High availability web browser access to servers

• US 6,134,588 A
• Filed: 11/12/1997
• Issued: 10/17/2000
• Est. Priority Date: 11/12/1997
• Status: Expired due to Term

• Current Assignee: Activision Publishing Incorporated (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: Guenthner, Timothy John, Lawlor, Francis Daniel, Schmitt, Charles Rudolph
• Primary Examiner(s): Maung, Zarni
• Assistant Examiner(s): MOEZZI, MANZAR
• Application Number:
  US08/968,037
• Time in Patent Office: 1,070 Days
• Field of Search: 709/226, 709/229, 709/228, 707/9, 395/600
• US Class Current: 709/226
• CPC Class Codes: G06F 16/95 Retrieval from the web
• CPC Class Codes: H04L 61/4511 using domain name system [DNS]
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1019 Random or heuristic server selection

Abstract

One or more policies are implemented at a Web browser to enhance access to Web servers that host content requested by the browser. When the browser issues a request, a name service returns a list of IP addresses that may service that request. The list is configured as "random" or "ordered" according to a given naming convention or other local policy, and IP addresses are selected from the list at random or in order (as the case may be) until a connection to an appropriate server is obtained. The browser remembers (for a given time period) which IP addresses have failed so that those addresses are not repeatedly tried. The browser'"'"'s "timeout" period is also selectively varied depending on the type of list returned from the name service.

Claims

1. A method of communication in a computer network comprising at least one client, a plurality of servers, and a nameserver, where in response to a request issued from the browser, a list of server addresses is returned from the nameserver, the method operative in the client and, comprising the steps of:

favoring a given server address over other server addresses in the list based on a given policy;

attempting to establish a connection from the client machine to a server identified by the given server address;

if, during a timeout period, the connection to the server identified by the given server address cannot be established, restricting use of the given server address for a given time period; and

attempting to establish a connection from the client machine to a second server identified by at least one of the other server addresses in the list.

2. The method as described in claim 1 wherein the given policy establishes the list as a random list and the given server address is an address selected from the list at random.

3. The method as described in claim 2 wherein the other server address is selected from the list at random.

4. The method as described in claim 2 wherein prior to selecting a given server address, the method includes the steps of:

determining whether a given first time period has elapsed since the random list was last retrieved from the nameserver; and

if the given first time period has elapsed, re-fetching the list from the nameserver.

5. The method as described in claim 1 wherein the given policy establishes the list as an ordered list and the given server address is a first address in the list.

6. The method as described in claim 5 wherein the other server address is a next address in the ordered list.

7. The method as described in claim 5 wherein prior to selecting a given server address, the method includes the steps of:

determining whether a given time period has elapsed since the ordered list was last retrieved from the nameserver; and

if the given time period has elapsed, re-fetching the list from the nameserver.

8. The method as described in claim 1 wherein the timeout period is variable.

9. The method as described in claim 8 further including the step of decreasing the timeout period if, during an attempt to establish a connection, other server addresses in the list remain unused.

10. A computer program product in a computer-readable medium for use in a Web client connectable in a computer network having a plurality of servers and a name service, comprising:

means responsive to selection of a hypertext reference for issuing a request to the name service and receiving in return a list of server addresses;

means responsive to the issuing means for favoring a given server address over other server addresses in the list based on a given policy; and

means for restricting use of a given server address on the list for a given time period.

11. The computer program product as described in claim 10 wherein the restricting means is responsive to a failure to establish a connection from the client machine to a server identified by the given server address during a given timeout period.

12. The computer program product as described in claim 11 wherein the restricting means further includes means for varying the given timeout period.

13. The computer program product as described in claim 10 wherein the given policy configures the list as a random list and the given server address is an address selected at random.

14. The computer program product as described in claim 10 wherein the given policy configures the list as an ordered list and the given server address is a first address on the list.

