Re: NSE Script http-methods.nse

[Vlatko Kosturjak <kost () linux hr>]

On 02/19/2010 06:47 AM, David Fifield wrote:
> On Mon, Feb 01, 2010 at 12:25:37PM -0700, David Fifield wrote:
> > On Sat, Jan 30, 2010 at 09:11:01PM +0100, Bernd Stroessenreuther wrote:
> > > > Fyodor had some ideas for enhancement of the script, such as making it
> > > > default and only printing "interesting" methods in non-verbose mode.
> > >
> > > Even if I'm not sure why some methods should be more interesting then others, 
> > > I'm prying. You surely know much better then I, what nmap users find 
> > > interesting or useful...  
> The uninteresting set {GET, HEAD, POST, OPTIONS, TRACE} just comes from
> a quick observation of a handful of web servers. I welcome suggestions
> of methods to be removed from or added to the set.

Not suggestion of methods, but some time ago, I've written a quick and 
straightforward paper (2 pages long) about HTTP methods enumeration 
(known and unknown), so maybe it can help you for ideas:
http://kost.com.hr/dl/enumerating-http-methods.pdf

It might be old, but I think it still have all tricks to get the HTTP 
methods available/enabled. So, maybe you want to take a look for 
inspiration.

There is also some Perl code as PoC:
http://kost.com.hr/dl/http-get-methods-1.1.tar.gz

Some of the tricks can be used for HTTP fingerprinting as well. I think 
HTTP fingerprinting would perfectly fit into Nmap set of scripts, but 
with all known HTTP server enumeration tricks (analyze additional 
response headers returned by HTTP server to different requests, analyze 
order of HTTP headers, etc). Maybe entry job for some GSOC student?

Kost
_______________________________________________
Sent through the nmap-dev mailing list
http://cgi.insecure.org/mailman/listinfo/nmap-dev
Archived at http://seclists.org/nmap-dev/