From owner-freebsd-questions@FreeBSD.ORG Wed Aug 23 13:29:42 2006 Return-Path: X-Original-To: questions@freebsd.org Delivered-To: freebsd-questions@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0BACF16A5B1 for ; Wed, 23 Aug 2006 13:29:42 +0000 (UTC) (envelope-from nalists@scls.lib.wi.us) Received: from mail.scls.lib.wi.us (mail.scls.lib.wi.us [198.150.40.25]) by mx1.FreeBSD.org (Postfix) with ESMTP id AFCEF43D46 for ; Wed, 23 Aug 2006 13:29:41 +0000 (GMT) (envelope-from nalists@scls.lib.wi.us) Received: from [172.26.2.238] ([172.26.2.238]) by mail.scls.lib.wi.us (8.13.1/8.13.1) with ESMTP id k7NDTXoB006330; Wed, 23 Aug 2006 08:29:33 -0500 (CDT) (envelope-from nalists@scls.lib.wi.us) Message-ID: <44EC583D.3010409@scls.lib.wi.us> Date: Wed, 23 Aug 2006 08:29:33 -0500 From: Greg Barniskis User-Agent: Thunderbird 1.5.0.5 (Windows/20060719) MIME-Version: 1.0 To: Perry Hutchison References: <10608230648.AA14609@pluto.rain.com> In-Reply-To: <10608230648.AA14609@pluto.rain.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: gregb@scls.lib.wi.us, questions@freebsd.org Subject: Re: new 6.1 install will not boot X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 23 Aug 2006 13:29:42 -0000 Perry Hutchison wrote: >> Recommend you get a [bigger|second] disk if you can though, >> or housecleaning will be a constant chore. > > I got it more or less working, although not completely set up, and > then that 10GB disk died: click -- kerthunk -- click -- kerthunk > continuously, even after cycling power, even with only the power > connected :( > > After replacing it with a 160GB Hitachi, and reinstalling Windoze, > Linux, and FreeBSD (in that order, as before), I seem to be back at > square one -- FreeBSD won't boot -- but the details are different. > Partition Commander now has: > > Ptn size ----- type ----- 1st sector # of sectors > P0 250M FAT32 0x0B 63 514017 > P1 7M Linux ext2 0x83 514080 16065 > P2 41.99G Unix 0xA5 530145 88068330 > P3 85.75G Extended 0x0F 88598475 179831610 > L0 43.75G FAT32 0x0B 88598538 91763217 > L1 400M Linux swap 0x82 180361818 819252 > L2 41.60G Linux ext2 0x83 181181133 87248952 > > Sysinstall had not commented about the geometry with the 10GB disk, > but it did this time; and as suggested I let it do what it wanted. > The Dell BIOS will not tell me what it thinks the geometry is -- it > just says the drive is EIDE -- so I have no direct way of verifying > sysinstall's geometry; however the first BIOS partition is a > working FAT32 and per the instructions that "should" be enough for > sysinstall to have gotten it right. (The second BIOS partition is > a Linux /boot, which also works.) > > The install appeared to succeed, and the FreeBSD boot manager does > successfully boot Windoze and Linux, but all attempts to boot FreeBSD > from the hard disk fail. > > The following was transcribed by hand, so there might be some typos; > and I've added some notes to the right of the lsdev output. I've > also confirmed, using the loader's "ls", that there is no visible > file named 'kernel' in the root directory, nor anywhere under /boot, > /rescue, or /sbin. Where is it supposed to come from, and how do > I get it where it needs to be without reinstalling the whole thing > *yet again*? > > ==================================================================== > > > F1 DOS > F2 Linux > F3 FreeBSD > > Default: F3 > > BTX loader 1.00 BTX version is 1.01 > Consoles: internal video/keyboard > BIOS drive A: is disk0 > BIOS drive C: is disk1 > BIOS 640kB/195584kB available memory > acpi: bad RSDP checksum (210) > > FreeBSD/i386 bootstrap loader, Revision 1.1 > (root@opus.cse.buffalo.edu, Sun May 7 03:20:03 UTC 2006) > Loading /boot/defaults/loader.conf > Unable to load a kernel! > / > can't load 'kernel' > > Type '?' for a list of commands, 'help' for more detailed help. > OK lsdev > cd devices: > disk devices: > disk0: BIOS drive A: > disk1: BIOS drive C: > disk1s1: FAT32 # C: > disk1s2: ext2fs # Linux /boot > disk1s3a: FFS # FreeBSD / > disk1s3b: swap > disk1s3d: FFS # FreeBSD /var > disk1s3e: FFS # FreeBSD /tmp > disk1s3f: FFS # FreeBSD /usr > disk1s4: Unknown fs: 0xf # contains FAT32 D:, Linux swap and / > pxe devicde: > OK Well, you're at least as far as having the disk sliced up in a workable way, or the bootstrap wouldn't start at all. This jumps out as not only being bad, but happening right before meltdown. > acpi: bad RSDP checksum (210) Have you got the latest Dell BIOS for this hardware? If not you may be SOL if they don't support this hardware any more. I expect the GX1 is well past Dell's official EOL, but they may still have the files downloadable on their support site. It might not help anyway. The alternative to making an old ACPI implementation work right is to try to work around its shortcomings by trying to boot around the problem or reconfigure the underlying system to eliminate the root cause of the conflict. Searching http://www.google.com/bsd for "bad RSDP checksum" turns up that this is a recurring issue on older Dells, along with wildly disparate pokes at causes and solutions including: toggling ACPI support on/off in the BIOS workarounds for funky RAID cards replacing the video card with a different model Ain't low level hardware troubleshooting grand? I expect there are also boot loader command line options you can try to coax the system to start with hardware "as is" but I've never had to resort to that; the Handbook or others on the list are going to be more helpful than I can on that point. -- Greg Barniskis, Computer Systems Integrator South Central Library System (SCLS) Library Interchange Network (LINK) , (608) 266-6348