From owner-freebsd-net Sun Oct 6 6:36:28 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 7FDCD37B401 for ; Sun, 6 Oct 2002 06:36:26 -0700 (PDT) Received: from smtp.cwnt.com (smtp.cwnt.com [192.116.246.129]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7044D43E3B for ; Sun, 6 Oct 2002 06:36:11 -0700 (PDT) (envelope-from ido@cwnt.com) X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3 content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Subject: RE: Help with net.inet.ip.intr_queue_maxlen Date: Sun, 6 Oct 2002 16:35:57 +0300 Message-ID: X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: Help with net.inet.ip.intr_queue_maxlen Thread-Index: AcJszPAH6N3JMPkrTveC1FpT4NosUQAbpyfA From: "Ido Barnea" To: "Steve Francis" , 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 > -----Original Message----- > From: Steve Francis [mailto:steve@expertcity.com] > Sent: Sunday, October 06, 2002 3:13 AM > To: freebsd-net@freebsd.org > Subject: Help with net.inet.ip.intr_queue_maxlen >=20 >=20 > Can someone help me with net.inet.ip.intr_queue_maxlen tuning? >=20 > Firstly, its the "size of the IP input queue", per the source. >=20 > So does that mean after the NIC has received the packet, the interupt > from the NIC has been processed and the packet retrieved from the NIC, > then the packet is placed in this queue, before the IP stack=20 > looks at it? > i.e. its unrelated to interupt coalescing or polling, or NIC > performance, as they have already occurred in order to put the packet > into the queue. Yes? >=20 Absolutely correct > I am getting incrementing net.inet.ip.intr_queue_drops at around 8,000 > pps (increasing drops at rate of 10 or so per second.) > Yet, if my statement above about what the queue is, is=20 > correct, then it > just means that the system was busy doing stuff before it had a chance > to process the incoming packets, so there was no room for new ones to > enter the queue. But as the system was only 50% busy, then if=20 > I increase > the input queue, I should be able to avoid these drops, correct? At > least until the system gets a lot busier. >=20 Correct again > Is there a sane upper recommended limit to the queue length? >=20 Basically there shouldn't be a problem incrementing the queue length, = since the space is not allocated in advance. So, you should experiment with few values until the drop rate drops to a = reasonable value. The only problem I can think of is that you (or an attacker) can exhaust = your mbuf pool if you allow the queue length to become too large. You can check out the state of your mbuf poll using = 'netstat -m' > Or am I way off base here? > Thanks >=20 >=20 > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-net" in the body of the message >=20 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message