Hitachi

Hitachi Advanced Database Application Development Guide


4.5 Handling unresponsive application programs

A problem such as a communication error, temporary failure (including power outage), or disk failure might cause an application program to stop responding. Such an unresponsive application program might delay other application programs and command processing.

Specify the following operand in the client definition in order to minimize the effects of an unresponsive application program:

Specify in this operand the maximum amount of time to wait for a response after a processing request has been sent from the HADB client to the HADB server. If there is no response from the HADB server within the specified time, a timeout error whose SQLCODE is -732 (KFAA30732-E) is returned to the application, processing of the SQL statement is canceled, and the transaction is rolled back. Then, the application is disconnected from the HADB server.

The following figure shows the wait time monitoring procedure based on the specification of the adb_clt_rpc_sql_wait_time operand.

Figure 4‒3: Wait time monitoring procedure based on the specification of the adb_clt_rpc_sql_wait_time operand

[Figure]

Explanation:

This example monitors the wait time until a response is received after a processing request has been issued from the HADB client to the HADB server. For example, if 600 seconds is specified in the adb_clt_rpc_sql_wait_time operand, a wait time of 600 seconds is set for the monitoring interval. Therefore, as a guideline, specify a wait time at least equal to the longest SQL processing time. Specify a realistic value that most likely will indicate when the application program has become unresponsive.

For details about the adb_clt_rpc_sql_wait_time operand, see the explanation of the adb_clt_rpc_sql_wait_time operand in 2.2.2 Operands related to application program status monitoring.