Method and system for transparently failing over a computer name in a server cluster

• US 6,243,825 B1
• Filed: 04/17/1998
• Issued: 06/05/2001
• Est. Priority Date: 04/17/1998
• Status: Expired due to Term

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Vert, John D., Lucovsky, Mark, Gamache, Rod
• Primary Examiner(s): Lim, Krisna
• Application Number:
  US09/062,392
• Time in Patent Office: 1,145 Days
• Field of Search: 709/227, 709/228, 709/239, 709/235, 714/4
• US Class Current: 714/4.4
• CPC Class Codes: H04L 41/00 Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
• CPC Class Codes: H04L 67/01 Protocols
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/10015 Access to distributed or replicated servers, e.g. using brokers
• CPC Class Codes: H04L 67/1034 Reaction to server failures by a load balancer
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]
• CPC Class Codes: H04L 9/40 Network security protocols

Abstract

A method and system for transparently failing over a computer name with a legacy application running in a server cluster. When the application is set for failing over in the cluster, the cluster software locates a virtual computer name on which the application is dependent, and associates the virtual name with the application. When the application runs and requests its computer name from the system, the system looks for the virtual computer name, and, if detected, the system returns the virtual computer name to the application as the computer name. The virtual computer name thus moves with the application regardless of the machine on which it is running in the cluster. When the application is not set for failing over in the cluster, the system instead returns the actual machine name.

Claims

1. A method of returning a computer identity to an application independent of the machine on which the application is running, comprising:

providing a machine for running the application, the machine having a first computer identity;

associating a second computer identity with the application;

receiving a request from the application to return a computer identity to the application;

determining that the application is set for failing over in a cluster environment; and

returning the second computer identity to the application in response to the request.

2. The method of claim 1 wherein associating the second computer identity with the application includes searching a dependency tree for a virtual computer name.

3. The method of claim 1 wherein associating the second computer identity with the application includes writing a virtual computer name into a block of information associated with the application.

4. The method of claim 3 wherein determining if the application is set for failing over includes checking for the virtual computer name in the block of information.

5. The method of claim 1 wherein associating the second computer identity with the application includes searching a dependency tree for a virtual Internet protocol address.

6. The method of claim 1 further comprising creating a child process of the application, receiving a request to return a computer identity to the child process, and returning the second computer identity to the child process in response to the request.

7. The method of claim 1 wherein associating the second computer identity with the application includes providing a selection mechanism for defining a virtual server for the application.

8. A computer-readable medium having computer executable instructions for performing the method of claim 1.

9. A method of selectively returning a computer identity to an application, comprising:

(a) providing a machine for running the application, the machine having a first computer identity;

(b) determining if the application is to be failed over in a server cluster, and if the application is to be failed over, associating with the application a second computer identity that is independent of the machine;

(c) running the application;

(d) receiving a request from the application to return a computer identity to the application, and in response, determining if the second computer identity is associated with the running application; and

(i) if the second computer identity is associated with the running application, returning the second computer identity to the application in response to the request, and

(ii) if the second computer identity is not associated with the running application, returning the first computer identity to the application.

10. The method of claim 9 wherein determining if the application is to be failed over in a server cluster comprises running a custom dynamic link library.

11. The method of claim 9 wherein associating the second computer identity with the application includes searching a dependency tree for a virtual computer name.

12. The method of claim 9 wherein associating the second computer identity with the application includes writing a virtual computer name into a block of information associated with the application.

13. The method of claim 12 wherein determining if the second computer identity is associated with the running application includes checking for the virtual computer name in the block of information.

14. The method of claim 9 wherein associating the second computer identity with the application includes searching a dependency tree for a virtual Internet protocol address.

15. The method of claim 9 further comprising creating a child process of the application, receiving a request to return a computer identity to the child process, and if the second computer identity is associated with the running application, returning the second computer identity to the child process in response to the request.

16. The method of claim 9 wherein associating with the application the second computer identity includes providing a selection mechanism for defining a virtual server for the application.

17. The method of claim 9 wherein determining if the application is to be failed over in a server cluster comprises checking a property of the application.

