From owner-freebsd-questions@FreeBSD.ORG Fri Oct 8 10:52:42 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 6C37216A4CE for ; Fri, 8 Oct 2004 10:52:42 +0000 (GMT) Received: from sferics.mongueurs.net (sferics.mongueurs.net [81.80.147.197]) by mx1.FreeBSD.org (Postfix) with ESMTP id ACE2843D1D for ; Fri, 8 Oct 2004 10:52:41 +0000 (GMT) (envelope-from david@landgren.net) Received: from [127.0.0.1] (exo.bpinet.com [81.80.147.206]) by sferics.mongueurs.net (Postfix) with ESMTP id 88E40A98B for ; Fri, 8 Oct 2004 12:52:40 +0200 (CEST) Message-ID: <41667173.1050401@landgren.net> Date: Fri, 08 Oct 2004 12:52:35 +0200 From: David Landgren Organization: The Lusty Decadent Delights of Imperial Pompeii User-Agent: Mozilla Thunderbird 0.8 (Windows/20040913) X-Accept-Language: en-us, en MIME-Version: 1.0 To: freebsd-questions@freebsd.org Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: Sluggishness on a 6 CPU HP-6000 Netserver 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: Fri, 08 Oct 2004 10:52:42 -0000 Hello, I have just installed an HP-6000r Netserver with 2Gb RAM and 6 P-III 700MHz processors. I cvsupped src (4.10-STABLE) yesterday and built the world and kernel. I've stripped out just about everything else I can from the kernel: usb, firewire, apm, serial port, //l port, PCMCIA, other RAID/SCSI/network drivers. The kernel has dummynet support. I have tried option HZ=x, for x=1000, 200, 100 but network response is sluggish. I haven't specifically referred to any dummynet features in my ipfw ruleset. When I type quickly I can run several characters ahead of the display, which jerks along behind. Display output, e.g. when compiling the kernel, is also jerky. From the console the response is crisp. Both the switch to which the box is connected (a Cisco Catalyst) and the fxp network driver are hard-coded to 100baseTX full-duplex. netstat -m: 70/480/26624 mbufs in use (current/peak/max): 66 mbufs allocated to data 4 mbufs allocated to packet headers 64/128/6656 mbuf clusters in use (current/peak/max) 376 Kbytes allocated to network (1% of mb_map in use) 0 requests for memory denied 0 requests for memory delayed 0 calls to protocol drain routines It must be something, but I have no idea what to look {at,for}/what to tune. It's the first time I've built dummynet, first time I've used a box with more than two CPUs and I wonder if there might be some sort of contention happening. Thanks for any clues I can use. David