Description
An error appears in the interface when attempting to create a VM backup.
To view details:
- In the Virtual Machines section, select the VM.
- Enter Parameters → Information → Last 10 events.
- Click on the Failed status in the Event column. This displays the Task Information tab with the detailed error text:
Unhandled exception: Error message: Creating snapshot fail
Command exited with message error: unsupported configuration: nothing selected for snapshot
The error occurs if the VM snapshot already exists.
Solution
To resolve the issue:
- Connect to the node server via SSH.
-
Get a list of virtual disks:
virsh domblklist <libvirt_domain>
CommentExample output:
# virsh domblklist 2698_test-alma8 Target Source ------------------------------------------------ vda /vm2/2711_test-alma8.2698_test-alma8_snapshot
The output shows that the VM already has a snapshot.
-
Compare the modification dates of the virtual disk master file and the snapshot. Determine which file was modified later:
stat <virtual_disk_file> stat <snapshot_file>
Example output for the virtual disk master file. See Modify value:
# stat /vm2/2711_test-alma8 File: /vm2/2711_test-alma8 Size: 2419458048 Blocks: 4725400 IO Block: 4096 regular file Device: 902h/2306d Inode: 23330831 Links: 1 Access: (0644/-rw-r--r--) Uid: ( 107/ qemu) Gid: ( 107/ qemu) Access: 2023-11-20 12:04:25.826960539 +0200 Modify: 2023-11-20 12:06:15.420142110 +0200 Change: 2023-11-20 12:06:15.420142110 +0200 Birth: 2023-10-26 08:38:45.574566975 +0300
- The next steps depend on which file was modified later and is relevant:
-
if the disk file is more relevant, delete the snapshot:
rm <snapshot_file>
-
If the more relevant file is the snapshot file:
-
Merge the file data:
virsh blockcommit <libvirt_domain> vda --active --verbose --pivot
CommentExample reply for a successful merge:
# virsh blockcommit 2698_test-alma8 vda --active --verbose --pivot Block commit: [100 %] Successfully pivoted
-
Delete the snapshot:
rm <snapshot_file>
-
-
- Repeat the backup.