diff --git a/nova/conf/workarounds.py b/nova/conf/workarounds.py
index 26b834b47ef8..ea68b8e739ab 100644
--- a/nova/conf/workarounds.py
+++ b/nova/conf/workarounds.py
@@ -422,6 +422,19 @@ backend.
 Related options:
 
 * :oslo.config:option:`DEFAULT.vif_plugging_timeout`
+"""),
+    cfg.BoolOpt('enable_qemu_monitor_announce_self',
+                default=False,
+                help="""
+If it is set to True the libvirt driver will  try as a best effort to send
+the announce-self command to the QEMU monitor so that it generates RARP frames
+to update network switches in the post live migration phase on the destination.
+
+Please note that this causes the domain to be considered tainted by libvirt.
+
+Related options:
+
+* :oslo.config:option:`DEFAULT.compute_driver` (libvirt)
 """),
 ]
 
diff --git a/nova/virt/libvirt/driver.py b/nova/virt/libvirt/driver.py
index f86ad2e1617b..2c6dab389162 100644
--- a/nova/virt/libvirt/driver.py
+++ b/nova/virt/libvirt/driver.py
@@ -10410,6 +10410,26 @@ class LibvirtDriver(driver.ComputeDriver):
         """
         self.unplug_vifs(instance, network_info)
 
+    def _qemu_monitor_announce_self(self, instance):
+        """Send announce_self command to QEMU monitor.
+
+        This is to trigger generation of broadcast RARP frames to
+        update network switches. This is best effort.
+        """
+        if not CONF.workarounds.enable_qemu_monitor_announce_self:
+            return
+
+        LOG.info('Sending announce-self command to QEMU monitor',
+                 instance=instance)
+
+        try:
+            guest = self._host.get_guest(instance)
+            guest.announce_self()
+        except Exception:
+            LOG.warning('Failed to send announce-self command to QEMU monitor',
+                        instance=instance)
+            LOG.exception()
+
     def post_live_migration_at_destination(self, context,
                                            instance,
                                            network_info,
@@ -10425,6 +10445,7 @@ class LibvirtDriver(driver.ComputeDriver):
         :param block_migration: if true, post operation of block_migration.
         """
         self._reattach_instance_vifs(context, instance, network_info)
+        self._qemu_monitor_announce_self(instance)
 
     def _get_instance_disk_info_from_config(self, guest_config,
                                             block_device_info):
diff --git a/nova/virt/libvirt/guest.py b/nova/virt/libvirt/guest.py
index 54e63e42928c..53080e41f0b4 100644
--- a/nova/virt/libvirt/guest.py
+++ b/nova/virt/libvirt/guest.py
@@ -48,6 +48,12 @@ if ty.TYPE_CHECKING:
 else:
     libvirt = None
 
+try:
+    import libvirtmod_qemu
+except ImportError:
+    libvirtmod_qemu = None
+
+
 LOG = logging.getLogger(__name__)
 
 VIR_DOMAIN_NOSTATE = 0
@@ -632,6 +638,10 @@ class Guest(object):
         """Switch running live migration to post-copy mode"""
         self._domain.migrateStartPostCopy()
 
+    def announce_self(self):
+        libvirtmod_qemu.virDomainQemuMonitorCommand(
+            self._domain._o, 'announce_self', 1)
+
     def get_job_info(self):
         """Get job info for the domain
 
diff --git a/releasenotes/notes/announce-self-post-live-migration-936721b1ab887514.yaml b/releasenotes/notes/announce-self-post-live-migration-936721b1ab887514.yaml
new file mode 100644
index 000000000000..f0ec39f196bd
--- /dev/null
+++ b/releasenotes/notes/announce-self-post-live-migration-936721b1ab887514.yaml
@@ -0,0 +1,8 @@
+---
+features:
+  - |
+    Added a new configuration option ``[workarounds]/enable_qemu_monitor_announce_self``
+    that when enabled causes the Libvirt driver to send a announce_self QEMU
+    monitor command post live-migration. Please see `bug 1815989 <https://bugs.launchpad.net/nova/+bug/1815989>`_
+    for more details. Please note that this causes the domain to be considered
+    tainted by libvirt.