From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Wed, 20 Jun 2001 05:54:54 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Wed, 20 Jun 2001 05:54:43 -0400 Received: from firewall.sch.bme.hu ([152.66.215.213]:1664 "EHLO singular.sch.bme.hu") by vger.kernel.org with ESMTP id ; Wed, 20 Jun 2001 05:54:40 -0400 Date: Wed, 20 Jun 2001 11:56:36 +0200 (CEST) From: =?ISO-8859-2?Q?Kajt=E1r_Zsolt?= To: linux-kernel@vger.kernel.org cc: rusty@rustcorp.com.au Subject: Re: Iptables ipt_unclean bug? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org > In message > you write: > > > > Hi all! > > > > I think it's possible to hang the kernel useing isic 0.05 > > (www.packetfactory.net/Projects/ISIC/), when there's a unclean match > > in iptables rules. > Thanks for the bug report. I've just done an audit of the unclean > code: patch against 2.4.5 is below. There were some bad thinkos there, > two fatal. 2 infinite loops ;) After patching now everything seems to be ok. But there must be some other bug somewhere in TCP/IP stack or af_packet (or my machine is just crap ;) ), because after ~5 mins I can still hang my kernel on loopback WITHOUT iptables... Will check which one is on fault on my home network... -soci-