From owner-freebsd-stable@FreeBSD.ORG Sat Mar 5 17:15:12 2005 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4BB7C16A4CE for ; Sat, 5 Mar 2005 17:15:12 +0000 (GMT) Received: from rproxy.gmail.com (rproxy.gmail.com [64.233.170.202]) by mx1.FreeBSD.org (Postfix) with ESMTP id BA0CB43D1D for ; Sat, 5 Mar 2005 17:15:11 +0000 (GMT) (envelope-from phil.brennan@gmail.com) Received: by rproxy.gmail.com with SMTP id 34so531188rns for ; Sat, 05 Mar 2005 09:15:11 -0800 (PST) DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:reply-to:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:references; b=Eo4GW3n8bYABz9GIWUevfVnOAgeHw3dKdpDjAUbmwz2gTMWdm01mDH3dm9Ojs4V6BRVkRrcsE3VQxgEtCsBHlPy3j14Dmsba3gQHpk35H0KX2NcKxZBORc56ana2y2Ep9qogyZAutDxOzt7Q+ml/JPTwM5GRoD8hT/1P4QCe5Ys= Received: by 10.38.8.56 with SMTP id 56mr23497rnh; Sat, 05 Mar 2005 09:15:11 -0800 (PST) Received: by 10.38.179.65 with HTTP; Sat, 5 Mar 2005 09:15:11 -0800 (PST) Message-ID: Date: Sat, 5 Mar 2005 17:15:11 +0000 From: Phil Brennan To: Vlad In-Reply-To: Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit References: cc: freebsd-stable@freebsd.org Subject: Re: performance under heavy load X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: Phil Brennan List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 05 Mar 2005 17:15:12 -0000 True, on that linux machine its more io load than anything else. But anyway, the problem was caused by fortune -l going into an infinite loop, and a script running this every 5 minutes. On Sat, 5 Mar 2005 11:58:08 -0500, Vlad wrote: > > On a highly linux machine, you lose all control of the machine past a load of about 6 - 10. > > to be fair, I should note that as admin / user of few tens of servers > running both systems, I can assure you that if your linux "loses > control" with LA ~ 10, then something is seriously wrong with that > server and it's not because of the linux (rather it's hardware or > wrong kernel configuration). I had cases of LA climbing over 150 on > linux machine - it was extremely slow but I could get it back to life > w/o need for reboot. > > On Sat, 5 Mar 2005 16:28:09 +0000, Phil Brennan wrote: > > Hi, I'd just like to give some credit to the freebsd developers for a > > job well done. > > A user on our system ( freebsd 5.2.1 smp ) managed with a runaway > > script to start up 500 intensive processes, raising the load average > > to about 200. > > We managed to remotely, over ssh get a somewhat responsive session and > > kill the offending processes. Yes, I know we shouldn't have let it > > happen in the first place, by putting in proper user limits and all > > that, but it was amazing that the machine still worked. We thought > > we'd have to reboot. Even with a load of nearly 200, the machine was > > still able to serve web pages :) > > Once the load came down past 60, the system feltl fully responsive again. > > On linux, we would have had to reboot in this situation. On a highly > > linux machine, you lose all control of the machine past a load of > > about 6 - 10. This just further vindicates my decision to use freebsd > > for this service. ( Its a shell server with about 100 active users, > > apache, nfs, mysql, ldap ). Just wanted to share a success story :) > > Regards, > > > > Philip Brennan > > _______________________________________________ > > freebsd-stable@freebsd.org mailing list > > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > > To unsubscribe, send any mail to "freebsd-stable-unsubscribe@freebsd.org" > > > > -- > Vlad >