Traceback when RUG destroys router neutron has deleted, but instance is already gone

I believe this gets triggered if the Rug finds a router has been deleted in neutron and goes to delete the router appliance, but the instance has already been deleted.

b/python2.7/dist-packages/oslo_messaging/_drivers/amqpdriver.py:392
2015-08-06 13:58:33.801 INFO akanda.rug.state.3c7adcdb-f9b9-4688-9e62-8f36a8edc9fe:11384:p00:t00 [req-8db40f2e-2186-461f-85fe-c5ee77577a1a None None] Destroying router neutron has deleted
2015-08-06 13:58:33.801 ERROR akanda.rug.state.3c7adcdb-f9b9-4688-9e62-8f36a8edc9fe:11384:p00:t00 [req-8db40f2e-2186-461f-85fe-c5ee77577a1a None None] StopInstance.execute() failed for action: delete
2015-08-06 13:58:33.801 TRACE akanda.rug.state.3c7adcdb-f9b9-4688-9e62-8f36a8edc9fe Traceback (most recent call last):
2015-08-06 13:58:33.801 TRACE akanda.rug.state.3c7adcdb-f9b9-4688-9e62-8f36a8edc9fe File "/opt/stack/akanda-rug/akanda/rug/state.py", line 438, in update
2015-08-06 13:58:33.801 TRACE akanda.rug.state.3c7adcdb-f9b9-4688-9e62-8f36a8edc9fe worker_context,
2015-08-06 13:58:33.801 TRACE akanda.rug.state.3c7adcdb-f9b9-4688-9e62-8f36a8edc9fe File "/opt/stack/akanda-rug/akanda/rug/state.py", line 280, in execute
2015-08-06 13:58:33.801 TRACE akanda.rug.state.3c7adcdb-f9b9-4688-9e62-8f36a8edc9fe self.instance.stop(worker_context)
2015-08-06 13:58:33.801 TRACE akanda.rug.state.3c7adcdb-f9b9-4688-9e62-8f36a8edc9fe File "/opt/stack/akanda-rug/akanda/rug/instance_manager.py", line 309, in stop
2015-08-06 13:58:33.801 TRACE akanda.rug.state.3c7adcdb-f9b9-4688-9e62-8f36a8edc9fe if not nova_client.get_instance_by_id(self.instance_info.id_):
2015-08-06 13:58:33.801 TRACE akanda.rug.state.3c7adcdb-f9b9-4688-9e62-8f36a8edc9fe AttributeError: 'NoneType' object has no attribute 'id_'
2015-08-06 13:58:33.801 TRACE akanda.rug.state.3c7adcdb-f9b9-4688-9e62-8f36a8edc9fe