From owner-freebsd-net Fri Nov 1 11:17:45 2002 Delivered-To: freebsd-net@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id BCD7C37B401 for ; Fri, 1 Nov 2002 11:17:43 -0800 (PST) Received: from mailgw2a.lmco.com (mailgw2a.lmco.com [192.91.147.7]) by mx1.FreeBSD.org (Postfix) with ESMTP id BE70443E91 for ; Fri, 1 Nov 2002 11:17:42 -0800 (PST) (envelope-from rick.norman@lmco.com) Received: from emss01g01.ems.lmco.com ([129.197.181.54]) by mailgw2a.lmco.com (8.11.6/8.11.6) with ESMTP id gA1JHbu02924; Fri, 1 Nov 2002 14:17:39 -0500 (EST) Received: from CONVERSION-DAEMON.lmco.com by lmco.com (PMDF V6.1-1 #40643) id <0H4W00F01W688L@lmco.com>; Fri, 01 Nov 2002 11:17:32 -0800 (PST) Received: from lmco.com ([129.197.20.43]) by lmco.com (PMDF V6.1-1 #40643) with ESMTP id <0H4W00DGUW59RT@lmco.com>; Fri, 01 Nov 2002 11:15:09 -0800 (PST) Date: Fri, 01 Nov 2002 11:08:55 -0800 From: rick norman Subject: Re: Problem in High Speed and Long Delay with FreeBSD To: mallman@grc.nasa.gov Cc: Don Bowman , "'Fran.Lawas-Grodek@grc.nasa.gov'" , Luigi Rizzo , Cindy.Tran@grc.nasa.gov, freebsd-net@FreeBSD.ORG Message-id: <3DC2D147.8357C198@lmco.com> MIME-version: 1.0 X-Mailer: Mozilla 4.79 [en] (WinNT; U) Content-type: text/plain; charset=us-ascii Content-transfer-encoding: 7BIT X-Accept-Language: en References: <200211011909.OAA63340@guns.lerc.nasa.gov> Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Are you sure you're not hitting the top of the pipe and bouncing around in congestion avoidance ? Unless your window size limits your bw at exactly the correct amount, you'll never get the steady state bw you want. Mark Allman wrote: > > Have you checked that both sides are negotiating SACK? > > No SACK in 4.1. But, there is no loss in th connection. > > > And both sides are negotiating a window scale option sufficiently > > large? (sounds like you need a window scale option of at least 5 > > bits?) > > We're seeing a shift of 6. > > > And the socket-buffer to ttcp is actually being set as large > > as you think? (perhaps run 'ktrace' or 'truss' on ttcp and look > > for an error on the setsockopt). > > We hacked ttcp to run getsockopt() to tell us if the kernel did not > honor our setsockopt() request. All looks fine. > > Thanks! Other ideas? > > allman > > -- > Mark Allman -- BBN/NASA GRC -- http://roland.grc.nasa.gov/~mallman/ > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-net" in the body of the message -- "I'm a-goin' to stay where you sleep all day Where they hung the jerk that invented work In the Big Rock Candy Mountains" wk: 408 742 1619 rick.norman@lmco.com hm: 650 726 0677 rnorman@ikaika.com cell: 650 303 3877 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message