the Win7 and Win10 is probably easier - and you wouldn’t even bother with the ISO itself, but just copy the files to a windows share. Then, if you can just get WinPE to work (there are advanced copies out there now), then use that to just get to the share and launch the setup.exe, far more efficient than loading a huge ISO into the RAM of a machine that may or may not have it to spare.

But, for linux based ISOs, you’d mount as read-only with loopback, and then you’d point to the kernel and init after it’s mounted via NFS. I think you will find the below articles helpful.

http://www.cyberciti.biz/tips/how-to-mount-iso-image-under-linux.html

http://ipxe.org/appnote/ubuntu_live

@Sebastian-Roth I see. A coincidence this was the IP address of my old server, so that is the reason I asked.

@Tom-Elliott I’m on a roll with these questions with obvious answers…sigh…thanks again Tom!

All snapins do run as the “SYSTEM” user, it’s how the fog client is designed.

What exactly does bitlocker need to operate? I’m new to bitlocker.

OK that makes sense.

Hi all, thx a lot for your Help! Got it workin now!

Hi @Sebastian-Roth

That fixed it 👍 Thanks for your and Tom’s assistance.

Not that I know of…

@Tom-Elliott Done - looks like its working now 🙂

@Frank I don’t know if an update is stable until somebody updates and lets me know if prior issues are corrected for.

It’s one of the “cons” i guess of being on Trunk.

While I try my darndest to keep things as functional as I can, with changes can come problems.

@danilopinotti So the admin network and the academic network are on two different broadcast domains, and maybe even further separated logically via Vlans.

Then you just need a FOG server with two network interfaces (and connect each one to each network). You’d have two storage groups, and in each of those, one master node each, pointing to the exact same directory, but a different interface and IP for each (for the two network interfaces). I think it could work.

Or, just totally separate out the admin images and academic images. Have a storage group for each, and put each one on it’s appropriate interface and IP, and give them each a separate folder.

But Tom is probably right, the location plugin would be far more simple - just have two fog servers, one just a storage node for academic. Set each up as a location, and maybe even give each their own storage group, and make each the master of their group.

Lots of options.

But - the thing that I’m most concerned with is keeping the domain information straight. You’ll really, really need to pay attention to that - or you might find all of your computers joined to one or the other domains… and that wouldn’t be good at all. So there’s a whole lot of possibility for a really large failure with this setup.

@Sebastian-Roth Thank you.

@Wayne-Workman Code added.

@Rico Maybe this helps: https://bbs.archlinux.org/viewtopic.php?id=179929

Seams like some kind of weird partclone thing… You might need to create an empty output file first.

As well you want to read this: https://sourceforge.net/p/clonezilla/discussion/Clonezilla_live/thread/4db4b2bd/ (partclone.restore seams to be deprecated and you should use partclone.<fstype>)

@Wayne-Workman I think it’s very close indeed, and do realize there are a lot of things that needed to be changed in order to make gpt/uefi/windows 10 work properly. I’m not really complaining as I’ve used many different coding languages, except php, and completely understand the complexities with trying to perfect a program, but we should have a better idea than “hopefully sometime”. It’s just a little discouraging sometimes.

@greenfog @dylan This has been in our Wiki for quite a while: https://wiki.fogproject.org/wiki/index.php?title=Troubleshoot_MySQL

From my personal experience I would recommend updating to the latest trunk version of FOG. It is much more advanced and I can capture and deploy W10 images without issues.

You need to install “subversion” and then download the latest package with

svn checkout https://svn.code.sf.net/p/freeghost/code/trunk

From there you install as usual. But be aware that updating might not always work flawlessly so I would test it in a non-productive environment and consider a clean install of the trunk version.

@Wayne-Workman said:

@plegrand said:

It’s not our dns servers

You don’t have to use them. You can make your own and direct all hosts to use yours.

I install dynamique DNS with bind and it works fine. Thanks again

@Wayne-Workman thank you for support.
Works like a charm.
Regards