From mboxrd@z Thu Jan 1 00:00:00 1970 From: Imre Gergely Subject: Re: hashlimit module parameters Date: Tue, 27 Sep 2005 18:06:59 +0300 Message-ID: <43396013.1060201@astral.ro> References: <43329209.8040607@astral.ro> <20050922215104.GU26520@sunbeam.de.gnumonks.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: Harald Welte Return-path: To: Netfilter Development Mailinglist In-Reply-To: <20050922215104.GU26520@sunbeam.de.gnumonks.org> 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 hi how does this hashlimit behave above rate 5000/sec ? and how it is supposed to behave? i tried something like this: iptables -A limit_udp -p udp -m hashlimit --hashlimit 5000/sec --hashlimit-mode dstip --hashlimit-name limit_udp -j RETURN; iptables -A limit_udp -p udp -m limit --limit 1/sec -j LOG --log-prefix='limit_udp: ' i have some dstip to with traffic is around 500pps, but it still appears in the logs, that means the second rule matches, while it shouldn't according to hashlimit 5000/sec. and another thing. if i try with a rate above 5000 (say 6k, 8k) it always puts 10k in. why is that? thx. config: iptables-1.3.2, kernel 2.6.13, dual Intel Xeon CPU 2.40GHz (pls reply in private too.) Harald Welte wrote: > On Thu, Sep 22, 2005 at 02:14:17PM +0300, Imre Gergely wrote: > >>hi >> >>i was wondering about something. i want to use this hashlimit module, >>and the doc says the --hashlimit parameter is just like the one in the >>limit match, but when i want to do >> >>/usr/local/iptables-1.3.3-20050921/sbin/iptables -A INPUT -s 1.1.1.1 -m >>hashlimit --hashlimit ! 5/sec --hashlimit-mode srcip --hashlimit-name >>test -j ACCEPT >> >>it gives me an error: >> >>iptables v1.3.3-20050921: hashlimit does not support invert >>Try `iptables -h' or 'iptables --help' for more information. >> >>is there a way to specify an invert to this ? > > > I don't think so, sorry :( Seems like I forgot to take care of limit > inversion. Patches welcome. >