Fault diagnostic distributed processing method and system

• US 4,991,174 A
• Filed: 03/28/1989
• Issued: 02/05/1991
• Est. Priority Date: 09/08/1983
• Status: Expired due to Term

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Shiraha, Takeshi, Miyamoto, Shoji, Mori, Kinji
• Primary Examiner(s): Fleming, Michael R.
• Application Number:
  US07/329,558
• Time in Patent Office: 679 Days
• Field of Search: 371/15, 371/5, 371/16, 371/7, 371/11, 371/8, 371/13, 371/22, 371/2, 371/9, 371/34, 371/38, 371/8.1, 371/8.2, 371/9.1, 371/11.1, 371/11.2, 371/11.3, 371/15.1, 371/16.1, 364/200, 364/900
• US Class Current: 714/10
• CPC Class Codes: G06F 11/0709 in a distributed system consisting of a plurality of standalone computer nodes, e.g. clusters, client-server systems
• CPC Class Codes: G06F 11/079 Root cause analysis, i.e. error or fault diagnosis in a hardware test environment G06F11/22; in a software test environment G06F11/36
• CPC Class Codes: G06F 11/1625 in communications, e.g. transmission, interfaces
• 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 11/327 Alarm or error message display
• CPC Class Codes: H04L 12/437 Ring fault isolation or reconfiguration for SDH/SONET ring networks H04J3/085

Abstract

In a distributed processing system having a plurality of interconnected subsystems of equal level, each subsystem operates to diagnose faults in other subsystems and protects its own subsystem based on the diagnosis of the faults of the other subsystems. The subsystems may be network control processors connected to common signal transmission lines, each of which carries major and minor loop check messages used to detect the faults in the other network control processors and constitutes a bypass route to protect its own unit when the fault occurs. If a transient fault occurs in the systems, an indication of the degree of the transient fault is stored and a possibility that the transient fault will change to a permanent fault is determined based on a time variation of the degree of the fault. If it is determined that the transient fault will change to a permanent fault, it is indicated to a man-machine system. In this manner, a loop transmission system which can be readily prediction-diagnosed for a fault is provided. In a second embodiment, means for calculating a feedback rate of a signal sent out to the transmission line and means for calculating the degree of fault on the transmission line based on the feedback rate calculated by the calculation means are provided.

Claims

1. A method for processing data in a distributed processing system having a plurality of subsystems interconnected through at least one transmission path, said method comprising the steps of:

sending, in each subsystem, a diagnosing signal for diagnosing a fault on a transmission route including at least one other subsystem and said transmission path;

receiving, in each subsystem, a signal relating to said diagnosing signal from a transmission route;

detecting the presence of a fault based on the signal relating to said diagnosing signal received in each subsystem;

suppressing, in each subsystem, signal transmission to said transmission route which contains said fault upon detection of the fault;

sending, in each subsystem, a signal representing the fault to said transmission path; and

determining, in at least one subsystem, a fault location based on said fault representing signal.

2. A method for processing data in a distributed processing system having a plurality of subsystems interconnected through at least one transmission path, said method comprising the steps of:

sending, in at least one subsystem, an address train when a power is supplied to said system; and

receiving, in other subsystems, said address train;

in each subsystem receiving said address train, sending to said transmission path said address train including information relating to the address of the receiving subsystem; and

receiving said address train in at least one subsystem, and recognizing the address construction in said system by means of the received address train.

3. A method for processing data according to claim 2, wherein said step of recognizing includes a step of detecting duplication of an address in said subsystem.

4. A method for processing data according to claim 2, wherein said address train is circulated twice on the transmission path.

5. A method for processing data according to claim 3, wherein said address train is circulated twice on the transmission path.

6. A method for processing data in a distributed processing system having a plurality of subsystems interconnected through at least one transmission path, said method comprising the steps of:

sending, in each subsystem, a signal for diagnosing a fault on a transmission route including at least one other subsystem and said transmission path;

receiving, in each subsystem, a signal related to said diagnosing signal from a transmission route;

detecting a fault based on the signal relating to said diagnosing signal received in each subsystem;

suppressing, in each subsystem, signal transmission to said transmission route containing said fault upon detection of the fault;

sending, in each subsystem, a signal for detecting recovery of the fault to said transmission route after detection of the fault;

receiving, in each subsystem, a signal relating to a recovery detecting signal from said transmission path;

detecting the recovery of the fault based on the signal relating to the recovery detecting signal received in each subsystem; and

sending, in each subsystem, a signal representing said recovery of the fault to said transmission path upon detection of the recovery.

7. A method for processing data according to claim 6, which further comprises a step of determining, in at least one subsystem, a recovery location based on said recovery representing signal.

9. A method for processing data in a distributed processing system having a plurality of subsystems interconnected through at least one transmission path, said method comprising the steps of:

sending, in each subsystem, a signal for diagnosing a fault on a transmission route including at least one other subsystem and said transmission path;

receiving, in each subsystem, a signal relating to said diagnosing signal from said transmission route;

detecting a fault based on the signal relating to said diagnosing signal received in each subsystem;

sending, in each subsystem, a signal representing the fault to said transmission path; and

determining, in at least one subsystem, a fault location based on said fault representing signal.

