Method and system for reliably de-allocating resources in a networked computing environment

• US 7,386,615 B1
• Filed: 05/10/2002
• Issued: 06/10/2008
• Est. Priority Date: 05/10/2002
• Status: Expired due to Term

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Surlaker, Kapil, Saha, Debashis, Gollapudi, Sreenivas, Chidambaran, Lakshminarayanan
• Primary Examiner(s): Thomson; William
• Assistant Examiner(s): Hoang; Phuong N
• Application Number:
  US10/144,246
• Time in Patent Office: 2,223 Days
• Field of Search: 718/104, 709226-227, 709/222, 714/4, 714/2, 714/48, 714/100
• US Class Current: 709/226
• CPC Class Codes: G06F 9/5011 the resources being hardware resources other than CPUs, Servers and Terminals
• CPC Class Codes: H04L 43/0811 by checking connectivity
• CPC Class Codes: H04L 67/14 Session management for real-time applications in data packet communications networks H04L65/1066

Abstract

A system and method is disclosed that allows resources to be shared among multiple entities, but which can appropriately release system resources after the failure or shutdown of the calling entity. In one approach, a monitoring entity is available to check for session failures. If a session failure is detected, the session is identified in a shared list that is accessible to other related entities. The related entities can be configured to piggyback a message to an appropriate node to kill, shutdown, or release resources associated with the failed session. Alternatively, a related entity can be specifically initiated to send a message to the appropriate node to kill, shutdown, or release the resources.

Claims

1. A method for deallocating a remote resource after termination of a local resource associated with the remote resource, the local resource residing on a local computing node, the remote resource residing on a remote computing node, in which one of the remote resource and the local resource was initiated by the other, comprising:

a) monitoring, at the local computing node, for termination of the local resource residing on the local computing node;

b) determining, at the local computing node, if an abnormal condition is detected wherein termination of the local resource due to the abnormal condition does not result in expected deallocation of the remote resource, in which the remote resource resides on the remote computing node, and one of the local resource and the remote resource was initiated by the other;

c) identifying information about the abnormal condition in one or more lists;

d) checking the one or more lists to identify the remote resource that should be deallocated;

e) causing, by the local computing node, a deallocation message to be sent to the remote computing node to deallocate the remote resource; and

f) causing, by the local computing node, the remote computing node to deallocate the remote resource.

2. The method of claim 1 in which the step (d) is performed by an entity that sends a message to the remote computing node.

3. The method of claim 2 in which the deallocation message is piggybacked to the message.

4. The method of claim 1 in which step (a) is performed by a process monitor.

5. The method of claim 1 in which the one or more lists is maintained in a shared memory region.

6. The method of claim 1 in which the local resource is locally processed by an entity within a first process pool and the deallocation message can be sent only by any process within the first process pool.

7. The method of claim 1 in which a monitoring process performs step (d).

8. The method of claim 7 in which the monitoring process periodically performs step (d).

9. The method of claim 7 in which the monitoring process instructs another process to perform step (e).

10. The method of claim 7 wherein the monitoring process performs step (e).

11. The method of claim 1 wherein the local resource is locally processed by an entity within a first process pool and the deallocation message can be sent by any process on the local computing node.

12. The method of claim 1, further comprising:

adding the abnormal condition to the one or more lists, wherein the abnormal condition is not identified in the one or more lists.

13. The method of claim 1, further comprising:

identifying the remote computing node or the remote resource initiated by the local resource in the one or more lists; and

causing an instruction for releasing the remote resource initiated by the local resource to be sent to the remote computing node.

14. The method of claim 1, in which the information about the abnormal condition comprises a first information of the remote computing node.

15. The method of claim 1, further comprising:

checking the one or more lists to identify a second remote resource residing on the remote computing node which is not initiated by the abnormal condition but should be deallocated; and

causing the remote computing node to deallocate the second remote resource.

16. A method for coordinating the lifetime of a remote session at a remote computing node to the lifetime of a local session at a local computing node, in which the remote session is de-coupled from a specific connection between the remote computing node and the local computing node, comprising:

a) monitoring, at the local computing node, for termination of the local session residing on the local computing node, in which the local session was initiated by a process in a process pool;

b) determining if the termination of the local session was abnormal so as to cause the remote session not to have appropriately terminated, in which the remote session resides on the remote computing node and was initiated by the local session; and

c) causing by the local computing node a deallocation message to be sent to the remote computing node to release the inappropriately terminated remote session associated with the abnormally terminated local session, in which the action of causing a deallocation message to be sent is performed by a process at the local computing node.

17. The method of claim 16 further comprising the step of terminating the remote session.

18. The method of claim 17 further comprising the step of deallocating resources associated with the remote session.

19. The method of claim 16 in which the deallocation message is piggybacked to another message sent to the remote computing node.

