Disk array apparatus and method for controlling the same

A disk array apparatus comprising:

• a dual controller for performing data input/output processing of disk drives in response to a data input/output request from a host system,
• wherein said dual controller includes:
• a main controller for performing the data input/output processing,
• a Fibre Channel (FC) protocol controller for interface connection to the host system via a front interface connection FC, and for interface connection to the disk drive via a back interface connection FC,
• two FC-Arbitrated Loops (ALs) for loop connection to the disk drives,
• a connection switching circuit for switching connection of one selected FC-AL, of the two FC-ALs, to the back interface connection Fibre Channel,
• a plurality of switching circuits for connecting the respective disk drives to the FC-ALs, and
• an enclosure controller for controlling the switching circuits;
• wherein the main controller includes:
• a memory for storing control information for specifying a failure of at least one of the disk drives, and failure information of the faulty disk drive,
• a first circuit for specifying, from the failure information, the faulty disk drive in which a time-out failure has occurred, and issuing an instruction to retry a control command from the enclosure controller to the faulty disk drive,
• a second circuit for outputting a control signal from the enclosure controller to the faulty disk drive in order to hard reset the disk drive if the faulty disk drive does not recover from the failure by normally executing the retry control command,
• wherein the hard reset initializes a chip that includes an electronic circuit of the disk drive,
• wherein the dual controller judges whether the disk array apparatus is in a warning state or not, and if the disk array apparatus is in the warning state, the dual controller hard resets a suspected faulty disk drive and issues a spin-up command to the suspected faulty disk drive, thereby recovering the suspected faulty disk drive,
• wherein the warning state is a state in which any one of the disk drives is already in a locked-out state,
• wherein a suspected faulty disk drive is a disk drive that has sent an error message or a disk drive from which an acknowledgement has not been received within a specified period of time, and
• wherein the hard reset includes a power on reset of a disk drive.

A disk array apparatus comprising:

• a plurality of disk drives;
• and
• a control unit for performing data input/output processing of the disk drives in response to a data input/output request from a host system;
• wherein said control unit includes:
• a memory for storing control information for specifying a failure of at least one of the disk drives, and failure information of the faulty disk drive;
• a circuit for specifying, from the failure information, the disk drive in which a time-out failure has occurred, and issuing an instruction to retry a control command to the disk drive;
• a circuit for outputting a control signal to the faulty disk drive in order to lock-out the disk drive if the disk drive does not recover from the failure;
• a circuit for receiving a Small Computer System Interface (SCSI) Enclosure Services (SES) command including a hard reset command and for outputting a reset signal according to the received SES command to the faulty disk drive in order to hard reset the faulty disk drive;
• wherein the hard reset initializes a chip that includes an electronic circuit of the disk drive,
• wherein the control unit judges whether the disk array apparatus is in a warning state or not, and if the disk array apparatus is in the warning state, the control unit hard resets a suspected faulty disk drive and issues a spin-up command to the suspected faulty disk drive, thereby recovering the suspected faulty disk drive,
• wherein the warning state is a state in which any one of the disk drives is already in a locked-out state,
• wherein a suspected faulty disk drive is a disk drive that has sent an error message or a disk drive from which an acknowledgement has not been received within a specified period of time,
• wherein during the hard reset period of the faulty disk drive, the control unit recovers, from information stored on the other disk drives, storage data on the read-accessed faulty disk drive and provides the recovered storage data to the host system, and
• wherein the hard reset includes a power on reset of a disk drive.

A method for controlling a disk array apparatus including a plurality of disk drives and a control unit for performing data input/output processing of the disk drives in response to a data input/output request from a host system, the method comprising the steps of:

• specifying a failure of at least one of the disk drives;
• storing failure information of the faulty disk drive;
• judging from the failure information that a time-out failure has occurred;
• instructing the disk drive, in which the time-out failure has occurred, to retry a control command;
• hard resetting the faulty disk drive if the faulty disk drive does not recover from the failure by normally executing the retry control command,
• wherein the hard resetting initializes a chip that includes an electronic circuit of the disk drive,
• wherein the control unit judges whether the disk array apparatus is in a warning state or not, and if the disk array apparatus is in the warning state, the control unit hard resets a suspected faulty disk drive and issues a spin-up command to the suspected faulty disk drive, thereby recovering the suspected faulty disk drive,
• wherein the warning state is a state in which any one of the disk drives is already in a locked-out state,
• wherein a suspected faulty disk drive is a disk drive that has sent an error message or a disk drive from which an acknowledgement has not been received within a specified period of time, and
• wherein the hard resetting includes a power on resetting of a disk drive.