Re: Ncat proposal: new environment variables

[David Fifield <david () bamsoftware com>]

On Sat, Jul 20, 2013 at 02:19:03AM +0200, Jacek Wielemborek wrote:
> I had a meeting with David this Tuesday and was assigned to implement
> the proposal from my previous post, without ncatin/ncatout FD numbers
> yet (I will probably be adding them next week). So, here's a list of
> environment variables that I introduced to the
> --exec/--sh-exec/--lua-exec child processes (along with a quote from
> the docs written by me):
>
>  * NCAT_REMOTE_ADDR - the remote address - in connect mode, it's the
> one we're connecting to, in listen mode - it's the one that connected
> to us,
>  * NCAT_REMOTE_PORT - the remote port number, where "remote" means the
> same as above,
>  * NCAT_LOCAL_ADDR - our local address used for
> establishing/receiving the connection,
>  * NCAT_LOCAL_PORT - our local port number,
>  * NCAT_PROTO - the protocol name - TCP, UDP or SCTP,

For some background on these choices for variable names, we took some
guidance from existing inetd programs
(https://www.gnu.org/software/inetutils/manual/html_node/Inetd-Environment.html)
and ucspi-tcp (http://cr.yp.to/ucspi-tcp/environment.html).

These programs set variable names like these:
 * TCPREMOTEIP
 * TCPREMOTEPORT
 * TCPLOCALIP
 * TCPLOCALPORT
 * PROTO

We chose to use different variable names. The main reason for this is
that we wanted the names to remain the same no matter the protocol being
used--so you can use the same script without having to change
TCPREMOTEIP to UDPREMOTEIP or SCTPREMOTEIP.

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