Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1161075AbVKDFm7 (ORCPT ); Fri, 4 Nov 2005 00:42:59 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1161076AbVKDFm7 (ORCPT ); Fri, 4 Nov 2005 00:42:59 -0500 Received: from willy.net1.nerim.net ([62.212.114.60]:51720 "EHLO willy.net1.nerim.net") by vger.kernel.org with ESMTP id S1161075AbVKDFm6 (ORCPT ); Fri, 4 Nov 2005 00:42:58 -0500 Date: Fri, 4 Nov 2005 06:30:44 +0100 From: Willy Tarreau To: JaniD++ Cc: linux-kernel@vger.kernel.org Subject: Re: Reboot problem. Message-ID: <20051104053044.GC11266@alpha.home.local> References: <028901c5e0d7$52fab640$0400a8c0@dcccs> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <028901c5e0d7$52fab640$0400a8c0@dcccs> User-Agent: Mutt/1.5.10i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1589 Lines: 42 Hello, On Fri, Nov 04, 2005 at 01:33:01AM +0100, JaniD++ wrote: > Hello list, > > Is there any way to force reboot after this: > > Nov 3 21:31:39 192.168.2.50 kernel: ------------[ cut here ]------------ > Nov 3 21:31:39 192.168.2.50 kernel: kernel BUG at mm/highmem.c:183! > Nov 3 21:31:39 192.168.2.50 kernel: invalid operand: 0000 [#1] > Nov 3 21:31:39 192.168.2.50 kernel: SMP > Nov 3 21:31:39 192.168.2.50 kernel: Modules linked in: netconsole > Nov 3 21:31:39 192.168.2.50 kernel: CPU: 3 (...) > Nov 3 21:31:39 192.168.2.50 kernel: Code: e8 08 06 00 00 89 c7 e9 38 ff ff > ff 55 89 e5 53 83 ec 04 89 c3 b8 80 6c 68 c0 e8 3e > Nov 3 21:31:39 192.168.2.50 kernel: <0>Fatal exception: panic in 5 seconds > > At this point the system is freez, and only reset can help. It should have rebooted, but the system is too instable to be able to do so. In this case, the only thing that can help is a hardware watchdog. Possibly, you motherboard includes a chipset with a watchdog that you can simply enable by loading the module and having a simple daemon to ping it (I have one which takes 12 kB of RAM and which tries mallocs, forks and FS accesses). If the daemon stops pinging the hardware watchdog for too long, the chipset will simply assert the RESET signal and the system will reboot. > Thanks > > Janos Regards, Willy - 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/