From owner-freebsd-hackers@FreeBSD.ORG Fri Apr 11 05:26:11 2003 Return-Path: Delivered-To: freebsd-hackers@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 2737B37B401; Fri, 11 Apr 2003 05:26:11 -0700 (PDT) Received: from haldjas.folklore.ee (Haldjas.folklore.ee [193.40.6.121]) by mx1.FreeBSD.org (Postfix) with ESMTP id 0809343FBD; Fri, 11 Apr 2003 05:26:10 -0700 (PDT) (envelope-from narvi@haldjas.folklore.ee) Received: from haldjas.folklore.ee (localhost [127.0.0.1]) by haldjas.folklore.ee (8.12.3/8.11.3) with ESMTP id h3BCQ8UE087811; Fri, 11 Apr 2003 15:26:08 +0300 (EEST) (envelope-from narvi@haldjas.folklore.ee) Received: from localhost (narvi@localhost)h3BCQ7r1087808; Fri, 11 Apr 2003 15:26:07 +0300 (EEST) Date: Fri, 11 Apr 2003 15:26:07 +0300 (EEST) From: Narvi To: Andreas Klemm In-Reply-To: <20030411081330.GA918@titan.klemm.apsfilter.org> Message-ID: <20030411152026.I75698-100000@haldjas.folklore.ee> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: freebsd-hackers@freebsd.org Subject: Re: Solaris 2.x compat. system accounting, sarcheck for fbsd (commercial) X-BeenThere: freebsd-hackers@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Technical Discussions relating to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 11 Apr 2003 12:26:11 -0000 On Fri, 11 Apr 2003, Andreas Klemm wrote: > On Fri, Apr 11, 2003 at 06:01:03PM +1000, Peter Jeremy wrote: > > On Fri, Apr 11, 2003 at 08:08:23AM +0200, Andreas Klemm wrote: > > >I wrote some performance monitoring utilities and need for > > >example a parameter wait for I/O, that I didn't find in our > > >tools, did I overlook something ? > > > > "wait for I/O" is a System-V state that BSD doesn't measure. > > One of the grey[er] beards may be able to answer why SysV does > > and BSD doesn't consider "wait for I/O" a state worth measuring. > > Would be interested to hear why, since I think its an interesting > parameter. At least I know of some sun consultants who are the > opineon, that this is an important parameter. > > If I understood the parameter right it should bring you a > value, who measures how often (in percentage) the CPU had > to wait for I/O operations. Not only disk, additionally > network I/O as well (from hearsay) but Im not 100% sure about > it. So theoretically it should bring to you a value, if you > are I/O bound. Togehter with other values (disk I/O and Network > stats it should be easy to see, if you have a problem with your > disks or with your network). > Well, you can get pretty pathetic results where a program causes a significant amount of time to be spent waiting for I/O and the rest of teh system crawls. One would hope you just can't get the same situation on freebsd. > BTW, are there some automatic tools around for BSD, that > automate the process of telling you, where you have problems > with your machine or what kernel parameters need an update ?? > > Andreas /// > > -- > Andreas Klemm - Powered by FreeBSD 4.7 > Need a magic printfilter today ? -> http://www.apsfilter.org/ >