Nmap Development mailing list archives

Re: [nmap-svn] r26641 - nmap


From: Fyodor <fyodor () insecure org>
Date: Wed, 5 Oct 2011 19:35:24 -0700

On Fri, Sep 30, 2011 at 11:59:40AM +0200, Luis MartinGarcia. wrote:
On 09/20/2011 01:37 AM, commit-mailer () insecure org wrote:

Since the symbol ":" is not allowed in hostnames or IPv4 addresses,
can't we make Nmap assume "-6" by default when one of the targets
contains ":"? I don't see any case where a user would use ":" in a
target if he doesn't mean IPv6. Even in the remote case that we wanted
to support syntaxes like "insecure.org:443", we could always check for
more than one colon, since IPv6 addresses have a least two (eg. fe80::1).

That is a good idea, but I'm worried that it might cause more user
confusion.  If we could always determine via heuristics what sort of
address to scan, I think it would be worth doing.  But there will
always be common ambiguous cases (e.g. dual stack hostnames like
scanme.nmap.org) where we don't know the user's preferred protocol
unless the user tells us.  So we might as well train them to use -6
when they need IPv6.  And what would we do if the user specified an
IPv6 address and an IPv4 one?  Also, target specifiers given from
standard input ("-iL -") could be a problem.

Also, most other networking tools don't seem to do this sort of
auto-detection.  E.g. we have ping and ping6, traceroute and
traceroute6, etc.  So we should probably be consistent unless we find
a good reason for differing.

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


Current thread: