From owner-freebsd-stable@FreeBSD.ORG Sat Sep 24 14:18:13 2005 Return-Path: X-Original-To: freebsd-stable@freebsd.org Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id F3A5716A41F for ; Sat, 24 Sep 2005 14:18:12 +0000 (GMT) (envelope-from b.candler@pobox.com) Received: from thorn.pobox.com (thorn.pobox.com [208.210.124.75]) by mx1.FreeBSD.org (Postfix) with ESMTP id A704C43D48 for ; Sat, 24 Sep 2005 14:18:12 +0000 (GMT) (envelope-from b.candler@pobox.com) Received: from thorn (localhost [127.0.0.1]) by thorn.pobox.com (Postfix) with ESMTP id E87B5B6 for ; Sat, 24 Sep 2005 10:10:17 -0400 (EDT) Received: from billdog.local.linnet.org (dsl-212-74-113-66.access.uk.tiscali.com [212.74.113.66]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by thorn.sasl.smtp.pobox.com (Postfix) with ESMTP id C5FC13F70 for ; Sat, 24 Sep 2005 10:10:17 -0400 (EDT) Received: from brian by billdog.local.linnet.org with local (Exim 4.50 (FreeBSD)) id 1EJAuL-0000Na-B5 for freebsd-stable@freebsd.org; Sat, 24 Sep 2005 15:21:41 +0100 Date: Sat, 24 Sep 2005 15:21:41 +0100 From: Brian Candler To: freebsd-stable@freebsd.org Message-ID: <20050924142141.GB1236@uk.tiscali.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.4.2.1i Subject: geom_mirror and ggatec/d safe for filesystem replication? X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 24 Sep 2005 14:18:13 -0000 Consider the following scenario: - Machine 1 has a UFS filesystem on a block device. - It is mirrored to a block device on machine 2, using geom_mirror and ggatec/ggated - Machine 2 mounts this filesystem read only. My question is: is this last step safe? When an update occurs on machine 1, certain disk blocks will change "under the nose" of machine 2's mounts. Will all necessary caches be invalidated? (e.g. block caches, inode caches) Or could this result in nasty behaviour, up to and including panics? If so, is there a safer or better way of performing real-time filesystem replication? Thanks, Brian Candler.