Take the 2-minute tour ×
Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

I think I've cocked up a smidge, so asking for some advice.

I have 4 drives in my machine: sda - OS sdb-d - sdb1, sdc1, sdd1 - swap sdb2, sdc2, sdd2 - LVM

Total logical volume size 2.7Tb

What I've done in the past is to upgrade the OS on sda, and during the install process it has spotted the LVM, and let me specify the mount point.

Last week, I set about this, and all seemed well, moving over to PCLinuxOS, until it came to the restart - sdb2 reported as an LVM partition, but no data present. So the LVM wouldn't mount.

The sdx2 partitions are all around 923Gb.

Although I have a backup of the majority of the data, there is some that has been created in the mean time that I would like to recover.

Hexediting /dev/sdx2 I can see the data

I don't have a metadata backup for the partitions

I have tried using the PCLinuxOS LVM manager, but that shows an empty non-accessible area.

Does anyone have any advice/suggestions for me to recover the data?

I've tried using photorec, and whilst that is finding lots of good data, I have no idea what the correct filenames should be...

I have also worked through a number of google articles, but they all assume that you have a backup of the .vg file to be able to recreate the metadata

$pvs
PV        VG       Fmt  Attr PSize   PFree
/dev/sdb2 LogVol00 lvm2 a-   923.51g 923.51g
/dev/sdc2 LogVol00 lvm2 a-   923.51g 923.51g
/dev/sdd2 LogVol00 lvm2 a-   923.51g 923.51g

$pvscan
PV /dev/sdb2 VG LogVol00 lvm2 [923.51 GiB / 923.51 GiB free]
PV /dev/sdc2 VG LogVol00 lvm2 [923.51 GiB / 923.51 GiB free]
PV /dev/sdd2 VG LogVol00 lvm2 [923.51 GiB / 923.51 GiB free]
Total: 3 [2.71 TiB] / in use: 3 [2.71 TiB] / in no VG: 0 [0 

--- Volume group ---
  VG Name               LogVol00
  System ID             
  Format                lvm2
  Metadata Areas        3
  Metadata Sequence No  3
  VG Access             read/write
  VG Status             resizable
  MAX LV                0
  Cur LV                0
  Open LV               0
  Max PV                0
  Cur PV                3
  Act PV                3
  VG Size               2.71 TiB
  PE Size               4.00 MiB
  Total PE              709254
  Alloc PE / Size       0 / 0   
  Free  PE / Size       709254 / 2.71 TiB
  VG UUID               OaOPee-jkxz-d06i-fRi9-ODs0-672J-dxqqly

]

share|improve this question
    
What's the output of pvs and pvscan ? –  Darth Android May 16 '13 at 16:15
    
The current output shows pvs PV VG Fmt Attr PSize PFree /dev/sdb2 LogVol00 lvm2 a- 923.51g 923.51g /dev/sdc2 LogVol00 lvm2 a- 923.51g 923.51g /dev/sdd2 LogVol00 lvm2 a- 923.51g 923.51g pvscan PV /dev/sdb2 VG LogVol00 lvm2 [923.51 GiB / 923.51 GiB free] PV /dev/sdc2 VG LogVol00 lvm2 [923.51 GiB / 923.51 GiB free] PV /dev/sdd2 VG LogVol00 lvm2 [923.51 GiB / 923.51 GiB free] Total: 3 [2.71 TiB] / in use: 3 [2.71 TiB] / in no VG: 0 [0 ] that is where I am now - this is after following some of the suggestions that I googled –  Si Kellow May 16 '13 at 19:35
    
gah - I don't seem to be able to get the reply to format clearly –  Si Kellow May 16 '13 at 19:35
    
I formatted your information into the question. The good news is it looks like it automatically found all your physical volumes and activated them successfully. All your data is there. I imagine the shortcoming is that the volume group, while active, is not actually mounted. –  Darth Android May 16 '13 at 20:07
    
sadly not - those are the recreated pv's that I made following a piece about recreating the PV with no metadata - then I got to the bit that did a vcfgrestore and I havent got the .vg files - if i dhex /dev/sdb2 sdc2 and sdd2 I can see data, so at least I haven't completely mullered, but looking at the metadata tables, my attempts have overwritten any original metadata. –  Si Kellow May 16 '13 at 21:08

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.