From mboxrd@z Thu Jan 1 00:00:00 1970 From: Joel Newkirk Subject: Re: Problem with DNAT Date: Sun, 10 Nov 2002 22:38:58 -0500 Sender: netfilter-admin@lists.netfilter.org Message-ID: <200211101421.20026.netfilter@newkirk.us> References: <200211091837.00935.ss310636@inf.tu-dresden.de> <200211091518.22273.netfilter@newkirk.us> <200211092136.46643.ss310636@inf.tu-dresden.de> Reply-To: netfilter@newkirk.us Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: <200211092136.46643.ss310636@inf.tu-dresden.de> Errors-To: netfilter-admin@lists.netfilter.org List-Help: List-Post: List-Subscribe: , List-Id: List-Unsubscribe: , List-Archive: Content-Type: text/plain; charset="us-ascii" To: Steffen Schoenwiese , netfilter@lists.netfilter.org On Saturday 09 November 2002 03:36 pm, Steffen Schoenwiese wrote: > > > I have some problems with iptables 1.2.7a. Some of my rules which > > > worked well with iptables 1.2.5 produce an error message "iptables: > > > Invalid argument". The specific rules look like this: > > > > > > iptables -t nat -A OUTPUT -p tcp -d -j DNAT --to : > > > > > > I also tried to use --to-destination instead of --to, same error > > > message. All my other rules work perfect, only the DNAT ones get th= is > > > error. Does anyone have a possible solution? > > > > All nat table rules, or are SNAT and MASQ accepted? Do you have a in= smod > > or modprobe iptable_nat? I don't know your prior/current setup, but = it > > might have been compiled in before, a module now. > > SNAT doesn't work either, MASQ works. iptable_nat is loaded. The system= I'm > running is a new SuSE 8.1 if that helps. The same rule is accepted if it's prerouting? What about without the port= redirection? SUSE8.1 is kernel 2.4.19, right? This shouldn't apply then but... =46rom the "Known Bugs" file in the 1.2.7a distribution: 1) NAT in the OUTPUT chain only works since kernel 2.4.18. However, there is a patch for previous kernels in patch-o-matic, called the 'local-nat.patch'. This patch adds a CONFIG_NF_IP_NAT_LOCAL kernel co= nfig option. I'm still running 1.2.5, so I can't even try this locally right now. Sor= ry. j