From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mx1.redhat.com (ext-mx16.extmail.prod.ext.phx2.redhat.com [10.5.110.21]) by int-mx12.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id r042uq9m032713 for ; Thu, 3 Jan 2013 21:56:53 -0500 Received: from hoptical.rarforge.com (hoptical.rarforge.com [207.115.88.84]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id r042uonr029093 for ; Thu, 3 Jan 2013 21:56:51 -0500 Received: from localhost (localhost [127.0.0.1]) by hoptical.rarforge.com (Postfix) with ESMTP id 284A7AA244 for ; Thu, 3 Jan 2013 18:56:50 -0800 (PST) Received: from hoptical.rarforge.com ([127.0.0.1]) by localhost (hoptical.rarforge.com [127.0.0.1]) (amavisd-new, port 10026) with ESMTP id JAXVVGJkk4IF for ; Thu, 3 Jan 2013 18:56:49 -0800 (PST) Received: from [10.0.0.111] (salish.linuxjunkie.com [207.115.66.170]) by hoptical.rarforge.com (Postfix) with ESMTPSA id 48EBFA096F for ; Thu, 3 Jan 2013 18:56:49 -0800 (PST) Message-ID: <50E644EF.6060209@rarforge.com> Date: Thu, 03 Jan 2013 18:56:47 -0800 From: Rob MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Subject: [linux-lvm] LVM snapshot with Clustered VG Reply-To: lists@rarforge.com, 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"; format="flowed" To: linux-lvm@redhat.com Is there a way to convert an active LV on a clustered VG to an exclusive lock without deactivating it first? Or some way of taking a snapshot of a clustered LVM? I know I have read snapshots are not possible on CLVM, but the information seems outdated. ha node1: /dev/vg_clus/logvol_clus ha node2: /dev/vg_clus/logvol_clus 1) I can deactivate the the entire vg on node1 and node2 with 'vgchange -an vg_clus' and activate with -ay. Just stating cluster vg is working. 2) I can deactivate locally on node2 with 'vgchange -aln vg_clus' , it shows inactive on node2 and active still on node1 - After deactivating, I cannot get exclusive lock on node1 with 'vgchange -aey vg_clus' I get an error locking on node1 3) I can get exclusive an lock if I deactivate the VG on both nodes and then run 'vgchange -aey vg_clus' on either node. The lvcreate snapshop works, but again, it requires deactivating the vg, which isn't an option in production. Setup: Centos 6.3 (current) cman+pacemaker dual primary drbd for HA KVM w/ live migration disk: lvm -> drbd -> clvm -> kvm virt Currently I am taking a lvm snapshot of the drbd lvm backing device, but it's not ideal ( as there are other issues ) Thanks for the time! -Rob