From owner-freebsd-questions@FreeBSD.ORG Tue Jan 27 05:10:54 2009 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 61A0D106564A for ; Tue, 27 Jan 2009 05:10:54 +0000 (UTC) (envelope-from freebsd@sopwith.solgatos.com) Received: from sopwith.solgatos.com (pool-173-50-231-101.ptldor.fios.verizon.net [173.50.231.101]) by mx1.freebsd.org (Postfix) with ESMTP id B49868FC13 for ; Tue, 27 Jan 2009 05:10:53 +0000 (UTC) (envelope-from freebsd@sopwith.solgatos.com) Received: by sopwith.solgatos.com (Postfix, from userid 66) id 9ACCEB650; Mon, 26 Jan 2009 21:09:20 -0800 (PST) Received: from localhost by sopwith.solgatos.com (8.8.8/6.24) id BAA04110; Tue, 27 Jan 2009 01:23:18 GMT Message-Id: <200901270123.BAA04110@sopwith.solgatos.com> To: freebsd-questions@freebsd.org In-reply-to: Your message of "Sun, 25 Jan 2009 22:47:29 -0900." <200901252247.29775.fbsd.questions@rachie.is-a-geek.net> Date: Mon, 26 Jan 2009 17:23:18 +0000 From: Dieter Subject: Re: swap_pager complaints but not using swap 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: Tue, 27 Jan 2009 05:10:54 -0000 In message <200901252247.29775.fbsd.questions@rachie.is-a-geek.net>, Mel writes: > On Sunday 25 January 2009 04:28:47 Dieter wrote: > > >>>> AMD64 FreeBSD 7.0 2 GiB main memory > > >>>> > > > So the machine doesn't normally use swap much at all, but messing with > > the large ISO apparently kicked something out of memory, and the disk > > with the swap partition was already busy writing at the other end of > > the disk. > > > > Do you consider writing a large file to disk a "truly unspeakable" load? > > Just curious, since you're running 7.0: SCHED_4BSD or SCHED_ULE? And if you > have a chance to change it, does the scenario persist? kern.sched.name: 4BSD Right now I'm trying to collect enough info to fix the Seagate 2700.11 firmware fubar. Once that is resolved, I plan to update to 7.1. IIRC 7.1 defaults to ULE. If not I can try switching. But my understanding is that the scheduler pays little if any attention to I/O, a holdover from the days when CPU was the main thing that needed rationing. These days it is just the opposite. CPUs have gotten much much faster, but I/O is only slightly faster.