Re: What would we want in a new Netcat/Hping?

[Andreia Gaita <shana.ufie () gmail com>]

Well I submitted to do the ncat rewrite... but didn't come even close
to the ideas you have for it, guess I don't stand much of a chance eh
:p
Anyway...

> Command line interface. I'm not convinced that a GUI is necessary, but you 
> can always write one if you want. 

It was never the purpose of netcat to be used as a gui... I've been
finding the weirdest uses for it all over the net since SoC started,
and it's all command line. :)

- Must support all Netcat 1.10 options, except for port scanning. 
Yeah, port scanning goes against the purpose of the tool, I think :p

> Connection redirection 
> SSL support 
> Inetd-like Ability 
> Ability to limit incoming connections 
> Simple HTTP and/or SOCKS proxy support. 
> Connection brokering. 

Methinks that these ideas are just awesome, and that some or all of
these features are over the purpose of the tool, and should be done
outside the core tool. Ncat's core should have the purpose to connect
to somewhere and shuffle data around. We want to put in as much
functionality in it as possible, of course, but keep it small and
simple at the same time, so where to draw the line? For me, anything
that handles data specifically should be done outside in a
filter/plugin, so that ncat doesn't stray from the purpose of
shuffling meaningless (to it) data around.
So multiple connection hability is cool to put in the tool, but data
handling, such as redirecting it to have connection brokering,
application spawning or whatever done to the data goes outside in a
plugin. Data encryption can be done on the side, so SSL should be done
on the side too?

Of course, I'm a sucker for abstraction, so I may be going over the
top with this, so what do you guys think?

> Must support and be tested under Linux, Windows, and one of the following: 
> Mac OS X, Free-, Net-, or Open-BSD, and Solaris. There should be no barrier 
> towards supporting most of the other systems, though no need to test on all of 
> those initially. That is what users are for :).

BTW, I've been trying to compile nmap on debian and ubuntu, and I
always get a configure error when compiling nbase... anyone get that?

shana
gone crazy, back soon, leave message


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