From owner-freebsd-smp@FreeBSD.ORG Thu Aug 21 03:33:27 2003 Return-Path: Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id ACA8316A4BF; Thu, 21 Aug 2003 03:33:27 -0700 (PDT) Received: from klima.physik.uni-mainz.de (klima.Physik.Uni-Mainz.DE [134.93.180.162]) by mx1.FreeBSD.org (Postfix) with ESMTP id B794C43FBF; Thu, 21 Aug 2003 03:33:26 -0700 (PDT) (envelope-from ohartman@klima.physik.uni-mainz.de) Received: from klima.physik.uni-mainz.de (klima.physik.uni-mainz.de [134.93.180.162])h7LAXPJD018115; Thu, 21 Aug 2003 12:33:25 +0200 (CEST) (envelope-from ohartman@klima.physik.uni-mainz.de) Date: Thu, 21 Aug 2003 12:33:25 +0200 (CEST) From: "Hartmann, O." To: John Baldwin In-Reply-To: Message-ID: <20030821122912.Q17320@klima.physik.uni-mainz.de> References: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: smp@FreeBSD.org Subject: Re: 5.1-R-p2 crashes on SMP with AMI RAID and Intel 1000/Pro X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 21 Aug 2003 10:33:28 -0000 On Thu, 14 Aug 2003, John Baldwin wrote: Dear John. Today I got this panic (wrote the text down from screen). Hope it helped a bit: Fatal trap 12 : page fault while in kernel mode cpuid = 1; lapic.id = 00000000 fault virtual address = 0x24 fault code = supervisor read, page not present instruction pointer = 0x8:0xc01b29db stack pointer = 0x10:0xe8ff3b70 frame pointer = 0x10:0xe8ff3b84 code segment = base 0x0, limit 0xfffff, type 0x1b = DPL 0, pres 1, def 32, gran 1 processor eflags = interrupt enabled, resume, IOPL = 0 current process = 27510 (bunzip2) trap number = 12 panic: page fault cpuid = 1, lapic.id = 00000000 boot() called on cpu#1 syncing disks, buffers remaining ... panic: absolutely cannot call smp_ipi_shutdown with interrupts already disabled cpuid = 1; lapic.id = 00000000 boot() called on cpu#1 Uptime 1d20h18m55s pfs_vncache_unload(): 6 entried remaining Fatal double fault: eip = 0xc03134ic esp = 0xe8ff1ff8 ebp = 0xe8ff2014 cpuid = 1, lapic.id = 00000000 panic: double fault cpuid = 1, lapic.id = 00000000 boot() called on cpu#1 Uptime: 1d20h18m55s pfs_vncache_unload(): 6 entries remaining The crash occured when I tried to 'truss' the following command: root: /usr/local/samba/lib: truss -fae -o /tmp/afrestore afrestore -v -p "/usr/homes/kurs*" -C / truss have problems to start, I need to start a trace many times befor it works, I get weird error messages, like this: truss: PIOCWAIT top of loop: Input/output error We use afbackup for backups. Afbackup does not work on this specific SMP machine under FreeBSD 5.1-p2/SMP or UP, but it works fine on a UP machine running FBSD 5.1-p2 or -CURRENT. :> :>On 14-Aug-2003 Hartmann, O. wrote: :>> On Wed, 13 Aug 2003, Hartmann, O. wrote: :>> :>> Dear John. :>> :>> I go to try an UP kernel today, but I make no promises. :>> :>> I changed today these three config line in the running, but :>> crashing kernel: :>> :>> options HZ=1000 :>> options PQ_CACHESIZE=256 :>> options AUTO_EOI_1 :>> :>> I commented them all out because I suspect them causing trouble. :>> But this kernel does not boot anymore, it gets stuck at the same place :>> (showing amrd0 raid is ok) as the other ones. :>> :>> The nasty thing is I can not downgrade to 4.8 because I used UFS2, :>> so I'm highly interested in getting this weird TYAN Thunder 2500 SMP :>> system to work. :>> :>> I remember myself that the first occurence of those massive IRQ problems :>> occured after one of the AMI/LSI RAID BIOS updates, which where necessary :>> due to problems with IBM harddrives. Please be aware that I have the newest :>> BIOS installed (but the problems where present with the previous ones also). :>> :>> I'll send you dmesg output with UP kernel if it works. I remember myself about :>> a kernel boot option to force the loader to tell the kernel to boot :>> more verbose. Can you tell me this option? I think it will be more suitable to get a :>> verbose kernel log than this short one ... :> :>boot -v will give you the verbose boot output. If you have another machine :>available you can boot using a serial console (type 'set console=comconsole' :>in the loader and the console will move to COM1 at 9600 bps) and log that output :>which wilh let you send it to the list. :> :>-- :> :>John Baldwin <>< http://www.FreeBSD.org/~jhb/ :>"Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ :> -- MfG O. Hartmann ohartman@mail.physik.uni-mainz.de ------------------------------------------------------------------ Systemadministration des Institutes fuer Physik der Atmosphaere (IPA) ------------------------------------------------------------------ Johannes Gutenberg Universitaet Mainz Becherweg 21 55099 Mainz Tel: +496131/3924662 (Maschinenraum) Tel: +496131/3924144 (Buero) FAX: +496131/3923532