Tom Herbert <[email protected]> wrote: > This is an update to the host to network signaling draft. Note the name > is now properly host2netsig.
> Other changes include: * Add suggestion that signals could be in a
> namespace managed by IANA and allow vendors to define their own signals
I found section 3 vs section 4 a bit confusing.
Many of the things in section 3 seemed to be actually existing mechanisms.
for instance:
3.4. Traffic flow analysis
[I-D.cc-v6ops-wlcg-flow-label-marking] describes a mechanism to mark
packets of flows with information that identifies the application or
user that is sending packets.
seems to be a different host to network signaling mechanism.
Section 4 seems to be more about what didn't work :-)
I just want to re-iterate section 4.1.1:
* Stateful devices can be an anonymous single points of failure in
the network path. For instance, stateful devices can break
individual connections mid-flow due to state eviction.
The incumbent telco in Canada has long used ECMP in a stateful way that
basically
always breaks ssh connections that last longer than a few minutes. They only
fixed this when HTTP/1.1 became predominant. Even sending keepalives did not
help.
"* They are IPv6 specific, there is no equivalent support in IPv4."
seems like a feature for IPv6, not a problem :-)
We have many ways for embedding IPv4 inside IPv6, if needed.
Do you really need to make such a long argument for IPv6 Hop-by-Hop?
I think that this document is really a kind of merge Requirements and
Architecture. Maybe it will also be a Roadmap to other documents?
--
Michael Richardson <[email protected]>, Sandelman Software Works
-= IPv6 IoT consulting =- *I*LIKE*TRAINS*
signature.asc
Description: PGP signature
_______________________________________________ Int-area mailing list [email protected] https://www.ietf.org/mailman/listinfo/int-area