18. A computer-readable medium having computer executable instructions for performing the method of claim 9.

19. In a server cluster, a method of returning a computer name to an application, comprising:

receiving a request to run the application on a machine in the cluster;

establishing a process environment block associated with the application;

searching a dependency tree for a virtual computer name related to the application;

writing the virtual computer name into the process environment block;

running the application;

receiving a request from the application to return a computer name thereto;

detecting the virtual computer name in the process environment block; and

returning the virtual computer name to the application as the computer name.

20. The method of claim 19 further comprising failing the application on the machine, and running the application on another machine in the cluster.

21. The method of claim 20 further comprising, establishing a new process environment block on the other machine in the cluster and associating the new process environment block with the application, searching the dependency tree for the virtual computer name related to the application, writing the virtual computer name into the new process environment block, receiving a request from the application to return a computer name thereto, detecting the virtual computer name in the new process environment block, and returning the virtual computer name to the application as the computer name.

22. The method of claim 19 further comprising creating a child process of the application, the child process inheriting the process environment block of the application, receiving a request to return a computer name to the child process, and returning the virtual computer name in the process environment block to the child process in response to the request.

23. The method of claim 19 further comprising receiving a virtual computer name, and storing the virtual computer name in the dependency tree.

24. The method of claim 19 further comprising determining if the application is to be failed over in a server cluster.

25. The method of claim 24 wherein determining if the application is to be failed over comprises checking a property of the application.

26. A computer-readable medium having computer executable instructions for performing the method of claim 19.

27. In a server cluster, a system for returning a computer name to an application, comprising:

a dependency tree maintained in the cluster for the application, the dependency tree including a virtual computer name in a dependent relationship with the application;

a block of information associated with the application;

a mechanism for searching the dependency tree to locate the virtual computer name and for writing the virtual computer name into the block of information; and

an interface for receiving a request from the application to return a computer name thereto, the interface returning the virtual computer name in the block of information to the application as the computer name.

28. The system of claim 27 further comprising a child process of the application, the child process inheriting the block of information associated with the application, and a mechanism for returning the virtual computer name to the child process when the child process requests the return of a computer name thereto.

29. The system of claim 27 further comprising a selection mechanism for defining a virtual computer name for the application.

30. The system of claim 27 wherein the interface further detects the virtual computer name in the block of information before returning the virtual computer name to the application.

31. The system of claim 27 wherein the interface is an application programming interface.

32. The system of claim 27 wherein the application has a property set therefor, and wherein the mechanism is responsive to the property to search the dependency tree and to write the virtual computer name into the block of information.

33. In a server cluster, a method of failing over a computer name with an application, comprising:

(a) relating the application to a virtual computer name;

(b) on a first machine of the cluster;

(i) associating the application with the virtual computer name;

(ii) running the application;

(iii) detecting a request for a computer name from the application;

(iv) returning the virtual computer name in response to the request; and

(v) failing the application;

and

(c) on a second machine of the cluster;

(i) associating the application with the virtual computer name;

(ii) running the application;

(iii) detecting a request for a computer name from the application, and

(iv) returning the virtual computer name in response to the request.

34. The method of claim 33 wherein relating the application to a virtual computer name comprises storing the virtual computer name in a dependency tree.

35. The method of claim 33 wherein each step of associating the application with the virtual computer name comprises writing the virtual computer name into a process environment block associated with the application.

36. The method of claim 33 wherein each step of returning the virtual computer name in response to the request comprises reading the virtual computer name from the process environment block associated with the application.

37. A computer-readable medium having computer executable instructions for performing the method of claim 33.

38. A method of selectively returning a computer identity to a running application, comprising:

(a) running the application on a machine having a first computer identity;

(b) receiving a request from the application to return a computer identity to the application, and in response;

(i) determining if the application is set for failing over in a server cluster, and

(1) if the application is set for failing over, returning a second computer identity that is independent of the machine in response to the request; and

(2) if the application is not set for failing over, returning the first computer identity in response to the request.

39. The method of claim 38 wherein the machine is node of a server cluster, wherein the application is set for failing over in the server cluster and has the second computer identity returned thereto, and further comprising, failing over the application to another node of the server cluster, including providing the second computer identity to the other node.

40. The method of claim 38 wherein determining if the application is set for failing over in a server cluster includes evaluting information in a process environement block associated with the application.

41. The method of claim 38 wherein the second computer name is maintained in a dependency tree.

42. A computer-readable medium having computer executable instructions for performing the method of claim 38.