Security Incidents mailing list archives

Re: Request For Comments from Firewall Community


From: Martin Hoz <mhoz () gama fime uanl mx>
Date: Tue, 17 Jul 2001 21:16:44 -0600

William:

Probably another forum for discussing this is the INCIDENTS
forum.

http://www.securityfocus.org/forums/incidents/intro.html

Their goals are precisely about how to track attacks and and how
to deal with these...

Since this is a matter that not only implies about just
"reporting" and "blocking" IP's/subnets, it could be worth
to try with an audience more used about dealing with done
attacks as well as protecting against them (i.e. The Incidents
forum and the firewalls forum).

I like the idea... :-)

Regards.

- Martín.

-- 
Martin H. Hoz-Salvador
EX-A-IEC, EX-A-FIME
http://gama.fime.uanl.mx/~mhoz

"Gimme a firewall sandwich with packet filter bread and 
fast ethernet mustard. No pickles, please.” - A. A.
""'Firewall sandwich with load balancers' sounds good; I'll
 order two with extra mayonaise and a Coca Cola" - C. R. Wilson


William Bartholomew wrote:

I am making this suggestion based on Ron DuFresne's [dufresne () winternet com]
email "You're on your own now" last week. I am open to everyone's thoughts
on whether it is a good idea or not or any suggestions that you may have.
This has been written very quickly and is intended as an overview of the
idea only.

Background

As firewall administrators we have all seen a variety of attempts against
the networks that we have been hired to protect, these attacks range from
stab-in-the-dark probes to pinpoint purposeful attacks. However, a large
number of attacks come from a small number of IP addresses, and furthermore
most of the attacks are aimed at wide-ranges of IP addresses and as such
affect more than one company and as such affect more than just one firewall
administrator. The majority of these attacks are harmless (attacks agains
ports such as 23, 111, 135, 137, and 139) as they are usually stopped by our
border-firewalls.

Suggestion

What is needed is a facility to allow us to coordinate our efforts to
protect ourselves from that small number of IP addresses that probe our
networks CONSTANTLY. The most effective way for us to do this I believe is a
service similar to ORBS, but instead of tracking open mail relays it tracks
IP addresses that are known threats. Ideally the system would allow IP
addresses to be split into categories and then firewall administrators could
download pre-written rule sets for various firewalls to block the IP
addresses in the categories that they select. A system such as this would
quickly reduce the effectiveness of attacks from these IP addresses as
numerous firewalls would effectively block these ranges even before the
attack is attempted against them.

Potential Problems

* System could be abused by people blocking good IP addresses
  - Possible Soln: Firewall administrators have to register to block IP
addresses
  - Possible Soln: Multiple requests to block the same IP address have to be
received before the rule is activated
  - Possible Soln: Log files have to be provided

* Too many rules may be generated
  - Possible Soln: Rules may be split into categories and firewall
administrators can choose which categories they wish to use
  - Possible Soln: Rules in certain categories may expire after a certain
time period

* Spoofing of IP addresses
  - Possible Soln: Protected somewhat by the measures shown under "System
could be abused by people blocking good IP addresses"
  - Possible Soln: Rules may be removed by successful application from the
owner of the IP address/range

Any thoughts on this would be greatly appreciated.

Kind Regards
William Bartholomew


----------------------------------------------------------------------------


This list is provided by the SecurityFocus ARIS analyzer service.
For more information on this free incident handling, management 
and tracking system please see:

http://aris.securityfocus.com


Current thread: