From what I can see right now we're pretty safe from mappers. Since
the number of possible connections increases exponentially and the
maximum mapping speed increases linearly (each node only allows x
incomming connections) so long as the network is big it will be
impossible to map.
I'm going to add support for fine-grained keyspace mapping to fnmap
and re-run the whole thing. If I manage to find any usefull patterns
it may be possible to only map a subset of the possible
connections. If this can be done reliably, and that subset increases
linearly, we're in trouble again. :)
Also combinding the attack of figuring out the nodes that initally get
some data and a mapping operation may prove usefull. Once I've got
fine-grained keyspace mapping I'll try that attack. Even if I can't
find any usefull patterns this should be a pretty easy attack due to
the low number of inital nodes.
One thing we should do is make locally inserted data ignore the
routing tables. Just send it out totally randomly. We could even go as
far to have the first x hops do this.
--
Need some Linux help or custom C(++) programming? Drop me a line and
I'll see what I can do. Resume at http://retep.tripod.com/resume.html
GCS d s+:-- a--- C++++ UL++++ P L+++ E W++ N- o K- w-- O- M
V- PS+ PE+ Y+ PGP+++ t 5 X R+ tv-- b+ DI+ D++ G e- h! r-- y--
[EMAIL PROTECTED] http://www.petertodd.ca
PGP signature
