Multiple sup swap mechanism

• US 4,894,828 A
• Filed: 12/22/1987
• Issued: 01/16/1990
• Est. Priority Date: 12/22/1987
• Status: Expired due to Term

• Current Assignee: Amdahl Corporation (Fujitsu Limited)
• Original Assignee: Amdahl Corporation (Fujitsu Limited)
• Inventors: Novy, Robert, Chimenti, Moreno A., Guyon, Richard
• Primary Examiner(s): Smith, Jerry
• Assistant Examiner(s): BEAUSOLIEL JR, ROBERT W
• Application Number:
  US07/136,409
• Time in Patent Office: 756 Days
• Field of Search: 371/7, 371/8, 371/9, 371/11, 371/12, 371/62, 371/68, 364/200 MS File, 364/900 MS File, 364/184, 364/187, 364/132
• US Class Current: 714/11
• CPC Class Codes: G06F 11/20 using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
• CPC Class Codes: G06F 11/202 where processing functionality is redundant redundant communication control functionality G06F11/2005, redundant storage control functionality G06F11/2089
• CPC Class Codes: G06F 11/2736 using a dedicated service processor for test

Abstract

In a data processing system having a main processor and a plurality of service processors providing service functions to the main processor, including a first service processor providing primary service functions and a second service processor providing secondary service functions. The main processor performs failure detection on the first processor, and upon the detection of a failure, initiates a switch between the first and the second service processors.

Claims

1. In a data processing system having at least one main processor performing system tasks and a plurality of service processors, including a first service processor performing primary service functions and a second service processor performing secondary service functions, a method for providing service facilities from said service processors, comprising the steps of:

using a main processor to detect a failure indication in said first service processor; and

upon detection of a failure indication, transferring said primary service functions from said first service processor to said second service processor.

2. A method as in claim 1 wherein said step of using the main processor to detect a failure indication in said first service processor further comprises the step of monitoring time durations of service functions performed by said first service processor and generating said failure indication if at least one service function is not completed within a predetermined time period.

3. A method as in claim 2, further including the step of using the main processor to initiate a reset in said first service processor upon detection of said failure indication.

4. A method as in claim 3, further including the step of transferring said secondary service functions to said first service processor after said reset.

5. A method as in claim 1 wherein said step of using the main processor to detect a failure indication in said first service processor includes the steps of sending an inquiry to said first service processor and generating said failure indication if an acknowledgement thereof is not received by said main processor within a predetermined time period.

6. A method as in claim 5, further including the step of using the main processor to initiate a reset to said first service processor upon detection of said failure indication.

7. A method as in claim 6, further including the step of transferring said secondary service functions to said first service processor after said reset.

8. In a data processing system having at least one main processor performing system tasks and a plurality of service processors, including a first service processor performing primary service functions and a second service processor performing secondary service functions, a method for providing service facilities from said service processors, comprising the steps of:

monitoring in said main processor time durations of service functions performed by said first service processor and generating a failure indication if at least one service of said functions is not completed within a predetermined time period;

sending an inquiry to said first service processor and generating a failure indication if an acknowledgement thereof is not received within a predetermined time period; and

upon detection of said failure indication, the steps of;

transferring said primary service functions from said first service processor to said second service processor;

initiating a reset in said first service processor; and

transferring said secondary service functions to said first service processor after said reset.

9. In a data processing system having at least one main processor performing system tasks and a plurality of service processors, including a first service processor performing primary service functions and a second service processor performing secondary service functions, apparatus for providing service facilities from said service processors, comprising:

first means in said main processor for detecting a failure indication in said first service processor; and

second means responsive to said failure indication for transferring said primary service functions from said first service processor to said second service processor.

10. The apparatus as in claim 9 wherein said first means comprises means for monitoring time durations of service functions performed by said first service processor.

11. The apparatus as in claim 10 further including means in said main processor for initiating a reset in said first service processor if at least one service function is not completed by said first service processor within a predetermined time period.

12. The apparatus as in claim 11 further including means in said main processor for transferring said secondary functions to said first service processor after said reset.

13. The apparatus is in claim 9 wherein said first means comprises means for sending an inquiry to said second service processor and means for monitoring for an acknowledgement thereof.

14. The apparatus as in claim 13 further including means in said main processor for initiating a reset in said first service processor if said acknowledgement is not received within a predetermined time period.

15. The apparatus as in claim 14 further including means in said main processor for transferring said secondary functions to said first service processor after said reset.

16. In a data processing system having at least one main processor performing system tasks and a plurality of service processors, including a first service processor performing primary service functions and a second service processor performing secondary service functions, apparatus for providing service facilities from said service processors, comprising:

first means in said main processor for detecting failure indications in said first service processor, comprising;

means for monitoring time durations of service functions performed by said first service processor and for generating a failure indication if at least one service function is not completed within a predetermined time period; and

means for sending an inquiry to said second service processor and means for monitoring for an acknowledgement of said inquiry and for generating a failure indication if said acknowledgement is not received within a predetermined time period;

second means responsive to said failure indication from said first means for transferring said primary service functions from said first service processor to said second service processor;

means for initiating a reset in said first service processor; and

means for transferring said secondary functions to said first service processor after said reset.

17. In a data processing system having a first main processor and a second main processor performing system tasks, a first service processor performing primary service functions and a second service processor performing secondary service functions, a method for improving the availability of service facilities from said service processors, comprising the steps of:

cross-coupling said service processors to said main processors;

assigning a time limit to at least a selected one of said primary service functions;

upon requesting said selected primary service function from said first service processor, generating a first failure indication if a completion response of said selected primary service function is not received within said time limit from said first processor;

periodically polling said first processor for a second response;

generating a second failure indication if a response is not received for said polling;

responsive to one of said first and second failure indications, determining whether said second service processor is operational; and

if said second service processor is operational, the steps of;

transferring said primary service functions to said second service processor;

initiating a reset to said first service processor; and

transferring said secondary service functions to said second service processor.

18. In a data processing system having at least one main processor performing system tasks and a plurality of service processors, including a first service processor performing primary service functions including monitoring of said main processing system and a second service processor performing secondary service functions including monitoring of said first service processor, a method comprising the steps of:

monitoring in said main processor system service functions performed by said first service processor and generating a failure indication if at least one service function is not performed,

transferring said service functions from one of said service processors to another one of said service processors upon detecting said failure indication.