From mboxrd@z Thu Jan 1 00:00:00 1970 From: Patrick McHardy Subject: Re: traffic shaping and fixed per-packet overhead Date: Thu, 21 Jul 2005 20:28:28 +0200 Message-ID: <42DFE94C.4050307@trash.net> References: Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit Cc: netfilter-devel@lists.netfilter.org Return-path: To: Matthias Urlichs In-Reply-To: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: netfilter-devel-bounces@lists.netfilter.org Errors-To: netfilter-devel-bounces@lists.netfilter.org List-Id: netfilter-devel.vger.kernel.org Matthias Urlichs schrieb: > Hi, > > I have an IPIP tunnel through a DSL line (PPPoE) which I need to > traffic-shape / flow control. Unfortunately, the tunnel has a fixed > per-packet overhead, which means that small packets overrun the > bandwidth allowed for the tunnel and get dropped, while (if I set the > allowed bandwidth to too-small values) the wire will be idle when > I send large packets. > > (I'm using HTB.) > > I can think of a couple of solutions ... > - Use bit-banging in the outer filter instead of nice rules on the tunnel. > (OK, that's a dumb idea from a useability POV.) > - Extend the HTB code to allow specifying a per-packet overhead. > - Hack the kernel to propagate a packet's netfilter MARK. Totally unrelated to netfilter-devel, but: # tc qdisc add htb help ... mpu minimum packet size used in rate computations overhead per-packet size overhead used in rate computations