--reason and --packet-trace with ipv6

[Paul Jenkins]

Stop me if this is the wrong place to post these but in addition to the
DNS error and -sP times with ipv6, this morning I noticed something
else.

Command line

Nmap -6 -n -v --reason -sP -iL c:\ipv6.txt -oN c:\blahblahblah.txt

Nmap -6 -n -v --reason --packet-trace -sP -iL c:\ipv6.txt -oN
c:\blahblahblah.txt

The first set of options run with out a hitch on CentOS 5.3, WinXP, and
Win 2003, in short no issues all 20 hosts nmap reports as up.

The second set of options runs fine on the Linux box, reporting all 20
hosts up, however on the 2 windows boxes I receive a bunch of "CONN
(0.31020s) TCP localhost > [target ipv6 address] => Unknown error" and
then reports only 3 hosts up.

So all things equal adding --packet-trace seems to break the reporting
that all 20 hosts are indeed up, Wireshark shows identical packet
conversations. It appears that when the remote target refuses a
connection under the --packet-trace option it reports the host as down.

Thoughts?

Thanks,
-Paul

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

[David Fifield]

On Wed, Sep 30, 2009 at 08:50:20AM -0400, Paul Jenkins wrote:

> Stop me if this is the wrong place to post these but in addition to the
> DNS error and -sP times with ipv6, this morning I noticed something
> else.
>
> Command line
>
> Nmap -6 -n -v --reason -sP -iL c:\ipv6.txt -oN c:\blahblahblah.txt
>
> Nmap -6 -n -v --reason --packet-trace -sP -iL c:\ipv6.txt -oN
> c:\blahblahblah.txt
>
> The first set of options run with out a hitch on CentOS 5.3, WinXP, and
> Win 2003, in short no issues all 20 hosts nmap reports as up.
>
> The second set of options runs fine on the Linux box, reporting all 20
> hosts up, however on the 2 windows boxes I receive a bunch of "CONN
> (0.31020s) TCP localhost > [target ipv6 address] => Unknown error" and
> then reports only 3 hosts up.
>
> So all things equal adding --packet-trace seems to break the reporting
> that all 20 hosts are indeed up, Wireshark shows identical packet
> conversations. It appears that when the remote target refuses a
> connection under the --packet-trace option it reports the host as down.

That is a very strange error. Does it happen every time that you add
--packet-trace? If so, please send me Wireshark captures and -d3 logs of
running with and without --packet-trace.

Does the same thing happen with a -sT scan with IPv4 targets?

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