10. A method for processing data in a distributed processing system having a plurality of subsystems interconnected through at least one transmission path, said method comprising the steps of:

sending, in each subsystem, a signal for detecting recovery of a fault on a transmission route including at least one other subsystem and said transmission path;

receiving, in each subsystem, a signal relating to said recovery detecting signal from said transmission route;

detecting the recovery of the fault based on the signal relating to the recovery detecting signal received in each subsystem;

sending, in each subsystem, a signal representing said recovery of the fault to said transmission route upon detection of the recovery; and

determining, in at least one subsystem, a recovery location based on said recovery representing signal.

8. A method for processing data according to one of claims 6 or 10, which further comprises a step of reinstating the signal transmission on said transmission route containing said fault upon detection of the fault recovery.

11. A method for processing data in a distributed processing system having a plurality of subsystems interconnected through at least one transmission path, said method comprising the steps of:

sending, in each subsystem, a signal for diagnosing a fault on a transmission route including other at least one subsystem and said transmission path;

receiving, in each subsystem, a signal relating to said diagnosing signal from said transmission route;

detecting a fault based on the signal relating to said diagnosing signal received in each subsystem;

storing, in each subsystem, a history of fault detection; and

generating, in each subsystem, a signal for controlling signal transmission to said transmission route based on the stored history of fault detection.

12. A distributed processing system having a plurality of subsystems interconnected through at least one transmission path, comprising:

means in each subsystem for sending a diagnosing signal for diagnosing a fault on a transmission route including at least one another subsystem and said transmission path;

means in each subsystem for receiving a signal relating to said diagnosing signal from a transmission route;

means for detecting the presence of a fault based on the signal relating to said diagnosing signal received in each subsystem;

means in each subsystem for suppressing signal transmission to said transmission route which contains said fault upon detection of the fault;

means in each subsystem for sending a signal representing the fault to said transmission path; and

means in at least one subsystem for determining a fault location based on said fault representing signal.

13. A distributed processing system having a plurality of subsystems interconnected through at least one transmission path, comprising:

means in at least one subsystem for sending an address train on said transmission path when power is supplied to said system; and

means in other subsystems for receiving said address train;

means in each subsystem receiving said address train for sending to said transmission path said address train including information relating to the address of the receiving subsystem; and

means for receiving said address train in at least one subsystem and for recognizing the address construction in said system by means of the received address train.

14. A distributed processing system according to claim 13, wherein said means for receiving said address train in at least one subsystem and for recognizing the address construction in said system operates to detect duplication of an address in said address train.

15. A distributed processing system according to claim 14, wherein said address train is circulated twice on the transmission path.

16. A distributed processing system according to claim 13, wherein said address train is circulated twice on the transmission path.

17. A distributed processing system having a plurality of subsystems interconnected through at least one transmission path, comprising:

means in each subsystem for sending a signal for diagnosing a fault on a transmission route including at least one other subsystem and said transmission path;

means in each subsystem for receiving a signal related to said diagnosing signal from a transmission route;

means for detecting a fault based on the signal relating to said diagnosing signal received in each subsystem;

means in each subsystem for controlling signal transmission to said transmission route containing said fault upon detection of the fault;

means in each subsystem for sending a signal for detecting recovery of the fault to said transmission route after detection of the fault;

means in each subsystem for receiving a signal relating to a recovery detecting signal from said transmission path;

means for detecting the recovery of the fault based on the signal relating to the recovery detecting signal received in each subsystem; and

means in each subsystem for sending a signal representing said recovery of the fault to said transmission path upon detection of the recovery.

18. A distributed processing system according to claim 17, which further comprises means in at least one subsystem, for determining a recovery location based on said recovery representing signal.

19. A distributed processing system according to one of claims 17 or 21, which further comprises means for reinstating the signal transmission on said transmission route containing said fault upon detection of the fault recovery.

20. A distributed processing system having a plurality of subsystems interconnected through at least one transmission path, comprising:

means in each subsystem for sending a signal for diagnosing a fault on a transmission route including at least one other subsystem and said transmission path;

means in each subsystem for receiving a signal relating to said diagnosing signal from said transmission route;

means for detecting a fault based on the signal relating to said diagnosing signal received in each subsystem;

means in each subsystem for sending a signal representing the fault to said transmission path; and

means in at least one subsystem for determining a fault location based on said fault representing signal.

21. A distributed processing system having a plurality of subsystems interconnected through at least one transmission path, comprising:

means in each subsystem for sending a signal for detecting recovery of a fault on a transmission route including at least one other subsystem and said transmission path;

means in each subsystem for receiving a signal relating to said recovery detecting signal from said transmission route;

means for detecting the recovery of the fault based on the signal relating to the recovery detecting signal received in each subsystem;

means in each subsystem for sending a signal representing said recovery of the fault to said transmission route upon detection of the recovery; and

means in at least one subsystem for determining a recovery location based on said recovery representing signal.

22. A distributed processing system having a plurality of subsystems interconnected through at least one transmission path, comprising:

means in each subsystem for sending a signal for diagnosing a fault on a transmission route including at least one other subsystem and said transmission path;

means in each subsystem for receiving a signal relating to said diagnosing signal from said transmission route;

means for detecting a fault based on the signal relating to said diagnosing signal received in each subsystem;

means in each subsystem for storing a history of fault detection; and

means in each subsystem for generating a signal for controlling signal transmission to said transmission route based on the stored history of fault detection.