From owner-freebsd-current@FreeBSD.ORG Fri Feb 6 03:38:07 2004 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 1893B16A4CF for ; Fri, 6 Feb 2004 03:38:07 -0800 (PST) Received: from avocet.mail.pas.earthlink.net (avocet.mail.pas.earthlink.net [207.217.120.50]) by mx1.FreeBSD.org (Postfix) with ESMTP id 37EA043D1D for ; Fri, 6 Feb 2004 03:38:05 -0800 (PST) (envelope-from chris@behanna.org) Received: from user183.net446.oh.sprint-hsd.net ([65.40.131.183] helo=192.168.1.2) by avocet.mail.pas.earthlink.net with esmtp (Exim 3.33 #1) id 1Ap4JA-00074H-00 for freebsd-current@freebsd.org; Fri, 06 Feb 2004 03:38:04 -0800 From: Chris BeHanna Organization: Western Pennsylvania Pizza Disposal Unit To: freebsd-current@freebsd.org Date: Fri, 6 Feb 2004 06:38:02 -0500 User-Agent: KMail/1.5.4 References: In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200402060638.02548.chris@behanna.org> X-Mailman-Approved-At: Fri, 06 Feb 2004 05:12:59 -0800 Subject: Re: Processes blocked on getblk or ufs X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: chris@behanna.org List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 06 Feb 2004 11:38:07 -0000 On Thursday 05 February 2004 15:58, Robin P. Blanchard wrote: > > I have a machine sitting at the ddb prompt in exactly this > > state. Anyone know how to dig something useful out of the > > current state the buffer cache is in? I know how to find out > > what vnodes are locked but don't know how to go from that to > > see what underlying buffers are causing the problem. > > > > I should be able to cause a dump of the machine as well but > > it has 2Gb of memory in it so transferring the dump to > > someone other than me who could look at it might be painful... > > Well...I went ahead an induced a dump (where the 2650 has been wedged in > "getblk" since last installworld). Anyone want to take a look at it. It's a > 1024M dump compressed to 171M. Since others are unlikely to have the exact same sources that you have, it'd be best if you could do some preliminary poking yourself, since your sources match your kernel. If you know any C, printing the contents of variables referenced on the line that triggered the crash would be helpful. gdb -k /usr/obj/usr/src/sys/YOURKERNEL/kernel.debug /var/crash/vmcore.0 gdb> bt gdb> frame frame-number-just-before-the-trap-code gdb> p some-interesting-variable -- Chris BeHanna Software Engineer (Remove "bogus" before responding.) chris@bogus.behanna.org Turning coffee into software since 1990.