From owner-freebsd-isp Mon Jan 12 03:06:15 1998 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id DAA28780 for isp-outgoing; Mon, 12 Jan 1998 03:06:15 -0800 (PST) (envelope-from owner-freebsd-isp) Received: from hq.seicom.net (mail.seicom.net [194.97.200.30]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id DAA28768 for ; Mon, 12 Jan 1998 03:06:08 -0800 (PST) (envelope-from udesign!lukas@hq.seicom.net) Received: from udesign.UUCP by hq.seicom.net (8.8.5/8.8.3) with UUCP id LAA23541; Mon, 12 Jan 1998 11:57:09 +0100 (CET) Received: by reactor.design.de id m0xrhiY-000BijC (Debian Smail-3.2 1996-Jul-4 #2); Mon, 12 Jan 1998 12:07:42 +0100 (MET) Message-ID: <19980112120742.56306@reactor> Date: Mon, 12 Jan 1998 12:07:42 +0100 From: Lukas Wunner To: "Jeffrey J. Mountin" Cc: Lukas Wunner , =?iso-8859-1?Q?=22Luis_E=2E_Mu=F1oz=22?= , freebsd-isp@FreeBSD.ORG Subject: Re: [fbsd-isp] Designing for a very large ISP References: <3.0.5.32.19980103121611.007af8f0@pop.can <3.0.5.32.19980103121611.007af8f0@pop.cantv.net> <19980104141146.32430@reactor> <3.0.3.32.19980112021334.006fe3dc@mixcom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.88 In-Reply-To: <3.0.3.32.19980112021334.006fe3dc@mixcom.com>; from Jeffrey J. Mountin on Mon, Jan 12, 1998 at 02:13:34AM -0600 Sender: owner-freebsd-isp@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, > Quite a few PPro boards have 5 PCI slots and Tyan's S1571 is one Pentium > board with 5 PCI slots. 5 slots is beyond the PCI specification so the last slot is always non-busmastering which makes it practically useless for any serious use. So we can forget about the 5th PCI slot on these boards and assume they also have 4 *usable* slots. > Many of the replies to this thread mention memory capacity and depending on > what you want to use either 384, 512, or 1024 Mb can be used. Not sure why > this is a factor, since in dealing PC based ISP's load sharing and > distribution of services is why it works and is cost effective *and* scalable. Sure. So how do you propose one should distribute an innd process among several machines? Your suggestion does not work out in this case *at all*. > Since you have a Tyan, what about the S1668 Titan Pro for the > PPro with 5 PCI slots and up to 1 Gb of memory? Have you actually *tested* if this board will grok 1GB of memory or are you just saying this out of the blue, based on the specs published by Tyan? If the latter is the case, let me assure you that Tyan's specs aren't worth a penny when it comes to accurateness. The Tomcat III board is advertised to support 512MB of RAM, which would be sufficient for our news machine. However, in contrast to the specs, it actually only supports a maximum of 256MB of RAM which is not sufficient. I have tested this with lots of different types of SIMMs and with several permutations of these SIMMs to no avail. If the machine has been turned off for a while and is cold, it will (unreliably) report either 256MB or 288MB. In all other cases, it will report 256MB on bootup. Michael Beckmann has reported a similar problem: he didn't even manage to get the Tomcat III to work with 256MB of RAM (the system didn't run too stable). Downgrading to 192MB would solve the problem. As a general rule of thumb, with PCs you will almost always be able to get 128MB to work reliable, more than 128MB is a matter of luck, more than 256MB is practically impossible. I had to learn this the hard way. > Pardon my ignorance, but I've always thought of SGI boxes as grahpics > designers dreams, not a server for an ISP. I've dealt with them on > occasion (regretfully) and know a grahpics designer that now has his own > SGI to use and couldn't be happier, but I'd pass. Of the three German news boxes in the upper 50 ranks in Top1000 12/97 (www.freenix.fr/top1000), two are SGI-based systems (fu-berlin.de and newsfeed.ecrc.net). No comment. > If the board can handle 128 Mb SIMMs, the Titan pro must since 8 * 128 = 1 > Gb and either EDO or EDO/ECC are comparable in price. Sure, go ahead and test it. > Plenty of alternatives to the SGI. Probably not. Lukas. -- lukas wunner unix, internetworking and security engineer lukas@wunner.de LW26-RIPE http://www.wunner.de/~lukas/ Funkmodems mit 2.4GHz FAQ http://www.wunner.de/~lukas/funk/