Method and system for automated handling of resolvable and non-resolvable errors in execution of system management flows consisting of system management tasks

• US 7,840,857 B2
• Filed: 09/18/2007
• Issued: 11/23/2010
• Est. Priority Date: 09/25/2006
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Breiter, Gerd, Rost, Steffen, Spatzier, Thomas, Maass, Ruediger
• Primary Examiner(s): Iqbal; Nadeem
• Application Number:
  US11/856,759
• Publication Number:
  US 20080077935A1
• Time in Patent Office: 1,162 Days
• Field of Search: 714 8- 13, 714/15, 714/16, 714/20, 714/26, 714/27, 714/31, 714/37, 714/39, 714/48, 714/49, 714/51, 714/57
• US Class Current: 714/48
• 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 11/0715 in a system implementing multitasking multitasking per se G06F9/46
• 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
• CPC Class Codes: G06F 11/28 by checking the correct order of processing G06F11/08 - G06F11/26 take precedence; monitoring patterns of pulse trains H03K5/19

Abstract

The present invention provides a method and system for automated handling of resolvable and non-resolvable errors in the execution of system management flows by enhancing Workflow Engines (30) by an Error Handling component (40) and by adding a supportive Error Handling Layer (60, 90, 100) to invoked system management tasks which serves as the counterpart to the Workflow Engine'"'"'s Error Handling component (see FIG. 2). The additional functionality of the task-provided Error Handling Layer (90, 70, 100) is accessible for the Workflow Engine via extended web services interfaces (91, 71, 101). The Workflow Engine'"'"'s Error Handling component (40) and the task-provided Error Handling Layer allow for the definition of a standard protocol between the Workflow Engine and invoked tasks (31) for automated error handling.

Claims

1. A method for automated handling of resolvable and non-resolvable errors in system management flows consisting of system management tasks, wherein said system management tasks (10-12) are provided by system management applications to perform certain operations on the managed IT infrastructure (1), wherein said system management tasks (10-12) provide interfaces (13-15) via said system management tasks which are invoked (20) during the execution of the system management flow definition by a Workflow Engine (30),

characterized by the steps of:

executing a system management flow definition (35) by said Workflow Engine (30), wherein said system management flow definition additionally includes an ErrorHandlingStrategy attribute for determining the processing mode of the system management flow in the case of errors, wherein said processing mode can be “

compensate”

, or “

forced”

,

intercepting execution of a task as defined in said system management flow definition by said Workflow Engine (30),

invoking Error Handling component (40) via “

invokeTask interface”

(41), wherein said Error Handling component (40) uses an error handling interface with an Error Handling Layer (90, 70, 100) for invocation of system management tasks and for error handling, wherein said Error Handling Layer (90, 70, 100) uses a Processing Context Store (80) storing processing state information of invoked tasks,

wherein said Error Handling component (40) performs the steps of;

invoking a system management task as defined by said task via the system management task'"'"'s startProcessing operation (72) of said error handling interface by said Error Handling Component (40) and storing processing state information of said invoked task in a Processing Context Store (68), wherein said Processing Context Store (68) contains information of each task being invoked including a unique task ID, a system management flow definition ID to which said task belongs to, processing state information of said invoked task, wherein said processing state can be started, completed, or compensated, and contains information for resuming processing of said invocated task,

receiving error condition via the “

suspendProcessing”

operation (44) of said error handling interface from said Error Handling Layer (90, 70, 100),

switching to the processing mode as defined in the “

ErrorHandlingStrategy”

, and

processing said system management flow in a processing mode as defined in said “

ErrorHandlingStrategy”

accordingly.

2. The method according to claim 1, wherein said switching to said processing mode as defined in said “

ErrorHandlingStrategy”

is accomplished automatically without prior evaluation of said error.

3. The method according to claim 1, wherein said error condition includes a detailed error description.

4. The method according to claim 2, wherein said evaluation is accomplished by an internal or external Error Resolving Device (32) by using said error description.

5. The method according to claim 3, wherein said Error Resolving Device (32) is an Expert system.

6. The method according to claim 1, further including the steps of:

receiving a non-error condition via the “

delegateProcessing”

operation (43) of said error handling interface from said Error Handling Layer if the processing is done,

giving back control to said Workflow Engine for continuing execution of the subsequent system management task as defined in said system management flow definition (35).

