Mapping IPv4 to IPv6
-
I have a couple of questions:
a) how does pfSense handle addresses of the type::FFFF:123.45.67.89
and
::123.45.67.89Are these routeable? Are there IPv6 tunnel service providers that will create a tunnel for such address, given one has a proper assignment of the corresponding IPv4 address block?
b) are there IPv6 addresses along the lines of
::ABCD:123.45.67.89:EF01
or something like that, where ABCD is some fixed (unknown to me, if this type of address exists) reserved prefix just like in the ::FFFF:123.45.67.89 case the FFFF, which is followed by a mapped IPv4 address, and followed with e.g. an extra 16-bit of unused address space, such that if someone now has e.g. a class-C IPv4 network, it could be transformed into a mixed mode IPv4/IPv6 network in this way:
With IPv6 connectivity, the network only sees IPv6 addresses, so every IPv6 addressable host can be reached.
In the time after the final demise of IPv4, that's all that remains, and the bragging right that the address space was derived from one of the original IPv4 assignments.
With IPv4 connectivity, only the addresses in the 123.45.67.0/24 (or whatever netmask applies) are reachable, however, each of these addresses could be used by a NAT router with a private /16 network behind it, whereby say we have a NAT router at 123.45.67.89 which uses a private network of 10.89.0.0/16 then a host behind that NAT router with an address of 10.89.12.34 would be mapped into the IPv6 address space by the router as
::ABCD:123.45.67.89.12.34
An IPv4 device in the 123.45.67.0/24 range that's not a NAT router would simply waste the extra bits at the end, and thus be reachable from the IPv6 net under a whole bunch of different, but equivalent addresses.This would allow even non IPv6 aware devices to fully participate in the IPv6 net by a translation/mapping router, get these devices out from behind NAT into the IPv6 address space, would automatically assign IPv6 address space to anyone who already has assigned IPv4 address space, etc.
