Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S263322AbTEVVPA (ORCPT ); Thu, 22 May 2003 17:15:00 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S263332AbTEVVPA (ORCPT ); Thu, 22 May 2003 17:15:00 -0400 Received: from tux.rsn.bth.se ([194.47.143.135]:62654 "EHLO tux.rsn.bth.se") by vger.kernel.org with ESMTP id S263322AbTEVVO6 (ORCPT ); Thu, 22 May 2003 17:14:58 -0400 Subject: Re: kernel panic on heavily loaded 2.4.20 router From: Martin Josefsson To: Frank Dekervel Cc: linux-kernel@vger.kernel.org In-Reply-To: <200305222147.39687.kervel-lkml@drie.kotnet.org> References: <200305222147.39687.kervel-lkml@drie.kotnet.org> Content-Type: text/plain Content-Transfer-Encoding: 7bit Organization: Message-Id: <1053638881.9483.75.camel@tux.rsn.bth.se> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.2.4 Date: 22 May 2003 23:28:01 +0200 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1719 Lines: 40 On Thu, 2003-05-22 at 21:47, Frank Dekervel wrote: > hello, > > a linux-2.4.20 (vanilla) machine we use as a router/iptables firewall > crashes once a month in what seems to be the IP stack > (see the lspci/lsmod/ksymoops output below). Does somebody know > what the cause could be ? > Module Size Used by Not tainted > ipt_limit 960 19 (autoclean) > ip_conntrack_ftp 3776 0 (unused) > ipt_state 608 4 (autoclean) > ipt_LOG 3232 13 (autoclean) > iptable_nat 14452 1 (autoclean) > ip_conntrack 16716 3 (autoclean) [ip_conntrack_ftp ipt_state iptable_nat] > iptable_filter 1728 1 (autoclean) > ip_tables 10624 7 [ipt_limit ipt_state ipt_LOG iptable_nat iptable_filter] > serial 42336 0 (autoclean) > ne2k-pci 4832 2 > 8390 5952 0 [ne2k-pci] > via-rhine 11912 1 > mii 2288 0 [via-rhine] > 3c59x 24712 1 Kernel 2.4.20 contains a bug in conntrack (memory-corruption) that can be triggered when using conntrack helpers. It's fixed in 2.4.21-rc2 so I suggest you try that. The crashes you see might be the result of this bug, or maybe not. I suggest you send this report to linux-net@vger.kernel.org and ask if it looks like a real bug or if it's memory-corruption. But you really should upgrade your kernel. -- /Martin - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/