I finally got to test this today. It appears to be
Hostname/bind
Reboot
Waits for a complete reboot (this is good for be because this is where my GPO installs happens) then after the boot process completes the snapin runs. In this case running a script to install AutoCAD.
@Sebastian-Roth
By check do you mean to check for binding? The issue then is that it’s not going to try to run again if it wasn’t bound at the time; as the snapin will have technically exited fine having run it’s check and done nothing. Unless there is a way for a batch file to exit and re-arm the snapin to be executed at a later time. If I can determine the order I can work out the timing of when each thing lands.
Hi,
I have a snaping (batch script) that I want to run on my machines right after imaging, but it needs to run after the machine is bound to AD. What is the order of operations with regard to namechange, binding, and snapins? Also there are Group Policies running on the machine after the first boot from binding, how does that play in with snapins?
That’s weird. I’m definitely on master branch (at least according to #git branch) but I’m still running MySQL.
What is the error on Ubuntu? I’m running on 18.04.3 LTSB and upgraded a while back and haven’t been having issues. What should I be looking out for?
In the meantime I would suggest writing a powershell script to unbind and rebind and you could push it out as a snapin after executing the name change. I don’t know how many of these you are doing and how coordinated you are with the machines so YMMV
Right, I upload a .bat and use the batch snap-in template
If the system is already bound to AD and the share is on an AD bound system it should still work. I push AutoCAD this way for one-offs (as a snapin). Just a UNC path with no net use.
IF EXIST "C:\Program Files\Autodesk\Revit 2020" goto _End
"\\SHARE\GPO Installs\Revit2020\Img\Setup.exe" /W /q /I \\SHARE\GPO Installs\Revit2020\Img\REVIT2020.ini /language en-us
:_End
FOG is agnostic to the difference. You could go pata spinning disk to sata ssd and it won’t matter as long as FOS can boot on your system and see the drives, and as long as the OS has the drivers for the system, or they are available to it.
The Share is read/change for everyone, but NTFS permissions is only read and execute for everyone.
Mind you I have the script in my snapins, but I usually use it as part of group policy and its a startup script with logic to look for an existing install.
The way I do it, is that the script goes into FOG but the install files live on a different share that I can call from a UNC path. so the guts to my script are:
\\installersshare\office\setup.exe /configure \\installersshare\office\config.xml
I can safely assume my devices are all running the latest version so I created a group policy to replace the settings.json with a fixed copy on a network share. Which is fine for what I need and I will just update my base images to reflect the correct setting.
Thanks for the help
Is there a way to report on the client version that my currently deployed machines are running? I can push a new settings.json with group policy but I don’t want to have incorrect settings within it and I can’t guarantee that all my users have the same version.
Recently moved to using HTTPS and now clients with both existing and newly installed fogclient are not communicating with the server. If I install the client on a new machine it does not show up as a pending host in fog server. Running 1.5.6.4 dev build.
Do I need to specify HTTPS in the fog URL in the installer?
I’m calling it spam. Especially with that oddly spaced URL at the bottom. Flagged for your consideration.
Still getting errors on iPXE with the new build
(note to anyone with this issue, I only did a buildipxe.sh on dev-branch and that didnt work, a full reinstall on dev-branch fixed it)
Gotta go pick up my kids. I’ll take another swing at this tomorrow.