From: Henrik Nordstrom Subject: Re: Handling Port-Unreachable Response to UDP Date: 28 Jul 2003 16:48:43 +0200 Sender: netfilter-devel-admin@lists.netfilter.org Message-ID: <1059403723.12254.10.camel@henrik.marasystems.com> References: <3F252AAE.7000109@variosecure.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Cc: netfilter-devel@lists.netfilter.org Return-path: To: Tim Burress In-Reply-To: <3F252AAE.7000109@variosecure.net> Errors-To: netfilter-devel-admin@lists.netfilter.org List-Help: List-Post: List-Subscribe: , List-Unsubscribe: , List-Archive: List-Id: netfilter-devel.vger.kernel.org m=E5n 2003-07-28 klockan 15.52 skrev Tim Burress: > We've been looking at a problem in which a client sends a UDP packet to= =20 > a server on an unused port. This generates an ICMP port-unreachable=20 > packet (as usual), but we find that this same ICMP packet can then be=20 > replayed back to the client over and over. Apparently it gets through=20 > netfilter because our standard rules allow RELATED traffic. Explain the context of "replayed back to the client over and over". > We're wondering if an appropriate response to this situation would be t= o=20 > destroy the conntrack entry for the UDP connection after receiving the=20 > first port-unreachable? Sounds reasonable to me, but may as you said cause a significant load impact on connntrack if this client floods you... --=20 Henrik Nordstrom MARA Systems AB