From owner-freebsd-net@FreeBSD.ORG Thu Mar 21 02:14:06 2013 Return-Path: Delivered-To: freebsd-net@FreeBSD.org Received: from mx1.freebsd.org (mx1.FreeBSD.org [8.8.178.115]) by hub.freebsd.org (Postfix) with ESMTP id 576BAEF3 for ; Thu, 21 Mar 2013 02:14:06 +0000 (UTC) (envelope-from markd-freebsd-net@bushwire.net) Received: from smtp1.bushwire.net (f5.bushwire.net [199.48.133.46]) by mx1.freebsd.org (Postfix) with SMTP id 0A5A1120 for ; Thu, 21 Mar 2013 02:14:05 +0000 (UTC) Received: (qmail 98970 invoked by uid 1001); 21 Mar 2013 02:14:04 -0000 Delivered-To: qmda-intercept-freebsd-net@FreeBSD.org DomainKey-Signature: a=rsa-sha1; q=dns; c=simple; s=2004; d=bushwire.net; b=4onIC8ERdOjI6nDt30lJrVNSv44idZQ6g5q1xaABK+OlZE1T0UE+NN76jLJec5of; Comments: DomainKeys? See http://en.wikipedia.org/wiki/DomainKeys DomainKey-Trace-MD: h=13; b=17; l=C18R71D32M65F47T27S73R65?69M17C39C27I81; Comments: QMDA 0.3 Received: (qmail 98963 invoked by uid 1001); 21 Mar 2013 02:14:04 -0000 Date: 21 Mar 2013 02:14:04 +0000 Message-ID: <20130321021404.98962.qmail@f5-external.bushwire.net> From: "Mark D" To: freebsd-net@FreeBSD.org Subject: Re: Best way for an app to accept traffic on 30,000+ interfaces? References: <20130321005959.98706.qmail@f5-external.bushwire.net> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 21 Mar 2013 02:14:06 -0000 On 20Mar13, Juli Mallett allegedly wrote: > Well, the easiest thing is to add 30k aliases to your Ethernet > interface (you may hit a limit, not sure) and then just listen on > INADDR_ANY (or the IP6 equivalent), assuming you don't mind listening > to other addresses you have configured as well. The application can > always decide to close an incoming connection that wasn't going to one > of those 30k IP addresses. Agreed that INADDR_ANY will fix the socket count. But what of the interface/alias count? Is 300K ok? How about 3M aliases? I'll spin up an instance and try it out, but I'm a little worried that there might be something non-linear or some threshold limit that won't necessarily be exposed by a modicum of adhoc testing. Mark.