Enable recovering the management servers after BOTH of them are down by writing all state to disk, if persistence is configured.