First failure data capture

• US 7,080,287 B2
• Filed: 07/11/2002
• Issued: 07/18/2006
• Est. Priority Date: 07/11/2002
• Status: Expired due to Fees

• Current Assignee: LinkedIn Corporation (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: Salem, Hany A.
• Primary Examiner(s): Beausoliel, Robert
• Assistant Examiner(s): Manoskey, Joseph D
• Application Number:
  US10/195,181
• Publication Number:
  US 20040024726A1
• Time in Patent Office: 1,468 Days
• Field of Search: 714/25, 714/26, 714/30, 714/38, 706/45
• US Class Current: 714/38.11
• CPC Class Codes: G06F 11/0706 the processing taking place on a specific hardware platform or in a specific software environment
• 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 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40

Abstract

An example of a solution provided here comprises: selecting an incident that requires further processing, capturing data associated with said incident, outputting captured data, and outputting a description for said captured data, whereby problem-solving is promoted. Another example comprises: providing runtime features for data capture, selecting an incident that requires further processing, capturing data on a thread that encounters said incident, and outputting captured data. In some cases, such a solution might include comparing a current incident to known incidents, and if a match is found, retrieving information that is relevant to said current incident. In some cases, such a solution might include taking recovery action or corrective action in response to said incident. Methods for handling errors, systems for executing such methods, and instructions on a computer-usable medium, for executing such methods, are provided.

Claims

1. A method for handling errors, said method comprising:

detecting an occurrence of an incident in a thread executing in a runtime environment of a data processing system;

providing incident information regarding the incident to an incident filter, wherein the incident information identifies an incident type;

comparing, in the incident filter, the incident type of the incident to a list of incident types requiring further processing;

selecting the incident for further processing if the incident has an incident type that matches an incident type in the list of incident types requiring further processing;

identifying a diagnostic engine associated with the thread, wherein each thread in the runtime environment has a separate associated diagnostic engine; and

determining if data capture for the incident is to be performed based on a list of recent incidents processed by the identified diagnostic engine, wherein if the incident is selected for further processing and data capture for the incident is to be performed, the method further comprises;

capturing data associated with said incident;

outputting captured data; and

outputting a description for said captured data, whereby problem-solving is promoted.

2. The method of claim 1, further comprising:

keeping code for said capturing substantially separate from in-line paths.

3. The method of claim 1, wherein the list of incident types that require further processing is maintained in the identified diagnostic engine.

4. The method of claim 1, wherein determining if data capture for the incident is to be performed based on a list of recent incidents processed by the identified diagnostic engine further comprises:

preventing multiple invocation of said capturing data for a same incident based on the list of recent incidents processed by the identified diagnostic engine.

5. The method of claim 1, wherein said capturing further comprises at least one action chosen from:

producing a thread dump;

producing a process dump; and

producing a core dump.

6. The method of claim 1, wherein said capturing further comprises:

parsing exceptions; and

based on said parsing, selecting data for capture.

7. The method of claim 1, wherein said capturing further comprises:

performing operations on multiple servers.

8. The method of claim 1, further comprising:

customizing said capturing based on said incident'"'"'s characteristics.

9. The method of claim 1, further comprising:

comparing a current incident to known incidents; and

if a match is found, retrieving information that is relevant to said current incident.

10. The method of claim 1, further comprising:

taking corrective action in response to said incident.

11. The method of claim 1 further comprising:

creating a diagnostic module;

registering said diagnostic module with the identified diagnostic engine; and

making data available to said diagnostic module.

12. The method of claim 1, further comprising:

bundling said incident in a persistent form that gathers data from multiple components.

13. The method of claim 1, wherein said capturing further comprises:

initiating boundaries for an incident;

capturing general data for said incident,

logging data from components that are involved with said incident; and

bundling output when processing is complete.

14. The method of claim 13, wherein said initiating boundaries further comprises at least one action chosen from:

associating an incident with an output stream; and

creating an output stream for said incident.

15. The method of claim 13, wherein said capturing general data further comprises capturing at least one item chosen from:

date;

timestamp;

classpath;

class loader; and

configuration settings.

16. The method of claim 1, further comprising:

suppressing said capturing during certain states.

17. The method of claim 1, further comprising:

temporarily logging data that can be retrieved later if an incident occurs.

18. The method of claim 1, further comprising:

comparing a current incident to known incidents; and

if a match is found, retrieving information that is relevant to said current incident.

19. The method of claim 1, further comprising:

taking corrective action in response to said incident.

