The host works without issue and running an instance from either nova or virt-manager can be done without issue.  Additionally, I can attach volumes from local stroage to the same VM via virt-manager. It is just the attachment of a Cinder volume to a Nova instance that exhibits the exception and only to one of the hypervisor hosts.

Do you think it may be an issue with the hypervisor host name, e.g. the hostname alone is displayed in the logs but the FQDN is shown in 'hypervisor-list'?

The following is nova.conf.  Variables in '<>' have been replaced to conceal their actual.  While '<PASSWORD>' is used below in multiple places, these are different; however, '<CONTROLLER_IP>' is the same as there is only 1 controller.

[DEFAULT]
debug = True
log_dir = /var/log/kolla/nova
state_path = /var/lib/nova
osapi_compute_listen = <CONTROLLER_IP>
osapi_compute_listen_port = 8774
osapi_compute_workers = 5
metadata_workers = 5
metadata_listen = <CONTROLLER_IP>
metadata_listen_port = 8775
firewall_driver = nova.virt.firewall.NoopFirewallDriver
allow_resize_to_same_host = true
compute_driver = libvirt.LibvirtDriver
my_ip = <CONTROLLER_IP>
instance_usage_audit = True
instance_usage_audit_period = hour
compute_monitors = nova.compute.monitors.cpu.virt_driver
transport_url = rabbit://openstack:<PASSWORD>@<CONTROLLER_IP>:5672
use_cow_images = false
resume_guests_state_on_host_boot = true

[api]
use_forwarded_for = true

[conductor]
workers = 5

[vnc]
novncproxy_host = <CONTROLLER_IP>
novncproxy_port = 6080
server_listen = <CONTROLLER_IP>
server_proxyclient_address = <CONTROLLER_IP>
novncproxy_base_url = http://<CONTROLLER_IP>:6080/vnc_auto.html

[oslo_middleware]
enable_proxy_headers_parsing = True

[oslo_concurrency]
lock_path = /var/lib/nova/tmp

[glance]
api_servers = http://<CONTROLLER_IP>:9292
num_retries = 1
debug = False

[cinder]
catalog_info = volumev3:cinderv3:internalURL
os_region_name = RegionOne
auth_url = http://<CONTROLLER_IP>:35357
auth_type = password
project_domain_name = Default
user_domain_id = default
project_name = service
username = cinder
password = <PASSWORD>

[neutron]
metadata_proxy_shared_secret = <SHARED_SECRET>
service_metadata_proxy = true
auth_url = http://<CONTROLLER_IP>:35357
auth_type = password
project_domain_name = Default
user_domain_id = default
project_name = service
username = neutron
password = <PASSWORD>
region_name = RegionOne
valid_interfaces = internal

[cache]
backend = oslo_cache.memcache_pool
enabled = True
memcache_servers = <CONTROLLER_IP>:11211

[keystone_authtoken]
www_authenticate_uri = http://<CONTROLLER_IP>:5000
auth_url = http://<CONTROLLER_IP>:35357
auth_type = password
project_domain_id = default
user_domain_id = default
project_name = service
username = nova
password = <PASSWORD>
memcache_security_strategy = ENCRYPT
memcache_secret_key = <SECRET_KEY>
memcached_servers = <CONTROLLER_IP>:11211

[libvirt]
connection_uri = qemu+tcp://<CONTROLLER_IP>/system
virt_type = kvm

[upgrade_levels]
compute = auto

[oslo_messaging_notifications]
transport_url = rabbit://openstack:<PASSWORD>@<CONTROLLER_IP>:5672
driver = messagingv2
topics = notifications,notifications_designate

[privsep_entrypoint]
helper_command = sudo nova-rootwrap /etc/nova/rootwrap.conf privsep-helper --config-file /etc/nova/nova.conf

[guestfs]
debug = False

[wsgi]
api_paste_config = /etc/nova/api-paste.ini

[scheduler]
max_attempts = 10
discover_hosts_in_cells_interval = -1

[placement]
auth_type = password
auth_url = http://<CONTROLLER_IP>:35357
username = placement
password = <PASSWORD>
user_domain_name = Default
project_name = service
project_domain_name = Default
region_name = RegionOne
os_interface = internal

[notifications]
notify_on_state_change = vm_and_task_state

[barbican]
auth_endpoint = http://<CONTROLLER_IP>:5000

[compute]
consecutive_build_service_disable_threshold = 0