From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mx3.redhat.com (mx3.redhat.com [172.16.48.32]) by int-mx1.corp.redhat.com (8.12.11.20060308/8.12.11) with ESMTP id k7O6svV8007417 for ; Thu, 24 Aug 2006 02:54:57 -0400 Received: from mail-svr1.cs.utah.edu (mail-svr1.cs.utah.edu [155.98.64.241]) by mx3.redhat.com (8.13.1/8.13.1) with ESMTP id k7O6shwr014444 for ; Thu, 24 Aug 2006 02:54:43 -0400 Received: from localhost (localhost [127.0.0.1]) by mail-svr1.cs.utah.edu (Postfix) with ESMTP id A295565007F for ; Thu, 24 Aug 2006 00:54:37 -0600 (MDT) Received: from mail-svr1.cs.utah.edu ([127.0.0.1]) by localhost (mail-svr1.cs.utah.edu [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 22322-04 for ; Thu, 24 Aug 2006 00:54:35 -0600 (MDT) Received: from trust.cs.utah.edu (trust.cs.utah.edu [155.98.65.28]) by mail-svr1.cs.utah.edu (Postfix) with ESMTP id E18A8650071 for ; Thu, 24 Aug 2006 00:54:35 -0600 (MDT) Received: from localhost (localhost [127.0.0.1]) by trust.cs.utah.edu (Postfix) with ESMTP id B58FB3F71 for ; Thu, 24 Aug 2006 00:54:35 -0600 (MDT) Date: Thu, 24 Aug 2006 00:54:35 -0600 (MDT) From: Prashanth Radhakrishnan Message-ID: MIME-Version: 1.0 Subject: [linux-lvm] Snapshot of an LVM snapshots? Reply-To: LVM general discussion and development List-Id: LVM general discussion and development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , List-Id: Content-Type: TEXT/PLAIN; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-lvm@redhat.com Hi LVM Gurus, >From what I've read around, it is not possible to take a snapshot of an LVM snapshot. Is this something that is worked upon currently as a future enhancement? If not, why not? To give you a background, we're working on a periodic (xen) VM checkpointing project, using LVM for VM disk checkpointing. As we take several snapshots of the same volume, the write throughput reduces drastically. I conjecture this is because, for each write to the (base) volume, each of the snapshots' mappings is checked to decide whether to copy the blocks to the snapshot or not. It seems that this can be helped by: - taking a snapshot of the latest snapshot and having the writes go to the new snapshot or - making LVM use only the latest snapshot for CoW as writes go to the base volume. Any suggestions/pointers greatly appreaciated. thanks, Prashanth