> > The issue is if several freenet nodes operate in a small block of changing
> > IP addresses.  Its very likely that one node will occupy the IP address of
> > a previous node.  This would cause the node contacting it to believe that
> > its been subverted or at least changed its key.  
> 
> No, the node would say "The node I'm trying to connect to (identified by
> it's pk) is no longer at this address" and behave just like if the
> connection had failed (ie, look up ARK or drop reference).
Exactly.  My point was you can't necessarily assume that it indicates
subversion, because of these blocks of dynips.

> If all connections use PK authentication there is absolutely no need to
> get all up and over about the fact that somebody might have been trying to
> subvert the connection - since such an attack can't succeed, there is no
> need to take any extra precautions. This is just like there is no reason
> for a node to get more excited about bad CHK data then a broke connection
> - if the attack is as productive as hitting brick wall with your fists
> then let mallory has his fun.
Perhaps.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: 
<https://emu.freenetproject.org/pipermail/devl/attachments/20000818/24830fd7/attachment.pgp>

Reply via email to