20:01:04 <jborean93> #startmeeting Ansible Windows Working Group
20:01:04 <zodbot> Meeting started Tue Aug  6 20:01:04 2019 UTC.
20:01:04 <zodbot> This meeting is logged and archived in a public location.
20:01:04 <zodbot> The chair is jborean93. Information about MeetBot at http://wiki.debian.org/MeetBot.
20:01:04 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
20:01:04 <zodbot> The meeting name has been set to 'ansible_windows_working_group'
20:01:11 <jhawkesworth> hey jborean93
20:01:16 <jborean93> sorry about last week I was sick and decided to sleep in a bit
20:01:19 <jborean93> hey
20:01:27 <jhawkesworth> no worries hope you are all well again now
20:01:28 <jborean93> #chair jhawkesworth Shachaf92
20:01:28 <zodbot> Current chairs: Shachaf92 jborean93 jhawkesworth
20:02:06 * jhawkesworth checks agenda
20:02:13 <jborean93> I believe it's just me today, nitzmahone is travelling to London so travel time and timezones may not be the best for him for a few weeks
20:02:45 <jhawkesworth> oh yeah don't envy him the jetlag
20:03:20 <jhawkesworth> 8 hours the 'wrong' way.
20:03:34 <jhawkesworth> did we get to here? https://github.com/ansible/community/issues/420#issuecomment-516565410
20:04:24 <Shachaf92> one above i think
20:04:25 <jborean93> I believe https://github.com/ansible/community/issues/420#issuecomment-516376057 is first
20:04:35 <jborean93> #topic https://github.com/ansible/community/issues/420#issuecomment-516376057 win_data_deduplication
20:04:51 <jborean93> this sounds familiar, didn't we have a module that did something like this already in a PR?
20:05:16 <Shachaf92> maybe because if the the issues in find?
20:05:19 <jhawkesworth> I don't recall one from before
20:06:12 <jborean93> ah wait that was `win_compact` or something
20:06:53 <jborean93> need to add it to the list of things to review and go through it when I have time
20:07:05 <jhawkesworth> yeah you are right.. https://github.com/ansible/ansible/pull/56475#issuecomment-492756926
20:07:41 <jborean93> has tests so just needs a standard review unless someone has something they want to add let's move on
20:08:19 <jhawkesworth> ok
20:08:23 <jborean93> #topic https://github.com/ansible/community/issues/420#issuecomment-516565410
20:08:37 <Shachaf92> more reviews
20:09:56 <jborean93> should https://github.com/ansible/ansible/pull/57902/files#diff-30c11df7691ee0c89aa924eaf5a58396R48 include removable drives?
20:10:11 <jborean93> I would suggest we just use Local disks (3)
20:10:16 <jhawkesworth> https://github.com/ansible/ansible/pull/57902 still looks good to me.  dont have any s2019 to test against at the moment though so seems reasonable to wait for feed back from @spectator
20:10:52 * jhawkesworth cannot remember last time I defragged a drive
20:11:00 <jborean93> it's automatic since Vista days
20:11:17 <Shachaf92> Change to only local drives?
20:11:34 <jborean93> that's my suggestion, what do you think
20:11:43 <Shachaf92> If im not mistaken /C included non local and that is why i added it
20:12:08 <Shachaf92> since we want it to act as /C would have acted
20:12:34 <jborean93> have you confirmed that?
20:13:31 <Shachaf92> on my laptop it did include a removable disk i inserted
20:14:18 <Shachaf92> The problem was it didn;t include c: if im not mistaken
20:14:24 <Shachaf92> anything else worked
20:14:40 <Shachaf92> everything*
20:14:50 <jhawkesworth> only remaining spinning disks I have are external drives now.  If we are going to have it, not sure we should enforce arbitrary limits.
20:15:36 <jborean93> I suppose I should just test it out myself as well see if I can replicate that issue on 2019
20:16:20 <jborean93> For https://github.com/ansible/ansible/pull/58887, we should be adding a warning if the path does not exist
20:17:01 <jborean93> the other 2 just needs the standard review
20:17:50 <Shachaf92> does teh linux setup add a warning?
20:18:19 <jborean93> not sure but that's a change in behaviour to the existing windows setup module which is a stable interface
20:18:51 <jborean93> I would even argue that silently ignoring a missing file is wrong if that's how the Python setup module works
20:19:30 <Shachaf92> k, ill add a warning
20:19:48 <jhawkesworth> local facts are optional though
20:20:53 <jborean93> it's still a change in behaviour. I also dislike silent ignores
20:21:27 <jborean93> Makes it hard for people to debug issues when they act up, e.g. having a warning would tell the suer that the path they set does not exist so they can see that the path was set properly but the issue exists on their host
20:22:38 <jhawkesworth> ok fair enough as long as its only a warning if you have specified a local facts path
20:22:54 <Shachaf92> added it to the PR now
20:22:58 <jborean93> cool
20:23:09 <jborean93> looks like the next one is just a ping for the data deduplication module
20:23:17 <jhawkesworth> nice
20:23:21 <jborean93> and chopraaa is not here to talk about his last comment
20:23:26 <jhawkesworth> yep, already discussed
20:23:46 <jhawkesworth> hard for chopraa due to timezones
20:23:52 <jborean93> yea
20:23:57 <Shachaf92> wanted to add something, 2 actually
20:24:23 <jborean93> fire away
20:24:37 <Shachaf92> this - https://github.com/ansible/ansible/issues/59829, and soem issues on the win_domain_user one regarding the samaccountname
20:24:58 <jborean93> let's start with 59829
20:25:05 <Shachaf92> k
20:25:36 <Shachaf92> i think it should be a request to backport a feature to 2.8 but no one replied there
20:26:03 <jborean93> feel free ot backport, the PR that fixed this is definitely a bug fix and should be backported
20:26:15 <jborean93> the only thing you need with a backport is a changelog entry
20:26:53 <jborean93> also there's no need to request a backport, just do one and the release manager will either accept or reject it when they go to merge the backports
20:26:53 <Shachaf92> there is one in the PR, ill backport later
20:27:01 <Shachaf92> k
20:27:29 <jborean93> the main thing we don't backport is PRs that add new options
20:27:30 <Shachaf92> btw CI is acting up right now
20:27:56 <jborean93> define acting up?
