Re: DNS bug, or is it my config in 2.3Stable1

On Tue, 1 Feb 2000, Dancer wrote:

> Henrik Nordstrom wrote:
>
> > Merton Campbell Crockett wrote:
> >
> > > Except for SunOS and Solaris, the interpretation of 0.0.0.0 as a broadcast
> > > address was corrected long ago. The correct interpretation of 0.0.0.0 is
> > > "this host".
> >
> > Says who?
> >
> > * Havent seen this in any RFC, only references to 0.0.0.0 being an
> > obsolete limited broadcast address.
> > * It is not mentioned in SUSv2 from what I can tell..
>
> I don't know if it's written explicitly anywhere. But the semantics do support
> it's use as 'any address' or 'any address here'. bind() for example, uses it
> this way to bind to all addresses on a host. I seem to recall a couple other
> similar usages which would support the impression that 0.0.0.0 would mean 'this
> host'.
>
> Really speaking though, it's just a magic cookie. A reserved address
> interpreted in a variety of ways. Certainly there is some software out there
> that treats it as 'no address'.

RFC 791 that defines the Internet Protocol (IP) is the first to identify the
address 0.0.0.0 to mean "this network and this host". In case, there were
questions regarding this interpretation, it was clarified in Internet Host
Requirements -- Communications Layers, RFC 1122. This has been elevated
from a proposed standard to a full standard and published as STD 0003.

Unfortunately, I can't recall which RFC that defines its use within DNS. I
believe it was one that Mockpetris authored or was one of the co-authors.

Merton Campbell Crockett
Received on Wed Feb 02 2000 - 08:15:42 MST