collecting information on operating statuses inside and outside the computer system, on a process executed in the computer system, and on at least one of records of events that have been generated in the computer system;

storing correlation information indicating a correlation among the collected pieces of information;

at least one of: detecting a failure that has occurred in a service provided by the computer system from the collected information with reference to the stored correlation information and creating a recovering process for the detected failure; and inputting a recovering process executed by the computer system;

evaluating an effect and a side-effect given to the computer system through execution of at least one of the created process and the inputted process with reference to the correlation information;

determining at least one of a need, an order, and time of execution of the at least one process whose effect and side-effect are evaluated;

canceling, in one of a case where the effect given to the computer system through the execution of the at least one process is smaller than a predetermined threshold, and a case where the side-effect given to the computer system through the execution of the at least one process is larger than a predetermined threshold, the execution of the at least one process; and

performing one of a re-creation of the process and a request for a re-input of the process.