From owner-freebsd-questions Tue Apr 30 02:43:49 1996 Return-Path: owner-questions Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id CAA23926 for questions-outgoing; Tue, 30 Apr 1996 02:43:49 -0700 (PDT) Received: from MediaCity.com (root@easy1.mediacity.com [205.216.172.10]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id CAA23921 for ; Tue, 30 Apr 1996 02:43:47 -0700 (PDT) Received: (from brian@localhost) by MediaCity.com (8.6.11/8.6.9) id CAA06515; Tue, 30 Apr 1996 02:44:09 -0700 From: Brian Litzinger Message-Id: <199604300944.CAA06515@MediaCity.com> Subject: Re: IP aliases and delays To: angio@aros.net (Dave Andersen) Date: Tue, 30 Apr 1996 02:44:08 -0700 (PDT) Cc: freebsd-questions@FreeBSD.org In-Reply-To: <199604300641.AAA26399@terra.aros.net> from Dave Andersen at "Apr 30, 96 00:41:22 am" Reply-To: brian@MediaCity.com X-Mailer: ELM [version 2.4ME+ PL11 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-questions@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk Dave Andersen wrote: > > Why bother? FreeBSD allows you to alias an arbitrary number of addresses > > to an interface, so dummy interfaces are completely unnecessary. > > On the stupid question front, I'm wondering if anyone's actually looked > at the kind of degredation you get with packet response times on aliased > interfaces? > > To wit: On a linux 1.2.13 with aliasing, the 130th alias takes ~3ms > longer to ping on a local ethernet than the interface address itself. > (4.5ms vs 1.5ms). I don't have enough aliases on the freebsd boxes to > test this (and they're 'production' so I'm not in the mood to play with > them. :), so I'm curious if someone's actually looked at it? ping True Ip: 1.592ms 157th alias: 2.556ms Machine is an P5-75 with DEC 21040 ethernet interface. Brian Litzinger brian@mediacity.com