Well the server did a kernel dump last night while doing the pvmove. Resulting in one logical volume being corrupted in the process. Not a big deal as the data is on tape. However now lvdisplay still lists the pvmove0 logical volume. So how do I clean up what was left behind?
-David On 5/31/05, Benjamin Smee <[EMAIL PROTECTED]> wrote: > lo, > > On Tuesday 31 May 2005 17:49, David Miller wrote: > > I've been running LVM2 now for some time but recently I have not > > been able to update device-mapper or lvm2 without vgscan failing to > > find all the pv's. So I'm stuck using lvm2-2.00.25 and device-mapper > > 1.00.19-r2. > > Now one thing to note is that a few of my pv's were added as the > > entire device rather than a partition. Could this be throwing a newer > > version for a loop since it may first be looking for an LVM partition > > before looking for a vg metadata? Or have I just overlooked something > > that I need to do in order to migrate my current vg's up to the newer > > versions? > > There is nothing that you need to do to "migrate" your lvm2 partitions to a > newer format, whatever problem that you have will be something local not > generic as many people are successfully using the latest versions of both > with no problems. You could start off running the vgscan in debug mode and > letting us know the relevant output, and if that fails try running it with a > strace or similar util to see precisely what is happening. Other pertinant > questions are things like, are you using udev? with persistant devices (ie > the tarball option in the rc.conf) what version of baselayout are you using, > do you have devfs enabled? > > regards, > > -- > Benjamin Smee (strerror) > 497F 5E98 1FA0 C313 EA0B 08C7 004A 66ED 448B E78C > > > -- [email protected] mailing list
