--- David Frechette <[EMAIL PROTECTED]> wrote:

> Hello,
> 
> Recently I'm receiving the following 2 notifications
> in the report of a single host:
> 
> 12213:
> "The remote host might be vulnerable to a sequence
> number approximation bug, which may allow an attacker
> to send spoofed RST packets to the remote host and
> close established connections."
> 
> 4259:
> "The TCP initial sequence number of the remote host
> are incremented by random positive values.
> Good!"
> 
> IMHO, these notifications are contradictory. I've
> rescanned the system multiple times, but these 2
> notification keep showing up. Is there an explanation
> for this?

Yes.  Read the cross-reference the first one mentions.  The latter
purely relates to how the ISN is calculated and if it could be
guessed.  If it could, then one could perform a blind spoof attack
to take advantage of IP-level access controls (e.g. a service
relying upon host.allow entries).  The prior refers to the ability
to send an RST packet to an established TCP connection because of
how implementations deal with window sizes and allowed ranges of
RST values.  This mainly affects long-lived services, such as BGP,
that maintain a long TCP connection and suffer from having it drop.

Is the nmap plugin (14259 btw) statement good?  If you don't know
what ISN calculations are and the difference between that and the
12213 plugin, then no; it requires a priori information.  Once you
understand ISNs and what the output is stating, then IMHO, it's a
good statement.

12213 reference:
http://www.securityfocus.com/bid/10183/solution

ISN reference:
http://www.faqs.org/rfcs/rfc1948.html

> Thanks in advance,

Hope this helped and that I wasn't being condescending,

Jon



        
                
__________________________________ 
Yahoo! Mail - PC Magazine Editors' Choice 2005 
http://mail.yahoo.com
_______________________________________________
Nessus mailing list
[email protected]
http://mail.nessus.org/mailman/listinfo/nessus

Reply via email to