Re: RFC: Nping requirements and user interface

[Fyodor <fyodor () insecure org>]

On Fri, May 22, 2009 at 10:36:33AM +0300, Toni Ruottu wrote:
> > o Should be very useful and functional just for normal ping purposes
> >   (whether the user requests ICMP echo request, TCP connect() ping,
> >   UDP, ARP ping, etc.). This means providing useful stats, common
> >   ping options, etc. 
>
> What should nping do when it is given nothing, but an IP address? What
> are the default flags/options going to be? I guess the defaults should
> be somewhat alike to what regular ping does, but maybe not too similar.

Good question.  My instinct is that it should be very similar to the
normal "ping" utility: send ICMP echo request packets at 1 second
intervals, until the user quits with ^C (then print stats before
exiting).  That way we start with known, expected behavior.  And then
all of nping's great options will be available to modify this
behavior.

But if it turns out that people's desired usage is a lot different
than traditional ping, we aren't locked into that default.  For
example, maybe we would want to just do 1 probe by default and then
quit if we get a response or timeout.  Or maybe the default could be
5.  Or we could even default with a different type of probe.  For
example, David has just done some empirical tests to determine which
single probe type is most effective.  But we should only vary from the
"ping" default where we have strong justifications for doing so.

Cheers,
-F

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