20:28:30 <Shachaf92> doesn't work lol
20:28:33 <Shachaf92> at all
20:28:55 <jborean93> like it isn't being triggered
20:28:57 <Shachaf92> seems like it can't clone repos
20:29:01 <jborean93> ahh yea
20:29:14 <jborean93> I think GitHub is having issues right now, there have been reports of merging not working
20:29:44 <jborean93> just need to give it time and trigger a new runonce that passes
20:30:20 <Shachaf92> yeah, just thought i would let you guys know
20:31:07 <jborean93> thanks
20:31:14 <jborean93> what was the win_domain_user issue you wanted to talk about?
20:31:58 <Shachaf92> sec ill post the links
20:32:06 <jhawkesworth> this one? https://github.com/ansible/ansible/issues/59485#issuecomment-514438027
20:32:37 <Shachaf92> yeah and there is anotehr one
20:32:58 <Shachaf92> https://github.com/ansible/ansible/issues/44757
20:33:25 <Shachaf92> and another https://github.com/ansible/ansible/issues/45298
20:33:41 <jborean93> ah that old chestnut
20:33:50 <jborean93> so sounds like 2 issues
20:34:04 <jborean93> 1. the defaults result in a user that doesn't comply with MS guidelines
20:34:04 <Shachaf92> 2 of the 3 are samaccountname stuff the other one is another issue, i wanna get your input and maybe work on this on the weekend
20:34:18 <jborean93> that 2. we cannot easily rename an account
20:34:41 <jborean93> unless I'm misreading soemthing
20:36:09 <jhawkesworth> sounds like 2 could be fixed with examples showing you have to state=absent and state=present to change, perhaps (not module I am using, sorry)
20:36:40 <jborean93> that's not the best idea as it would result in a completely different SID,  breaking any mappings that exist for that user in the domain
20:36:56 <jhawkesworth> fair enough
20:37:01 <jborean93> 1 should be easy to fix by changing the behaviour
20:37:20 <jborean93> 2 is a lot harder as we need a way to uniquely identify the account so we cna change the sAMAccountName attribute
20:39:02 <jhawkesworth> would having an optional parameter which is used to uniquely identifier user by some attribute be an option?
20:39:20 <jhawkesworth> (i know so little about AD, sorry if this is a daft idea)
20:39:34 <Shachaf92> we can change the module to use the SID / GUID
20:39:57 <Shachaf92> some say that if they use the GUID and not the name to chagne the sam it works
20:40:02 <jborean93> I think having the optional parameter `id` that takes in the SID or GUID might be a better idea
20:40:12 <jborean93> the question is what does `name` relate to
20:40:25 <jborean93> and whether we should allow `name` to be the ID or SID of the account
20:42:19 <jhawkesworth> if this comments is correct, looks like name is a 'friendly name' rather than any kind of unique identifier: https://github.com/ansible/ansible/issues/45298#issue-472029975
20:42:28 <jborean93> everytime we call `Set-ADUser` we should be either passing in the suer object or use the GUID to reference the account. It's only when we create or get the user should `name` be used
20:43:02 <jborean93> looking at the code we do `Set-ADUser -Identity $username` where `$username` is the `name` module option
20:43:05 <jborean93> that is wrong
20:43:29 <Shachaf92> ok so ill do the changes and well see how it goes
20:43:45 <jborean93> sounds good, I'm sure there will be a few iterations you will have to go through
20:44:22 <Shachaf92> yep
20:44:41 <Shachaf92> i have a list of stuff to work on, been a busy time at work, haven't gotten around to 90%
20:45:17 <jborean93> sounds like my life right now, haven't been able to focus on the WIndows stuff as much as I want to
20:45:38 <jhawkesworth> everybody's busy
20:46:25 <Shachaf92> yep, ill have a short vacation soon and i think ill give it a push then
20:46:25 <jhawkesworth> I was going to say, I need to step back from ansible contribution for a bt.   Lots going on away from the keyboard at the moment.  I'll be back in the saddle in September though
20:47:00 <jborean93> I've got potential Jury Duty so I might be forced to step away from the keyboard
20:47:13 <jhawkesworth> thanks, appreciate all the work you are doing on PRs Shachaf92
20:47:24 <jborean93> yep they have been fantastic
20:48:55 <Shachaf92> no problem
20:48:59 <Shachaf92> a pleasure
20:50:24 <jborean93> cool unless there is nothing else to discuss let's get back onto our day
20:50:43 <Shachaf92> and night ;)
20:50:51 <Shachaf92> nothing here
20:51:03 <jhawkesworth> thanks all, have a good one
20:51:16 <jborean93> cool, thanks all
20:51:18 <jborean93> #endmeeting