Distributed transaction processing control

• US 20050228834A1
• Filed: 06/13/2005
• Published: 10/13/2005
• Est. Priority Date: 12/18/2002
• Status: Active Grant

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Shinkai, Yoshitake
• Granted Patent: US 7,587,397 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 11/1474 in transactions G06F16/20 takes precedence
• CPC Class Codes: G06F 9/466 Transaction processing
• CPC Class Codes: Y10S 707/99938 Concurrency, e.g. lock management in shared database

Abstract

A distributed transaction processing system includes a master server (coordinator) and slave servers (participants). The master server and the slave servers create log file that indicates progress of a transaction. The log file is stored in a shared disk accessible from all the servers. When a fault occurs in the master server, the master server can continue a local transaction processing, which has been interrupted by the fault, after recovery from the fault by referring to the log file. When the fault occurs in any one of the slave servers, the master server can perform fault recovery of a transaction in which the faulty server is involved.

Claims

1. A server in a distributed transaction processing system in which a master server performs a distributed transaction processing in cooperation with a plurality of slave servers that performs a local transaction processing, the server having both a function of the master server and a function of a slave server, comprising:

a log-data recording unit that records progress of any one of the distributed transaction processing and the local transaction processing as log data in a shared disk that is accessible from both the master server and the slave servers; and

a fault recovery unit that performs a fault recovery based on the log data.

2. The server according to claim 1, wherein the log-data recording unit includes

a first log-data recording unit that records progress of the distributed transaction processing when the server is functioning as the master server;

and

a second log-data recording unit that records progress of the local transaction processing when the server is functioning as the slave server.

3. The server according to claim 1, wherein the fault recovery unit resumes the local transaction processing after recovery from a fault that has occurred in the server functioning as the slave server.

4. The server according to claim 2, wherein the progress of the distributed transaction processing is represented by a commitment request log, and the progress of the local transaction processing is represented by a prepared log and a commitment receive log, wherein

upon successfully performing the local transaction processing, the server functioning as the slave server records the prepared log,

upon receiving success replies from all of the servers functioning as the slave servers, the server functioning as the master server records the commitment request log before sending a commitment request to the all of the servers functioning as the slave servers, and

upon receiving the commitment request, the server functioning as the slave server records the commitment receive log.

5. The server according to claim 4, wherein

the commitment request log includes a transaction number for uniquely identifying a distributed transaction processing and a slave server number for uniquely identifying the slave server, and

the prepared log and the commitment receive log respectively include the transaction number, the master server number, and a result of the local transaction processing.

6. The server according to claim 4, wherein

when the server is functioning as the slave server and a fault occurs in the master server, the fault recovery unit commits the local transaction processing if the commitment request log has been recorded by the master server and aborts the local transaction processing otherwise, and

when the server is functioning as the master server and a fault occurs in the slave server, the fault recovery unit aborts the distributed transaction processing.

7. The server according to claim 4, wherein when the server is functioning as the slave server, the fault recovery unit commits the local transaction processing, of which the prepared log has been recorded by the server, if the commitment request log has been recorded by the master server, and aborts the local transaction processing otherwise.

8. A computer-readable recording medium that stores a computer program for a server in a distributed transaction processing system in which a master server performs a distributed transaction processing in cooperation with a plurality of slave servers that performs a local transaction processing, wherein the computer program causes the server having both a function of the master server and a function of a slave server to execute:

recording progress of any one of the distributed transaction processing and the local transaction processing as log data in a shared disk that is accessible from both the master server and the slave servers; and

performing a fault recovery based on the log data.

9. The computer-readable recording medium according to claim 8, wherein the recording includes

recording progress of the distributed transaction processing when the server is functioning as the master server;

and

recording progress of the local transaction processing when the server is functioning as the slave server.

10. The computer-readable recording medium according to claim 8, wherein the performing includes resuming the local transaction processing after recovery from a fault that has occurred in the server functioning as the slave server.

11. The computer-readable recording medium according to claim 9, wherein the progress of the distributed transaction processing is represented by a commitment request log, and the progress of the local transaction processing is represented by a prepared log and a commitment receive log, wherein

upon successfully performing the local transaction processing, the server functioning as the slave server records the prepared log,

upon receiving success replies from all of the servers functioning as the slave servers, the server functioning as the master server records the commitment request log before sending a commitment request to the all of the servers functioning as the slave servers, and

upon receiving the commitment request, the server functioning as the slave server records the commitment receive log.

12. The computer-readable recording medium according to claim 11, wherein

the commitment request log includes a transaction number for uniquely identifying a distributed transaction processing and a slave server number for uniquely identifying the slave server, and

the prepared log and the commitment receive log respectively include the transaction number, the master server number, and a result of the local transaction processing.

13. The computer-readable recording medium according to claim 11, wherein

when the server is functioning as the slave server and a fault occurs in the master server, the performing includes committing the local transaction processing if the commitment request log has been recorded by the master server and aborting the local transaction processing otherwise, and

when the server is functioning as the master server and a fault occurs in the slave server, the performing includes aborting the distributed transaction processing.

14. The computer-readable recording medium according to claim 11, wherein when the server is functioning as the slave server, the performing includes committing the local transaction processing, of which the prepared log has been recorded by the server, if the commitment request log has been recorded by the master server, and aborting the local transaction processing otherwise.

15. A processing method for a server in a distributed transaction processing system in which a master server performs a distributed transaction processing in cooperation with a plurality of slave servers that performs a local transaction processing, the server having both a function of the master server and a function of a slave server, comprising:

recording progress of any one of the distributed transaction processing and the local transaction processing as log data in a shared disk that is accessible from both the master server and the slave servers; and

performing a fault recovery based on the log data.

16. The processing method according to claim 15, wherein the recording includes

recording progress of the distributed transaction processing when the server is functioning as the master server;

and

recording progress of the local transaction processing when the server is functioning as the slave server.

17. The processing method according to claim 15, wherein the performing includes resuming the local transaction processing after recovery from a fault that has occurred in the server functioning as the slave server.

18. The processing method according to claim 16, wherein the progress of the distributed transaction processing is represented by a commitment request log, and the progress of the local transaction processing is represented by a prepared log and a commitment receive log, wherein

upon successfully performing the local transaction processing, the server functioning as the slave server records the prepared log,

upon receiving success replies from all of the servers functioning as the slave servers, the server functioning as the master server records the commitment request log before sending a commitment request to the all of the servers functioning as the slave servers, and

upon receiving the commitment request, the server functioning as the slave server records the commitment receive log.

19. The processing method according to claim 18, wherein

the commitment request log includes a transaction number for uniquely identifying a distributed transaction processing and a slave server number for uniquely identifying the slave server, and

the prepared log and the commitment receive log respectively include the transaction number, the master server number, and a result of the local transaction processing.

20. A distributed transaction processing system comprising:

a plurality of slave servers that perform a local transaction processing;

a master server that performs a distributed transaction processing in cooperation with the local transaction processing; and

a shared disk that is accessible from both the master server and the slave servers, wherein

the master server and the slave servers respectively includes

a log-data recording unit that records progress of any one of the distributed transaction processing and the local transaction processing as log data in the shared disk; and

a fault recovery unit that performs a fault recovery based on the log data.