[SOLVED] Failed migration, lack of unused storage on desc server.

[kcybulski]

Hi,

I have proxmox 6.04 cluster, upgraded from 5.4.
When I try to do offline migration of VM between servers I get error:

storage 'local-ssd' is not available on node 'proxmox3'

The local-ssd storage is not used by this VM.
What can cause this error ?

VM configuration and full log below.

2019-07-22 15:06:37 starting migration of VM 105 to node 'proxmox3' (192.168.168.229)
2019-07-22 15:06:38 ERROR: Failed to sync data - storage 'local-ssd' is not available on node 'proxmox3'
2019-07-22 15:06:38 aborting phase 1 - cleanup resources
2019-07-22 15:06:38 ERROR: migration aborted (duration 00:00:01): Failed to sync data - storage 'local-ssd' is not available on node 'proxmox3'
TASK ERROR: migration aborted

cat /etc/pve/qemu-server/105.conf

gent: 1
balloon: 1024
bios: ovmf
bootdisk: scsi0
cores: 4
efidisk0: local:105/vm-105-disk-1.qcow2,size=128K
hotplug: disk,network,usb
ide2: none,media=cdrom
memory: 4096
name: SV-ADMIN
net0: virtio=00:15:5d:a8:f6:09,bridge=vmbr0
numa: 0
onboot: 1
ostype: l26
scsi0: local:105/vm-105-disk-0.qcow2,cache=writeback,discard=on,size=50G
scsi1: local:105/vm-105-disk-3.qcow2,cache=writeback,discard=on,size=127G
scsihw: virtio-scsi-pci
smbios1: uuid=f592e65e-ebe5-4e3e-8168-6940ef120ceb
sockets: 1
vga: qxl

 

[dietmar]

My guess is that storage local-ssd was previously used, and there are still (unused) disks on that storage.

What is the output of

# pvesm list local-ssd | grep 105

 

[kcybulski]

pvesm list local-ssd | grep 105
local-ssd:105/vm-105-disk-0.qcow2 qcow2 136365211648 105

I can't delete the the disk from local-ssd

Cannot remove image, a guest with VMID '105' exists!
You can delete the image from the guest's hardware pane

But in Hardware tab of VM there is no unassigned disks

 

[dietmar]

Try using the command line tool:

# pvesm free local-ssd:105/vm-105-disk-0.qcow2

 

[kcybulski]

The migration started now. Thanks.

 

[ITWarrior]

Dear future Googlers. I too had this issue, except that my disks are ZFS based:

root@node03:~# pvesm list node03-MZ | grep 107
node03-MZ:vm-107-disk-0 raw     images     85899345920 107
root@node03:~#

I'd tried several commands similar to the above (and from other threads), but none seem to do the job:

root@node03:~# pvesm free node03-MZ:107/vm-107-disk-0
unable to parse zfs volume name '107/vm-107-disk-0'
root@node03:~# pvesm free node03-MZ:107/vm-107-disk-0.raw
unable to parse zfs volume name '107/vm-107-disk-0.raw'
root@node03:~# pvesm free node03-MZ:107/vm-107-disk-0 raw
400 too many arguments
pvesm free <volume> [OPTIONS]
root@node03:~# pvesm free node03-MZ:/107/vm-107-disk-0
unable to parse zfs volume name '/107/vm-107-disk-0'
root@node03:~# pvesm free node03-MZ:/107/vm-107-disk-0.raw
unable to parse zfs volume name '/107/vm-107-disk-0.raw'
root@node03:~# pvesm free node03-MZ:107/vm-107-disk-0 raw
400 too many arguments
pvesm free <volume> [OPTIONS]
root@node03:~#

You have to drop the VM ID off, like this:

root@node03:~# pvesm free node03-MZ:vm-107-disk-0                                                                                                                                                                                         
Removed volume 'node03-MZ:vm-107-disk-0'

 

[JamesT]

Dear future Googlers. I too had this issue, except that my disks are ZFS based:

You have to drop the VM ID off, like this:

root@node03:~# pvesm free node03-MZ:vm-107-disk-0                                                                                                                                                                                        
Removed volume 'node03-MZ:vm-107-disk-0'

Came here to say this, but you've already done it. I should have read further down it would have saved me 10 mins. Good on you mate.

 

[aghadjip]

Dear future Googlers. I too had this issue, except that my disks are ZFS based:

root@node03:~# pvesm list node03-MZ | grep 107
node03-MZ:vm-107-disk-0 raw     images     85899345920 107
root@node03:~#

I'd tried several commands similar to the above (and from other threads), but none seem to do the job:

root@node03:~# pvesm free node03-MZ:107/vm-107-disk-0
unable to parse zfs volume name '107/vm-107-disk-0'
root@node03:~# pvesm free node03-MZ:107/vm-107-disk-0.raw
unable to parse zfs volume name '107/vm-107-disk-0.raw'
root@node03:~# pvesm free node03-MZ:107/vm-107-disk-0 raw
400 too many arguments
pvesm free <volume> [OPTIONS]
root@node03:~# pvesm free node03-MZ:/107/vm-107-disk-0
unable to parse zfs volume name '/107/vm-107-disk-0'
root@node03:~# pvesm free node03-MZ:/107/vm-107-disk-0.raw
unable to parse zfs volume name '/107/vm-107-disk-0.raw'
root@node03:~# pvesm free node03-MZ:107/vm-107-disk-0 raw
400 too many arguments
pvesm free <volume> [OPTIONS]
root@node03:~#

You have to drop the VM ID off, like this:

root@node03:~# pvesm free node03-MZ:vm-107-disk-0                                                                                                                                                                                        
Removed volume 'node03-MZ:vm-107-disk-0'

Still helping folks with this years later mostly commenting so i can find this later if needed