VM from template can't start too.
Cluster version 4.2 (LibgfApiSupported=true).
VM rXX-TEST is down with error. Exit message: internal error: qemu unexpectedly closed the monitor: [2020-10-13 10:29:08.503719] E [MSGID: 108006] [afr-common.c:5323:__afr_handle_child_down_event] 0-169-gluster-volume02-replicate-0: All subvolumes are down. Going offline until at least one of them comes back up.
[2020-10-13 10:29:08.503926] I [io-stats.c:4027:fini] 0-169-gluster-volume02: io-stats translator unloaded
2020-10-13T10:29:10.346661Z qemu-kvm: -drive file=gluster://10.169.2.227:24007/169-gluster-volume02/a706d3a9-c73d-474e-9544-6ca9491037d2/images/cf44e622-3f88-4dff-a883-ac18a1ff29f5/94ad4bcf-f56a-4695-9b8e-3340dd3d7665,file.debug=4,format=qcow2,if=none,id=drive-ua-cf44e622-3f88-4dff-a883-ac18a1ff29f5,serial=cf44e622-3f88-4dff-a883-ac18a1ff29f5,werror=stop,rerror=stop,cache=none,aio=native: Could not read L1 table: Input/output error.
rpm -qa gluster*
glusterfs-client-xlators-7.7-1.el7.x86_64
glusterfs-rdma-7.7-1.el7.x86_64
glusterfs-api-7.7-1.el7.x86_64
glusterfs-libs-7.7-1.el7.x86_64
glusterfs-fuse-7.7-1.el7.x86_64
glusterfs-geo-replication-7.7-1.el7.x86_64
glusterfs-server-7.7-1.el7.x86_64
glusterfs-cli-7.7-1.el7.x86_64
glusterfs-7.7-1.el7.x86_64
glusterfs-events-7.7-1.el7.x86_64
Setting the value to LibgfApiSupported=false for a new cluster (version 4.3) solve the problem with starting the VM.
Is there a workaround to fix the LibgfApi?