From owner-freebsd-current@FreeBSD.ORG Sun Jul 25 21:46:12 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 BF02716A4CE for ; Sun, 25 Jul 2004 21:46:12 +0000 (GMT) Received: from gw.catspoiler.org (217-ip-163.nccn.net [209.79.217.163]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6888443D48 for ; Sun, 25 Jul 2004 21:46:12 +0000 (GMT) (envelope-from truckman@FreeBSD.org) Received: from FreeBSD.org (mousie.catspoiler.org [192.168.101.2]) by gw.catspoiler.org (8.12.11/8.12.11) with ESMTP id i6PLk4LP055736; Sun, 25 Jul 2004 14:46:09 -0700 (PDT) (envelope-from truckman@FreeBSD.org) Message-Id: <200407252146.i6PLk4LP055736@gw.catspoiler.org> Date: Sun, 25 Jul 2004 14:46:04 -0700 (PDT) From: Don Lewis To: nakal@web.de In-Reply-To: <1090606572.2670.15.camel@klotz.local> MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii cc: freebsd-current@FreeBSD.org Subject: Re: Problems with interrupts on -CURRENT? X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 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: Sun, 25 Jul 2004 21:46:12 -0000 On 23 Jul, Martin wrote: > > Hi, > > I have a kernel from Mon Jul 19 01:06:15 CEST 2004 and it's > terribly slow, especially while executing applications in > parallel. > > Following symptoms: > - mouse cursor is non-responsive for 3-5 seconds Is it always like this, or only after a period of inactivity? If this only happens after a period of activity, it could be caused by the system paging heavily and paging out the X11 server. > - hard-disk is slow and is unusually active > (portupgrade needs about 15 minutes till > it arrives compiling first port) Do "vmstat 10" or "iostat 10" show an unusual amount of disk actvitiy? If so, the system could be slow because the disk is saturated. Is there anything that triggers this behaviour? Do you see a lot of activity when the system should be idle? If you are seeing a lot more disk activity than normal, there is some other problem besides interrupts. > - syncer is slowly counting vnodes down and needs > a few seconds till it arrives at 0 This is normal. The syncer now syncs the file systems before it shuts down instead of immediately shutting down and deferring the file system sync to a later stage of the system shutdown.