[linux-lvm] data loss because of mixing LVM and LVM2

[Jürgen Lange]

Sorry, I forgot the attachment,


Hello,

I made the error in using pvmove and vgreduce from a SuSE 9.2 bootdisk
(kernel 2.6.x) and LVM2 on a LVM created by SuSE 8.2 (kernel 2.4.20) and LVM.
in the attached file lvm_crash.txt I add the exact output from vgscan, 
vgmknodes, vgimport, pvdata and pvdisplay. (the stderr output of the same 
commands is in the other file lvm_crash_2.txt)

Basically vgscan reports:
  vgscan -- reading all physical volumes (this may take a while...)
  vgscan -- found active volume group "sysvg"
  vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get data of
                volume group "systemvg" from physical volume(s)
  vgscan -- "/etc/lvmtab" and "/etc/lvmtab.d" successfully created
(I created another VG, sysvg, in the meantime.)

I believe that all PEs (i.e. all of my lost data) are on my /dev/hda7, as 
pvdisplay on the other PVs of my VG systemvg always used to report:
"Allocated PE 0".
However, the VG "systemvg" is not recognised anymore.
I left the /dev/hda7 partition untouched - but I repartitioned the disk where 
the other (presumebly empty) PVs were on.

I put the last systemvg.conf file in the attachment as well (from probably 
before the vgreduce actually (not) happened).

Do you see a chance in recovering my data?
Thanks a lot,

-- 
Regards,
J. Lange
-------------- next part --------------
A non-text attachment was scrubbed...
Name: lvm.tgz
Type: application/x-tgz
Size: 19715 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/linux-lvm/attachments/20050127/c6c7b836/attachment.bin>