Re: Why does nmap send multiple probes to the same port?

[Andreas Ericsson <ae () op5 se>]

Casey Williams wrote:
> On 04:37 Thu 01/12/06 Jan    , Andreas Ericsson wrote:
>
> > Casey Williams wrote:
> >
> > > ***pure conjecture ahead***
> > >
> > > So I started wondering how nmap was able to be so reliable and yet SO fast 
> > > (kudos!).  Then I realized that it may be sending these "retry" probes to 
> > > hosts that it doesn't know the status of and that it didn't get a reply 
> > > from.
> > >
> > > I've been reading the nmap source and stepping through it, but I haven't 
> > > verified that this is correct.  Am I on the right track?  If not, would 
> > > this tactic be unrecommended?
> >
> > You're missing the point a bit. The error messages means that nmap 
> > didn't send those probes. It tried to, but the system told it there was 
> > already a connection attempt in progress.
>
>
> I'm slightly off topic since I'm talking about SYN scans instead of connect() scans, and I'm not 
> certain how your reply would apply to these.  (If it does apply to SYN scans, I apologize, I'm fairly 
> new to this level of TCP/IP programming and I suppose I need to read more :))

It doesn't. I thought you were referring to the error messages shown 
above. SYN scan requires root privileges and utilizes raw sockets or raw 
ethernet frames. Those packets are sent anyway, so long as there's no 
error on the socket level.

> In my case, when I try "nmap -sS -P0..." and I sniff the traffic that gets generated from that scan, 


Then you see what's actually sent. The original poster just read the 
debug messages output by nmap.


> I've noticed more than one probe gets sent to the same port on some of the hosts under certain 
> circumstances.  I shouldn't see these "extra" probes in the packet capture if NMap didn't actually 
> send them should I?


Nopes, that's true. Or rather, you see them as soon as the kernel has 
made them available to the pcap interface. There's no guarantee that the 
packets are actually being sent over the wire (although the kernel tries 
to and in 99.999999% of the cases they are). The ethernet specification 
is a bit loose.


>  Is this the expected behavior, or am I imagining things? :)

It's expected.

-- 
Andreas Ericsson                   andreas.ericsson () op5 se
OP5 AB                             www.op5.se
Tel: +46 8-230225                  Fax: +46 8-230231


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