Re: NMAP Identity obscuring

[Cameron L Palmer <palmer74 () pacbell net>]

I too was thinking about this from a all ports closed situation except ssh.
Setting the number sequence to comply with RFC 1948 (I think that is the
number) has been a long time practice.  But in my experimentation (limited)
I have had success in preventing NMAP from getting the identity of my
firewall.  This is more likely the fact that it doesn't respond to most
things, and SSH only from specific IP addresses. So this tends not to give
NMAP an opportunity to "ask" any telling questions.  But most of my ndd
tuning is more from a performance standpoint.  The identity obscuring is
sort of an after thought.

Here are the parameters I'm currently using on the gateways:

ndd -set /dev/arp arp_cleanup_interval 60000
ndd -set /dev/hme adv_100fdx_cap 1
ndd -set /dev/hme adv_autoneg_cap 0
ndd -set /dev/ip ip_forward_directed_broadcasts 0
ndd -set /dev/ip ip_forward_src_routed 0
ndd -set /dev/ip ip_forwarding 0
ndd -set /dev/ip ip_ignore_redirect 1
ndd -set /dev/ip ip_ire_flush_interval 60000
ndd -set /dev/ip ip_ire_pathmtu_interval 600000
ndd -set /dev/ip ip_respond_to_address_mask_broadcast 0
ndd -set /dev/ip ip_respond_to_echo_broadcast 0
ndd -set /dev/ip ip_respond_to_timestamp 0
ndd -set /dev/ip ip_respond_to_timestamp_broadcast 0
ndd -set /dev/ip ip_send_redirects 0
ndd -set /dev/ip ip_strict_dst_multihoming 1
ndd -set /dev/qfe adv_100fdx_cap 1
ndd -set /dev/qfe adv_autoneg_cap 0
ndd -set /dev/tcp tcp_close_wait_interval 60000
ndd -set /dev/tcp tcp_conn_req_max_q 1024
ndd -set /dev/tcp tcp_conn_req_max_q0 4096
ndd -set /dev/tcp tcp_fin_wait_2_flush_interval  67500
ndd -set /dev/tcp tcp_recv_hiwat 65535
ndd -set /dev/tcp tcp_slow_start_initial 2
ndd -set /dev/tcp tcp_xmit_hiwat 65535

At the end of /etc/system

set tcp:tcp_conn_hash_size=16384

set rlim_fd_max=32768


Any comments?

cameron palmer.
----- Original Message -----
From: "Ofir Arkin" <ofir () itcon-ltd com>
To: <cameron_douglas () email msn com>; <of () securityfocus com>;
<lance () spitzner net>
Cc: <nmap-hackers () insecure org>
Sent: Wednesday, November 22, 2000 04:47
Subject: RE: NMAP Identity obscuring


