Persistent stateful component-based applications via automatic recovery

• US 7,093,162 B2
• Filed: 06/29/2004
• Issued: 08/15/2006
• Est. Priority Date: 09/04/2001
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Lomet, David B., Barga, Roger S.
• Primary Examiner(s): Beausoliel, Robert
• Assistant Examiner(s): Wilson, Yolanda L
• Application Number:
  US10/879,324
• Publication Number:
  US 20040250157A1
• Time in Patent Office: 777 Days
• Field of Search: None
• US Class Current: 714/15
• CPC Class Codes: G06F 11/1438 Restarting or rejuvenating
• CPC Class Codes: G06F 11/1471 involving logging of persistent data for recovery
• CPC Class Codes: G06F 11/1482 by means of middleware or OS functionality

Abstract

Persistent components are provided across both process and server failures, without the application programmer needing take actions for component recoverability. Application interactions with a stateful component are transparently intercepted and stably logged to persistent storage. A “virtual” component isolates an application from component failures, permitting the mapping of a component to an arbitrary “physical” component. Component failures are detected and masked from the application. A virtual component is re-mapped to a new physical component, and the operations required to recreate a component and reinstall state up to the point of the last logged interaction is replayed from the log automatically.

Claims

1. A method of providing recovery in a computer system, comprising:

intercepting events comprising at least one of a plurality of method calls, component activations, and component deactivations;

logging the events during execution of stateful piece-wise deterministic (PWD) components having a component state;

detecting a failure of one of the components based on an error detected during execution;

replaying the logged events to transparently and automatically recover the component state; and

verifying that the component state has not already been recovered prior to recovering the component state.

2. The method of claim 1, further comprising:

intercepting non-deterministic events;

providing a policy mechanism that, upon interception of the events, permits the logging of the intercepted events;

providing an error handler that is invoked when an error is detected, such error handler being capable of masking the error from the component.

3. The method of claim 1, further comprising:

logging information about how the component is mapped to a first set of resources such that the component can be re-mapped to a different set of resources should a system failure result in a termination of the first set of resources, said re-mapping enabling scalability and load balancing.

4. The method of claim 3, wherein the information includes a logical identifier for the component that identifies logged state information for the component.

5. The method of claim 1, further comprising exactly-once execution comprising:

uniquely identifying messages at a sender;

detecting an original message and at least one duplicate message at a receiver;

eliminating the at least one duplicate message; and

providing a reply to the at least one duplicate message that is the same as a reply to the original message.

6. The method of claim 1, further comprising:

notifying an error handler when the error is detected;

contacting a recovery manager to recover the component that has failed;

re-trying an interaction that failed at the time of the failure; and

returning the result of the re-tried interaction to the component.

7. The method of claim 1, further comprising:

retrieving information related to the failed call; and

identifying the component based on the retrieved information.

8. The method of claim 1, further comprising:

updating references to the component with an interface pointer corresponding to the component; and

logging the recovery of the component.

9. The method of claim 1, further comprising intercepting the method call and response, and invoking a policy on at least one of the call and the response.

10. The method of claim 1, wherein the computer system comprises a client and a server, and detecting the failure of the component comprises the client or server detecting the failure, and further comprising a component runtime associated with the client invoking an error handler associated with the client and providing information related to the failed call to the error handler.

11. The method of claim 10, further comprising identifying a recovery manager responsible for the component and passing the information to the recovery manager.

12. A system for providing recovery in a computer system, comprising:

a storage device that logs events comprising at least one of a plurality of method calls, component activations, and component deactivations during execution of stateful piece-wise deterministic (PWD) components having a component state;

a component runtime that receives the events from the storage device;

a error handler that receives the events from the component runtime and identifies the component based on the events; and

a recovery manager that receives the events from the storage device and replays the events to transparently and automatically recover the component state, wherein the recovery manager logs the recovery of the component state in the storage device, verifies that the component state has not already been recovered prior to recovering the component state, and communicates to the error handler additional information about the component.

13. The system of claim 12, wherein the error handler is notified of an error and contacts the recovery manager to recover the component state.

14. The system of claim 12, wherein the component runtime intercepts the at least one method call and invokes a policy on the at least one method call.

15. The system of claim 12, further comprising an activation hook that logs information in the storage device that is sufficient to re-create the component.

16. A computer readable medium having computer-executable instructions for performing the steps comprising:

intercepting events comprising at least one of a plurality of method calls, component activations, and component deactivations;

logging the events during execution of at least one stateful piece-wise deterministic (PWD) component having a component state in a computer system;

detecting a failure of one of the components based on an error detected during execution;

replaying the logged events to transparently and automatically recover the component state; and

verifying that the component state has not already been recovered prior to recovering the component state.

17. The computer readable medium of claim 16, having further computer-executable instructions for:

intercepting non-deterministic events;

providing a policy mechanism that, upon interception of the events, permits the logging of the intercepted events;

providing an error handler that is invoked when an error is detected, such error handler being capable of masking the error from the component.

18. The computer readable medium of claim 16, having further computer-executable instructions for:

logging information about how the component is mapped to a first set of resources such that the component can be re-mapped to a different set of resources should a system failure result in a termination of the first set of resources, said re-mapping enabling scalability and load balancing.

19. The computer readable medium of claim 18, wherein the information includes a logical identifier for the component that identifies logged state information for the component.

20. The computer readable medium of claim 16, having further computer-executable instructions for providing exactly-once execution comprising:

uniquely identifying messages at a sender;

detecting an original message and at least one duplicate message at a receiver;

eliminating the at least one duplicate message; and

providing a reply to the at least one duplicate message that is the same as a reply to the original message.

21. The computer readable medium of claim 16, having further computer-executable instructions for:

notifying an error handler when the error is detected;

contacting a recovery manager to recover the component that has failed;

re-trying an interaction that failed at the time of the failure; and

returning the result of the re-tried interaction to the component.

22. The computer readable medium of claim 16, having further computer-executable instructions for:

retrieving information related to the failed call; and

identifying the component based on the retrieved information.

23. The computer readable medium of claim 16, having further computer-executable instructions for:

updating references to the component with an interface pointer corresponding to the component; and

logging the recovery of the component.