Black Ice small segment size FTP attack caused by FX-scanner

From: Curt Wilson (netw3_securityat_private)
Date: Thu Dec 05 2002 - 15:02:15 PST

  • Next message: Chris Adams: "Re: Incident tracking database"

    
     ('binary' encoding is not supported, stored as-is)
    Recently saw something different in my Black Ice logs recently. AdvICE 
    says that this particular attack is related to an old problem in FW-1 and 
    PIX reported by John McDonald and Thomas Lopatic in 2000 (see 
    http://www.securityfocus.com/bid/979) wherein packets destined for an FTP 
    server behind a vulnerable PIX or FW-1 using a small segment size and 
    specially crafted PASV arguments (similar to the FTP bounce attack) could 
    be used to exploit other services (Solaris 2.6 tooltalk was used in the 
    bid 979 example).
    
    Severity	 timestamp (GMT)	 issueId	 issueName	 
    intruderIp	 intruderName	 victimIp	 victimName	 
    parameters	 count	 responseLevel	 intruderPort	 victimPort	 
    packetFlags
    
    4	 2002-12-04 07:32:53	2000316	 TCP small segment size	 
    12.37.34.75	 mail.omnisys-inc.com	 131.xxx.xx.xxx	 	 
    port=21|57&flags=S&options=maxseg:1460;bad_length:80	8	 A
    	21855	21	 0x26c06
    
    This particular attacker, coming in from mail.omnisys-inc.com, and the 
    signature of their scan looks very much like the FX-Scanner (fx-
    tools.net) mentioned on Incidents recently- see 
    http://online.securityfocus.com/archive/75/299560/2002-11-10/2002-11-16/0 
    for more discussion on this.
    
    The pattern of my attacker is as follows:
    
    Two ICMP pings using the data "hello???"
    Six SYNs for HTTP (firewalled)
    Six SYNs for TCP 57 (evidently because this port is usually closed)
    Six SYNs for TCP 21 (FTP)
    
    The MSS is 1460 bytes, and Ethereal says "Maximum segment size (option 
    length = 80 bytes says option goes past end of options)" in the TCP 
    options section. From what I recall, 1460 is a common MSS over PPP and 
    Ethernet links, but it looks like this scanner indicates 1460 but is 
    actually trying to use 80 instead, similar to John McDonalds discussion 
    where he set the MTU to 100.
    
    Is anyone aware of any newer vulnerabilities that are being exploited by 
    this technique?
    
    Curt Wilson
    Netw3 Security Research
    www.netw3.com
    
    ----------------------------------------------------------------------------
    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
    



    This archive was generated by hypermail 2b30 : Thu Dec 05 2002 - 19:27:58 PST