[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: old GSE idea



Mostly just thinking out loud here (if this is still permitted)...

On Thursday, April 17, 2003, at 10:58  AM, Christian Huitema wrote:
Even if we solved MAC address collision on a single link, you have to
remember that a host can be multi-homed to several interfaces. Whatever
we do to support the survival of TCP connections should also work when
the traffic moves from interface A to interface B, e.g. from 802.11 to
GPRS. In such cases, the two interfaces are likely to belong to
different sites, and are like to use different interface identifiers
(since they indeed are different interfaces). The solution of "just
using 64 bits" will not work there.

-- Christian Huitema

If you define the multi-homed 'site' to be the host then I think it will.

and

On Thursday, April 17, 2003, at 12:35 PM, Christian Huitema wrote:
Sure. Another issue is privacy. I don't necessarily want different ISP
to be able to correlate that two traffic flows do in fact originate from
the same hosts. In fact, privacy advocates would go ballistic if we
mandated that.
No reason (other than efficiency) that a globally unique ID must be permanently assigned to an interface. If you can verify global uniqueness in a reasonably rapid fashion, you just auto-generate a new globally unique ID per 'session'.

Rgds,
-drc