Re: Proposed new scheme for resolving the system hostname
Previously Thomas Hood wrote:
> On Thu, 2005-06-16 at 21:07 +0200, Wichert Akkerman wrote:
> > I'm not quite sure what you mean here. If you mean the system should
> > always use a fixed hostname
>
> I don't mean that. I mean that when you do gethostbyname() on the
> system hostname, the h_name you get back is either the system hostname
> or the system hostname with a domain name suffix.
Ok, that makes sense.
> A minority of Debian systems are standalone. Of those, some might be
> networked in the future and so are no different from what I called
> "mobile systems" in my first message. In any case, the scheme I
> proposed works as well for Gilligan's Island systems as it does for
> others.
It does not work for systems that use a custom resolver though. And I
think we do ship with a few (if memory serves me correctly OOo and
mozilla both have one).
> > The concept of 'primary NIC' does not exist.
>
> This concept goes back a long way and can be found in old UNIX and Sun
> manuals. If a system has one or more permanently installed NICs with
> statically assigned IP addresses then one of these is regarded as the
> primary one and its address is the one associated with the system
> hostname in /etc/hosts.
We live in a different century now. I don't think the concept of a
single primary NIC is very useful.
> > 'defaults' is a highly confusing name
>
> It's the best name that has been suggested so far.
fake_hostname would be more intuitive I think.
> Yes, we'll have to add a section to the release notes.
Release notes should include it, but it should also be in all user
manuals, FAQs and other documents.
Wichert.
--
Wichert Akkerman <wichert@wiggy.net> It is simple to make things.
http://www.wiggy.net/ It is hard to make things simple.
Reply to: