From owner-freebsd-current  Sun Dec  5  1:40:15 1999
Delivered-To: freebsd-current@freebsd.org
Received: from uni4nn.gn.iaf.nl (osmium.gn.iaf.nl [193.67.144.12])
	by hub.freebsd.org (Postfix) with ESMTP
	id 67D2415147; Sun,  5 Dec 1999 01:38:05 -0800 (PST)
	(envelope-from wilko@yedi.iaf.nl)
Received: from yedi.iaf.nl (uucp@localhost)
	  by uni4nn.gn.iaf.nl (8.9.2/8.9.2) with UUCP id KAA26933;
	  Sun, 5 Dec 1999 10:24:06 +0100 (MET)
Received: (from wilko@localhost)
	by yedi.iaf.nl (8.9.3/8.9.3) id AAA30805;
	Sun, 5 Dec 1999 00:45:12 +0100 (CET)
	(envelope-from wilko)
From: Wilko Bulte <wilko@yedi.iaf.nl>
Message-Id: <199912042345.AAA30805@yedi.iaf.nl>
Subject: Re: Mounting one FS on more than one system
In-Reply-To: <199912042044.MAA05073@flamingo.McKusick.COM> from Kirk McKusick at "Dec 4, 1999 12:44:43 pm"
To: mckusick@flamingo.McKusick.COM (Kirk McKusick)
Date: Sun, 5 Dec 1999 00:45:12 +0100 (CET)
Cc: msmith@FreeBSD.ORG, match@elen.utah.edu, current@FreeBSD.ORG
X-Organisation: Private FreeBSD site - Arnhem, The Netherlands
X-pgp-info: PGP public key at 'finger wilko@freefall.freebsd.org'
X-Mailer: ELM [version 2.4ME+ PL43 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-freebsd-current@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG

As Kirk McKusick wrote ...

> Mounting on more than one system is generally problematical unless
> you are willing to have all systems read-only. The problem is cache
> coherence between the machines. If one changes a block, the other
> machines will not see it. Basically, this is why we have the NFS
> filesystem. That lets a disk be mounted on one machine, but shared
> out to others. If you wanted to write a protocol that would allow
> for multiple machines, then you would need to have some central
> coordinator running some sort of coherency protocol with a complexity
> akin to that of NFS.

I wonder how Tru64 is doing it. IIRC V5.0 Tru64 can do a cluster filesystem.
A CFS must have solved the coherency issue in some way.

Older revs had distributed raw devices (for Oracle and the like) but that
had all I/O go through one cluster member that did all the I/O for that DRD.
The I/O from the other cluster members was done via the Memory Channel to
the DRD-serving machine. 

Interesting..

Wilko
-- 
|   / o / /  _  	Arnhem, The Netherlands	  - Powered by FreeBSD -
|/|/ / / /( (_) Bulte 	WWW : http://www.tcja.nl  http://www.freebsd.org


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message