From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S965074AbVKBPMY (ORCPT ); Wed, 2 Nov 2005 10:12:24 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S965076AbVKBPMY (ORCPT ); Wed, 2 Nov 2005 10:12:24 -0500 Received: from mx2.mail.elte.hu ([157.181.151.9]:36259 "EHLO mx2.mail.elte.hu") by vger.kernel.org with ESMTP id S965074AbVKBPMX (ORCPT ); Wed, 2 Nov 2005 10:12:23 -0500 Date: Wed, 2 Nov 2005 16:12:42 +0100 From: Ingo Molnar To: =?utf-8?B?UGF3ZcWC?= Sikora Cc: linux kernel mailing list , Rusty Russell , netfilter-devel@lists.netfilter.org Subject: Re: [2.6.14-rt1] slowdown / oops. Message-ID: <20051102151242.GA23809@elte.hu> References: <200511021420.28104.pluto@agmk.net> <20051102134723.GB13468@elte.hu> <20051102135516.GA16175@elte.hu> <20051102140025.GA17385@elte.hu> <20051102142533.GA18453@elte.hu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20051102142533.GA18453@elte.hu> User-Agent: Mutt/1.4.2.1i X-ELTE-SpamScore: 0.0 X-ELTE-SpamLevel: X-ELTE-SpamCheck: no X-ELTE-SpamVersion: ELTE 2.0 X-ELTE-SpamCheck-Details: score=0.0 required=5.9 tests=AWL autolearn=disabled SpamAssassin version=3.0.4 0.0 AWL AWL: From: address is in the auto white-list X-ELTE-VirusStatus: clean Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org more updates: NETFILTER_DEBUG catches the situation too - the problem seems to be wrong reference counts on the skb. I've also added a few printouts of ct fields: NF_IP_ASSERT: include/linux/netfilter_ipv4/ip_conntrack.h:329(ip_conntrack_put) NF_IP_ASSERT: net/ipv4/netfilter/ip_conntrack_core.c:346(destroy_conntrack) NF_IP_ASSERT: net/ipv4/netfilter/ip_conntrack_core.c:346(destroy_conntrack) BUG: NAT ct f0d6cf10 still hashed! .. list empty ORIG: 0, REPLY: 0 .. expecting: 0 .. timeout pending: 1 .. ct_general: 0 [f0d6cf10] .. nfct: f0d6cf10 [] dump_stack+0x13/0x20 (12) [] check_hashes+0x13b/0x150 (48) [] destroy_conntrack+0xd8/0x1f0 (32) [] ip_local_deliver+0x182/0x270 (40) [] ip_rcv+0x2f5/0x4e0 (64) [] netif_receive_skb+0x15d/0x1e0 (52) [] rtl8139_rx+0x1b7/0x340 (80) [] rtl8139_poll+0x58/0x110 (40) [] net_rx_action+0x72/0x140 (24) [] ksoftirqd+0xb9/0x140 (40) [] kthread+0x94/0xa0 (28) [] kernel_thread_helper+0x5/0xc (138035228) i.e. we have ct->timeout still pending but the reference count is 0? That cannot be right. Ingo