System and method for failure detection and recovery

• US 5,771,343 A
• Filed: 02/14/1996
• Issued: 06/23/1998
• Est. Priority Date: 02/14/1996
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: Sterling Commerce, Inc. (International Business Machines Corporation)
• Inventors: Mellott, Elizabeth P., Batka, Charles J., Hafner, Edward A.
• Primary Examiner(s): Beausoliel, Jr., Robert W.
• Assistant Examiner(s): LE, DIEU MINH T
• Application Number:
  US08/600,013
• Time in Patent Office: 860 Days
• Field of Search: 395/182.02, 395/182.14, 395/182.21, 364/229.4, 364/230.4, 364/262.2, 364/264, 364/265, 364/266.5, 364/281.9, 364/285
• US Class Current: 714/4.4
• CPC Class Codes: G06F 11/0757 by exceeding a time limit, i.e. time-out, e.g. watchdogs
• CPC Class Codes: G06F 11/1438 Restarting or rejuvenating

Abstract

According to one embodiment of the present invention, a transaction processing system is provided comprising at least one server and a mainframe. Recovery process for restarting a batch is also provided comprising failure detection process for indicating a batch failure, and a startup means responsive to the failure detection process for starting failed batch on the mainframe. In a particular embodiment of the present invention, a batch failure may be indicated by a server'"'"'s failure to respond to a request or by a step within a batch requiring longer than a predetermined length of time to complete. Further, in alternative embodiments of the present invention, recovery process may run on the mainframe, the server processing the batch, or a secondary server.

Claims

1. A transaction processing system for processing transactions and providing failure detection and recovery comprising:

a mainframe computer;

at least one remote server, said at least one remote server in communication with said mainframe computer;

said at least one remote server processing transactions sent by the mainframe computer;

recovery means for detecting a failure occurring on said at least one remote server and for restarting said transactions affected by said failure;

wherein said recovery means comprises a failure detection means for detecting failures occurring upon said at least one server and a startup means for restarting said transactions affected by said failure;

wherein said failures comprise batch failures and remote server failures; and

wherein said failure detection means indicates a failure when the processing of said batch requires longer than a predetermined amount of time.

2. The transaction processing system of claim 1 wherein said transactions affected by said failure are restarted on said mainframe computer.

3. The transaction processing system of claim 1 wherein said transactions affected by said failure occurring on a first remote server are restarted on a second remote server.

4. The transaction processing system of claim 1 wherein said failure detection means indicates a failure when a batch processing process fails to respond.

5. The system of claim 1 wherein said failure detection means detects failures using a status table.

6. The system of claim 5 wherein said status table indicates the completion of batch segments within interchanges.

7. The system of claim 1 further comprising:

a communications manager;

an initiator;

at least one inbound process; and

at least one outbound process.

8. The system of claim 1 further comprising middleware allowing communication between said mainframe computer and one or more of said remote servers.

9. The system of claim 8 further comprising a network for communication between said mainframe and said at least one remote server.

10. The system of claim 9 wherein said network comprises an Ethernet network.

11. The system of claim 1 comprising a single remote server.

12. The system of claim 1 further comprising a customer system in communication with said mainframe computer.

13. A transaction processing system for processing individual transactions and batches of transactions and providing failure detection and recovery comprising:

a main frame computer;

at least one remote server, said at least one server in communication with said mainframe computer;

said at least one remote server processing transaction batches sent by the mainframe computer;

manager means running said at least one remote server for managing processing of said transaction batches;

failure detection means for detecting a failure of batch processing when the processing of said batch requires longer than predetermined amount of time on said at least one remote server, said failure detection means notifying said manager means in the event of a failure; and

said manager means further comprising startup means responsive to said failure detection means for starting sad failed batch on said mainframe computer.

14. The system of claim 13 wherein said manager means is further operable to shut down a remote server in the event that a failure occurs on said remote server.

15. The system of claim 13 wherein said failure detection means comprises a process running on said remote at least one server.

16. The system of claim 13 further comprising a duplicate manager means running on said mainframe computer.

17. The system of claim 13 further comprising at least one inbound process running on each of said remote servers.

18. The system of claim 17 further comprising a duplicate inbound process running on said mainframe computer for each inbound process running on said remote servers.

19. The system of claim 13 further comprising at least one outbound process running on each of said remote servers.

20. The system of claim 19 further comprising a duplicate outbound process running on said mainframe computer for each outbound process running on said remote servers.

21. The system of claim 13 further comprising means for detecting failures within said failure detection means.

22. The system of claim 21 wherein said means for detecting failures within said failure detection means queries a status table indicative of the status of each remote server in order to detect failures within said failure detection means.

23. A method for detecting and restarting a failed batch in a transaction processing system, said transaction processing system comprising a mainframe computer and at least one remote server in communication with said mainframe computer, said method comprising the steps of:

initiating a recovery process on said at least one remote server;

sending a batch process from the mainframe computer to the remote server;

starting the batch process on said at least one remote server;

determining a failure of said batch process the processing of said batch require longer than a predetermined amount of time; and

in the event said batch process has failed, restarting said batch process on said mainframe computer.