Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1750974AbVJBF2Z (ORCPT ); Sun, 2 Oct 2005 01:28:25 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1750975AbVJBF2Z (ORCPT ); Sun, 2 Oct 2005 01:28:25 -0400 Received: from mail.collax.com ([213.164.67.137]:28815 "EHLO localhost.localdomain") by vger.kernel.org with ESMTP id S1750973AbVJBF2Y (ORCPT ); Sun, 2 Oct 2005 01:28:24 -0400 Message-ID: <433F6FF3.1040706@trash.net> Date: Sun, 02 Oct 2005 07:28:19 +0200 From: Patrick McHardy User-Agent: Debian Thunderbird 1.0.6 (X11/20050802) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Radoslaw Szkodzinski CC: linux-kernel@vger.kernel.org, Netfilter Development Mailinglist Subject: Re: 2.6.13-rc2+ - problem with DHCP References: <433EBBEC.4050203@gorzow.mm.pl> <433ECE42.2070400@trash.net> <433F0228.6000304@gorzow.mm.pl> In-Reply-To: <433F0228.6000304@gorzow.mm.pl> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1515 Lines: 38 Radoslaw Szkodzinski wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Patrick McHardy wrote: > | > | Are you sure? The patch was supposed to fix problems with DHCP clients > | using regular UDP sockets for sending DHCP requests. Which client are > | you using? > | > > udhcpcd, version 0.9.9-pre (Gentoo ebuild > net-misc/udhcp-0.9.9_pre20041216-r1, no crazy optimisations, stock init > script, IP release disabled) I can't reproduce the problem, it reliably works for me using pump, dhclient or udhcpc. One thing I've noticed is that udhcpc 0.98 doesn't set the interface up itself and fails if it is down. Please make sure that it is up in your tests. > 2.6.13, 2.6.14-rc1 (up to the patch) both work fine. > 2.6.14-rc2 and 2.6.14-rc3 do not. (they can't discover IP address) > The window is between that commit and rc2. > (about 180 changesets) > > I only suspect that patch, it could be something else but I highly doubt > it. I'll check the current kernel with the patch backed out when I have > to restart. Thanks. You mentioned you're setting up your ruleset after DHCP, which means the patch can't be responsible because the codepath is never taken for DHCP queries, so you probably need to do a binary search over the remaining 180 changesets. - 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/