[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

[linux-lvm] Lost a VG - please help



I have just upgraded a machine running Redhat
9 to Whitebox Linux 3.  Unfortunately, it
appears to have killed/destructively
reconfigured my LVM setup.

The original config consisted of:

/dev/hda1	
	- Regular ext3 Linux partition mounted as /boot
/dev/hda2	
	- LVM partition containing:
	vg_system
		- VG with several LVs containing the OS.

A RAID5 md device, /dev/md0, consisting of:
	/dev/hdc1
	/dev/hde1
	/dev/hdg1
	/dev/hdi1
	/dev/hdk1

/dev/md0 was configured as a single VG,
vg_data.  This VG had about 5 LVs.

Before installing the new system, I
'vgexport'ed vg_data as indicated in the LVM
HOWTO
(http://www.tldp.org/HOWTO/LVM-HOWTO/recipemovevgtonewsys.html):

(Unmount all relevant filesystems)

'vgchange -an vg_data'
          
'vgexport vg_data'


I "upgraded" by replacing /dev/hda with a new
drive and installing WBL straight onto the
new drive.  The LVM config I used for the new
drive was:

/dev/hda1
	- Regular ext3 Linux partition mounted as /boot
/dev/hda2	
	- LVM partition containing:
	vg_wbl
		- VG with a single LV containg the OS

Unfortunately, the VG that was on /dev/md0 -
vg_data - appears to have disappeared and
/dev/md0 appears to have been "merged" into
vg_wbl.  The commands for importing the old
VG do not give the results I was expecting:

[root terminus root]# pvscan
pvscan -- reading all physical volumes (this
may take a while...)
pvscan -- ACTIVE   PV "/dev/md0"  of VG
"vg_wbl" [447.14 GB / 447.14 GB free]
pvscan -- WARNING: physical volume
"/dev/hde1" belongs to a meta device
pvscan -- WARNING: physical volume
"/dev/hdc1" belongs to a meta device
pvscan -- ACTIVE   PV "/dev/hda2" of VG
"vg_wbl" [37.16 GB / 25.40 GB free]
pvscan -- total: 4 [484.32 GB] / in use: 4
[484.32 GB] / in no VG: 0 [0]

[root terminus root]#

I do not think /dev/md0 should be a PV of
vg_wbl....

Unsurprisingly, the next step of the import
didn't work:

[root terminus root]# vgimport vg_data /dev/md0
vgimport -- ERROR "pv_read(): multiple
device" reading physical volume "/dev/md0"

[root terminus root]#


I'd appreciate any help people can offer
recovering my data.  It would appear the
470Gb VG that was on /dev/md0 has disappeared
in a puff of electrons and /dev/md0 has
somehow been "sucked in" to the new vg_wbl VG
I created while installing the OS.  I'm sure
I'll need to provide more info, but I'm
unsure as to what it is - please don't
hesistate to ask.

Regards,
CS

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]