diff --git a/ironic/conductor/manager.py b/ironic/conductor/manager.py
index ba4c971069..d6371545ec 100644
--- a/ironic/conductor/manager.py
+++ b/ironic/conductor/manager.py
@@ -2898,9 +2898,12 @@ class ConductorManager(base_manager.BaseConductorManager):
     @periodics.periodic(spacing=CONF.conductor.send_sensor_data_interval,
                         enabled=CONF.conductor.send_sensor_data)
     def _send_sensor_data(self, context):
-        """Periodically sends sensor data to Ceilometer."""
+        """Periodically collects and transmits sensor data notifications."""
+
+        filters = {}
+        if not CONF.conductor.send_sensor_data_for_undeployed_nodes:
+            filters['provision_state'] = states.ACTIVE
 
-        filters = {'associated': True}
         nodes = queue.Queue()
         for node_info in self.iter_nodes(fields=['instance_uuid'],
                                          filters=filters):
diff --git a/ironic/conf/conductor.py b/ironic/conf/conductor.py
index ae7ed983cb..adecee58d6 100644
--- a/ironic/conf/conductor.py
+++ b/ironic/conf/conductor.py
@@ -125,6 +125,16 @@ opts = [
                 help=_('List of comma separated meter types which need to be'
                        ' sent to Ceilometer. The default value, "ALL", is a '
                        'special value meaning send all the sensor data.')),
+    cfg.BoolOpt('send_sensor_data_for_undeployed_nodes',
+                default=False,
+                help=_('The default for sensor data collection is to only '
+                       'collect data for machines that are deployed, however '
+                       'operators may desire to know if there are failures '
+                       'in hardware that is not presently in use. '
+                       'When set to true, the conductor will collect sensor '
+                       'information from all nodes when sensor data '
+                       'collection is enabled via the send_sensor_data '
+                       'setting.')),
     cfg.IntOpt('sync_local_state_interval',
                default=180,
                help=_('When conductors join or leave the cluster, existing '
diff --git a/ironic/tests/unit/conductor/test_manager.py b/ironic/tests/unit/conductor/test_manager.py
index de282fb328..a38efd1939 100644
--- a/ironic/tests/unit/conductor/test_manager.py
+++ b/ironic/tests/unit/conductor/test_manager.py
@@ -5720,6 +5720,10 @@ class SensorsTestCase(mgr_utils.ServiceSetUpMixin, db_base.DbTestCase):
         self.assertEqual(number_of_workers,
                          mock_spawn.call_count)
 
+    # TODO(TheJulia): At some point, we should add a test to validate that
+    # that a modified filter to return all nodes actually works, although
+    # the way the sensor tests are written, the list is all mocked.
+
 
 @mgr_utils.mock_record_keepalive
 class BootDeviceTestCase(mgr_utils.ServiceSetUpMixin, db_base.DbTestCase):
diff --git a/releasenotes/notes/send-sensor-data-for-all-nodes-a732d9df43e74318.yaml b/releasenotes/notes/send-sensor-data-for-all-nodes-a732d9df43e74318.yaml
new file mode 100644
index 0000000000..a55c43b53d
--- /dev/null
+++ b/releasenotes/notes/send-sensor-data-for-all-nodes-a732d9df43e74318.yaml
@@ -0,0 +1,19 @@
+---
+features:
+  - |
+    Adds a ``[conductor]send_sensor_data_for_undeployed_nodes`` option to
+    enable ironic to collect and transmit sensor data for all nodes
+    for which sensor data collection is available. By default, this option
+    is not enabled which aligns with the prior behavior of sensor data
+    collection and transmission where such data was only collected
+    if an ``instance_uuid`` was present to signify that the node has been or
+    is being deployed. With this change set to ``True``, operators may be able
+    to identify hardware in a faulty state through the sensor data and take
+    action before an instance workload is deployed.
+fixes:
+  - |
+    Fixes an issue where nodes in the process of deployment may have metrics
+    data collected and transmitted during the deployment process which
+    may erroneously generate alarms depending on the operator's monitoring
+    configuration. This was due to a database filter relying upon the
+    indicator of an ``instance_uuid`` as opposed to the state of a node.