From owner-freebsd-current@FreeBSD.ORG Fri Aug 14 13:47:06 2009 Return-Path: Delivered-To: current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id EFD4B106568C for ; Fri, 14 Aug 2009 13:47:06 +0000 (UTC) (envelope-from ianf@clue.co.za) Received: from inbound01.jnb1.gp-online.net (inbound01.jnb1.gp-online.net [41.161.16.135]) by mx1.freebsd.org (Postfix) with ESMTP id 821F98FC43 for ; Fri, 14 Aug 2009 13:47:06 +0000 (UTC) Received: from [41.161.16.10] (helo=clue.co.za) by inbound01.jnb1.gp-online.net with esmtpsa (TLS-1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.63) (envelope-from ) id 1Mbx7Y-0001cb-1y; Fri, 14 Aug 2009 15:47:04 +0200 Received: from localhost ([127.0.0.1] helo=clue.co.za) by clue.co.za with esmtp (Exim 4.69 (FreeBSD)) (envelope-from ) id 1Mbx7k-0001zP-UW; Fri, 14 Aug 2009 15:47:16 +0200 To: Ruslan Ermilov From: Ian FREISLICH In-Reply-To: <20090814120602.GA64565@edoofus.dev.vega.ru> References: <20090814120602.GA64565@edoofus.dev.vega.ru> <20090814080044.A93661@maildrop.int.zabbadoz.net> X-Attribution: BOFH Date: Fri, 14 Aug 2009 15:47:16 +0200 Message-Id: Cc: "Bjoern A. Zeeb" , FreeBSD current mailing list Subject: Re: panic in netisr (I think) + DUMP FAILED (ERROR 6) X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 14 Aug 2009 13:47:07 -0000 Ruslan Ermilov wrote: > On Fri, Aug 14, 2009 at 10:47:05AM +0200, Ian FREISLICH wrote: > > Do you have any idea why kernel dumps are not working as shown in > > the OP? > > > There was a message, "Attempt to write outside dump device boundaries.", > that explains that. Search the mailing list archives where I explained > the underlying bug with current versions of FreeBSD (in short page maps > get updated while we're writing a dump, due to interrupts still enabled). And then, this bug is probably why my dumps have been unusable since about 12 months now? These (2) servers have a continuous interrupt rate in excess of 40k/s, network related, so interrupts during dumping are highly likely. Reading that thread I'm given to believe that the problem exists only for mini-dumps. I've had crashes without cores with (and without) mini-dumps enabled, but I've only recently started capturing console output, so I'll see what happens on its next crash. They reliably crash at least once a week. Ian -- Ian Freislich