| 摘要 |
Even if business processing is transaction, a standby-type cluster system copes with double faults in a short time while holding down the standby computer cost to the minimum. An execution server monitoring thread in a standby server periodically monitors an operation state of an execution server. Upon detecting time-out as a result of the monitoring, the thread resets the execution server, stops load balancing to an IP address of the execution server, recovers transaction of the execution server, and returns to monitoring of the operation state. If normality of the execution server is detected as the result of monitoring and an operation state in an execution server management table is "stop," the thread starts the load balancing to the execution server and returns to the monitoring of the operation state.
|
