RE: PI/metro/geo [Re: The state of IPv6 multihoming development]

["Michel Py" <michel@arneill-py.sacramento.ca.us>]

Peter,

>> I've seen forged RSTs to take out the BGP session also,
>> but the earliest attack I saw involved somone stealing
>> a BGP TCP session and then injecting  false routing
>> information. Operators who don't have BGP TCP MD5
>> deployed are at serious operational risk these days.

> Peter R. Tattam wrote:
> I thought this was a man-on-the-side attack, not
> man-in-the-middle

I am no expert in attack classification, but can you explain why? I have
done that myself once in the lab, and it was a MITM as far as I am
concerned: Get in the middle, intercept the traffic from the mark to the
peer and vice versa, and inject yours instead.

Michel.