Re: CVS commit: src/bin/hostname

["Roland C. Dowdeswell" <elric%imrryr.org@localhost>]

On Wed, Jul 31, 2013 at 04:26:40AM +0000, David Holland wrote:
>

> On Tue, Jul 30, 2013 at 11:13:57AM +0100, Roland C. Dowdeswell wrote:
>  > > If you *want* to use a FQDN as your hostname I see no problem with it
>  > > (apparently it makes Kerberos work better/at all) but I don't think it
>  > > should be required, expected, or even encouraged.
>  > 
>  > Well, I think that while we ship software that works in a particular
>  > way, we should note the requirements in the likely sorts of places that
>  > people will look for documentation.  Postfix as an example expects that
>  > either gethostname(2) returns a FQDN or that /etc/postfix/main.cf has
>  > the myhostname or mydomain defined.
> 
> Nope; it will also do a DNS lookup on the gethostname result. Not that

It certainly doesn't on NetBSD or in the official Postfix distribution.
Debian patches it to do this which is why you might have seen this
behaviour.

>  > I'd suggest that we simply state that some
>  > software that we ship either in base or pkgsrc works a bit better
>  > if you use a hostname which meets a number of basic requirements
>  > such as being an FQDN.
> 
> Except that this is not, in my experience at least, actually true.

There are two examples of such software that we ship, namely Postfix
and Kerberos.  I'm failing to see why we wouldn't want to document
that a bit more explicitly.  Perhaps it would be better for people
to just discover on their own how the tools work.

Also, we document that a hostname is a ``network name'' as we simply
return the hostname in uname(3) in the nodename field.  If you want
to assert that a hostname should not be documented to be a network
name, then we should likely create a different value to represent
nodename for the purposes of uname(3).

--
    Roland Dowdeswell                      http://Imrryr.ORG/~elric/