This may be a bug, or a possible way we did not think how it works so let me explain this!
What we have been doing with this client is putting an IPIP tunnel bridging it, in v2.9, and bridging the Ethernet port of a hotspot network. This network gets its DHCP from across the bridge at the hotspot controller (non MT).
So, we have HOTSPOT NETWORK ↔ IPIP BRIDGE ↔ HOTSPOT CONTROLLER
This works fine, as the hotspot controller gets MACs and issues out DHCP.
In version 3 we are trying to standardize with L2TP tunnels. Same as above, but now we are wishing to only use L2TP tunnels. In Version three, what it seams to happen is the MTU is not changed across the bridge between the Ethernet and the L2TP Tunnel. What seams to be happening is packets are entering the MT v3rc5 with a packet size of 1500 bytes as they should, when they are bridged with a IPIP tunnel in version 2.9, it seams the packets are broken up into whatever size is necessary the IPIP Tunnel. Even if we set this size to 1000 bytes, it works fine.
In version 3rc5, it seams it that the L2TP tunnel does not split the packets up, i.e. we get the black hole router effect, as anything above the size of the L2TP tunnel does not get out. i.e. some websites come up, etc.
Is this the way it should work, is there a setting that I am missing? Etc?
Dennis