ERROR IDENTIFICATION

• US 20110320873A1
• Filed: 06/24/2010
• Published: 12/29/2011
• Est. Priority Date: 06/24/2010
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Bryant, Barbara J., Abrams, Robert M., Durand, Donald T., Ng, Ming-Yin
• Granted Patent: US 8,762,783 B2
• Time in Patent Office: Days
• US Class Current: 714/37
• CPC Class Codes: G06F 11/3636 by tracing the execution of the program
• CPC Class Codes: G06F 11/366 using diagnostics G06F11/0703 takes precedence

Abstract

A method of identifying errors in a computing system operation is provided and includes identifying that a certain record of interest in system trace information has a number of entries that exceeds a predefined number and inferring from the excessive number of entries that a work unit associated with the certain record of interest is affected by an error.

Claims

1. A method of identifying errors in a computing system operation, the method comprising:

identifying that a certain record of interest in system trace information has a number of entries that exceeds a predefined number; and

inferring from the excessive number of entries that a work unit associated with the certain record of interest is affected by an error.

2. The method according to claim 1, wherein the system trace information is provided in a copy of a system trace table provided for each processor in the computing system.

3. The method according to claim 1, further comprising sorting and organizing the system trace information according to a system hierarchy.

4. The method according to claim 3, wherein the sorting and organizing comprises:

arranging the system trace information by job, task and count; and

constructing a system trace information array according to the job, the task and the count.

5. The method according to claim 1, wherein the entries of interest comprise those entries not removed as noise or discarded as relating to special processors and/or being representative of branch and/or other miscellaneous events.

6. The method according to claim 1, wherein the predefined number is updateable.

7. The method according to claim 1, further comprising reporting the error to an operator.

8. The method according to claim 1, further comprising issuing a cancel command to cancel the work unit.

9. A method of identifying execution loops in a computing system, comprising:

obtaining system trace information;

sorting and organizing the system trace information;

identifying that a certain type of record in the system trace information has a first predefined number of entries;

removing noise entries and discarding certain other entries; and

in an event a number of remaining entries is greater than a second predefined number, inferring that a work unit associated with the certain type of record is in a loop.

10. The method according to claim 9, wherein the system trace information is provided in a copy of a system trace table.

11. The method according to claim 10, wherein the system trace table is provided for each processor in the computing system.

12. The method according to claim 9, wherein the sorting and organizing comprises arranging the system trace information according to a system hierarchy.

13. The method according to claim 9, wherein the sorting and organizing comprises arranging the system trace information by job, task and count in an array.

14. The method according to claim 9, wherein the certain other entries are related to special processors and/or being representative of branch and/or other miscellaneous events.

15. The method according to claim 9, wherein the first and second predefined numbers are updateable.

16. The method according to claim 15, further comprising evaluating an accuracy of the inferring and reducing or increasing the first or second predefined numbers accordingly.

17. The method according to claim 9, further comprising reporting the loop to an operator.

18. The method according to claim 9, further comprising issuing a cancel command to cancel the work unit.

19. The method according to claim 18, further comprising issuing a force command to force cancel the work unit in an event the cancel command is ineffective.

20. A non-transitory computer readable medium having executable instructions stored thereon, which, when executed cause a processing unit of a computing system to perform a method of identifying errors in an operation of the computing system, the method comprising:

identifying that a certain record of interest in system trace information has a number of entries that exceeds a predefined number; and

inferring from the excessive number of entries that a work unit associated with the certain record of interest is affected by an error.