From mboxrd@z Thu Jan 1 00:00:00 1970 From: Esteban Subject: Re: [NEW TARGET] MARKCB connection marking balancing Date: 03 Mar 2003 12:31:11 -0300 Sender: netfilter-devel-admin@lists.netfilter.org Message-ID: <1046705471.2728.150.camel@debian> References: <3E62AF18.8010000@zobniow.priv.pl> <200303031653.47644.fabrice@netfilter.org> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Cc: netfilter-devel@lists.netfilter.org Return-path: To: fabrice@netfilter.org In-Reply-To: <200303031653.47644.fabrice@netfilter.org> 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 what is that nth? i couldnt find in iptables man page. where can i find more information? thanks, ps: is there any diference between using -j SNAT --to-source and -j MASQUERADE? ive got static ip address on my internet connection? cause ive got my internal network masqueraded and i try to use fwmarks in preroutinng to route to another gw i have (iproute2)..and it works (packets goes from internal to destination over the gw i choose) but they dont come back because nat is not applyed and ofcourse, internal packets are not routeable packets (10.0.0.0/24)..any idea? that would be a smooth solution for balance traffic. On Mon, 2003-03-03 at 05:53, Fabrice MARIE wrote: > > Dzien Dobre Maciek, > > On Monday 03 March 2003 09:25, Maciek Zobniow wrote: > > Hi all! > > This is my first message to this list and I want to annouce a first > > version of my new netfilter module for connection tracking balancing. > > First maybe I will try to describe what this target should do: > > I made this as a solution for balancing connections from one LAN, > > betewen a few providers links, but I think that is possible to find > > another good usages for this target. > > Idea is quite simply: each connection (yes, connection not packet) which > > arrived to this target is marked. It recive one mark which is use for > > marking all packets from this connection. Marks are sharing by choosen > > algorithm (for now I implemented RR, WRR and special one- basing on > > amout of data for each mark for last 1000 packets). > > [...] > > Just out of curiosity, how different is that from using the nth match > to do RR SNATting ? > > [ > i.e, Taken from the example section of the patch: > > iptables -t nat -A POSTROUTING -o eth0 -m nth --counter 7 --every 3 --packet 0 -j SNAT --to-source 10.0.0.5 > iptables -t nat -A POSTROUTING -o eth0 -m nth --counter 7 --every 3 --packet 1 -j SNAT --to-source 10.0.0.6 > iptables -t nat -A POSTROUTING -o eth0 -m nth --counter 7 --every 3 --packet 2 -j SNAT --to-source 10.0.0.7 > > This example evenly splits connections between the three SNAT addresses. > ] > > Using nth, we use the fact that only the first packet of connections need to be SNATed. > This extension to the nth original patch was made by Richard. I haven't tried it myself yet though > as I have only one ISP :-) > > Have a nice day, > > Fabrice. > -- > Fabrice MARIE > > "Silly hacker, root is for administrators" > -Unknown > >