From mboxrd@z Thu Jan 1 00:00:00 1970 From: Miles Fidelman Subject: Re: ceph for small cluster? Date: Mon, 31 Dec 2012 09:14:29 -0500 Message-ID: <50E19DC5.303@meetinghouse.net> References: <50E0B457.70200@meetinghouse.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from server1.neighborhoods.net ([207.154.13.48]:40564 "EHLO server1.neighborhoods.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751081Ab2LaOOb (ORCPT ); Mon, 31 Dec 2012 09:14:31 -0500 Received: from localhost (localhost.localdomain [127.0.0.1]) by server1.neighborhoods.net (Postfix) with ESMTP id BB958CC0C3 for ; Mon, 31 Dec 2012 09:14:30 -0500 (EST) Received: from server1.neighborhoods.net ([127.0.0.1]) by localhost (server1.neighborhoods.net [127.0.0.1]) (amavisd-new, port 10024) with LMTP id gQyZa+WnlGct for ; Mon, 31 Dec 2012 09:14:29 -0500 (EST) Received: from new-host-2.home (pool-71-184-234-147.bstnma.fios.verizon.net [71.184.234.147]) by server1.neighborhoods.net (Postfix) with ESMTPSA id 9C462CC03C for ; Mon, 31 Dec 2012 09:14:29 -0500 (EST) In-Reply-To: Sender: ceph-devel-owner@vger.kernel.org List-ID: Cc: ceph-devel Matt, Thanks for the comments. A follow-up if I might (inline): Matthew Roy wrote: > What I'm not doing that you'd need to test is running VMs on the same > servers as storage. I'd be careful about mounting RBD volumes on the > OSDs, you can run into kernel deadlock trying to write out things to > physical disk when trying to write to the mounted volume. Mounts > inside VMs should be okay. I was thinking of running pinning one CPU to DomO and running the OSD from there, and mounting RBD volumes only in DomUs. And leaving a bit of disk space outside the OSD for booting and Dom0. Which raises another question: how are you combining drives within each OSD (raid, lvm, ?). Thanks again, Miles -- In theory, there is no difference between theory and practice. In practice, there is. .... Yogi Berra