The warning was triggered during _sync_power_states periodic task getting vm_power_state of SHUTDOWN from the baremetal IPMI driver. This will happen if the IPMI driver is not able to check the power state of the node, because get_info() defaults to believing the power_state is SHUTDOWN.
I believe the correct default returned when the IPMI driver isn't able to determine the power state is NOSTATE, which would lead to a warning being logged, and nothing else. SHUTDOWN is clearly the wrong default.
The warning was triggered during _sync_power_states periodic task getting vm_power_state of SHUTDOWN from the baremetal IPMI driver. This will happen if the IPMI driver is not able to check the power state of the node, because get_info() defaults to believing the power_state is SHUTDOWN.
https:/ /github. com/openstack/ nova/blob/ stable/ grizzly/ nova/virt/ baremetal/ driver. py#L362
def get_info(self, instance): node_by_ instance_ uuid(inst_ uuid) manager( node=node, instance=instance) SHUTDOWN
...
node = _get_baremetal_
pm = get_power_
ps = power_state.
if pm.is_power_on():
ps = power_state.RUNNING
return {'state': ps,
Based on the behaviour of _sync_instance_ power_state, /github. com/openstack/ nova/blob/ stable/ grizzly/ nova/compute/ manager. py#L3773
https:/
I believe the correct default returned when the IPMI driver isn't able to determine the power state is NOSTATE, which would lead to a warning being logged, and nothing else. SHUTDOWN is clearly the wrong default.