While doing a little house cleaning on a VM, I discovered I was unable to reboot it. Basically, I had a logical volume that was rather large and wanted to remove it. I used the lvremove to remove the logical volume that spanned across two hard drives. There was no other logical volume on those drives. Shutdown the server and deleted the two drives. Rebooted the server, and the server would not start up.
I was able to boot into the emergency mode. pvdisplay revealed that it “Coudn’t find divice uuid …..” and making note of the VG Name for the volume group.
The solution.
vgreduce --removemissing --force offendingvolumegroup
The result of the command.
# vgreduce --removemissing --force offendingvolumegroup Couldn't find device with uuid xxxxxx-xxxx-xxxx-xxxx-xxxx-xxxx-xxxxxx. Removing partial LV LogVol00. Logical volume "LogVol00" successfully removed Wrote out consistent volume group offendingvolumegroup
pvdisplay didn’t reveal any issues.
Rebooted and the system was happy.
Source(s)
- https://serverfault.com/questions/319427/how-to-remove-bad-disk-from-lvm2-with-the-less-data-loss-on-other-pvs