The attack could create a buffer overflow which exectuted a reverse telnet 
connection. Since the sniffing interface does not have an IP address the 
reverse telnet connection will be sent out the management interface. This is 
only a theory...but it could easly happen. 

Neil








Can someone from XForce please take the time to consider all aspects of this 
recently announced exploit?  For example, organizations that have configured 
snort to run on a "stealth" interface only should not be impacted under any 
circumstances and therefore should not impetuously scramble to download and 
deploy the latest build.  

For example:  The section "It is also not necessary to know the network 
location of a Snort sensor. Exploit packets can be sent to any portion of a 
network upon which a target Snort sensor is listening" "A successful attack can 
either crash the Snort sensor, or lead to complete remote compromise."

Understandably this can potentially crash the snort daemon, cause the system to 
hang, or in a best case scenario generate a genuine buffer overflow and thereby 
render the system open to arbritary code.  However, providing the intruder 
crafts a framgemented RPC packet and targets this at any asset on the known 
network where there also happens to be snort IDS system running in stealth 
mode.  What possible advantage can the intruder have on following up with a 
attempting to take control of the compromised snort system? Additionally if 
snort if running in non-promiscuous mode, what are the possiblities of the 
intruder obtaining and highjacking a layer 2 address from the snort system?

It may be advantageous to inform the public about the full scope of this 
exploit as well as any potential restrictions or drawbacks.

Thanks

Tim Slighter




_______________________________________________
ISSForum mailing list
[EMAIL PROTECTED]

TO UNSUBSCRIBE OR CHANGE YOUR SUBSCRIPTION, go to 
https://atla-mm1.iss.net/mailman/listinfo

Reply via email to