The DNS is mainly used to find the address that corresponds to a given name. However, it is sometimes useful to obtain the name that corresponds to an IP address. This done by using the `PTR` (`pointer`) `RR`. The `RData` part of a `PTR` `RR` contains the name while the `Name` part of the `RR` contains the IP address encoded in the `in-addr.arpa` domain. IPv4 addresses are encoded in the `in-addr.arpa` by reversing the four digits that compose the dotted decimal representation of the address. For example, consider IPv4 address `192.0.2.11`. The hostname associated to this address can be found by requesting the `PTR``RR` that corresponds to `11.2.0.192.in-addr.arpa`. A similar solution is used to support IPv6 addresses :rfc:`3596`, but slightly more complex given the length of the IPv6 addresses. For example, consider IPv6 address `2001:1890:123a::1:1e`. To obtain the name that corresponds to this address, we need first to convert it in a reverse dotted decimal notation : `e.1.0.0.1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.a.3.2.1.0.9.8.1.1.0.0.2`. In this notation, each character between dots corresponds to one nibble, i.e. four bits. The low-order byte (`e`) appears first and the high order (`2`) last. To obtain the name that corresponds to this address, one needs to append the `ip6.arpa` domain name and query for `e.1.0.0.1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.a.3.2.1.0.9.8.1.1.0.0.2.ip6.arpa`. In practice, tools and libraries do the conversion automatically and the user does not need to worry about it.
None