Re: using --host-timeout with multiple hosts

[Ryan Giobbi <ryan () tgbemail com>]

Thanks

I think --host-timeout is doing exactly what it should but not what the
user might expect for smaller values.

If someone does a scan that takes 5 minutes on one host and then runs a
scan on 90 hosts they might think that 10 minutes is a reasonable timeout
since its twice what their initial scan took. The scan probably won't work
because when scanning 90 hosts it looks like nmap will spread out the scan
time and the hosts will start timing out, even though nmap will finish 90
hosts much faster than doing them one at a time.

For larger host timeout values, it shouldn't be a problem since the other
hosts will be done and that's probably a better use of --host-timeout
anyway.














On Tue, Nov 6, 2012 at 12:00 PM, David Fifield <david () bamsoftware com>wrote:

> On Tue, Nov 06, 2012 at 05:51:21AM -0500, Ryan Giobbi wrote:
> > When I use
> > nmap -sT <IP>* *--host-timeout 2m
> > the scan works and finishes
> >
> > However, when I scan more IPs
> > nmap -sT -iL <IP_list>* *--host-timeout 2m
> > the scan times out (for the same IP) even though it was run from the same
> > host as the first.
> >
> > Is this because nmap takes a longer time to scan some hosts when multiple
> > addresses are scanned?
>
> That's probably right. If scanning one host takes 1:30, and scanning two
> takes 2:10 (assuming they finish at the same time), then both will time
> out with a 2m timeout. We should perhaps prorate the passage of time
> according to the number of hosts being scanned, but that is different
> than what --host-timeout means now.
>
> David
_______________________________________________
Sent through the nmap-dev mailing list
http://cgi.insecure.org/mailman/listinfo/nmap-dev
Archived at http://seclists.org/nmap-dev/