Distributed session failover

• US 20070192326A1
• Filed: 02/01/2006
• Published: 08/16/2007
• Est. Priority Date: 02/01/2006
• Status: Active Grant

• Current Assignee: Oracle America, Inc. (Oracle Corporation)
• Original Assignee: Oracle America, Inc. (Oracle Corporation)
• Inventors: Sondhi, Ajay, Chu, Ching-Wen, Angal, Rajeev, Penumatsa, Subash
• Granted Patent: US 8,051,179 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 11/1482 by means of middleware or OS functionality
• CPC Class Codes: G06F 11/203 using migration
• CPC Class Codes: G06F 11/2035 without idle spare hardware

Abstract

A distributed session failover mechanism is disclosed for facilitating the replication and retrieval of session information. A first server, in a trusted network, providing a single sign-on (SSO) solution, stores session information pertaining to a particular client requesting services associated with the server. In order to provide session failover, the first server sends a copy of the session information to a bus mechanism, which is connected to one or more persistent repositories. When a second server attempts to validate the client, the second server may discover that the first server failed. The second server then requests a copy of the session information pertaining to the client from the bus mechanism. The bus mechanism retrieves the copy from a persistent repository and provides the copy to the second server.

Claims

1. A method performed by a first session server, comprising:

receiving a request from a client to access one or more services, wherein the request comprises information indicating a particular session that was previously established by a second session server;

determining that the second session server, which maintains a set of session information pertaining to the particular session, has failed;

in response to a determination that the second session server has failed, sending a request to a bus mechanism to obtain a copy of the set of session information, wherein the set of session information is stored in one or more persistent repositories, wherein the first session server is unaware of the one or more persistent repositories, and wherein the first session server does not interact with any of the one or more persistent repositories;

receiving a copy of the set of session information from the bus mechanism after the bus mechanism has retrieved the set of session information from at least one of the one or more persistent repositories; and

using the copy of the set of session information to determine whether the client should be allowed to access the one or more services.

2. The method of claim 1, wherein each of the one or more persistent repositories is a database.

3. The method of claim 1, wherein the bus mechanism comprises of at least two application programming interfaces and a set of one or more brokers.

4. The method of claim 1, wherein the bus mechanism comprises a plurality of bus mechanisms.

5. The method of claim 1, wherein the first server and the second server are logically located in different networks.

6. A machine implemented method for carrying out distributed session failover, comprising:

storing, by a first server, a particular set of session information pertaining to a particular session associated with a client;

providing, by the first server, the particular set of session information to a bus mechanism;

causing, by the bus mechanism, the particular set of session information to be stored in a set of one or more persistent repositories, wherein the first server is unaware of the set of one or more persistent repositories, and wherein the first server does not interact with any persistent repository in the set of one or more persistent repositories;

receiving, by a second server, a request from the client for a particular set of services associated with the second server;

determining, by the second server, that the first server, which has stored the particular set of session information, has failed;

sending, by the second server, a request to the bus mechanism to obtain a copy of the particular set of session information, wherein the second server is unaware of the set of one or more persistent repositories, and wherein the second server does not interact with any persistent repository in the set of one or more persistent repositories;

retrieving, by the bus mechanism, a copy of the particular set of session information from one of the set of one or more persistent repositories; and

providing, by the bus mechanism, the copy of the particular set of session information to the second server.

7. A machine readable medium, comprising:

instructions for causing one or more processors to receive a request from a client for a particular set of services associated with the first server, wherein a particular set of session information pertains to a particular session, which is associated with the client;

instructions for causing one or more processors to determine that a second server, which has stored the particular set of session information, has failed;

instructions for causing one or more processors to send a request to a bus mechanism to obtain a copy of the particular set of session information, wherein the copy of the particular set of session information is stored in a set of one or more persistent repositories, wherein the first server is unaware of the set of one or more persistent repositories, and wherein the first server does not interact with any persistent repository in the set of one or more persistent repositories; and

instructions for causing one or more processors to receive the copy of the particular set of session information from the bus mechanism after the bus mechanism has retrieved the copy of the particular set of session information from one of the set of one or more persistent repositories.

8. The machine readable medium of claim 7, wherein each of the one or more persistent repositories is a database.

9. The machine readable medium of claim 7, wherein the bus mechanism comprises of at least two application programming interfaces and a set of one or more brokers.

10. The machine readable medium of claim 7, wherein the bus mechanism comprises a plurality of bus mechanisms.

11. The machine readable medium of claim 7, wherein the first server and the second server are logically located in different networks.

12. A system, comprising:

a first session server;

a second session server; and

a bus mechanism, wherein;

the first session server receives a request from a client to access one or more services, wherein the request comprises information indicating a particular session that was previously established by the second session server;

the first session server determines that the second session server, which maintains a set of session information pertaining to the particular session, has failed;

in response to a determination that the second server has failed, the first session server sends a request to a bus mechanism to obtain a copy of the set of session information, wherein the set of session information is stored in one or more persistent repositories, wherein the first session server is unaware of the one or more persistent repositories, and wherein the first session server does not interact with any of the one or more persistent repositories;

the first session server receives a copy of the set of session information from the bus mechanism after the bus mechanism has retrieved the set of session information from at least one of the one or more persistent repositories; and

the first session server uses the copy of the set of session information to determine whether the client should be allowed to access the one or more services.