From mboxrd@z Thu Jan 1 00:00:00 1970 From: "David S. Miller" Subject: Re: [PATCH 2.6] optimization of ip_conntrack_proto_tcp:tcp_packet() Date: Tue, 30 Mar 2004 10:45:32 -0800 Sender: netfilter-devel-admin@lists.netfilter.org Message-ID: <20040330104532.6b5cba46.davem@redhat.com> References: <20040329103348.GC1528@sunbeam.de.gnumonks.org> <20040329201722.7a22cf7c.davem@redhat.com> <40694F1F.9070409@eurodev.net> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: netfilter-devel@lists.netfilter.org Return-path: To: Pablo Neira In-Reply-To: <40694F1F.9070409@eurodev.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 On Tue, 30 Mar 2004 12:42:39 +0200 Pablo Neira wrote: > as current tcp tracking doesn't perform so strong tcp transitions > checking, there won't be any problems. Joszef Kadlecsik is working on a > full featured tcp tracking system which will take care of this stuff, > but it still needs a bit of time. ... > Actually, I think that this change won't modify the current behaviour of > the tcp connection tracking system because the conntrack will be drop > sooner or later if a rst is received. You're right, in fact I keep getting reminded every few weeks about the lack of sequence number checking in conntrack so I should have known this :-) Thanks for the clarification.