I have a server that's recently been crashing occasionally in odd ways, so I decided to configure up Intel AMT on it in the hope of being able to reset it remotely when this happens. For better or for worse, in the current set up, this server has a dynamic IP address, but things are configured so that DNS entries are updated if a new IP address is assigned.
Since I made this change, I've been having issues with AMT and the host OS (Ubuntu) ending up with different IP addresses assigned via DHCP; inevitably, this manifests itself in the DNS entries pointing to the IP address for AMT rather than the host OS, making it rather challenging to access the server (thank goodness for mDNS or I might be out of luck!).
My understanding is that when the host OS is running, AMT is just supposed to use the same IP address, but it seems like it's actually doing its own DHCP requests for some reason.
Any ideas how to resolve this situation?
Edit: out of technical curiosity and for some practical reasons, I'd really rather know how to make this work with dynamic IPs, although static IPs would theoretically resolve the issue I think.
Best Answer
I think this is a normal behaviour. AMT operates independently of you operating system, so you may think of it as another device on your network and it should be connected to the network even when your server is powered off.
One way to do it is to use separate network adapter for your OS and leave embedded port to AMT.