This object reports the global deadtime value, the period during which a TACACS+ server that has failed to respond is skipped in favor of other configured servers before being retried. An admin tunes this to balance quickly failing over away from a downed server against not permanently abandoning a server that may have only had a brief outage. A longer deadtime avoids repeatedly retrying a server that's likely still down, at the cost of a slower return to using it once it recovers.