Nmap Development mailing list archives

Re: Strange errors with nmap 4.68


From: Nathan <nathan.stocks () gmail com>
Date: Thu, 11 Dec 2008 14:44:09 -0700

On Thu, Dec 11, 2008 at 1:10 PM, Brandon Enright <bmenrigh () ucsd edu> wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Thu, 11 Dec 2008 12:37:22 -0700
Nathan <nathan.stocks () gmail com> wrote:

Yes!  iptables is running.  But it's pretty bare.  I'm dropping all
inbound packets destined for ports 1-10,240, with an extra rule to
allow me to SSH in from my office.  But if I'm reading the error
right, it's complaining about a packet from port 57622 on the server
to 36343 on the target, neither of which are in the 1-10240 range
(???)  Here's the output if I run "iptables-save"

# Generated by iptables-save v1.4.0 on Thu Dec 11 12:31:44 2008
*filter
:INPUT ACCEPT [1670029810:498255753315]
:FORWARD ACCEPT [0:0]
:OUTPUT ACCEPT [4416085503:424141701772]
-A INPUT -s [my-office-ip-address] -p tcp -m tcp --dport 22 -j ACCEPT
-A INPUT -i lo -j ACCEPT
-A INPUT -p tcp -m tcp --dport 1:10240 -j DROP
COMMIT
# Completed on Thu Dec 11 12:31:44 2008

~ Nathan


On a completly unrelated side-note, running iptables with connection
tracking (the default) is a recipe for failure with Nmap.  Depending on
your kernel version, you'll want to look at either

"/proc/net/ip_conntrack" or use the conntrack-tools userspace utilities.

Once you've filled up your connection tracking table you'll be dropping
packets like crazy.

For one-off Nmap scans you should be fine, for lots of scanning though
connection tracking *must* be off.

Brandon

That sounds like exactly what's happening!  A quick "cat
/proc/net/ip_conntrack" spits out tons of connection tracking info.
Do I have to turn that off in my kernel config (i.e.
reconfigure/recompile my kernel) or is there a way to simply toggle it
on/off?  I'm currently on kernel 2.6.23.

~ Nathan

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


Current thread: