DynDNS
So far, this chapter has concentrated on the user gaining access to the network or application. There is, however, another important issue to consider—how to find the mobile equipment if the other side initiates the communication. As you will see in future chapters, there are a number of solutions that provide a stable anchoring point that can be used to find the current point of attachment or to direct all traffic to, but they require changes in the protocol stack. The standard way of informing "the Internet" where a certain host resides is by using the Domain Name System (DNS). DNS, after all, contains mappings from host names to IP addresses. So, if the entry in DNS is updated every time a host changes its point of attachment (and thus IP address), DNS information can be used to find the target IP address of a connection.
This is precisely what Dynamic DNS (DynDNS) is—a DNS server that is optimized for frequent updates of the mapping information. A number of implementations of DynDNS exist, often provided for free. For this to work, a DNS client that updates the current name to IP address mapping every time the host changes IP address is required.
Because of the distributed nature of DNS (it takes some time before DNS resolvers become aware of a change), DynDNS is not practical when DNS changes occur very frequently, in the order of magnitude of seconds.
Another point of concern is that rogue DNS updates can be used to redirect traffic. Unless the DNS updates can be authenticated—for example by using DNS Secure (DNSSEC)—this is a security problem. DNSSEC, however, is not yet widely in use.