I'm looking for some direction.
We have a virtual desktop environment (Citrix Xen Desktop) of about 6,000 desktops. These desktops have a LANDesk inventory/remote control (not SWD) agent on them. Occasionally, when using the LANDesk console to remote to a virtual desktop, LANDesk connects to a different desktop (not the one selected).
- The desktops are all pre-generated with unique device names (xendesk####).
- Those desktops all have unique MAC addresses as well.
- When the desktops are generated (using the Citrix tools) they are essentially "frozen" in the state of the "source" desktop, other than the unique name (which is tied to AD). I don't know if the desktops have had the LANDesk unique device ID stripped currently, but I'm pretty sure at one point they did.
- Once a week (give or take) the "master" image is unfrozen and edited, then refrozen. As the desktops that are derived from that master reboot, they get the new version of the master image.
Should we be stripping the LANDesk unique ID from the master image every time it is edited?
If we do that, every time a desktop reboots (potentially daily) it would generate a new unique LANDesk ID (but still have the same name and MAC address).
My fear is that thousands of devices have the same unique LANDesk ID and that could cause problems from a remote control standpoint.
My other fear is that if we strip the unique ID, I'll end up with (literally) hundreds of database entries with the same name and MAC address but unique LANDesk device IDs.
Any idea what the best approach is here?