Dear Vladimir,

"almost" is often enough :)

btw, it was about triggering the vuln, not about exploiting it.

Guido Landi

Vladimir '3APA3A' Dubrovin wrote:
> Dear Guido Landi,
>
> For DoS - yes, you can use existing file, but it's (almost) impossible
> to create reliable code excution exploit since you can not (fully)
> control return address, like required in JMP ESP technique used in this
> exploit.
>
> --Wednesday, September 2, 2009, 12:33:47 PM, you wrote to 3APA3A (at) SECURITY.NNOV (dot) RU [email concealed]:
>
> GL> no, MKDIR is *not* required, also write access is *not* required.
>
> GL> Assuming a directory with a name that starts with "A" exists and that is
> GL> at least 14 chars long, this pattern will trigger the overflow:
>
>
> GL> NLST [Ax206]*/../A*/../A*/../A*/../A*/../A*/../A*/../A*/\r\n
>
>
> GL> At least on win2k3. Therefore, the workarounds for kb975191 on
> GL> microsoft.com are wrong.
>
>
>
> GL> Guido Landi
>
> GL> Vladimir '3APA3A' Dubrovin wrote:
>>> Dear Thierry Zoller,
>>>
>>> I think yes, MKDIR is required. It should be variation of
>>> S99-003/MS02-018. fuzzer should be very smart to create directory and
>>> user both oversized buffer and ../ in NLST - it makes path longer than
>>> MAX_PATH with existing directory.
>>>
>>> --Monday, August 31, 2009, 8:21:12 PM, you wrote to
>>> full-disclosure (at) lists.grok.org (dot) uk [email concealed]:
>>>
>>>
>>> TZ> Confirmed.
>>>
>>> TZ> Ask yourselves why your fuzzers haven't found that one - Combination of
>>> TZ> MKDIR are required before reaching vuln code ?
>>>
>>>
>>>
>>>
>>>
>
> GL> _______________________________________________
> GL> Full-Disclosure - We believe in it.
> GL> Charter: http://lists.grok.org.uk/full-disclosure-charter.html
> GL> Hosted and sponsored by Secunia - http://secunia.com/
>
>

[ reply ]