From owner-freebsd-isp Thu Sep 6 16:26:42 2001 Delivered-To: freebsd-isp@freebsd.org Received: from inet03.citec.qld.gov.au (inet03.citec.qld.gov.au [203.5.10.10]) by hub.freebsd.org (Postfix) with ESMTP id A024237B401 for ; Thu, 6 Sep 2001 16:26:37 -0700 (PDT) Received: by inet03.citec.qld.gov.au; id JAA19670; Fri, 7 Sep 2001 09:26:34 +1000 (EST) Received: from citecub.citec.qld.gov.au( 131.242.4.98) by inet03.citec.qld.gov.au via smap (V2.0) id xma019555; Fri, 7 Sep 01 09:26:26 +1000 Received: from guru.citec.qld.gov.au by citecub.citec.qld.gov.au (SMI-8.6/SMI-SVR4) id JAA08297; Fri, 7 Sep 2001 09:26:26 +1000 Received: from localhost (sgcccdc@localhost) by guru.citec.qld.gov.au (8.9.3/8.9.3) with ESMTP id JAA75353; Fri, 7 Sep 2001 09:26:24 +1000 (EST) (envelope-from sgcccdc@citec.qld.gov.au) X-Authentication-Warning: guru.citec.qld.gov.au: sgcccdc owned process doing -bs Date: Fri, 7 Sep 2001 09:26:23 +1000 (EST) From: Colin Campbell To: Cc: Subject: Re: Mail Cluster Question In-Reply-To: <20010906124736.7C8BB37B403@hub.freebsd.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-isp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Hi, On Thu, 6 Sep 2001 brian.jackson@third-rail.net wrote: > Hi everyone, > > I've currently built an e-mail cluster with three machines running > qmail (and FreeBSD, obviously). Each machine is running pop and smtp, > and they are all "balanced" using round robin DNS. > > One of the machines is exporting it's /usr/home directory, and the > other two are mounting this directory on their /usr/home, so that no > matter which machine you hit you get your mail (not enough $$ for a > NetApp Filer...). > > Each of these machines has two disks mirrored running vinum (for a > separate thread, yes you can have two disks mirrored running vinum and > boot off either...), but I'm paranoid - what happens in a catastrophic > situation where the machine exporting /usr/home goes away? Has anyone > done anything similar? > > I'm thinking of just using a simple rsync script locally on each > machine, and then un-mounting / re-exporting / re-mounting file > systems, but this seems - well, complicated. The only way to guarantee a clean failover is to use dual-ported disks or a SAN. Everything else (eg a NAS) leaves you with a single point of failure until someone allows mirroring of nfs-mounted disks or provides a network block device (Linux terminology) that also works under vinum for example. Thiking about this very problem the other day got me wondering whether vinum would work on a vnode which was a remote file (I know vn won't do that yet, but it would certainly make these HA systems easier). Colin To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-isp" in the body of the message