Hello,
Recently I was forced to hard-reboot a server that is running Proxmox because it was not responsive (frozen). After it rebooted, however, all of the VM disks (which are stored on an LVM thin volume) are inaccessible, with an error like
However, all of the VM disks still show up in lvs -a:
I already tried to run lvconvert --repair, but I get an error WARNING: recovery of pools without pool metadata spare LV is not automated.
When I try to make the VM disks active (available state), I get an error:
However, it is still possible to make the LVM-thin volume itself (vg0/pve-data) active, just not any of the volumes inside the LVM-thin volume.
Can anyone help me with this?
Thanks!
Recently I was forced to hard-reboot a server that is running Proxmox because it was not responsive (frozen). After it rebooted, however, all of the VM disks (which are stored on an LVM thin volume) are inaccessible, with an error like
Code:
Could not open '/dev/vg0/vm-106-disk-1': No such file or directoryHowever, all of the VM disks still show up in lvs -a:
Code:
root@HETZNER-E3-NUE ~/thin-provisioning-tools/bin # lvs -a
LV VG Attr LSize Pool Origin Data% Meta% Move Log Cpy%Sync Convert
base-100001-disk-1 vg0 Vri---tz-k 10.00g pve-data
pve-data vg0 twi---tz-- 1.56t
pve-data_meta0 vg0 -wi-a----- 1.21g
pve-data_meta1 vg0 -wi-a----- 128.00m
pve-data_meta10 vg0 -wi-a----- 128.00m
pve-data_meta11 vg0 -wi-a----- 128.00m
pve-data_meta12 vg0 -wi-a----- 384.00m
pve-data_meta2 vg0 -wi-a----- 128.00m
pve-data_meta3 vg0 -wi-a----- 128.00m
pve-data_meta4 vg0 -wi-a----- 128.00m
pve-data_meta5 vg0 -wi-a----- 128.00m
pve-data_meta6 vg0 -wi-a----- 128.00m
pve-data_meta7 vg0 -wi-a----- 128.00m
pve-data_meta8 vg0 -wi-a----- 220.00m
pve-data_meta9 vg0 -wi-a----- 128.00m
[pve-data_tdata] vg0 Twi-a----- 1.56t
[pve-data_tmeta] vg0 ewi-a----- 384.00m
root vg0 -wi-ao---- 250.00g
swap vg0 -wi-ao---- 16.00g
vm-100-disk-1 vg0 Vwi---tz-- 20.00g pve-data
vm-100001-cloudinit vg0 Vwi---tz-- 8.00m pve-data
vm-10002-disk-1 vg0 Vwi---tz-- 15.00g pve-data
vm-10007-disk-1 vg0 Vwi---tz-- 50.00g pve-data
vm-102-disk-1 vg0 Vwi---tz-- 5.00g pve-data
vm-102-disk-2 vg0 Vwi---tz-- 50.00g pve-data
vm-103-disk-1 vg0 Vwi---tz-- 60.00g pve-data
vm-104-disk-1 vg0 Vwi---tz-- 55.00g pve-data
vm-105-disk-1 vg0 Vwi---tz-- 20.00g pve-data
vm-106-disk-1 vg0 Vwi---tz-- 50.00g pve-data
vm-107-disk-1 vg0 Vwi---tz-- 20.00g pve-data
vm-109-disk-1 vg0 Vwi---tz-- 20.00g pve-data
vm-110-cloudinit vg0 Vwi---tz-- 8.00m pve-data
vm-110-disk-1 vg0 Vwi---tz-- 15.00g pve-data
vm-110-disk-2 vg0 Vwi---tz-- 500.00g pve-data
vm-111-disk-1 vg0 Vwi---tz-- 20.00g pve-data
vm-111-disk-2 vg0 Vwi---tz-- 500.00g pve-data
vm-112-disk-1 vg0 Vwi---tz-- 60.00g pve-data
vm-112-disk-2 vg0 Vwi---tz-- 4.00m pve-dataI already tried to run lvconvert --repair, but I get an error WARNING: recovery of pools without pool metadata spare LV is not automated.
When I try to make the VM disks active (available state), I get an error:
Code:
root@HETZNER-E3-NUE ~/thin-provisioning-tools/bin # vgchange -ay vg0
device-mapper: reload ioctl on (253:14) failed: No data available
device-mapper: reload ioctl on (253:14) failed: No data available
device-mapper: reload ioctl on (253:14) failed: No data available
device-mapper: reload ioctl on (253:14) failed: No data available
device-mapper: reload ioctl on (253:14) failed: No data available
device-mapper: reload ioctl on (253:14) failed: No data available
device-mapper: reload ioctl on (253:14) failed: No data available
device-mapper: reload ioctl on (253:14) failed: No data available
device-mapper: reload ioctl on (253:14) failed: No data available
device-mapper: reload ioctl on (253:14) failed: No data available
device-mapper: reload ioctl on (253:14) failed: No data available
device-mapper: reload ioctl on (253:14) failed: No data available
device-mapper: reload ioctl on (253:14) failed: No data available
device-mapper: reload ioctl on (253:14) failed: No data available
device-mapper: reload ioctl on (253:14) failed: No data available
device-mapper: reload ioctl on (253:14) failed: No data available
device-mapper: reload ioctl on (253:14) failed: No data available
device-mapper: reload ioctl on (253:14) failed: No data available
device-mapper: reload ioctl on (253:14) failed: No data available
16 logical volume(s) in volume group "vg0" now activeHowever, it is still possible to make the LVM-thin volume itself (vg0/pve-data) active, just not any of the volumes inside the LVM-thin volume.
Can anyone help me with this?
Thanks!