PURPOSE: To improve the error processing efficiency and to increase the response speed in an information process utilizing a disk subsystem, by switching a path to a substitute one if available to perform the retry of an access in case a fault occurs when an access is given to a disk device in the disk subsystem from a host computer system.

CONSTITUTION: The occurrence of a fault is decided by a fault segmenting means 1. If it is decided that a disk control processor 110 has a fault, 1 is added to an error counter 140 of the processor 110 of a counter indicator table 5 by a count-up means 2. Then the processor 110 is separated from a host computer system 100. If it is supposed that a disk device 120 has a fault, 1 is added to an error counter 150. When the value of the counter 150 exceeds a prescribed level, a track file 130 or the device 120 is blocked Then a recovery indicator 180 is set at '1' and the path is switched to a substitute one if available by a substitute path switching means 4. Thus an access is given to the device 120 via a control processor 111.