On Tue, Mar 12, 2002 at 08:04:01AM +0100, gunther.kuhlmann@web.de wrote: > On Mon, Mar 11, 2002, Heinz J . Mauelshagen wrote: > > > > I had two hard disks, the first with three VG's, the second > > > with one VG (vg2). When the second hard disk started to fail, I > > > installed a third with VG vg4 and copied the data over. Then, > > > without deleting vg2, I unplugged hd 2. After reboot, the > volume > > > on vg2 mounted cleanly, although the disk was not available. It > > > correctly accessed the LV on vg4 on the new disk. Is that > correct? > > > Can I now use either of the two vg2/vg4 to access the LV on the > > > new disk? How do I get rid of one (and which) without risking > my > > > data? Can I just vgremove vg2 away? > > > > Gunther, > > > > the result depends on *how* you copied vg2 over to your new hard > drive. > > > > If you used for eg. dd to do an image copy of the partitions, > which contained > > vg2 on the flaky drive or a whole disk copy keeping sizes, > you've got > +duplicated > > content and can unplug the flaky device. > > If you ran vgscan, you saw your 'old' vg2 in the new drive, > which is > > perfectly fine. > > No, I used pvcreate, vgcreate, lvcreate to create vg4 and a LV on > it. And than > rsync to copy the data. That is why I am a bit confused by this > behaviour. > Should I reconnect hd2, then delete vg2 and disconnect hd2 again? No need to. You still have /dev/vg2/ device nodes and the new vg4 got the same internal number so that you can access it through /dev/vg2/*. Either rename inactive vg4 to vg2 if you want to keep it by "rm -r /dev/vg2;vgrename vg4 vg2" or change your fstab to use the /dev/vg4 nodes and remove /dev/vg2/. /dev/vg2/* is left over, because you didn't remove vg2. Regards, Heinz -- The LVM Guy -- > > > BTW: your configuration is a little bit extraordinary in terms > of how to use > > drives as PVs. Typically you shouldn't have multiple VG's on a > single drive > > as you seem to have it. > > Much rather you should have multiple drives in one VG. > > Yes, I have one HD with two VG's. It might be unusual, but I can't > see anything > wrong with it. I do not intend to span a VG over more than one > physical disk, > because I don't want to have _part_ of a VG to fail. That's when > it gets messy. > I'm not saying it's impossible to recover, I just don't want to > try. > Besides, my HD size is large compared to my LV sizes, that's why I > needn't > bother. > > Gunther > > -- > Dipl.-Ing. Gunther Kuhlmann > gunther_kuhlmann@mentorg.com > gunther-kuhlmann@easy-call.net > Tel.: +49 (0)40 / 75 11 84 40 > PGP: E6 BC 78 6B E6 09 C7 16 AB 5D 9A 9A D7 1C 01 FB > -- > > > ______________________________________________________________________________ > Ohne Risiko spielend zum Börsenguru werden. Haben Sie schon eine > Strategie? Spekulieren und tolle Preis gewinnen http://boersenspiel.web.de > > > _______________________________________________ > linux-lvm mailing list > linux-lvm@sistina.com > http://lists.sistina.com/mailman/listinfo/linux-lvm > read the LVM HOW-TO at http://www.sistina.com/lvm/Pages/howto.html *** Software bugs are stupid. Nevertheless it needs not so stupid people to solve them *** =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Heinz Mauelshagen Sistina Software Inc. Senior Consultant/Developer Am Sonnenhang 11 56242 Marienrachdorf Germany Mauelshagen@Sistina.com +49 2626 141200 FAX 924446 =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- _______________________________________________ linux-lvm mailing list linux-lvm@sistina.com http://lists.sistina.com/mailman/listinfo/linux-lvm read the LVM HOW-TO at http://www.sistina.com/lvm/Pages/howto.html