When using a RHEL maching to write the bootc image to a storage device, and then trying to use that device to boot another VM, the other VM fails to boot and become available on the network.

Steps to reproduce:

1. Run a RHEL PPC64LE machine in an IBM cloud PowerVS workspace using the available RHEL 9.4 AMI.
2. Setup subscriptions as needed
3. Create a new 100MB storage device and attach it to the VM, use scsi-rescan and multipath -l to attach and detect the name of the new device.
4. Use bootc-image-builder to write the RHEL 9.5 bootc image to a qcow2 file:

   sudo podman run --rm -it --privileged \
       --pull=newer \
       --security-opt label=type:unconfined_t \
       -v $HOME/bib-output:/output \
       -v /var/lib/containers/storage:/var/lib/containers/storage \
       quay.io/centos-bootc/bootc-image-builder:latest \
       --type qcow2 --use-librepo=True \
       registry.redhat.io/rhel9/rhel-bootc:9.5-1738698007
   

1. Use qemu-img to write the image to the attached storage device (DM device name will probably be different when reproducing):

    sudo qemu-img dd -O raw if=~/bib-output/qcow2/disk.qcow2 of=/dev/mapper/36005076813810264e800000000006eaa bs=$((8*64*1024))
   

1. Enable detaching the storage device:

    
   sudo blockdev --flushbufs /dev/mapper/36005076813810264e800000000006eaa
   
   for d in q r s t u v w x; do
     echo 1 >> /sys/block/sd$d/device/delete
   done
   

1. From the IBM cloud UI, detach the storage device and attach to a new PPC VM.
2. Start the new VM
3. Wait for it to come up and respond to ping