>>> I am sorry, I no longer share this opinion [...]  The next version of
>>> cerowrt will do translation from the external IPv6 address range to
>>> a static internal one (or ones, in the case of multiple egress
>>> gateways),

>> (Insert strong expression of disagreement here.  Use any means available
>> to convince Dave otherwise, including flattery, threats, demagoguery, ad
>> hominem attacks and photographs of cute animals.)

> Hahaha. Thanks juliusz! I have laughed far too little in the past few
> weeks.

But I'm dead serious, Dave.  I'm actually making my serious face right now.

Just looking at stuff I'm intimately familiar with, you're breaking:

  - the IPv6 PEX code in Transmission, which assumes that if you call
    connect on an IPv6 UDP socket, then getsockname will return
    a currently usable IPv6 address;
  - the IPv6 DHT code in Tranmission, which assumes that calling the same
    sequence as above will return a host's currently preferred IPv6
    address (Kademlia needs a bound socket, changing IPs per destination
    breaks the algorithm);
  - Matthieu's multipath Mosh, which puts the server's local IPv6
    addresses on the wire.

Dave, let's please find a way to solve the renumbering problem in a way
that doesn't break applications.

-- Juliusz

_______________________________________________
homenet mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/homenet

Reply via email to