Rémi -
(2) On requirements 1 and 3:
REQ-1: A NAT MUST have an "Endpoint-Independent Mapping"
behavior for DCCP.
REQ-3: If application transparency is most important, it is
RECOMMENDED that a NAT have an "Endpoint-independent
filtering"
behavior for DCCP. If a more stringent filtering behavior is
most important, it is RECOMMENDED that a NAT have an
"Address-dependent filtering" behavior.
These requirements are general and not specific to DCCP. Would
it
make sense to specify them in a separate RFC for NATs in general,
independent of any specific transport protocol?
Whether it's of any use depends on the connection model (or lack
thereof) of
the transport protocol. I don't want to presume that this would make
sense
for all future transport protocols. [...]
I don't agree. A reason for recommending endpoint-independent mapping
and
filtering is to enable applications to refer each other to a host
behind a NAT.
This is desirable independent of the transport protocol.
(3) On requirement 6:
REQ-6: If a NAT includes ALGs, it MUST NOT affect DCCP.
This requirement is not 100% clear. I am assuming it means:
"If a
NAT includes ALGs, the NAT MUST NOT affect DCCP packets that are
processed by one of those ALGs." Suggest to reword the
requirement
in this way.
This reads worse to me. An ALG cannot "process" DCCP packets if it
does not
affect in any way. There is already a IESG discuss on this. What
about this?
REQ-6: If a NAT includes ALGs, they MUST NOT affect DCCP.
Make it even clearer:
REQ-6: If a NAT includes ALGs, the ALGs MUST NOT affect DCCP.
- Christian
_______________________________________________
Ietf@xxxxxxxx
https://www.ietf.org/mailman/listinfo/ietf