tcp/routing question...

Bob Chiodini rchiodin at bellsouth.net
Tue Jun 7 16:08:54 UTC 2005


On Tue, 2005-06-07 at 11:53 -0400, Scot L. Harris wrote:
> On Tue, 2005-06-07 at 11:18, bruce wrote:
> > are you sure about this...??
> > 
> > here's my question...
> >  client (a)  --->>>> bank server (b)
> >  client (a)  <<<---  bank server (b)
> > 
> > if server b gets the data/information from 'a', server 'b should get ip
> > address 1.2.3.4, which is the real ip address of client 'a'.
> > 
> > is there away for a mitm server, to get in the middle, manipulate the data
> > from 'a' to 'b', send the data to 'b' and spoof the ip address to look as
> > though the data came from 'a'..
> > 
> > -bruce
> 
> Short answer yes.  The idea of a MITM attack is that somehow the
> attacker has inserted a system or redirected your systems traffic
> through a intermediate system.  The middle system acts as a proxy.  It
> can be capable of rewriting the packets going between the two systems
> under attack.  The middle system will handshake with each of the other
> systems and relay packets between so you won't know it is there.  At
> that point it will collect information or can modify the packets going
> through for what ever purpose.
> 
> The difficulty is in getting a system inserted into such a position.  It
> typically requires physically inserting a system in the path unless the
> attacker is able to mess with the end systems proxy settings and
> redirect things that way.  
> 
> 
> -- 
> Scot L. Harris
> webid at cfl.rr.com
> 
> "For the love of phlegm...a stupid wall of death rays.  How tacky can ya get?"
> - Post Brothers comics 
> 

Another possibility is a worm or virus that usurps the network stack to
manipulate the packets.  No physical machine would be needed in the
network path, but the results would be the same.  I guess this would be
sort of "man on the side".  IIRC this is how Cisco's VPN client works,
but in a good (at least not a bad) way.

Bob...




More information about the fedora-list mailing list