Dump/Backup for running containers on btrfs not working with mode=snapshot (umount: /mnt/vzsnap0/: not mounted)

[patrickh]

As the title says, I get errors when trying to backup LXC containers whose volumes are on btrfs storage. This happens on multiple proxmox instances, even on freshly installed without anything configured/changed.

Storage config:

btrfs: local-btrfs
    path /var/lib/pve/local-btrfs
    content iso,vztmpl,backup,images,rootdir

Container config (also happens without mp0):

arch: amd64
features: keyctl=1,mknod=1,nesting=1
hostname: s01.lan
memory: 2048
mp0: local-btrfs:100/subvol-100-disk-1.subvol,mp=/volumes,backup=1,size=1
net0: name=eth0,bridge=vmbr0,hwaddr=BC:24:11:60:2D:28,ip=dhcp,ip6=dhcp,type=veth
onboot: 1
ostype: debian
parent: vzdump
rootfs: local-btrfs:100/subvol-100-disk-0.subvol,size=0T
swap: 512
timezone: host
unprivileged: 1

vzdump Log:

root@pve01:~# vzdump --compress zstd --mode snapshot --storage local-btrfs 100
INFO: starting new backup job: vzdump 100 --storage local-btrfs --mode snapshot --compress zstd
INFO: filesystem type on dumpdir is '-' -using /var/tmp/vzdumptmp1018801_100 for temporary files
INFO: Starting Backup of VM 100 (lxc)
INFO: Backup started at 2024-02-12 13:48:56
INFO: status = running
INFO: CT Name: s01.lan
INFO: including mount point rootfs ('/') in backup
INFO: including mount point mp0 ('/volumes') in backup
INFO: found old vzdump snapshot (force removal)
INFO: backup mode: snapshot
INFO: ionice priority: 7
INFO: suspend vm to make snapshot
INFO: create storage snapshot 'vzdump'
umount: /mnt/vzsnap0/: not mounted.
command 'umount -l -d /mnt/vzsnap0/' failed: exit code 32
INFO: resume vm
INFO: guest is online again after 1 seconds
ERROR: Backup of VM 100 failed - cannot mount subvol snapshots for storage type 'btrfs'
INFO: Failed at 2024-02-12 13:48:57
INFO: Backup job finished with errors
INFO: notified via target `mail-to-root`
job errors

Using mode=snapshot, mode=stop works always and mode=snapshot only works with a stopped container.

Is there some configuration I have to change or is this a bug/missing implementation in the btrfs storage implementation?