Comment 4 for bug 2009849

After having deleted the snapshot, I created a new one using horizon, then spawned a new VM on the volume. The VM failed to be spawned with the same message as above ("Invalid backing file") and deleting the snapshot resulted in the same error. qemu-img info showed the information below:

[root@phoenix ea003e1a1bee1aaf573aafc40cd188ff]# qemu-img info volume-e67af022-0d5b-4253-9981-c8b34622e058.cf3311ad-05eb-42fd-9b9c-1e9012a4a1f3
image: volume-e67af022-0d5b-4253-9981-c8b34622e058.cf3311ad-05eb-42fd-9b9c-1e9012a4a1f3
file format: qcow2
virtual size: 8 GiB (8589934592 bytes)
disk size: 196 KiB
cluster_size: 65536
backing file: volume-e67af022-0d5b-4253-9981-c8b34622e058
backing file format: qcow2
Format specific information:
    compat: 1.1
    compression type: zlib
    lazy refcounts: false
    refcount bits: 16
    corrupt: false
    extended l2: false
[root@phoenix ea003e1a1bee1aaf573aafc40cd188ff]#

So it would appear that either cinder creates a snapshot file with a wrong backing filename (bare volume filename with no path), or the check it performs before spawning a VM or deleting the snapshot (expect /var/lib/cinder/mnt/.../volume-...)