Re: [Full-disclosure] Fwd: Google vulnerabilities with PoC

[Krzysztof Kotowicz <kkotowicz+fd@xxxxxxxxx>]

2014-03-14 20:28 GMT+01:00 Nicholas Lemonias. <lem.nikolas@xxxxxxxxxxxxxx>:

> Then that also means that firewalls and IPS systems are worthless. Why
> spend so much time protecting the network layers if a user can send any
> file of choice to a remote network through http...
>

No, they are not worthless per se, but of course for an user content
publishing service they need to allow file upload over HTTP/s. How far
those files are inspected and later processed is another question - and
that could lead to a vulnerability that you DIDN'T demonstrate.

You just uploaded a .sh file. There's no harm in that as nowhere did you
prove that that file is being executed. Similarly (and that has been
pointed out in this thread) you could upload a PHP-GIF polyglot file to a
J2EE application - no vulnerability in this. Prove something by overwriting
a crucial file, tricking other user's browser to execute the file as HTML
from an interesting domain (XSS), popping a shell, triggering XXE when the
file is processed as XML, anything. Then that is a vulnerability. So far -
sorry, it is not, and you've been told it repeatedly.


As for the uploaded files being persistent, there is evidence of that.  For
> instance a remote admin could be tricked to execute some of the uploaded
> files (Social Engineering).
>

Come on, seriously? Social Engineering can make him download this file from
pastebin just as well. That's a real stretch.

IMHO it is not a security issue. You're uploading a file to some kind of
processing queue that does not validate a file type, but nevertheless only
processes those files as video - there is NO reason to suspect otherwise,
and I'd like to be proven wrong here. Proven as in PoC.

_______________________________________________
Full-Disclosure - We believe in it.
Charter: http://lists.grok.org.uk/full-disclosure-charter.html
Hosted and sponsored by Secunia - http://secunia.com/