Re: [v6ops] Security: Unique IPv6 Prefix per Host

[Tom Herbert <[e-mail]>]

On Mon, Nov 13, 2017 at 12:30 AM, Mark Smith <[e-mail]> wrote:
>
>
> On 29 Oct. 2017 09:20, "Ted Lemon" <[e-mail]> wrote:
>
> On Oct 28, 2017, at 5:59 PM, Tom Herbert <[e-mail]> wrote:
>
> Some of this has been touched upon in the IDEAS discussions, but nothing
> formalized yet (identifier-locator use implied to eliminate as much topology
> on address as possible). I could spin a draft specifically on one time use
> addresses and privacy (won't make cutoff though).
>
>
> The problem is that now you have a requirement for substantial
> infrastructure to manage the ID/locator separation.   Would look a bit like
> Tor.   Who's going to pay for it?
>
>
> Actually, I think multipathing e.g. MPTCP goes a long way to solving that
> problem. The 32 bit token generated and used by subflows to identify their
> peer is a temporary host ID that is independent of the locators/addresses of
> the host.
>
> My understanding is that in "original" ID/locator separation, the host ID
> was expected to be fixed and bound to a host across many transport layer
> connections. Perhaps that's what makes it a hard problem.
>
Mark,

The idea of single use addresses is that a host can be assigned many
identifiers from which it can create a unique source address for each
connection. It is up to the infrastructure to direct packets for each
addresses to the right end host. It's true this is potentially a lot
of state for the infrastructure to manage, but the total number of
state entries would be no more than the number of NAT entries if that
were in use today so I don't believe it is impractical.

> Temporary, per application session host IDs are also better for privacy.
>
MPTCP can also leverage this as a means to effectively change the
source address of an existing connection in order to benefit privacy
for long lived connections.

Tom