From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alasdair G Kergon Subject: Re: [linux-lvm] create vg with same name as exported vg Message-ID: <20031010204113.L6588@uk.sistina.com> References: <1065050019.1267.20.camel@rich> <1065126760.1280.0.camel@rich> <20031010200150.K6588@uk.sistina.com> <1065813644.1286.16.camel@rich> Mime-Version: 1.0 Content-Disposition: inline In-Reply-To: <1065813644.1286.16.camel@rich>; from rich@storix.com on Fri, Oct 10, 2003 at 12:20:44PM -0700 Sender: linux-lvm-admin@sistina.com Errors-To: linux-lvm-admin@sistina.com Reply-To: linux-lvm@sistina.com List-Help: List-Post: List-Subscribe: , List-Unsubscribe: , List-Archive: Date: Fri Oct 10 14:42:01 2003 List-Id: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Rich Turner Cc: linux-lvm@sistina.com On Fri, Oct 10, 2003 at 12:20:44PM -0700, Rich Turner wrote: > LVM1 actually handled this very well. With LVM1 it would rename the VG, > appending "PV_EXP" to the VG name when exporting it. LVM2 still does that for lvm1-format VGs of course. And for lvm2-format VGs it simply sets a flag in the metadata to say it's exported. No, the issue is that LVM2 is using the not-guaranteed-to-be-unique VG name internally in places where it ought to be using the unique VG uuid. And vgimport should work by VG uuid (in case there are several exported VGs present all with the same name) and require a new post-imported name to be specified if there's already a live VG with the same name. Alasdair -- agk@uk.sistina.com