15. The computer program product as described in claim 10 further including:

means for determining whether a given time period has elapsed since the list was last retrieved from the name service; and

means responsive to the determining means for selectively re-fetching the list from the name service.

16. The computer program product as described in claim 10 wherein the computer program product is a browser.

17. The computer program product as described in claim 10 wherein the computer program product is a browser plug-in.

18. A computer for use as a client in a computer network having a plurality of Web servers and a name service, comprising:

a processor having an operating system;

a Web browser including means responsive to selection of a hypertext reference for issuing a request to the name service and receiving in return a list of server addresses; and

means associated with the Web browser for enhancing access to the plurality of Web servers, operative in the client and, comprising;

means responsive to the issuing means for favoring a given server address over other server addresses in the list based on a given policy; and

means for restricting use of a given server address on the list for a given time period.

19. The computer as described in claim 18 wherein the restricting means is responsive to a failure to establish a connection from the client machine to a server identified by the given server address during a given timeout period.

20. The computer as described in claim 19 wherein the restricting means further includes means for varying the given timeout period.

21. The computer as described in claim 18 wherein the given policy configures the list as a random list and the given server address is an address selected at random.

22. The computer as described in claim 18 wherein the given policy configures the list as an ordered list and the given server address is a first address on the list.

23. The computer program product as described in claim 18 further including:

means for determining whether a given time period has elapsed since the list was last retrieved from the name service; and

means responsive to the determining means for selectively re-fetching the list from the name service.

24. A method of enhancing Web browser access to a plurality of servers in a computer network, where in response to a request issued from the Web browser, a list of server addresses is returned from a name service, the method operative in the client and, comprising the steps of:

biasing a given server address over other server addresses in the list based on a given policy;

attempting to establish a connection from the client machine to a server identified by the given server address; and

if the connection cannot be established using the given server address, using other server addresses in the list to respond to the request.

25. The method as described in claim 24 wherein the given policy configures the list as a random list and the given server address is an IP address selected at random.

26. The method as described in claim 24 wherein the given policy configures the list as an ordered list and the given server address is a primary IP address.

27. The method as described in claim 24 further including the step of re-fetching the list from the name service prior to the biasing step if the list is older than a given age.

28. A method of enhancing Web browser access to a plurality of servers in a computer network, where in response to a request issued from the Web browser, a set of server addresses is returned from a name service, the method operative in the client and, comprising the steps of:

biasing a subset of the server addresses over other server addresses in the set, wherein each of the server addresses in the subset is a duplicate;

attempting to establish a connection from the client machine to a server identified by the server addresses in the subset; and

if the connection cannot be established, using other server addresses in the set to respond to the request.

29. A method of enhancing Web browser access to a plurality of servers in a computer network, where in response to a request issued from the Web browser, a list of server addresses is returned from a name service, the method operative in the client and, comprising the steps of:

biasing a given server address over other server addresses in the list based on a given policy;

attempting to establish a connection from the client machine to a server identified by the given server address; and

if the connection cannot be establish using the given server address, associating the given server address with a given status and using other server addresses in the list to respond to the request.

30. The method as described in claim 29 wherein the given status indicates that the given server address is a bad address.

31. The method as described in claim 29 further including the step of maintaining the given status for a predetermined time period during which the given server address is not used.

32. The method as described in claim 31 further including the step of altering the given status after the predetermined time period.

33. A method of communication in a computer network comprising at least one client, a plurality of servers, and a nameserver, where in response to a request issued from the browser, a list of server addresses is returned from the nameserver, the method operative in the client and, comprising the steps of:

favoring a given server address over other server addresses in the list based on a given policy;

attempting to establish a connection from the client machine to a server identified by the given server address during a timeout period selected as a function of a number of untried server addresses in the list.

34. The method as described in claim 33 further including the steps of:

during the timeout period, determining if the connection to the server identified by the given server address can be established; and

if not, attempting to establish a connection from the client machine to a second server identified by at least one of the other server addresses in the list.