@robza Running commands via UNC is not a great scenario, what I usually do is map a drive with net use and use that drive then delete the mapping.

Net use z: \\upow-8.intranet\public\Office2016_PL_32B\
z:\setup.exe /adminfile z:\Unattended.msp
net use z: /delete 

This would be a better and more reliable way to set this up. I always use Z: because it usually doesn’t conflict with other common mappings such as U or H for home/user drives.

What version of FOG are you using? With the trunk version and the new FOG client you could push the client via GP and the hosts will register automatically tot he server. If you are not on trunk then this is not possible.

I just tested this and you can search by the name of the image associated tot he host. So create a group, add member and search for “4000WIN7” and add all the results. Then go to image association in the group change to “4000WIN10” all associated hosts will then be updated

you could create a group and search for key words to add hosts then change the image association of the group. That will update each host even if you later delete the group the hosts will still have the new image association.

What OS is the server running on? Also you say you updated to 6211, what version where you previously on, was it trunk or an older version?

If you went to trunk from .32 or .33 you may have to go assign the OS tot the image definitions, as they were assigned to the machine before. This maybe true for 1.x also, but I am not sure as I skipped that version in all of my deployments.

@Wayne-Workman That makes perfect sense, I hadn’t considered that it might need to delete an existing file before uploading a new file. I think a check in the code to ensure a file is selected before going further would solve this. I am not a PHP or HTML coder at all so unfortunately I wouldn’t be able to help much with this. I can script in windows and a little it Linux but not sure that will do much in this case.

DHCP settings are explained here if you need anymore info on them,

https://wiki.fogproject.org/wiki/index.php/Modifying_existing_DHCP_server_to_work_with_FOG

Did you add option 67 in DHCP as undionly.kpxe? You need both 66 and 67

This is by design, a group cannot dynamically update hosts as they are added. Instead when you add machines you must then go in and apply those settings to the group and it will update each host. A group is more of a place holder that you can change settings to multiple computers at once, unfortunately you cannot have a group that auto applies settings each time you add it to the group. This is why the AD settings appear to not be saved in the group, when you set them it changes each host currently in the group, there is no place in the DB that it stores the information for the group as a whole.

@Tom-Elliott Needing a file is fine, I was just trying it without a file because I just had a single command I wanted to run so I figured why bother with a batch file. What it is doing it deleting all current snapin files in the snapin directory. If I click add and not have a file associated with the new snanin it will delay for a time and then error out. When I go back into my /opt/fog/snapins directory it is empty of files, the folders are still there with the certs but no snapin files.

On Ubuntu 14.04, FOG Trunk 6150

If I try to create a snapin without a file associated with it it deletes all the snapin files in my /opt/fog/snapin directory. I noticed that they had been deleted a couple weeks ago but wasn’t sure why. I had just tried to update to the latest trunk that day and assumed it was a bug in the update. But today I just tried to create a snapin without any file association and it took forever to error out and then didn’t create the snap in. I then went to add a file and found that the snapin directory was empty. I restored from a back up and tried again and got the same results.

I have a few commands that I need/want to run and instead of creating a batch file I just figured it was easier to add the commands as snapins. Turns out FOG doesn’t like that. I am fine with having to do a file upload as a batch file instead, but deleting the snapins is not a good result of accidentally forgetting to select a file.

@PageTown You don’t want to use NTFS on an NFS share. It is best if you reformat the drive to Ext4 and then go back through the steps outlined by @Wayne-Workman

@PageTown d you run chmod as sudo?

sudo chmod -R 777 /media/BFD/images

Did you set the pxe options in your DHCP server? OPtions 66 and 67 in Windows.

https://wiki.fogproject.org/wiki/index.php/Modifying_existing_DHCP_server_to_work_with_FOG

@GFm Now to carve out the time to update those over to Partclone. Just a quick FYI in case you are not aware of this. To convert them you don’t have to have the actual machine they go on. I found a spare machine and pulled the image down, and shut off the machine (DO NOT LET IT BOOT AFTER IMAGING) then changed the type in FOG and re-uploaded to the server. Doesn’t matter what hardware you put it on as it never boots into windows.

But on the same token if you have the correct hardware for them it is a perfect time to update the images, I am sure windows updates are way behind. lol

@GFm I think we are going to have to defer this to the @Developers and @Tom-Elliot in particular. We at least have a place for them to start at as it appears the legacy images are the issue and your new images are fine. The only work around that I can think of in the mean time would be to roll back to an earlier build and use it, until a fix is figured out.

As I mentioned before that I would highly recommend getting those images converted over as the benefits are great and the there is no real draw back other than the few min it takes to pull switch and push back up.

I would recommend if that works that you pull the image down to a machine and then change to Partclone and re-upload. That will convert the image and keep it more in sync with the trunk and future releases. Partclone is much better and faster.

@GFm I think know what is wrong now, I missed that part. You need to go on the FOG server to FOG Configuration>FOG Settings>General Settings and check the box for FOG_FORMAT_FLAG_IN_GUI and save. Then go into your image and change it to Partimage in the "Image Manager"drop down. This will switch that image to the legacy style from .3x and allow you to image the machine.

Can you try to image the Dell with the image that won’t work just to see if it will pull the image down? I know it might not boot correctly after imaging, I am just wanting to see if the dell will pull the image. This will get us a little closer to finding if it is the image or machine. If the Dell is able to get the image then I would try updating the BIOS on the Lenovos.

Is it only failing on this machine or do you have any others you can try? You could try to do a disk wipe task and then redeploy? This will wipe any existing odd partition table information off the drive.