Qemu to talk to glusterfs - fails?

[lejeczek]

Hi guys.

I have libvirtd to manage my VMs and it all works well 
including backend to the storage being glusterfs, but when I 
try something simple, for the first time, this:

-> $ qemu-img create -f qcow2 
gluster://swir.direct:24007/VMs/fcos-node1.qcow2 20G
Formatting 
'gluster://swir.direct:24007/VMs/fcos-node1.qcow2', 
fmt=qcow2 cluster_size=65536 extended_l2=off 
compression_type=zlib size=21474836480 lazy_refcounts=off 
refcount_bits=16
[2021-04-21 14:10:08.923166] E [MSGID: 108006] 
[afr-common.c:6074:__afr_handle_child_down_event] 
0-VMs-replicate-0: All subvolumes are down. Going offline 
until at least one of them comes back up.

Being a novice I assume I'm missing something obvious there, 
rather I'm looking at a bug. What is the issue here?
I'm on CentOS with glusterfs-8.4, libvirt-7 and qemu-kvm-5.2.0
many thanks, L.