In recent years, the information services are more and more important. Services need to use the server to provide computing, storage and other functionalities. People are becoming increasingly dependent on information system and service it provided. So service availability of the system has become the key to continue operation. When a large number of servers are facing disaster situations such as power outage or fire, administrators may only have time to shut down some key equipment. When it is not shut down in time, they can only let most of the equipment stay on in the disaster condition. In this research, we propose a management system that has the ability to detect the environment parameters on the server farm and shutdown servers automatically in a predetermined sequence based on the environmental situation. Manager is required to first define the importance of each server according to the ISMS rules. The environmental parameters are retrieved through SNMP or other interface from the equipment. The resulting system is a server farm monitoring system which can be implemented in a very low cost. It can help the manager to manage servers in the situation of server farm emergency response.