Analysis of errors within computer code

• US 20070050679A1
• Filed: 08/30/2005
• Published: 03/01/2007
• Est. Priority Date: 08/30/2005
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Reddy, Venkat
• Granted Patent: US 7,707,559 B2
• Time in Patent Office: Days
• US Class Current: 714/38
• CPC Class Codes: G06F 11/3688 for test execution, e.g. scheduling of test suites

Abstract

Where code execution results in an error attributable to multiple data elements, the code is revised so the error results from one element. Where execution improperly functions without error, and the elements contain required but missing properties, the code is improperly functioning in its required-property handling. Errors are organized into classes and error-causing elements are organized into independent sets corresponding to the classes. Elements that are not within any set are determined. Symptom, error, no-error, refined-symptom, and function databases can be employed during code execution analysis. Symptom database entries correspond to elements and indicate errors, or that no errors, are attributable to the elements. Error database entries correspond to elements resulting in errors upon execution. No-error database entries correspond to elements resulting in no errors upon execution. Refined-symptom database entries correspond to root cause elements of errors. Function database entries correspond to root cause elements of no errors.

Claims

1. A method for analyzing computer code comprising:

executing computer code over first data and second data, the first data being at least substantially independent of the second data, and the first data having a plurality of elements;

at least one of;

where execution of the computer code results in a given error due to more than one element of the first data, revising the computer code so that the given error results from just one element of the first data;

where execution of the computer code improperly functions and no error results from execution of the computer code, and one or more elements of the first data each contains a property that is required but that is missing, determining that the computer code is improperly functioning in handling at least one of the properties that are required; and

,

where execution of the computer code results in a plurality of errors due to error-causing elements of the first data, organizing the errors into a plurality of error classes and organizing the error-causing elements of the first data into a plurality of independent sets of error-causing elements, each independent set of error-causing elements corresponding to an error class.

2. The method of claim 1, further comprising, wherein execution of the computer code results in the errors due to the error-causing elements of the first data, determining redundant elements of the first data that are not members of any of the independent sets of error-causing elements.

3. The method of claim 1, further comprising, after executing the computer code over the first data and the second data,

constructing an error database and a no-error database, the error database containing a plurality of entries corresponding to the elements of the first data due to which execution of the computer code results in errors, and the no-error database containing a plurality of entries corresponding to the elements of the first data due to which execution of the computer code does not result in errors.

4. The method of claim 3, further comprising, after executing the computer code over the first and the second data and prior to constructing the error database and the no-error database,

constructing a symptom database having a plurality of entries, each entry corresponding to an element of the first data and indicating one of:

an identity of an error attributable to the element, and that no error is due to the element,

wherein the error and the no-error databases are constructed from the symptom database.

5. The method of claim 3, further comprising, after constructing the error database and the no-error database, for each error within the error database,

determining which of the elements of the first data causing the error is a root cause element of the error resulting from execution of the computer code;

and,

removing the entries of the error database that correspond to the elements of the first data that caused the error but that are not a root cause element of the error.

6. The method of claim 5, wherein execution of the computer code resulting in a given error due to more than one element of the first data comprises more than one element of the first data being a root cause element of the given error, and

wherein revising the computer code so that the given error results from just one element of the first data comprises revising the computer code so that the given error results from just one root cause element of the first data.

7. The method of claim 5, wherein execution of the computer code resulting in the errors due to the error-causing elements of the first data comprises execution of the computer code resulting in the errors due to the error-causing elements that are root cause elements of the errors, and

wherein organizing the errors into error classes and organizing the error-causing elements into the independent sets of error-causing elements comprises organizing the error-causing elements that are root cause elements of the errors into the independent sets of error-causing elements.

8. The method of claim 5, wherein execution of the computer code resulting in the errors due to the error-causing elements of the first data comprises execution of the computer code resulting in the errors due to the error-causing elements that are root cause elements of the errors, and

wherein determining the redundant elements of the first data that are not members of any of the independent sets of error-causing elements comprises determining the redundant elements of the first data that are not members of any of the independent sets of error-causing elements that are root cause elements.

9. The method of claim 3, further comprising, after constructing the error database and the no-error database, constructing a refined-symptom database, comprising:

copying the error database as the refined-symptom database;

for each error within the error refined-symptom database,

determining which of the elements of the first data causing the error is a root cause element of the error resulting from execution of the computer code; and

,

removing the entries of the refined-symptom database that correspond to the elements of the first data that caused the error but that are not a root cause element of the error.

