[AusNOG] My Predictions for the ISP Industry

Mark Newton newton at atdot.dotat.org
Thu Mar 15 14:21:47 EST 2012


On Thu, Mar 15, 2012 at 01:22:05PM +1100, Eric Pinkerton wrote:

 > One of the things that I haven't seen much discussed about is that
 > in turning on dual stack you are obviously increasing your attack
 > surface.

Your attack surface is probably already increased, because autotunnelling
protocols mean your users are quite probably already running IPv6 behind
your back without even realizing it.

I did a presso that touched on that at AusCERT last year.

 > of course because many of the current defensive measures are
 > behind the curve here. 

At a network level, why would the defensive measures be any differnent?
Mirror your IPv4 ACLs with IPv6 ACLs; your dual-stack firewall appliance
is probably already controlling both protocols. 

 > Also you can draw your own conclusions about the efficacy of current
 > LI solutions where IPV6 is concerned.  

Again: Why would they change?

 > There are number exploits specifically targeting weaknesses within
 > IPV6 and ICMP6 protocols, and research has put pay to earlier
 > misguided perceptions that IPV6 is inherently more secure that
 > its predecessor. 

I don't think anyone ever seriously believed that IPv6 was or could
be inherently more secure than its predecessor.  I mean, sure, 
there've been a lot of years featuring a lot of people who don't
know what they're talking about saying a lot of things, but nobody
has actually listened to them, right?

 > Add to this  an increase in the opportunity for configuration
 > mistakes and you start to appreciate why many organisations are
 > in  'wait and see what everyone else does' mode.

Ah, right -- so that's how professionals in this industry deal with
things they don't know how to do these days, is it?

Despair.

  - mark



More information about the AusNOG mailing list