periodic replica update should not work on share servers in error

[Carthaca]

VserverNotSpecified: Vserver name is absent in backend details. Please check whether Vserver was created properly.
  File "manila/share/manager.py", line 2873, in _share_replica_update
    replica_state = self.driver.update_replica_state(
  File "manila/share/drivers/netapp/dataontap/cluster_mode/drv_multi_svm.py", line 163, in update_replica_state
    return self.library.update_replica_state(context, replica_list,
  File "manila/share/drivers/netapp/dataontap/cluster_mode/lib_base.py", line 2912, in update_replica_state
    vserver, vserver_client = self._get_vserver(share_server=share_server)
  File "manila/share/drivers/netapp/utils.py", line 115, in trace_wrapper
    result = f(self, *args, **kwargs)
  File "manila/share/drivers/netapp/dataontap/cluster_mode/lib_multi_svm.py", line 104, in _get_vserver
    raise exception.VserverNotSpecified(msg)

driver cannot fix this, so it should not throw errors on each run and instead skip those as early as possible (sadly we don't have a dedicated replica_state or status we could use for that: maybe introduce error_creating? )

[Carthaca]

In one case the share server was in error, because the neutron subnet used was out of free IPs - there is no easy way out