ietf
[Top] [All Lists]

Re: utility of dynamic DNS

2002-03-01 02:00:03
 In your previous mail you wrote:

   Perhaps.  Certainly stable IP address is preferable to being
   constantly and needlessly renumbered all the time (although if the
   practice became more prevelant, the silver lining is that it would
   likely put an end to that abomination known as IP-address dependent
   license keys).

=> unfortunately not because one can answer you may use another
abomination known as NATs (sorry Keith to use this dirty word :-).

   So for static installations such as Cable Modems and
   DSL lines, I agree with you, and I wish network providers would
   provide stable IP addresses.
   
=> I agree and there is no technical reason to stay with dynamic IP
addresses.

   However, I much prefer DHCP plus DDNS to Mobile IP.  With Dynamic
   DNS, the security model is well understood; I need only inform a
   single host with whom which I have a trust relationship --- the DNS
   server for thunk.org --- that the DNS address for my laptop should be
   changed to 1.2.3.4.
   
=> I disagree because in order to make your box quickly reachable
after an address change you have to play with RR TTLs and DNS is
*not* designed to behave well with short TTLs...

   With Mobile IP, the security model seems to be (in order to avoid
   triangle routing),

=> the key point is again inside the parenthesis: your concern is
not about Mobile IP (as you can get a trust relationship with your
home agent) but about routing optimization of Mobile IP.

   The other, is as near as I can tell, completely and totally
   hopeless as far as being practical or deployable.
   
=> and Mobile IP without routing optimization as a replacement or
a complement to DDNS?

Thanks

Francis(_dot_)Dupont(_at_)enst-bretagne(_dot_)fr



<Prev in Thread] Current Thread [Next in Thread>