Remote processor crash recovery

• US 4,589,090 A
• Filed: 09/21/1982
• Issued: 05/13/1986
• Est. Priority Date: 09/21/1982
• Status: Expired due to Term

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Federico, Anthony M., Husted, Raymond R., Wilczek, Stephen P., Ziehm, Richard T., Edmunds, Michael E., Downing, Curtis B.
• Primary Examiner(s): Shaw, Gareth D.
• Assistant Examiner(s): Jaser, Samir
• Application Number:
  US06/420,965
• Time in Patent Office: 1,330 Days
• Field of Search: 364/200 MS File, 364/900 MS File, 355/14 R, 355/14 C, 371/7
• US Class Current: 714/10
• CPC Class Codes: G05B 19/0421 Multiprocessor system
• CPC Class Codes: G05B 19/0428 Safety, monitoring G05B19/0423 takes precedence
• CPC Class Codes: G05B 2219/24114 Continue program if crashed microprocessor, program module is not crucial
• CPC Class Codes: G05B 2219/25232 DCS, distributed control system, decentralised control unit
• CPC Class Codes: G05B 2219/25381 Restart program at predetermined position, crash recovery after power loss
• CPC Class Codes: G06F 11/1417 Boot up procedures

Abstract

The present invention is a multiprocessor machine control system in which the failure of one of the processors to reset can be ignored by the rest of the control system. In particular, a software crash or other abnormality on one of the processors will generate a reset procedure. If the processor cannot be reset, this will indicate a processor board failure such as a hardware failure. If the processor and its controlled elements are not crucial to the machine operation, then the control wll ignore the failed processor as though it were not in the control system, and continue with machine operation.

Claims

1. In a multiprocessor machine control for controlling the operation of a printing machine having a plurality of machine components cooperating with one another to produce images on support material, the machine control including a plurality of control boards, each of the control boards controlling the operation of a portion of the machine components, one of the components being non-essential for producing images on the support material, one of the plurality of control boards being a master control board connected by reset circuitry to the other control boards, the master control board including a memory for storing status information of the other control boards, the method of recovering from a malfunction involving the control boards comprising the steps of

(1) identifying a particular control board manifesting a malfunction,

(2) the master control board attempting to reset said particular control board via the reset circuitry,

(3) checking the operability of said particular control board,

(4) determining said particular control board is inoperable but non-essential for producing image on the support material,

(5) continuing operation of the machine to produce images on the support material without the availability of said particular control board and saving status information for said particular control board in said memory.

2. The method of claim 1 wherein the said particular control board is determined to be operable after rechecking said particular control board, including the step of conveying to the particular control board said status information lost while said particular control board was inoperable and stored in said memory.

3. The method of claim 2 wherein the particular control board is a display control board and the master control board includes a message store including the step of holding display messages in the message store until said particular control board is reset and operable.

4. A multiprocessor control for controlling the operation of a printing machine having a plurality of controlled elements including an image forming member cooperating with one another to provide images on support material, each of the controlled elements providing a machine function, some of said functions being essential to machine operation, other of said functions being nonessential to machine operation, the multiprocessor control comprising

a plurality of remote processors, the remote processors controlling portions of the controlled elements, one of the remote processors controlling elements providing machine functions nonessential to machine operation,

a master processor responsive to a control malfunction, said master processor including logic means to determine that said control malfunction is associated with said one of the remote processors,

a reset line interconnecting the master processor with said one of the remote processors,

reset circuitry electrically connected to said one of the remote processors and responsive to the master processor,

activation means for the master processor to activate the reset circuitry and attempt to reset said one of the remote processors,

means to indicate a failed reset of said one of the remote processors, and

means for the multiprocessor control to continue machine operation independent of said one of the remote processors to provide images on said support material.

5. A multiprocessor control for controlling the operation of a printing machine having a plurality of controlled elements, including an image forming member, cooperating with one another to provide images on support material, each of the controlled elements providing a machine function, some of the functions being essential for providing images on support material, other of said functions, including a display control for displaying machine conditions, being non-essential to provide images on support material, the multiprocessor control comprising

a plurality of remote processors, the remote processors controlling portions of the controlled elements, one of the remote processors controlling the display control,

a master processor responsive to a control malfunction, said master processor including logic means to determine that said control malfunction is associated with the display control,

reset lines interconnecting the master processor with said one of the remote processors,

reset circuitry electrically connected to said one of the remote processors and responsive to the master processor,

activation means for the master processor to activate the reset circuitry and attempt to reset said one of the remote processors,

means to indicate a failed reset of said one of the remote processors, and

means for the multiprocessor control to continue machine operation independent of said one of the remote processors and the display control.