Re: Get Outlook Web Access URL - revised and improved

[Dan <cdcdanielctin () gmail com>]

Hello David,


> So far this is my understanding: Outlook Web Access is a webmail
> application. But if someone is already configured to use Microsoft
> Exchange, they don't necessarily know the OWA URL to do webmail or
> configure push email.

Outlook Web Access is a webmail service of Microsoft Exchange Server
and indeed the users don't necessarily know the OWA URL.
Because I did not found any other program that retrieves the URL and
because nmap has the ability to detect the SSL encryption protocol and
then launch an encrypted session and automatically parallelizes
network operations, I've decided it would be a very good candidate for
implementing my script.

> The OWA application is commonly found at an MX
> domain plus "/exchange" or "/owa".

It is not at all uncommon for a exchange admin to configure a hostname
for the OWA that is not easily found through normal DNS queries in
order to limit possible attacks.

Although the url is a trivial information, you can't access the
webmail service or configure a push service without it.

> I still don't understand what the script is good for. Can you give a an
> example, start to finish, of someone having a problem, running the
> script, and solving the problem? Include output of the script (you can
> replace domains with "example.com" or whatever).

An example of usage is when a user with user () example com tries to
configure a push email service but the input url server isn't
validated. He tries to guess the url using the mx but the MX query
returns no hostname (the domain has no MX record). In this case I
believe this script will help in finding the OWA URL and using a
simple interface for nmap, the script can be used as a validation
tool.

> Is this situation likely to happen to many people?

Although the script was good practice, I did developed it to be used
and it is used by my and my colleagues.

> > -- function to perform web crawling using google's search engine
> > function search(spath)
> >  local request = http.get( 'www.google.com', '80', spath )
> >   for w in string.gfind(request.body, "\<span%s+class\=\"c\"\>%s?(([%w]*)([\.%w-]*)([\.%w]*))") do
> >     if not (exist(raw_subdomains, w)) then
> >         table.insert(raw_subdomains, w)
> >     end
> >   end
> > end
>
> Please remove this feature of searching Google. These kinds of automatic
> queries can be problematic, and scraping HTML is fragile. Also, for
> privacy reasons it's not good to send to Google the address of hosts
> you're scanning.

I've removed the function and all callbacks to it

> Something like this would be better done in a dedicated script for
> finding alternate domain names. (Maybe something like hostmap.nse.) Your
> OWA script could depend on such a dedicated script.

Tried hostmap.nse but it retrieves hostnames that resolve to the
target's IP address and it isn't useful to me because an Exchange
Server can have a different IP address then the domain name.

> > -- in a 301 response case (Moved Permanently)
> > -- returns the link refered in the body
> > function ref_link(body)
> >   for link in string.gfind(body, "<a [hH][rR][eE][fF]=\"(.*)\"") do
> >    return link
> >   end
> > end
>
> Use response.header.location instead. See http-title.nse for an example.

Done.


Thank you for your feedback.

Daniel.

Attachment: get-owa.nse
Description:

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