Remotely imaging a laptop using FOG

[JYost]

My company has offices in about 10 different countries around the world, one of which is in Tokyo, Japan. That particular office is relatively small (about 20 users) so we do not have on-site IT support there. We do, however, have FOG servers set up for each of our off-sites. Due to the language barrier there it has proved virtually impossible to walk a user through the process of booting to PXE, entering the desired hostname, etc. during the imaging setup. I’d like to know if there is a way to remotely do this via the web interface on their FOG server. Ideally the user would boot the device to PXE, and the rest would be done automatically using the settings in the Web interface.
Thank you in advance,
John

[Sebastian Roth]

@JYost said in Remotely imaging a laptop using FOG:

Due to the language barrier there it has proved virtually impossible to walk a user through the process of booting to PXE, entering the desired hostname, etc. during the imaging setup.

All this can be done in the web UI without having to PXE boot any of the hosts. The one thing you need to know is the MAC address of each host. With that you can simply register the 20 clients in the Web UI manually.

For actually deploying an image you surely need to make the hosts PXE boot but no further user intervention is needed bejond that.

[george1421]

Your wish list is kind of how FOG works today. Once the host is registered, you schedule a deploy from the web ui, when the target computer is pxe booted it will just image the machine. If you have the fog client on the golden image, it will rename the computer and deploy any application you have defined. All without human intervention.

Now in my company’s case we don’t use the fog client at all. Our computer names are calculated while imaging by a post install script. The unattend.xml script connects the computer to the domain in the proper OU. The appropriate drivers are installed based on the hardware model. All using the lite touch concepts. The user “can” start the deployment by PXE booting and then selecting Deploy Image, reimage themselves (they pick the image from the deploy image menu that we predefine for their computer). They just start it, walk away and then come back about 20 minutes later to a user ready computer. This is not typically how we do this, but its possible. We will normally schedule a deployment in the web ui and then have the IT tech boot the computer into pxe using the F12 boot menu.

So with your question, there must be something I’m missing in your request. Other than a language barrier, what is the root of your problem deploying images using FOG?

[Sebastian Roth]

@JYost said in Remotely imaging a laptop using FOG:

Due to the language barrier there it has proved virtually impossible to walk a user through the process of booting to PXE, entering the desired hostname, etc. during the imaging setup.

All this can be done in the web UI without having to PXE boot any of the hosts. The one thing you need to know is the MAC address of each host. With that you can simply register the 20 clients in the Web UI manually.

For actually deploying an image you surely need to make the hosts PXE boot but no further user intervention is needed bejond that.

[JYost]

@george1421 Being a relatively new FOG user I never even thought about going into the web interface and pre-configuring a client. Because the registration is done for me during the initial PXE phase where you enter the hostname, image ID#, etc. this is how I’ve been doing it thus far. Now that I’m looking at it I see exactly what you mean about going in and pre-creating the HOST. This will take care of my problem. Thank you very much for your help!!

[george1421]

@JYost said in Remotely imaging a laptop using FOG:

Now that I’m looking at it I see exactly what you mean about going in and pre-creating the HOST.

This is referred to as manually registering the host. The only bit of information that may be problematic is the target computer’s MAC address. If you have the mac address then you an just go into the web ui and add a new host and enter its name, image, mac address via the web ui. You won’t get the full inventory right away, but there will be enough information to pxe boot the target computer. If the fog client is already running on the target computer, when you create a deploy task the target computer will reboot, or power on if WOL is enabled on the target computer. But to do this, the target computer needs to be configured to boot through PXE to the FOG menu every time. If there is nothing to do the target computer will continue on booting from the local hard drive.

ALSO, if the fog client is installed on the target computer BUT the target computer is not registered in FOG the target computer will be added to the FOG server in the pending state. You will need to approve the target computer to have it added to the inventory. You will still need to set the image name but you should have a near hands off registration.