ovn: Log every time a hash ring node is touched

This may help us to understand when nodes are touched and by whom.
There's a suspicion the maintenance thread gets locked / dies / not
scheduled on time and this makes nodes go offline (losing ovsdb monitor
events in the process).

Change-Id: If8d6d2200807c6fa3ada151fe202110b8853bdc5

neutron/plugins/ml2/drivers/ovn/mech_driver/ovsdb/maintenance.py

@@ -1121,6 +1121,9 @@ class HashRingHealthCheckPeriodics:
         # NOTE(lucasagomes): Note that we do not rely on the OVSDB lock
         # here because we want the maintenance tasks from each instance to
         # execute this task.
+        LOG.debug(
+            'Touching Hash Ring node "%s" from periodic health check thread',
+            self._node_uuid)
         hash_ring_db.touch_node(self.ctx, self._node_uuid)
 
         # Check the number of the nodes in the ring and log a message in

neutron/plugins/ml2/drivers/ovn/mech_driver/ovsdb/ovsdb_monitor.py

@@ -844,6 +844,9 @@ class OvnIdlDistributedLock(BaseOvnIdl):
                 # the error and continue with processing the event
                 try:
                     ctx = neutron_context.get_admin_context()
+                    LOG.debug(
+                        'Touching Hash Ring node "%s" from IDL notify handler',
+                        self._node_uuid)
                     ovn_hash_ring_db.touch_node(ctx, self._node_uuid)
                 except Exception:
                     LOG.exception('Hash Ring node %s failed to heartbeat',