From owner-freebsd-questions@FreeBSD.ORG Wed Jun 9 02:21:18 2004 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 5762A16A4CE for ; Wed, 9 Jun 2004 02:21:18 +0000 (GMT) Received: from dan.emsphone.com (dan.emsphone.com [199.67.51.101]) by mx1.FreeBSD.org (Postfix) with ESMTP id 0F0E243D54 for ; Wed, 9 Jun 2004 02:21:18 +0000 (GMT) (envelope-from dan@dan.emsphone.com) Received: (from dan@localhost) by dan.emsphone.com (8.12.10/8.12.10) id i592L0Qx071405; Tue, 8 Jun 2004 21:21:00 -0500 (CDT) (envelope-from dan) Date: Tue, 8 Jun 2004 21:21:00 -0500 From: Dan Nelson To: Doug Hardie Message-ID: <20040609022100.GC46338@dan.emsphone.com> References: <20040609005314.81640.qmail@web40312.mail.yahoo.com> <341E98A6-B9B4-11D8-BD00-000393681B06@lafn.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <341E98A6-B9B4-11D8-BD00-000393681B06@lafn.org> X-OS: FreeBSD 5.2-CURRENT X-message-flag: Outlook Error User-Agent: Mutt/1.5.6i cc: FreeBSD Question List Subject: Re: Top Consistency X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 09 Jun 2004 02:21:18 -0000 In the last episode (Jun 08), Doug Hardie said: > I am running FreeBSD 4.6 and top does not show consistent data (at > least in my understanding). The cpu states line shows the percent of > time in user state. I would expect the percent processor used by all > the active processes to add up to something close to that. (single > processor machine). However, it never seems to come close. Often it > will show 25% user and the sums of the active processes utilizations > will be around 2%. Other times it will show 2% user and the sum of > the processes is over 10%. Is top wacky or is my understanding > wrong? The %WCPU and %CPU columns are weighted averages over ~60 seconds, while the "CPU states" row is an instantaneous snapshot, so they will almost never total up. -- Dan Nelson dnelson@allantgroup.com