[Full-disclosure] TCP Hijacking (aka Man-in-the-Middle)

[3APA3A]

 Valdis,  you  should  back  to  Cretaceous period, because Oliver talks
 about   man-in-the-middle   attack,   not  about  blind  TCP  spoofing.
 Randomized ISN doesn't protect against MitM.

--Thursday, October 25, 2007, 9:40:53 PM, you wrote to olivereatsolives at gmail.com:

VKve> On Thu, 25 Oct 2007 10:09:47 PDT, Oliver said:

>> I have been searching all over the place to find an answer to this question,
>> but Google has made me feel unlucky these last few days. I hope I could find
>> more expertise here. The burning question I have been pondering over is -
>> could TCP connections be hijacked both ways?

VKve> Quick summary:

VKve> Steve Bellovin pointed out the issue. 19<stone age>

VKve> Kevin Mitnick exploited it. 19<bronze age>

VKve> Steve wrote RFC1948, which basically said "Use randomized ISNs so the attacker
VKve> has to work harder at it". 1996.

VKve> A lot of vendors sort of implemented it. 1996-2000.

VKve> Michael Zalewski did a nice phase-space analysis and showed a lot of vendors
VKve> botched it. http://lcamtuf.coredump.cx/oldtcp/tcpseq.html 2000

VKve> A lot of vendors fixed their shit, but a lot didn't.
VKve> http://lcamtuf.coredump.cx/newtcp/ 2001.

VKve> You're now caught up to 6 years ago.


-- 
~/ZARAZA http://securityvulns.com/
Впрочем, важнее всего - алгоритм!  (Лем)