Disk storage system with fault tolerant media maintenance

A data storage and retrieval system having a processor in communication with a RAM and a nonvolatile data storage, comprising:

• means for executing a media maintenance program on a selected track of nonvolatile data storage;
• checkpoint storage means for storing, in a selected area of the nonvolatile storage, checkpoint data prior to writing control information on the selected track during an execution of the media maintenance program, the checkpoint data including an address in the media maintenance program for resuming execution in the event that the media maintenance program is aborted by a hardware fault;
• means for reading the checkpoint data for the media maintenance program from the selected area of the nonvolatile storage after the media maintenance program has been aborted by a hardware fault;
• RAM storage means for storing execution environment data as specified by the checkpoint data in the RAM to restore the execution environment for the media maintenance program after the media maintenance program has been aborted by a hardware fault;
• and
• means for resuming execution of the media maintenance program at the address in the checkpoint data after the media maintenance program has been aborted by a hardware fault.

A method of executing a media maintenance program on a selected track in a data storage and retrieval system having a processor, a RAM and a nonvolatile data storage, comprising the steps of:

• storing, in a selected nonvolatile storage area, checkpoint data prior to writing control information on the selected track during an execution of the media maintenance program, the checkpoint data including an address in the media maintenance program for resuming execution in the event that the media maintenance program is aborted,
• reading the checkpoint data from the selected nonvolatile storage area after media maintenance has been aborted by a fault;
• storing data from the checkpoint data in the RAM to restore an execution environment for the media maintenance program after media maintenance has been interrupted by a fault;
• and
• resuming execution of the media maintenance program, using the execution environment, at the address in checkpoint data after media maintenance has been interrupted by a fault.

A controller for a data storage and retrieval system having a processor in communication with a RAM, a nonvolatile data storage, and means for transmitting and receiving data to and from a computer, comprising:

• means for executing a media maintenance program on a selected track of nonvolatile data storage;
• means for storing, in a selected area of the nonvolatile storage, checkpoint data for the media maintenance program prior to writing control information on the selected track during an execution of the media maintenance program, the checkpoint data including an address in the media maintenance program for resuming execution in the event that the media maintenance program is aborted by a fault, and the control information including data indicative of whether an alternate track has been assigned for the selected track;
• means for storing task control block data for the media maintenance program and the control information for the selected track in a data buffer;
• means for reading the checkpoint data from the selected area of the nonvolatile storage after the media maintenance program has been aborted by a fault;
• means for storing data as specified by the checkpoint data in the RAM to restore the execution environment for the media maintenance program after the media maintenance program has been aborted by a fault;
• and
• means for resuming execution of the media maintenance program at the address in the checkpoint data after the media maintenance program has been aborted by a fault.

A method of executing a media maintenance program on a selected track in a data storage and retrieval system having a processor, a RAM and a nonvolatile data storage, comprising the steps of:

• storing, in a selected area of the nonvolatile storage, checkpoint data prior to writing control information on the selected track during an execution of the media maintenance program, the checkpoint data including an address in the media maintenance program for resuming execution in the event that the media maintenance program is aborted by a fault, and control information including data indicative of whether an alternate track has been assigned for the selected track,
• reading the checkpoint data from the selected area of the nonvolatile storage after media maintenance has been aborted by a fault;
• storing execution environment data as specified by the checkpoint data in the RAM to restore an execution environment for the media maintenance program after media maintenance has been aborted by a fault;
• and
• resuming execution of the media maintenance program, using the execution environment, at the address in checkpoint data after media maintenance has been aborted by a fault.