Comment 1 for bug 1783565

After some check and test, the only reason I can think of is the sleep time is not good, as
sleep time and service_down_time is the same, maybe sometime when we post evacuate action
the service record did not got updated.