System and method for failover

• US 7,293,192 B2
• Filed: 01/18/2007
• Issued: 11/06/2007
• Est. Priority Date: 11/24/2003
• Status: Active Grant

• Current Assignee: TSX Inc. (TMX Group Limited)
• Original Assignee: TSX Inc. (TMX Group Limited)
• Inventors: Morosan, Tudor, Allen, Gregory Arthur
• Primary Examiner(s): Bonzo; Bryce P
• Application Number:
  US11/624,467
• Publication Number:
  US 20070180302A1
• Time in Patent Office: 292 Days
• Field of Search: 714/6
• US Class Current: 714/6.3
• CPC Class Codes: G06F 11/1482 by means of middleware or OS functionality
• CPC Class Codes: G06F 11/1687 at event level, e.g. by interrupt or result of polling
• CPC Class Codes: G06F 11/2007 using redundant communication media
• CPC Class Codes: G06F 11/2028 eliminating a faulty processor or activating a spare
• CPC Class Codes: G06F 11/2041 with more than one idle spare processing component
• CPC Class Codes: G06F 11/2046 where the redundant components share persistent storage G06F11/2043 takes precedence
• CPC Class Codes: G06F 11/2071 using a plurality of controllers
• CPC Class Codes: G06F 11/2097 maintaining the standby controller/processing unit updated initialisation or re-synchronisation thereof G06F11/1658 and subgroups

Abstract

The present invention provides a novel system and method for failover. In an embodiment, a primary server and a backup server are available to a plurality of clients. Messages containing requests are processed by the primary server, while a mirror image of transaction records generated by the processing of those requests is maintained in volatile memory of both the primary server and the backup server. In this mode of operation, the volatile memory is periodically flushed to non-volatile memory.

Claims

1. A method for switching a client from a primary server to backup server comprising:

establishing a first connection from a client to a primary server;

sending a message from said client to said primary server via said first connection for processing by said primary server;

prior to receiving an expected response to said message at said client, experiencing a failure of said primary server at said client whereby said first connection fails;

requesting initiation of a second connection from said client to a backup server;

confirming, at said backup server, that said primary server has failed;

sending a recovery signal from said backup server to said client representing an acknowledgement of said request to initiate said second connection;

receiving said recovery signal at said client;

sending said message to said backup server via said second connection for processing by said backup server; and

,

if said backup server contains results of processing said message as previously processed by said primary server prior to failure of said primary server, then sending said results from said backup server to said client;

or

if said backup server does not contain results of processing said message by said primary server then processing said message at said backup server and sending said results thereof from said backup server to said client.

2. The method of claim 1 wherein said primary server is configured to mirror results of processing said message to said backup server.

3. The method of claim 1 wherein said backup server writes said results to a transaction log maintained in volatile storage and flushes said transaction log to non-volatile storage after processing each said message.

4. The method of claim 1 further comprising:

resolving said failure of said primary server such that said primary server becomes available;

determining at said backup server that said primary server is available at which point each of said servers writes said results to a transaction log and;

flushing said transaction logs to non-volatile storage on a periodic basis that is less frequent than the processing of each said message.

5. The method of claim 4 further comprising reversing designated roles of said backup server and said primary server such that said backup server is designated as the primary server and said primary server is designated as the backup server.

6. The method of claim 4 further comprising:

confirming, at said primary server and said backup server, that said primary server and said backup server are available;

announcing, at said backup server, to said client that said backup server is no longer available;

requesting re-initiation of said first connection from said client to a primary server;

sending a second recovery signal from said primary server to said client representing an acknowledgement of said request to re-initiate said first connection;

receiving said second recovery signal at said client; and

,

sending another message to said primary server via said first connection for processing by said primary server.

7. The method of claim 1 wherein said primary server and said backup server maintain a main link and a failsafe link therebetween and said backup server confirms said primary server has failed if a keep-alive signal is not present on either said main link or said failsafe link.

8. The method of claim 7 wherein said client does not experience said failure unless said primary server fails to communicate via said first connection, even if said backup server confirms said primary server has failed via said main link and said failsafe link.

9. A method for switching a client from a primary server to backup server comprising:

maintaining a first address at said client identifying said primary server;

maintaining a second address at said client identifying said backup server;

establishing a first connection from said client to a primary server;

sending a message from said client to said primary server via said first connection for processing by said primary server;

prior to receiving an expected response to said message at said client, experiencing a failure of said primary server at said client whereby said first connection fails;

requesting initiation of a second connection from said client to said backup server using said second address;

receiving a recovery signal from said backup server at said client; and

,

receiving results from processing of said message from said backup server;

wherein said results are received from storage of said backup server as previously stored by said backup server if said results were previously processed by said primary server prior to failure of said primary server; and

otherwise said results are received from storage of said backup server as previously stored by said backup server if said results were processed by said backup server after failure of said primary server, such that said client is unaware as to which of said servers processed said message.

10. A client machine configured to switch from communicating via a first connection with a primary server to a second connection with a backup server said client machine comprising an arrangement of at least one processor connected to storage devices and a network interface for communicating via said first connection and said second connection;

said processor configured to maintain a first address at said client identifying said primary server and to maintain a second address at said client identifying said backup server;

said processor further configured to establish a first connection from said client to a primary server and to send a message from said client to said primary server via said first connection for processing by said primary server;

said processor further configured to, prior to receiving an expected response to said message at said client, experiencing a failure of said primary server at said client;

said processor further configure to request initiation of a second connection from said client to said backup server using said second address;

said processor further configured to receive a recovery signal from said backup server and further configured to receive results from processing of said message from said backup server;

wherein said results are received from storage of said backup server as previously stored by said backup server if said results were previously processed by said primary server prior to failure of said primary server; and

otherwise said results are received at said processor from storage of said backup server as previously stored by said backup server if said results were processed by said backup server after failure of said primary server, such that said processor is unaware as to which of said servers processed said message.

11. The client machine of claim 10 wherein said primary server is configured to mirror results of processing said message to said backup server.

12. The client machine of claim 10 wherein said backup server writes said results to a transaction log maintained in volatile storage and flushes said transaction log to non-volatile storage after processing each said message.

13. The client machine of claim 10 wherein said primary server and said backup server maintain a main link and a failsafe link therebetween and said backup server confirms said primary server has failed if a keep-alive signal is not present on either said main link or said failsafe link.

14. The client machine of claim 13 wherein said processor does not experience said failure unless said primary server fails to communicate via said first connection, even if said backup server confirms said primary server has failed via said main link and said failsafe link.

15. A backup server connected to a primary server via a main link and connectable to a client machine via a second connection if a first connection between said primary server and said client machine fails;

said backup server comprising an arrangement of at least one processor connected to storage devices and a network interface for communicating via said main connection and said second connection;

said processor configured to receive and store mirrored results of messages sent from said client to said primary server that that are processed by primary server;

said processor configured to determine if said primary server has failed and to respond to a request from said client machine to process said message if said client machine determines that said primary server has failed to respond to a request to process said message;

said processor configured to return results stored in said backup server if said message was processed by said primary server and mirrored to said backup server prior to failure of said primary server;

said processor further configured to generate results of said messages that are not previously stored if said message was not processed by said primary server and mirrored to said backup server prior to failure of said primary server.