Question on address configuration

[Edward Lewis]

At 1:19 AM -0500 1/10/02, Robert Elz wrote:
>How one finds DNS servers has nothing at all to do with this...

Your points up to this are accurate.  But I wouldn't agree that there is no
relationship to finding DNS servers.

Reading the replies and thinking more about this situation (and why two of
us IPv6 newcomers found this), we realized that this problem is rather
DNS-centric.  Any other server out there can use autoconfig with the MAC
address and rely on DNS for name to number mapping.  If the MAC card on,
say, an SMTP server dies, a new one is installed and the DNS zone data
modified.

There are only two places a DNS server's IP has to be fixed.  One is in the
(pardon the UNIX-centricity here) /etc/resolve.conf file and in the
server's delegating zone data (e.g., what Verisign says about your .com NS
servers).  Solving the /etc/resolv.conf problem can be done a few different
ways (including putting recursive servers on site/local-addresses.  But if
the glue records can't be updated at the parent, this is a problem...

One unwritten assumption I make in designing a DNS set up is to run two
servers, one that is recursive (answering general questions) and another
(set) that is authoritative (answering only specific questions).  ("Why" is
for the DNS mail lists.)  So it seems natural to have some site-local
addressed servers.

However, this won't help me when I am on the road and dialing in via other
networks.  But then again neither will an anycast approach if name servers
restrict whom they permit to launch recursive queries.  (And thus we're
back to why I need to set the IP[v6] address.)  Hmmmm.

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Edward Lewis                                                NAI Labs
Phone: +1 443-259-2352                      Email: lewis@tislabs.com

Opinions expressed are property of my evil twin, not my employer.