Brian Utterback <brian.utterb...@oracle.com> wrote: >> However, what I don't understand is why an IPv6 address does not fit >> into a struct sockaddr, and why this fact is so badly documented. >> It took me a lot of time to find why my queried IPv6 addresses were >> truncated. >> > > It is a little tricky to be sure. Part of it was backwards compatibility > and history, part of it was providing the proper abstractions. The root > cause of course is that IPv6 addresses are longer than IPv4 addresses.
But a struct sockaddr has always been padded to 16 bytes, with a little care an IPv6 address and the associated info would have fitted in that. > If you use them properly, you should never have a problem with > truncation though. You would have to be trying to put a IPv6 address > into a IPv4 sockaddr for that to happen. I used a "struct sockaddr" as the parameter to those getinfo calls, and it turned out to be too small to return a sockaddr_in6. (while it is large enough for a sockaddr_in) However, all the examples I found always used that struct sockaddr, and it is also the type of the parameter in the declaration. And in fact it works fine with IPv4! Only after long hairpulling I found you need to custom-write a union of struct sockaddr and struct sockaddr_storage, and pass that to the functions. (or declare variables of type struct sockaddr_storage and cast the types all the time) _______________________________________________ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
