Hello all I have two sets of eIDE hard drives from earlier servers, one centos & one fedora. Both were LVM volumes with three or four physical disks, with ext3 fs. One disk, maybe even the boot one may even be missing, either from one or both sets and we do not know the disk order. I got these left from an earlier sysadmin who left the company & nobody know what's what. I need to recover data so I need to backup each of the disks to a disk image file first, then find block-size/recover lvm metadata on the volume, mount the volume to my current server and recover whatever files can be recovered manually. My question is: - what switches to be used with dd in creating the image, so that I retain the lvm disk data, for future recovery, if I mess up the recovery job? Can you please give the actual command assuming disk is connected at /dev/sdb? - How do I find the blocksize used by the disk & how do I get the meta-data and mount the disk to current lvm fs. Request please give me step by step process. I am finding the howtos & examples on the net very daunting as most of them have co-issues of raid etc. Mine is a plain lvm spanning multiple disks. I don't have expertise of this level & suddenly my boss is expecting me to perform a miracle. Please help. With best regards. Sanjay.
On Thu, 2011-02-03 at 16:28 +0530, Sanjay Arora wrote:> Hello all > > I have two sets of eIDE hard drives from earlier servers, one centos & > one fedora. Both were LVM volumes with three or four physical disks, > with ext3 fs. One disk, maybe even the boot one may even be missing, > either from one or both sets and we do not know the disk order. I got > these left from an earlier sysadmin who left the company & nobody know > what's what. > > I need to recover data so I need to backup each of the disks to a disk > image file first, then find block-size/recover lvm metadata on the > volume, mount the volume to my current server and recover whatever > files can be recovered manually. > > My question is: > > - what switches to be used with dd in creating the image, so that I > retain the lvm disk data, for future recovery, if I mess up the > recovery job? Can you please give the actual command assuming disk is > connected at /dev/sdb? > > - How do I find the blocksize used by the disk & how do I get the > meta-data and mount the disk to current lvm fs. > > Request please give me step by step process. I am finding the howtos & > examples on the net very daunting as most of them have co-issues of > raid etc. Mine is a plain lvm spanning multiple disks. > > I don't have expertise of this level & suddenly my boss is expecting > me to perform a miracle.man scanvg
also, if the disk is multi-disk, which is the case here, you can look into
/etc/lvm/backup for LVM meta data information.
----- Original Message -----
| Hello all
|
| I have two sets of eIDE hard drives from earlier servers, one centos &
| one fedora. Both were LVM volumes with three or four physical disks,
| with ext3 fs. One disk, maybe even the boot one may even be missing,
| either from one or both sets and we do not know the disk order. I got
| these left from an earlier sysadmin who left the company & nobody know
| what's what.
|
| I need to recover data so I need to backup each of the disks to a disk
| image file first, then find block-size/recover lvm metadata on the
| volume, mount the volume to my current server and recover whatever
| files can be recovered manually.
|
| My question is:
|
| - what switches to be used with dd in creating the image, so that I
| retain the lvm disk data, for future recovery, if I mess up the
| recovery job? Can you please give the actual command assuming disk is
| connected at /dev/sdb?
|
| - How do I find the blocksize used by the disk & how do I get the
| meta-data and mount the disk to current lvm fs.
|
| Request please give me step by step process. I am finding the howtos &
| examples on the net very daunting as most of them have co-issues of
| raid etc. Mine is a plain lvm spanning multiple disks.
|
| I don't have expertise of this level & suddenly my boss is expecting
| me to perform a miracle.
|
| Please help.
|
| With best regards.
| Sanjay.
| _______________________________________________
| CentOS mailing list
| CentOS at centos.org
| http://lists.centos.org/mailman/listinfo/centos
--
James A. Peltier
IT Services - Research Computing Group
Simon Fraser University - Burnaby Campus
Phone : 778-782-6573
Fax : 778-782-3045
E-Mail : jpeltier at sfu.ca
Website : http://www.sfu.ca/itservices
http://blogs.sfu.ca/people/jpeltier