20. A system for handling errors, said system comprising:

means for detecting an occurrence of an incident in a thread executing in a runtime environment of a data processing system;

means for providing incident information regarding the incident to an incident filter, wherein the incident information identifies an incident type;

means for comparing, in the incident filter, the incident type of the incident to a list of incident types requiring further processing;

means for selecting the incident for further processing if the incident has an incident type that matches an incident type in the list of incident types requiring further processing;

means for identifying a diagnostic engine associated with the thread, wherein each thread in the runtime environment has a separate associated diagnostic engine;

means for determining if data capture for the incident is to be performed based on a list of recent incidents processed by the identified diagnostic engine;

means for capturing data associated with said incident;

means for outputting captured data; and

means for outputting a description for said captured data, wherein said means for capturing data, means for outputting captured data, and means for outputting a description of said captured data operate if the incident is selected for further processing and data capture for the incident is to be performed.

21. The system of claim 20, wherein said means for capturing further comprises:

code that is substantially separate from in-line paths.

22. The system of claim 20, wherein the list of incident types that require further processing is maintained in the identified diagnostic engine.

23. The system of claim 20, wherein said means for determining if data capture for the incident is to be performed based on a list of recent incidents processed by the identified diagnostic engine further comprises:

means for preventing multiple invocation of said means for capturing data, for a same incident based on the list of recent incidents processed by the identified diagnostic engine.

24. The system of claim 20, wherein said means for capturing further comprises means for performing at least one action chosen from:

producing a thread dump;

producing a process dump; and

producing a core dump.

25. The system of claim 20, wherein said means for capturing further comprises:

means for parsing exceptions; and

means responsive to said parsing, for selecting data for capture.

26. The system of claim 20, wherein said means for capturing further comprises:

means for performing operations on multiple servers.

27. The system of claim 20, further comprising:

means for customizing said means for capturing, based on said incident'"'"'s characteristics.

28. The system of claim 20, further comprising:

means for comparing a current incident to known incidents; and

means for retrieving information that is relevant to said current incident, if a match is found.

29. The system of claim 20, further comprising:

means for taking corrective action in response to said incident.

30. A computer readable storage medium having computer-executable instructions for handling errors, said computer-executable instructions comprising:

means for detecting an occurrence of an incident in a thread executing in a runtime environment of a data processing system;

means for providing incident information regarding the incident to an incident filter, wherein the incident information identifies an incident type;

means for comparing, in the incident filter, the incident type of the incident to a list of incident types requiring further processing;

means for selecting the incident for further processing if the incident has an incident type that matches an incident type in the list of incident types requiring further processing;

means for identifying a diagnostic engine associated with the thread, wherein each thread in the runtime environment has a separate associated diagnostic engine;

means for determining if data capture for the incident is to be performed based on a list of recent incidents processed by the identified diagnostic engine;

means for capturing data associated with said incident;

means for outputting captured data; and

means for outputting a description for said captured data, wherein said means for capturing data, means for outputting captured data, and means for outputting a description of said captured data operate if the incident is selected for further processing and data capture for the incident is to be performed.

31. The computer readable storage medium of claim 30, wherein said means for capturing further comprises:

code that is substantially separate from in-line paths.

32. The computer readable storage medium of claim 30, wherein the list of incident types that require further processing is maintained in the identified diagnostic engine.

33. The computer readable storage medium of claim 30, wherein said means for determining if data capture for the incident is to be performed based on a list of recent incidents processed by the identified diagnostic engine further comprises:

means for preventing multiple invocation of said means for capturing data, for a same incident based on the list of recent incidents processed by the identified diagnostic engine.

34. The computer readable storage medium of claim 30, wherein said means for capturing further comprises means for performing at least one action chosen from:

producing a thread dump;

producing a process dump; and

producing a core dump.

35. The computer readable storage medium of claim 30, wherein said means for capturing further comprises:

means for parsing exceptions; and

means responsive to said parsing, for selecting data for capture.

36. The computer readable storage medium of claim 30, wherein said means for capturing further comprises:

means for performing operations on multiple servers.

37. The computer readable storage medium of claim 30, further comprising:

means for customizing said means for capturing, based on said incident'"'"'s characteristics.

38. The computer readable storage medium of claim 30, further comprising:

means for comparing a current incident to known incidents; and

means for retrieving information that is relevant to said current incident, if a match is found.

39. The computer readable storage medium of claim 30, further comprising:

means for taking corrective action in response to said incident.