Phillip Hallam-Baker <hallam(_at_)gmail(_dot_)com> wrote:
Since the one legacy protocol that has a dependency on IP address constancy
is FTP, it would seem to me to be much easier to upgrade FTP to remove the
dependency than to try to control the network.
There are other protocols hiding out there.
MATIP, RFC2351 (not from the IETF, obviously), allows either endpoint
to send a session-open setting some paramaters for the session, and to
confirm the session-open from the other side. So if both sides send
session-open, they're both supposed to ignore the session opened by
the endpoint with the lower-numbered IP.
In practice, what this seems to mean is that sometimes when you set
up new links, MATIP doesn't work, you get on the phone with the admins
at the other side, and you agree which one of you will send the
session open. Then you configure one endpoint never to send it.
The RFC was published in 1998, though I don't know when they actually
came up with the protocol. Its use is, I think, still increasing, as
more airline industry links are set up over TCP/IP networks.
I wouldn't have known about MATIP if I hadn't gotten a job related to
airline industry networking, and I'm sure there are other naive
protocols out there, designed by people or groups who were not that
familiar with the way of the Internet and do protocols their own ways.
[MATIP has a host of other problems that I won't mention :) ]
-- Cos
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf