On 4/27/2017 5:36 AM, Stewart Bryant wrote:
...
So this seems to me to be an argument of principle vs pragmatism, and
in general pragmatism has served the Internet well so far.
I think pragmatism is what Fred and Brian arguing for. It is certainly
the general approach that I support.
IMO, 1981bis should limit its advice to behavior in response to ICMPs, e.g.:
In reaction to a received ICMP message, a node MUST NOT reduce its estimate
of the Path MTU below the IPv6
minimum link MTU.
Doing otherwise presents too big an attack surface.
Any other advice on how an end system manages its idea of MTU or
otherwise overrides the 1981bis computation of path MTU should be
limited to host requirements documents, not this one.
Joe