High reliability computer diagnostics system

• US 5,163,052 A
• Filed: 10/12/1989
• Issued: 11/10/1992
• Est. Priority Date: 10/12/1989
• Status: Expired due to Fees

• Current Assignee: NCR Corporation
• Original Assignee: NCR Corporation
• Inventors: Johnston, Gary L., Evans, Eric A., Zakhariya, Ramiz H., Rusell, Joseph D., Kaino, Masami S.
• Primary Examiner(s): ATKINSON, CHARLES
• Application Number:
  US07/420,862
• Time in Patent Office: 1,125 Days
• Field of Search: 364/267, 364/267.4, 364/269.2, 371/18, 371/16.1, 371/9.1, 395/575
• US Class Current: 714/31
• CPC Class Codes: G06F 11/22 Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing
• CPC Class Codes: G06F 11/2736 using a dedicated service processor for test

Abstract

In a multiple-board computer system, diagnostics are primarily performed by highly reliable microcontroller units (MCUs) on each processor (and memory) board. Operated separately from the board central processing units so that board failure will not prohibit diagnostic routines, the MCUs are serially connected to a bus separate from the main bus. Redundant diagnostic processor board MCUs constantly monitor the other MCUs and the serial bus, and oversee the operation of an MCUs on a malfunctioning board. In an alternate embodiment the diagnostic processor boards are replaced by a microcomputer serving as a monitor.

Claims

1. A multiple-system board computer system capable of performing diagnostics comprising:

a plurality of processor system boards, each processor system board comprising at least one central processing unit (CPU) for controlling the operation of said processor system board, and at least one microcontroller unit (MCU) operable independently of said CPU for performing at least a portion of said diagnostics pertaining to the processor system board where the particular MCU resides;

at least a memory system board, comprising at least one MCU for performing at least a portion of said diagnostics pertaining to the memory system board where the particular MCU resides;

a main bus coupled to each said CPU of each said processor system board and to memory of each said memory system board, allowing communication between each said system board;

a separate diagnostic bus coupled to each said MCU for allowing communication between each said MCU; and

at least a first diagnostic processor system board coupled to said main bus and having at least one MCU coupled to said diagnostics bus for monitoring all of the MCUs and for controlling the operation of an MCU on a system board when that system board is determined by its MCU to be malfunctioning;

wherein diagnostics for each said system board may be self-performed by each said system board.

2. The multiple-board computer system in claim 1 wherein said MCU on said first diagnostic processor board also determines if said diagnostic bus is malfunctioning.

3. The multiple-system board computer system in claim 2 further comprising at least a second but redundant diagnostic processor system board coupled to said main and diagnostic buses, and operational in the same manner as said first diagnostic processor system board, except that said second diagnostic processor system board is only operational when said first diagnostic processor system board is malfunctioning.

4. The multiple-system board computer system in claim 1 further comprising at least a second but redundant diagnostic processor system board coupled to said main and diagnostic buses, and operational in the same manner as said first diagnostic processor system board, except that said second diagnostic processor system board is only operational when said first diagnostic processor system board is malfunctioning.