Re: [v6ops] draft-ietf-v6ops-ula-usage-recommendations - work or abandon?

[Nick Hilliard <[e-mail]>]

On 08/11/2015 19:47, Philip Homburg wrote:
> Call me old fashioned, but what NAT breaks is not what is traditionally
> called a protocol layering violation.

When you code layer 3 information at what we usually refer to as layer 7,
this is layering violation.

> So it is perfectly normal for protocols above the transport layer to pass
> transport layer addresses around. How else would a distributed application
> organise itself?
> 
> But NAT breaks that.

No, the combination of layering violations and transport identifier
translation causes this to break.  The root cause of the problem is the
layering violation mandated in the L7 protocol; the breakage trigger is
NAT.  The result is end user trouble.

> But NAT also breaks the core internet protocols. Suppose a host uses dynamic
> DNS

ddns is also a layering violation because it makes an assumption at layer 7
about what's going on in layer 3.

Nick