Hi @george1421
Sorry to tag you but whatever I have configured so far was based on a number of answers from you in the forums and hence hoping if you have some time to reply.
Thanks much

Hi,
I have a working fog server VM and a master/golden image captured from one of the Dell desktops. It is the same as the target PCs that I need to image.
We need to deploy Ubuntu desktop on 20+ Dell PCs (same model) in a lab, all currently have Windows.
The “deploy an image” instructions mention the following as I understand:

1. Power-on PC, go to BIOS and boot from pxe
2. From Fog menu, select “Perform Full Host Registration and Inventory” and follow the prompts to provide the info including hostname, fog username/password, etc. PC will then reboot.
3. Interrupt boot and go to BIOS again to boot from PXE. Image will then be deployed and the PC will reboot again.

This seems like a pretty manual process to follow for 20+ desktops. Specially as I have to interrupt boot twice on each PC, once for host registration and the 2nd time for image deployment. I don’t have much experience with pxe so I’m not sure if I can make any config changes that can help.
I also don’t know if there is a way to tell all PC’s to boot from pxe on their next reboot, so I guess the first manual boot interrupt is unavoidable.

Here’s what I was thinking to happen ideally:

1. Power-on each PC, go to BIOS and boot from pxe
2. All PCs will then get the image directly from the fog server and get deployed, without any manual intervention and ideally at the same time.
   I read some forum posts about the “Load and Go” process which uses the “Deploy Image” option from the menu and does not require the PCs to be registered. This is helpful but still requires me to choose the “Deploy Image” option on each PC, so its not completely automated.

The last issue is of hostname. We have our own DHCP/DNS server which has the configuration to provide the pxe info and the fog server. We also use the our dhcp/dns server to provide hostnames to all our VMs. I was hoping I would be able to do that with these desktops. I can register each PC’s mac in our dns/dhcp server so that it receives the hostname, but not sure if fog deployment will mess with it since the master image has its own hostname?

Can someone please provide some help/guidance.
Thanks so much.

I was not able to figure out capturing the image from a VM, I decided to use an actual desktop as the master/golden image. It made more sense as it would include the right hardware type, drivers, etc. I have captured the image finally. Now on to deployment

Just following up, if anyone has any suggestions or advice.
Thanks

Hi,
Fog newbie here.
My goal is to be able to image 20+ Dell desktops for a lab with Ubuntu 24 desktop version with our configs and software stack.

1. Minor clarification:
   I installed the fog server on an Ubuntu 24 VM successfully, but need to understand one thing:
   I cloned the git repo in /root and then ran the installer script from /root/fogproject/bin/installfog.sh. Everything finished successfully and I got the " * Setup complete" message at the end. I’m able to browse to the url and see the dashboard. But I also see that fog server also installed/populated stuff under /opt. I see ipxe stuff under /root/fogproject/utils and src directories but not under /opt.
   Q: Is this expected or did I mess something up?

2. Image capture issue:
   I created another VM for the ‘master’ image (on the same network as fog server) to use for capture, but didnt realize that the master image VM also needs to pxe boot. That’s a problem since we have a separate network for the desktops lab (with dhcp options 66, 67 added for fog), and this network is not present in our VMware environment hence I cannot put the master image on the target network that has been configured with the dhcp options. There is already another pxe server for the VM network so I cannot put the dhcp options on the VM network also.
   I read a bit about booting a usb, and understood some of it but not quite clearly. Plus in my case I can’t use a usb as the master image host is a VM. Then I read a bit about capturing an image with a fog client and was able to install the fog client on the master VM and tried to start a capture from the fog server ui, but it did not do anything. I also didn’t see any connections between the fog server and this master VM using netstat.
   Q: Is there any way I can capture the image of the master VM to use for the desktop deployments?

Thanks very much