From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261929AbTLUBBP (ORCPT ); Sat, 20 Dec 2003 20:01:15 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261936AbTLUBBO (ORCPT ); Sat, 20 Dec 2003 20:01:14 -0500 Received: from 12-211-67-128.client.attbi.com ([12.211.67.128]:411 "EHLO waltsathlon.localhost.net") by vger.kernel.org with ESMTP id S261929AbTLUBBJ (ORCPT ); Sat, 20 Dec 2003 20:01:09 -0500 Message-ID: <3FE4F0D3.2020904@comcast.net> Date: Sat, 20 Dec 2003 17:01:07 -0800 From: Walt H User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.5) Gecko/20031121 X-Accept-Language: en-us MIME-Version: 1.0 To: Nicklas Bondesson Cc: linux-kernel@vger.kernel.org Subject: Re: Error mounting root fs on 72:01 using Promise FastTrak TX2000 (PDC20271) References: In-Reply-To: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Nicklas Bondesson wrote: > Hi, > > Oh I forgot to say that i'm running RAID1 and it detects both drives > perfectly (I'm passing the IDE addresses to the kernel at boot time from the > lilo conf, see previous post). The system was reinstalled yesterday with two > brand new 80GB Western Digital disks (WD800JB-00DUA3). The thing is, I have > succesfully installed the TX2000 card with the native ATARAID drivers before > using two 30GB Maxtor (something) disks and kernel 2.4-20 - 2.4.23. I wonder > why I can't get it up and running now. It will only work with the pre > compiled kernel shipped with Debian 3.0 (2.4.18-bf2.4). I have tried all > sorts of kernel settings. Since I have got it to work before I think should > be able to do it again. > > Please advise. > > /Nicke Hmmm. I'm pretty sure that the partitions are enumerated the same way in 2.4.23 vs. 2.4.18. There was a change in the way IDE geometry was returned from the kernel that caused a hiccup in the pdcraid driver around 2.4.22 I think? My patch just tells the pdcraid to use an alternate method of finding the RAID superblock on each drive. Not sure what else might be the problem. Do you see the ataraid driver fire up (looking something like this): ataraid/d0: ataraid/d0p1 Drive 0 is 239372 Mb (33 / 0) Drive 1 is 239372 Mb (34 / 0) Raid1 array consists of 2 drives. Promise Fasttrak(tm) Softwareraid driver for linux version 0.03beta The only problem I've recently come upon with pdcraid, is when it detects just one of the drives and fails. I didn't think that should happen with raid1 though.