> Cameron,
>
> I have read the Article in the sysAdmin magazine, unfortunately it is not
> enough to fool me :)
>
> The article, for those who do not subscribed to SysAdmin, deals with
> changing parameters with
> Solaris so NMAP will not detect it. The article suggests changing the PMTU
> policy, and harden the sequence
> numbers sequencing.
>
> But, there are a lot of other wild ideas we can use that will reveal the
> Solaris box.
> Changing the most common identification parameters will not hide the other
> :)
> Sure, it will make the job harder, but not impossible.
>
>
> Ofir Arkin  [ofir () itcon-ltd com]
> Senior Security Analyst
> Chief of Grey Hats
> ITcon, Israel.
> http://www.itcon-ltd.com
>
> Founder
> http://www.sys-security.com
>
> "Opinions expressed do not necessarily
> represent the views of my employer."
>
>
> -----Original Message-----
> From: Cameron Palmer [mailto:cameron_palmer () hotmail com]
> Sent: Sunday, November 05, 2000 2:50 AM
> To: of () securityfocus com; ofir () itcon-ltd com; lance () spitzner net
> Cc: nmap-hackers () insecure org
> Subject: NMAP Identity obscuring
>
>
> I know we have seen the argument before, but the recent SysAdmin magazine
> has an article on Solaris security.  They recommend changing some NDD
> parameters to obscure the identity of Solaris from nmap.  They have some
> interesting points, which is essentially they aren't looking for that as
the
> sole form of protection of the machine but merely make Solaris conform to
> the RFCs instead of having its own quirks that give away too much
> information.  I would normally be dissuaded from security by obscurity
> arguments, but by taking out the things that make the OS unique and
conform
> to RFCs you do raise the ante as it were.  Additionally I've seeen some
> other good OS tuning parameters with NDD that help performance that are a
> good idea, like fixing your Quad card to having multiple MAC addresses
> instead of the single hostid.  Apparently you can gain a 40% speed boost
on
> a Checkpoint firewall. This came from the Checkpoint web site.  They have
a
> number of recommendations for security related changes.
>
> Any thoughts?
>
> cameron.
>
>
> From: Oliver Friedrichs <of () securityfocus com>
> To: Ofir Arkin <ofir () itcon-ltd com>, Lance Spitzner <lance () spitzner net>
> CC: nmap-hackers () insecure org
> Subject: RE: firewalk meets nmap - TTL (tested)
> Date: Sat, 04 Nov 2000 15:36:23 -0800
> MIME-Version: 1.0
> Received: from mta1.snfc21.pbi.net (mta1-pr) by sims1.snfc21.pbi.net (Sun
> Internet Mail Server sims.3.5.2000.01.05.12.18.p9) with ESMTP id
> <0G3I00609XSQ25 () sims1 snfc21 pbi net> for palmer74@sims-ms-daemon; Sat, 4
> Nov 2000 15:41:14 -0800 (PST)
> Received: from amy.insecure.org ([208.184.74.98]) by mta1.snfc21.pbi.net
> (Sun Internet Mail Server sims.3.5.2000.01.05.12.18.p9) with SMTP id
> <0G3I000N7XQ0PL () mta1 snfc21 pbi net> for palmer74 () sims1 snfc21 pbi net;
Sat,
> 04 Nov 2000 15:39:37 -0800 (PST)
> Received: (qmail 20825 invoked by uid 508); Sat, 04 Nov 2000 23:46:19
+0000
> Received: (qmail 20725 invoked from network); Sat, 04 Nov 2000 23:41:28
> +0000
> Return-path: <nmap-hackers-return-887-palmer74=pacbell.net () insecure org>
> Message-id: <10786F3AE30CD4118FAC00A0CC58F9F1015929@MAIL>
> X-Mailer: Internet Mail Service (5.5.2650.21)
> Precedence: bulk
> Delivered-to: mailing list nmap-hackers () insecure org
> Delivered-to: moderator for nmap-hackers () insecure org
> Mailing-List: contact nmap-hackers-help () insecure org; run by ezmlm
>
>  >Lance, we should automate this somehow. This is a cool thing.
>  >But again correct configuration will prevent this from happening.
>
> This is a really neat idea.  It should be easy to automate, if you
> add in some traceroute functionality to nmap to determine the hop
> where packets are being dropped (this would be the firewall), then
> you only need to specify an address on the internal network.  I think
> nmap could use UDP/TCP ACK/ICMP traceroute functionality anyways.
> And while your at it, make it parallel, send out 32 packets with
> incrementing ttl's at the very start.. none of this 1 hop at a time
> slowness.
>
> - Oliver
>
> --------------------------------------------------
> For help using this (nmap-hackers) mailing list, send a blank email to
> nmap-hackers-help () insecure org . List run by ezmlm-idx (www.ezmlm.org).
>
>
> _________________________________________________________________________
> Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.
>
> Share information about yourself, create your own public profile at
> http://profiles.msn.com.
>
>
>
> --------------------------------------------------
> For help using this (nmap-hackers) mailing list, send a blank email to
> nmap-hackers-help () insecure org . List run by ezmlm-idx (www.ezmlm.org).


--------------------------------------------------
For help using this (nmap-hackers) mailing list, send a blank email to 
nmap-hackers-help () insecure org . List run by ezmlm-idx (www.ezmlm.org).