20:00:00 #startmeeting Ansible Windows Working Group 20:00:00 Meeting started Tue Aug 31 20:00:00 2021 UTC. 20:00:00 This meeting is logged and archived in a public location. 20:00:00 The chair is jborean93. Information about MeetBot at http://wiki.debian.org/MeetBot. 20:00:00 Useful Commands: #action #agreed #halp #info #idea #link #topic. 20:00:00 The meeting name has been set to 'ansible_windows_working_group' 20:00:05 o/ 20:00:06 ahoy hoy 20:00:56 how goes it 20:01:31 not too bad, yourself? 20:01:37 p good 20:03:00 nitz is away on potential jury duty 20:03:47 πŸ‘©β€βš–οΈ 20:03:49 #topic open floor 20:04:02 I got nothin as usual, saying hi 20:04:11 how goes your hashicorp stuff? 20:04:16 how's your psrp stuff going? 20:04:54 haha, it's good, had to slow down a little the past few weeks, other obligations, but going well 20:05:23 I've taken a bit of a break unfortunately, was just getting too much of an overload thinking about it. I've been working on a few other things recently and am hoping to get back to the psrp stuff soon 20:05:31 almost all the functionality I wanted pulled out into utils is done, now further separating the tests for that stuff, moving along 20:05:43 that makes sense, it seemed like grueling work 20:06:03 it's just so many layers, I don't want to think about testing it all unfortunately 20:06:07 when it gets like that, I find myself just staring at the screen barely actually changing anything 20:06:19 good to step away for a bit 20:06:52 Yea that's what I was finding and thought a break would be best. I did finish the actual protocol side https://github.com/jborean93/psrpcore. It is hooked into the IO code as well but the latter needs a big tidy up and testing added 20:07:39 I've been working on a Python krb5 API to fill in some gaps with GSSAPI https://github.com/jborean93/pykrb5. Just waiting to hear back from PyPI on some namespace troubles before publishing that. 20:08:03 wow nice! 20:08:13 definitely ⭐ing both these 20:09:36 hey LowlyDBA ! we're in open floor, discussing some of jborean93 's python kerberos and psrp projects 20:09:39 I also have a POC for supporting compression with WinRM but I haven't hooked it into pypsrp yet. I don't even know if the performance hit is even worth it 20:09:45 howdy 20:11:03 hm that would be interesting to see, I suppose it could help on the ansible side with the module/utils transfer on very slow links 20:12:06 Hi! 20:12:38 btw that hackathon thing I did a while back to have Ansible powershell modules executing with `pwsh` via a connection and shell plugin, that was inspired (and used) by LowlyDBA , for his growing list of SQL server modules :) 20:12:40 hey 20:12:58 yea I can't promise anything but I am hopeful it might speed up some larger data transfers but will see how it is when it is integrated 20:13:11 ah nice 20:13:48 I mostly added support for compression to help with analysing live traffic in Wireshark 20:15:26 oh interesting, does that just keep the noise down by making the captures smaller? 20:16:44 it should reduce the amount of data that is transferred but as with all compression it's highly dependent on the data that is being sent. Sometimes it might have a decent reduction, other times it may not 20:17:16 right, but is that the main benefit in packet captures, just reducing the volume? or is there some other benefit to compression? 20:17:33 (as it related to traffic analysis I mean) 20:18:11 Ahh I see. The native client enabled compression itself so any large exchanges may be compressed making it difficult to analyse what each packet contains 20:18:23 By having something that can decompress the data I can then see what is exchanged 20:19:15 ahhhh gotcha 20:19:34 so it's MITM for compression 😜 20:20:07 that's the original purpose for creating it, I'm hoping it can find other uses though 20:20:15 cool cool 20:20:44 it might even be useful for my SMB library as it also can compression using Xpress + Huffman that WinRM usees 20:22:19 oo good stuff 20:23:16 I can't take full credit for that though, it's mostly ripped off of Microsoft's pwsh implementation and I just fixed a few bugs I came across 20:24:53 heh, it's good that open sourcing pwsh can lead to advantages for completely separate projects too 20:25:37 yea was definitely a nice win there for me, I wasn't going to try and implement a compression algorithm myself 20:26:40 haha reminds me of when I was a kid re-implementing `strcpy` functions and such in C/x86 asm 20:26:53 good learning experience.. not actually recommended to accomplish anything 20:27:17 I've yet to breach the assembly barrier, C is as low level as I've gone 20:27:50 ironically I did it a bunch way back, and now haven't touched in like 20 years 😬 20:28:16 for context, I did 16 and 32bit, and have never tried it on 64, so I'm.. rusty to say the least 20:29:33 sounds like fun 20:30:53 16 bit is... special, having to address memory in two different values. I've all but forgotten how to do that though. Anyway compilers are scary good these days, so πŸ€·β€β™‚οΈ 20:31:53 I've heard about the peculiarities of 16-bit, glad I never had to deal with it 20:33:54 it's mostly before my time as well, I think all I ever did in 16 bit was make 16 bit DOS stub executables that replace the generic "this application requires Windows" message when you tried to run win9x programs from DOS 20:33:58 πŸ‘΄ 20:34:16 heh 20:35:12 anyway.. drifting offtopic. I gotta run for today. 20:35:24 'til next time πŸ‘‹ 20:36:07 enjoy 20:36:17 thanks for joining LowlyDBA 20:36:23 #endmeeting