Re: VML Exploit vs. AV/IPS/IDS signatures

[nirvana <karmic_nirvana () yahoo com>]

Sanjay,
   
  It's not that bad as it sounds. This whole discussion was in the context of client-side scripting attacks and how 
commercial IPS/IDS solutions tackle them (and I am talking about the best-of-the-breed here, not confined to India). I 
wanted to show some disbelief on the fact as to how the network-style vendors  declare that they cover vulnerabilities 
like the recent VML one. In actuality, the solutions are very, very exploit-facing. But they become slightly effective 
with real-time exploit updation. IPS are not meant to fix these kind of vulnerabilites, is the point I am rhetorically 
stressing on.
   
  >I have heard ( I am not sure whether it is really true) that few products (I 
> don't even know the name!!) are simply dependent on open sources 
> like Snort or bleeding snort or nessus!!!!) for signatures.
   
  Well, let's not count those here. My focus was on the popular vendors which have good capabilities to protect against 
network attacks.
   
  >To overcome this (pathetic) situation, the solution, being suggested by 
> Pukhraj makes sense i.e. collect as much exploits as possible and 
> then try to analyze them and write signature. Here I want to add one 
> thing that once we have a number of exploits, we should, at least 
> now, try to understand the vulnerability based on the information 
> present in all the exploits and try to come out with a common 
> signature (or a common set of signatures).
   
  As an IPS developer in the past, my aim was to perfectly decode the network traffic, normalize the malicious elements 
and provide generic vulnerability-facing solutions. It did happen in most cases with exceptions like scripting attacks, 
obscure propriety binary protocols etc. Perfect decoding and hence vulnerability facing solutions are viable in almost 
all cases.
  
 
SanjayR <sanjayr () intoto com> wrote:
  Hi Aviv/Pukhraj & others:
As a security professional and researchers, our aim is to provide 
more in-depth information on intrusion (security) aspects, for 
example, some virus out-break, new windows vulnerability etc. Aviv is 
right by saying that signatures should match the vulnerability, not 
the exploits. Signature writing is a very responsible task and, of 
course, technical too. But unfortunately, there are not many people, 
who have required knowledge ( i m talking in context to india). but 
companies need people with the requirement of writing signatures. So, 
in this process, so called security professional start looking at 
exploits and write signatures (just to mention, I have seen few snort 
signatures that match "Shellcode part"!!!) . As Pukharaj mentioned 
that there are not many variants found in the wild, such signatures 
work and company and hired-security-professionals are happy. I have 
heard ( I am not sure whether it is really true) that few products (I 
don't even know the name!!) are simply dependent on open sources 
(like Snort or bleeding snort or nessus!!!!) for signatures. To 
overcome this (pathetic) situation, the solution, being suggested by 
Pukhraj makes sense i.e. collect as much exploits as possible and 
then try to analyze them and write signature. Here I want to add one 
thing that once we have a number of exploits, we should, at least 
now, try to understand the vulnerability based on the information 
present in all the exploits and try to come out with a common 
signature (or a common set of signatures).

regards
-Sanjay

At 11:07 AM 9/28/2006, Pukhraj Singh wrote:
> And you tell me how many of these variants you will actually find in
> the wild. Won't be a significant number I bet.
>
> Cheers!
> Pukhraj
>
> On 9/27/06, avivra wrote:
> > Hi,
> >
> > > i.e. I can't afford to buy "specialized" security tools/devices for
> > > "speclialized" attacks unless my company relies heavily on web/content
> > > services.
> >
> > So, you will buy "specialized" security tools like firewall or
> > Anti-Virus, but not web content filtering tool?
> >
> > > In our company, we established a information-sharing
> > > network with other security companies. So the real-time exploit-facing
> > > signatures were then subjected to live traffic, honeypots and countless
> > > variants; They seemed to work out pretty well.
> >
> > I would like to see how your real-time signatures get updated with the
> > randomization implemented in the new VML metasploit module. Your
> > "countless" exploit variants will become really innumerable.
> >
> > The problem is that the signatures are written for the exploit, and
> > not for the vulnerability.
> >
> > -- Aviv.

Sanjay Rawat
Security Research Engineer
INTOTO Software (India) Private Limited
Uma Plaza, Nagarjuna Hills
PunjaGutta,Hyderabad 500082 | India
Office: + 91 40 23358927/28 Extn 424
Website : www.intoto.com
Homepage: http://sanjay-rawat.tripod.com
Computer Security: A little delay to break into your network.
-- DSR






                                
---------------------------------
Want to be your own boss? Learn how on  Yahoo! Small Business.

_______________________________________________
Full-Disclosure - We believe in it.
Charter: http://lists.grok.org.uk/full-disclosure-charter.html
Hosted and sponsored by Secunia - http://secunia.com/