10. The method of claim 3, further comprising, after constructing the error database and the no-error database,

determining which of the elements of the first data causing no error is a root cause element of no error resulting from execution of the computer code;

and,

removing the entries of the no-error database that correspond to the elements of the first data that caused no error resulting from execution of the computer code but that are not a root cause element of no error resulting from execution of the computer code.

11. The method of claim 10, wherein execution of the computer code improperly functioning but resulting in no error comprises at least one of the elements of the first data that are root cause elements of no error resulting from execution of the computer code each having a property that is required but that is missing, and

wherein determining that the computer code is improperly functioning in handling at last one of the properties that are required comprises determining that the computer code is improperly functioning in handling at least one of the properties of the root cause elements that are required.

12. The method of claim 3, further comprising, after constructing the error database and the no-error database, constructing a function database, comprising:

copying the no-error database as the function database;

determining which of the elements of the first data causing no error is a root cause element of no error resulting from execution of the computer code; and

,

removing the entries of the function database that correspond to the elements of the first data that caused no error resulting from execution of the computer code but that are not a root cause element of no error resulting from execution of the computer code.

13. An analysis system comprising:

a fault injection mechanism to inject faults into configuration data for a target system, the configuration data having a plurality of elements;

a data collection mechanism to collect monitoring data on the target system upon execution of the target system based on the configuration data and customer-supplied data, the monitoring data including errors resulting from execution of the target system based on the configuration data having the faults injected thereinto; and

,

a data analysis mechanism to analyze the errors of the monitoring data collected by the data collection mechanism, wherein the data analysis mechanism performs at least one of;

recommend computer code of the target system to be revised, where execution of the computer code results in a given error due to more than one element of the configuration data, so that the given error instead results from just one element of the configuration data;

determine that the computer code of the target system is improperly functioning in handling of one or more properties of one or more elements of the configuration data that are required but that are missing, where execution of the computer code improperly functions but no error results; and

,

organize the errors into a plurality of error classes and organizing error-causing elements of the configuration data into a plurality of independent sets of error-causing elements, each independent set of error-causing elements corresponding to an error class, where execution of the computer code results in the errors due to the error-causing elements.

14. The analysis system of claim 13, wherein the data analysis mechanism further determines redundant elements of the configuration data that are not members of any of the independent sets of error-causing elements.

15. The analysis system of claim 13, further comprising a symptom database constructed by the data analysis mechanism and having a plurality of entries, each entry corresponding to an element of the configuration data and indicating one of:

an identity of an error attributable to the element, and that no error is due to the element.

16. The analysis system of claim 15, further comprising:

an error database constructed by the data analysis mechanism from the symptom database and containing a plurality of entries corresponding to the elements of the configuration data due to which execution of the computer code results in the errors; and

,

a no-error database constructed by the data analysis mechanism from the symptom database and containing a plurality of entries corresponding to the elements of the configuration data due to which execution of the computer code does not result in the errors.

17. The analysis system of claim 16, further comprising a refined-symptom database constructed by the data analysis mechanism from the error database and containing a plurality of entries corresponding to the elements of the configuration data that are root causes of the errors resulting from execution of the computer code.

18. The analysis system of claim 16, further comprising a function database constructed by the data analysis mechanism from the no-error database and containing a plurality of entries corresponding to the elements of the configuration data that are root causes of no errors resulting from execution of the computer code.

19. An article of manufacture comprising:

a computer-readable medium; and

,

means in the medium for executing computer code over data having a plurality of elements and, and for analyzing results of execution of the computer code by at least one of;

where execution of the computer code results in a given error due to more than one element of the data, recommending that the computer code be revised so that the given error results from just one element of the data;

where execution of the computer code improperly functions and no error results from execution of the computer code, and one or more elements of the data each contains a property that is required but that is missing, determining that the computer code is improperly functioning in handling at least one of the properties that are required; and

,

where execution of the computer code results in a plurality of errors due to error-causing elements of the data, organizing the errors into a plurality of error classes and organizing the error-causing elements of the data into a plurality of independent sets of error-causing elements, each independent set of error-causing elements corresponding to an error class.

20. The article of claim 19, further comprising, wherein the means is further for analyzing the results of execution of the computer code by, where execution of the computer code results in the errors due to the error-causing elements of the data, determining redundant elements of the data that are not members of any of the independent sets of error-causing elements.