing things liek deadlocks
20:03:40 <nitzmahone> Did the targeted retry thing work out, or still TBD?
20:03:59 <nitzmahone> (or find anything out from a checked/debug Windows build?)
20:04:18 <jborean93> Unfortunately I haven't found anything that could cause it, or at least not something we probably can help
20:04:56 <jborean93> I'm going to just implement a retry on a read timeout mechanism and have a way to acknowledge the final complete message to ensure that isn't lost
20:05:05 <nitzmahone> nice
20:05:21 <nitzmahone> Was this the one that was triggered by clock shifts, or something else?
20:05:43 <jborean93> Something else I have no idea, probably a specific update changing something critical
20:05:53 <jborean93> The clock shift is easily fixable on our side
20:06:19 <nitzmahone> Always fun when the mechanics are changing out the hydraulics while the plane's in flight ;)
20:06:26 <jborean93> pretty much
20:06:49 <jborean93> I do need to revisit `microsoft.ad` and the way it treats no explicit path being specified
20:07:19 <jborean93> I need to change the behaviour so that if `path` isn't found but an object was found by `identity` it won't try and move that to the default path. It should only move if an explicit path was provided
20:07:40 <nitzmahone> Yeah, that seems like the right thing
20:08:23 <nitzmahone> is that operation always including an object type filter?
20:08:44 <nitzmahone> (or is it up to the user?)
20:09:12 <jborean93> It uses whatever the AD equivalent models do so I assume it does
20:09:30 <jborean93> there is a generic "object" module though
20:09:36 <nitzmahone> (eg, it wouldn't blindly move a matching object from a user container to a machine container)
20:10:08 <nitzmahone> I can't remember if the lowest-level APIs will let you do that anyway
20:10:34 <jborean93> is there a user/machine container, I thought it was all just OUs and it didn't care
20:10:40 <nitzmahone> I seem to recall trashing DCs doing silly things in the early Win2000 days, but maybe they've tightened things up since then :D
20:11:55 <nitzmahone> IIRC there are some containers that have "expected" object types in them (certainly for things like Exchange, but even for the base schema stuff too in some cases)
20:12:41 <nitzmahone> The details escape me, I just remember irretreivably trashing a couple of test lab DCs many many years ago doing things that the API happily let me
20:15:12 <jborean93> If there is one then it'll be up to the cmdlet to stop it
20:16:40 <jborean93> Anywho, unless there's anything else I'llend the meeting
20:17:49 <jborean93> Thanks for coming
20:17:51 <jborean93> #endmeeting