7. The method according to claim 1, wherein said ErrorHandlingStrategy is valid for the entire system management flow.

8. The method according to claim 1, wherein said Error Handling Strategy attribute is defined on the top-level XML element of the BPEL system management flow definition if BPEL is used as the language for encoding system management flow definitions.

9. The method according to claim 1, further including the steps of:

invoking a system management task consisting of sub tasks via said start processing operation of said error handling interface,

creating for each sub task a processing state in said processing context database (80) by said Error Handling Layer (90, 70, 100),

logging said current processing state if an error occurs,

setting the processing state “

processing state suspended”

of said Error Handling Layer, and

invoking said Error Handling component via said suspendProcessing operation (44) of said error handling interface with an error condition.

10. The method according to claim 6, wherein in the case of switching from the normal processing mode to the processing mode “

compensate”

the following further steps are performed;

loading said previously stored process state information from said Processing Context Store (68),

invoking the last active task according to said process state information via said error handling interface (71),

compensating the work previously done by said task by using processing state information previously stored in said Process Context Store (80),

repeating those steps for all previously invoked tasks in the reverse order according to process state information loaded from said Process Context Store (68).

11. The method according to claim 6, wherein in the case of switching from the normal processing mode to the processing mode “

forced”

the following further steps are performed;

invoking the previously failed task according to the Process Context Store (68),

ignoring all subsequent errors in said previously failed task,

continuing said system management flow in said forced processing mode,

ignoring all errors in said subsequent tasks of said system management flow.

12. The method according to claim 11, wherein said error descriptions of all ignored errors are logged in said Processing Context Store (68).

13. A system for automated error handling of resolvable and non-resolvable errors in system management flows consisting of system management tasks, wherein said system management tasks (10-12) are provided by system management applications to perform certain operations on the managed IT infrastructure (1), wherein said system management tasks (10-12) provide interfaces (13-15) via said system management tasks to be get invoked (20) during the execution of the system management flow definition,

characterized by the means of:

a Workflow Engine (30) for executing a system management flow definition, wherein said system management flow definition additionally includes an ErrorHandlingStrategy attribute for determining the processing mode of the system management flow, wherein said processing mode can be “

compensate”

, or “

forced”

, wherein said Workflow Engine including;

means for intercepting execution of a task as defined in said system management flow definition,

means for invoking Error Handling component (40) via “

invokeTask interface”

(41),

an Error Handling Component which uses a Processing Context Store (68) to store processing state information, wherein said Processing Context Store (68) contains information of each task being invoked including a unique task ID, a system management flow definition ID to which said task belongs to, processing state information of said invoked task, wherein said processing state can be started, completed, or compensated, and contains information for resuming processing of said invocated task, and further, wherein said Error Handling component uses an error handling protocol with an Error Handling Layer (90, 70, 100) for invoked system management tasks, wherein said Error Handling Layer uses a Processing Context Store (80) to store processing state information, wherein said Error Handling component including;

means for invoking a system management task as defined by said system management flow definition via the system management task'"'"'s startProcessing interface operation (72),

means for receiving error condition via interface “

suspendProcessing”

(44) of said error handling interface from said Error Handling Layer, and

means for switching to the processing mode as defined in the “

ErrorHandlingStrategy”

.

14. The system according to claim 13, wherein said means for switching to said processing mode as defined in the “

ErrorHandlingStrategy”

includes means for prior evaluation that said error is non-resolvable.

15. The system according to claims 13, wherein said means for switching to said processing mode as defined in said “

ErrorHandlingStrategy”

is automatically used without prior evaluation of said error.

16. The system according to claim 13, wherein said error condition includes a detailed error description.

17. The system according to claim 14, wherein said means for evaluation is an internal or external Error Resolving Device (32).

18. The system according to claim 13, further including means for receiving a non-error condition via the

“

delegateProcessing”

operation (43) of said error handling interface from said Error Handling Layer if the processing is done and for giving back control to said Workflow Engine for continuing execution of the subsequent task as defined in said system management flow definition (35).