Summary of Carla's Reading Group 6/30, 7/3

In last Monday's discussion we
critiqued the paper on the Bluebox policy-based IDS,
which went pretty quickly. The main flaws were that
the authors did not go into how difficult it would be
to implement properly and did not give any test
results to show how well it works compared with other
IDS's. At today's meeting, one of the grad students,
Jacky, gave a talk about three OSPF attacks, the
Seq++, MaxAge, and  when a compromised router
advertises its cost to be 0 so that it becomes a sink.
He also showed measurements of the effects of these
attacks. The presentation was pretty clear and
interesting.