DetachVolume failure: Disk is not getting unplugged from Domain

[kasabe28]

Describe the bug
Disk is not getting unplugged from domain when DetachVolume() IRI call is triggered. Volume gets removed from store but not from DomainXML.

To Reproduce

1. Create a machine with 2 empty disks
2. Detach one emty disk from machine
3. Verify domain XML for machine - empty disk will not get removed

Expected behavior
Empty disk should be removed from Domain XML also.

[lukas016]

Detach logic for libvirtd works properly. Detach is asynchronous call too, so it cannot detach disk immediately

    <disk type='network' device='disk'>
     <driver name='qemu' type='raw' cache='none' io='threads'/>
     <source protocol='rbd' name='ceph/img_05979169209155bdb34087ec06ddc31bfb6eeadf5c47b3f9f4389304e678fc4' index='3'>
       <host name='2a00...' port='6789'/>
       <auth username='volume-rook-ceph--ceph'>
         <secret type='ceph' uuid='4013d936-3459-5fb5-815e-79a85f3f2c1a'/>
       </auth>
     </source>
     <target dev='vda' bus='virtio'/>
     <serial>oda-4127059624245cdc</serial>
     <alias name='ua-volume-b3MtZGlzaw'/>
     <address type='pci' domain='0x0000' bus='0x03' slot='0x00' function='0x0'/>
   </disk>
   <disk type='file' device='disk'>
     <driver name='qemu' type='raw'/>
     <source file='/home/.../.libvirt-provider/machines/cf5caf4b-9af8-4402-856f-2aae5b9281d3/volumes/libvirt-provider.ironcore.dev~empty-disk/ephe-disk/disk.raw' index='2'/>
     <backingStore/>
     <target dev='vdb' bus='virtio'/>
     <serial>odb-8d534764e409d8df</serial>
     <alias name='ua-volume-ZXBoZS1kaXNr'/>
     <address type='pci' domain='0x0000' bus='0x04' slot='0x00' function='0x0'/>
   </disk>
   <disk type='file' device='disk'>
     <driver name='qemu' type='raw'/>
     <source file='/home/../.libvirt-provider/machines/cf5caf4b-9af8-4402-856f-2aae5b9281d3/volumes/libvirt-provider.ironcore.dev~empty-disk/ephe-disk2/disk.raw' index='1'/>
     <backingStore/>
     <target dev='vdc' bus='virtio'/>
     <serial>odc-d3c005e425b513c9</serial>
     <alias name='ua-volume-ZXBoZS1kaXNrMg'/>
     <address type='pci' domain='0x0000' bus='0x05' slot='0x00' function='0x0'/>
   </disk>

after detach:

    <disk type='network' device='disk'>
      <driver name='qemu' type='raw' cache='none' io='threads'/>
      <source protocol='rbd' name='ceph/img_05979169209155bdb34087ec06ddc31bfb6eeadf5c47b3f9f4389304e678fc4' index='3'>
        <host name='2a00... port='6789'/>
        <auth username='volume-rook-ceph--ceph'>
          <secret type='ceph' uuid='4013d936-3459-5fb5-815e-79a85f3f2c1a'/>
        </auth>
      </source>
      <target dev='vda' bus='virtio'/>
      <serial>oda-4127059624245cdc</serial>
      <alias name='ua-volume-b3MtZGlzaw'/>
      <address type='pci' domain='0x0000' bus='0x03' slot='0x00' function='0x0'/>
    </disk>
    <disk type='file' device='disk'>
      <driver name='qemu' type='raw'/>
      <source file='/home/.../cf5caf4b-9af8-4402-856f-2aae5b9281d3/volumes/libvirt-provider.ironcore.dev~empty-disk/ephe-disk2/disk.raw' index='1'/>
      <backingStore/>
      <target dev='vdc' bus='virtio'/>
      <serial>odc-d3c005e425b513c9</serial>
      <alias name='ua-volume-ZXBoZS1kaXNrMg'/>
      <address type='pci' domain='0x0000' bus='0x05' slot='0x00' function='0x0'/>
    </disk>

Root cause for wrong behavior in this case is using wrong os image. Image isn't booted properly and vm is stuck in efi shell. Problem is efi shell doesn't process ACPI signal. These signals are required for detach volume (https://wiki.ubuntu.com/QemuDiskHotplug#Removing_disks_from_Guests). I didn't go extremely depth into problematic because detach of volume for wrong booted os is edge case.

[lukas016]

Minimally, we have wrong logic for update status of api.Machine in store. Main problem is deleting volume from status, which is still part of domain xml.

  "status": {
    "volumeStatus": [
      {
        "name": "ephe-disk2",
        "handle": "libvirt-provider.ironcore.dev/empty-disk/ephe-disk2",
        "state": "Attached",
        "size": 524288000
      }
    ],