From owner-freebsd-stable@FreeBSD.ORG Mon Oct 25 08:49:19 2010 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 8C9C8106566B; Mon, 25 Oct 2010 08:49:19 +0000 (UTC) (envelope-from petefrench@ticketswitch.com) Received: from constantine.ticketswitch.com (constantine.ticketswitch.com [IPv6:2002:57e0:1d4e:1::3]) by mx1.freebsd.org (Postfix) with ESMTP id 51D3D8FC15; Mon, 25 Oct 2010 08:49:19 +0000 (UTC) Received: from dilbert.rattatosk ([10.64.50.6] helo=dilbert.ticketswitch.com) by constantine.ticketswitch.com with esmtps (TLSv1:AES256-SHA:256) (Exim 4.69 (FreeBSD)) (envelope-from ) id 1PAIk1-0002XF-Ks; Mon, 25 Oct 2010 09:49:17 +0100 Received: from petefrench by dilbert.ticketswitch.com with local (Exim 4.72 (FreeBSD)) (envelope-from ) id 1PAIk1-000Bfy-Jb; Mon, 25 Oct 2010 09:49:17 +0100 Date: Mon, 25 Oct 2010 09:49:17 +0100 Message-Id: To: pjd@FreeBSD.org, to.my.trociny@gmail.com In-Reply-To: <20101022181301.GA2014@garage.freebsd.pl> From: Pete French Cc: freebsd-stable@freebsd.org Subject: Re: hast vs ggate+gmirror sychrnoisation speed 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: Mon, 25 Oct 2010 08:49:19 -0000 > What speed do you expect? IIRC from my tests, I was able to saturate > 1Gbit link with initial synchronization. Also note, that hast > synchronize only differences, and not the entire thing after crash or > power failure. I should probably have put some numbers in the original email, sorry! I am looking here at an initial syc of the entire mirror, not just the differences. ggate+gmirror gives me about 89 meg/second hast gives me about 44 meg/second The boxes have gigabit connectin between them - not directly cross connected, but they are plugged into the same switch. I have lagg doing LACP on them. But, as the setup is identical network-wise for each box then I don't think it's that. I have this is sysctl.conf kern.ipc.maxsockbuf=1048576 net.inet.tcp.sendspace=131072 net.inet.tcp.recvspace=131072 I will give the recompile with changed contants a try (maybe later today) - do I need to do this at both ends though, as I want to avoid downtime if possible. cheers, -pete.