Re: [Full-disclosure] Remote buffer overflow in httpdx

2009-10-16 Thread Freddie Vicious
Just saw this on Twitter, an MSF exploit published:
http://www.rec-sec.com/2009/10/16/httpdx-buffer-overflow-exploit/

On Fri, Oct 9, 2009 at 7:58 PM, pankaj...@gmail.com wrote:

 The addr value used is required to reach the ret instruction. The value
 used 0x63b8624f lies in idata segment of n.dll
 Note that in order to reach ret instruction,
 value at addr+0x0e0f should be non-zero for
 if(isset(client-serve.redirect)) to succeed  = 004069E1  CMP BYTE PTR
 DS:[EAX+0E0F],0
 and
 addr+0x0f24 should be writable for client-state = STATE_DONE to execute.
 = 00406AAF  MOV DWORD PTR DS:[EAX+0F24],0

 The other two addresses used are
 ret1 = 0x64f8134b (pop ret in core.dll) to pop addr and return to ret2
 ret2 = 0x7c874413 (jmp esp in kernel32.dll) to jump to shellcode following
 ret2.

 Though I am able to get a shell, the retn/offsets used are not universal.

 Thanks,
 Pankaj




-- 
Best wishes,
Freddie Vicious
http://twitter.com/viciousf
___
Full-Disclosure - We believe in it.
Charter: http://lists.grok.org.uk/full-disclosure-charter.html
Hosted and sponsored by Secunia - http://secunia.com/

Re: [Full-disclosure] Remote buffer overflow in httpdx

2009-10-12 Thread Freddie Vicious
Can't reproduce it too (XPSP3 En + httpdx 1.4.0)...

On Fri, Oct 9, 2009 at 8:49 AM, dr_...@hushmail.com wrote:

 this didn't seem to work for me. Test system XPSP3 + httpdx 1.4.0.

 Definitely causes a crash but the retn/offsets must not be universal?




-- 
Best wishes,
Freddie Vicious
http://twitter.com/viciousf
___
Full-Disclosure - We believe in it.
Charter: http://lists.grok.org.uk/full-disclosure-charter.html
Hosted and sponsored by Secunia - http://secunia.com/