From owner-freebsd-net@FreeBSD.ORG Tue Oct 6 06:38:55 2009 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 49EC8106566B for ; Tue, 6 Oct 2009 06:38:55 +0000 (UTC) (envelope-from rihad@mail.ru) Received: from mx38.mail.ru (mx38.mail.ru [94.100.176.52]) by mx1.freebsd.org (Postfix) with ESMTP id 031B88FC0C for ; Tue, 6 Oct 2009 06:38:54 +0000 (UTC) Received: from [217.25.27.27] (port=46837 helo=[217.25.27.27]) by mx38.mail.ru with asmtp id 1Mv3hF-000BbP-00; Tue, 06 Oct 2009 10:38:53 +0400 Message-ID: <4ACAE5FC.7040605@mail.ru> Date: Tue, 06 Oct 2009 11:38:52 +0500 From: rihad User-Agent: Mozilla-Thunderbird 2.0.0.22 (X11/20090706) MIME-Version: 1.0 To: Eugene Grosbein References: <20091005095600.GA73335@svzserv.kemerovo.su> <4AC9CFF7.3090208@mail.ru> <20091005110726.GA62598@onelab2.iet.unipi.it> <4AC9D87E.7000005@mail.ru> <20091005120418.GA63131@onelab2.iet.unipi.it> <4AC9E29B.6080908@mail.ru> <20091005123230.GA64167@onelab2.iet.unipi.it> <4AC9EFDF.4080302@mail.ru> <20091005140409.GB89194@svzserv.kemerovo.su> <4ACA02F7.6040409@mail.ru> <20091006042153.GA5857@svzserv.kemerovo.su> In-Reply-To: <20091006042153.GA5857@svzserv.kemerovo.su> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Spam: Not detected X-Mras: Ok Cc: freebsd-net@freebsd.org, Luigi Rizzo Subject: Re: dummynet dropping too many packets X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 06 Oct 2009 06:38:55 -0000 Eugene Grosbein wrote: > On Mon, Oct 05, 2009 at 07:30:15PM +0500, rihad wrote: > >>>> How do I investigate and fix this burstiness issue? >>> Please also show: >>> >>> sysctl net.isr >>> sysctl net.inet.ip.intr_queue_maxlen >> net.isr.swi_count: 65461359 >> net.isr.drop: 0 >> net.isr.queued: 32843752 >> net.isr.deferred: 0 >> net.isr.directed: -723075002 >> net.isr.count: -723074001 >> net.isr.direct: 1 >> >> net.inet.ip.intr_queue_maxlen: 50 > > Try to increase net.inet.ip.intr_queue_maxlen uptio 4096. > Apparently increasing the input queue had no effect on the output drops. Besides, net.inet.ip.intr_queue_drops has only reached 70 for 6 days of uptime. I'll soon be rebooting to change HZ 1000 -> 2000, and we'll see how that goes. There must be some global limit, not nmbclusters, not per-user output queues, not the interface input queue, but something else that triggers at around 360-370 mbit/s...