20. The method of claim 16 in which the local and remote sessions are database sessions.

21. The method of claim 16 further comprises:

maintaining one or more lists to track remote sessions that should be terminated.

22. The method of claim 21 in which a monitoring process periodically checks the one or more lists.

23. The method of claim 22 in which the monitoring process instructs another process to perform step (c).

24. The method of claim 22 wherein the monitoring process performs step (c).

25. A computer program product comprising a computer storage medium having executable code to execute a method for deallocating a remote resource after termination of a local resource associated with the remote resource, the local resource residing on a local computing node, the remote resource residing on a remote computing node, in which the remote resource was initiated by the local resource, the method comprising the steps of:

a) monitoring, at the local computing node, for termination of the local resource residing on the local computing node;

b) determining, at the local computing node, if an abnormal condition is detected wherein termination of the local resource due to the abnormal condition does not result in expected deallocation of the remote resource, in which the remote resource resides on the remote computing node, and one of the local resource and the remote resource was initiated by the other;

c) identifying information about the abnormal condition in one or more lists;

d) checking the one or more lists to identify the remote resource that should be deallocated;

e) causing, by the local computing node, a deallocation message to be sent to the remote computing node to deallocate the remote resource; and

f) causing, by the local computing node, the remote computing node to deallocate the remote resource.

26. The computer program product of claim 25 in which the deallocation message is piggybacked on to another message.

27. The computer program product of claim 25 in which the local resource is processed by an entity within a first process pool and the deallocation message can be sent only by any process within the first process pool.

28. The computer program of claim 25 in which a monitoring process performs step (d) and instructs another process to perform step (e).

29. The computer program product of claim 25 wherein a monitoring process performs step (d) and step (e).

30. A computer program product comprising a computer storage medium having executable code to execute a method for coordinating the lifetime of a remote session at a remote computing node to the lifetime of a local session at a local computing node, in which the remote session is de-coupled from a specific connection between the remote computing node and the local computing node, the method comprising the steps of:

a) monitoring, at the local computing node, for termination of the local session residing on the local computing node, in which the local session was initiated by a process in a process pool;

b) determining if the termination of the local session was abnormal such that the remote session has not appropriately terminated, in which the remote session resides on the remote computing node and was initiated by the local session;

c) causing by the local computing node a deallocation message to be sent to the remote computing node to release the remote session associated with the abnormally terminated local session, in which the action of causing a deallocation message to be sent is performed by a process at the local computing node.

31. The computer program product of claim 30 in which the deallocation message is piggybacked to another message sent to the remote computing node.

32. The computer program product of claim 30 in which one or more lists are maintained to track remote sessions that should be terminated.

33. The computer program product of claim 32 in which a monitoring process checks the one or more lists and instructs another process to perform step (c).

34. The computer program product of claim 32 wherein a monitoring process checks the one or more lists and performs step (c).

35. A system for deallocating a remote resource after termination of a local resource associated with the remote resource, the local resource residing on a local computing node, the remote resource residing on a remote computing node, in which the remote resource was initiated by the local resource, comprising:

a) means for monitoring, at the local computing node, for termination of the local resource residing on the local computing node;

b) a processor for determining if an abnormal condition is detected at the local computing node wherein termination of the local resource due to the abnormal condition does not result in expected deallocation of the remote resource, in which the remote resource resides on the remote computing node, and one of the local resource and the remote resource was initiated by the other;

c) means for identifying information about the abnormal condition in one or more lists;

d) means for checking the one or more lists to identify the remote resource that should be deallocated;

e) means for causing, by the local computing node, a deallocation message to be sent to the remote computing node to deallocate the remote resource; and

f) means for causing, by the local computing node, the remote computing node to deallocate the remote resource.

36. The system of claim 35 in which the deallocation message is piggybacked on to another message.

37. The system of claim 35 in which the means for checking the one or more lists and the means for causing a deallocation message to be sent are performed by different entities.

38. A system for coordinating the lifetime of a remote session at a remote computing node to the lifetime of a local session at a local computing node, in which the remote session is de-coupled from a specific connection between the remote computing node and the local computing node, the method comprising the steps of:

a) means for monitoring, at the local computing node, for termination of the local session residing on the local computing node, in which the local session was initiated by a process in a process pool;

b) a processor for determining if the termination of the local session was abnormal such that the remote session has not appropriately terminated, in which the remote session resides on the remote computing node and was initiated by the local session; and

c) means for causing by the local computing node a deallocation message to be sent to the remote computing node to release the remote session associated with the abnormally terminated local session, in which the action of causing a deallocation message to be sent is performed by a process at the local computing node.

39. The system of claim 38 in which the deallocation message is piggybacked on to another message.

40. The system of claim 38 in which the means for determining and the means for causing a deallocation message are performed by different entities.