|
|
|
|
General
|
|
Step Name
|
Type a name that describes the service in the flow.
|
|
Description
|
Type a description of the service in your flow (optional).
|
|
Step ID
|
A read-only field displaying the ID of the step in the flow.
|
|
Component ID
|
An internal identifier number generated automatically.
|
|
Component Type
|
A read-only field displaying the name of the step type.
|
|
Condition
|
Enter a condition to specify the flow's execution behavior. You can either click  to open the Expression Editor, or you can right-click on the component and select Condition from the context menu.
|
|
Processing mode
|
Select whether the step should be executed linear, parallel or stand-alone.
|
|
This is available only if the .NET utility is not the first step in a flow.
|
|
|
Wait for completion
|
Select whether the step should be run in Wait for Completion mode. The options are Yes or No.
|
|
This is available only when the .NET utility is a Linear step and if not the first step in the flow.
|
|
|
Advanced
|
|
Save point
|
Select Yes to indicate that the step is a Save Point in your flow.
|
|
Call logic flow
|
You can use this option if you want to call generic logic after the execution of a specific step. Once this step has been executed, the flow logic that you defined in the flow properties is called before the Server continues to the following steps.
The options are Yes or No (default).
|
|
Mapper
|
|
Error behavior
|
The step behavior if an error occurs at this step in a flow. Select one of the following from the drop-down list:
-
Continue (default): All mappings defined in the step are executed and the error policy defined is carried out at the end of the step. When the transaction is in Record level, a rollback has to be issued on that record and the Data Mapper continues to the next record. When no other error handling is defined, all errors caused during database operation, and not just the last error found, are written to the Monitor by the Data Mapper.
-
Exit: the step execution ends immediately and the error policy defined for the flow is executed. Any mapping defined after the point where the error occurs is not executed.
-
If you select Dynamic, the step execution ends immediately if the error flow ends in an error. If the error flow clears the error, the flow will continue. This is relevant for database Destinations only.
For a list of Data Mapper errors, click here.
|
|
DB Transaction Level
|
Select one of the following from the drop-down list:
-
Mapper (default): This executes the database transaction for all Database-type Data Mapper Destinations. All Data Mapper database modifications are part of the same transaction. The rollback procedure rolls back all database modifications performed by the Data Mapper. All records that were added or updated are rolled back. If the error behavior defined above is Continue, the error will be handled according to the defined settings and the Data Mapper will continue its operation with no rollback. If the error behavior defined above is Exit, and the error was not handled, all updated step records in all Destination databases will be rolled back.
-
Record: This runs each record as a separate transaction for all Database-type Data Mapper Destinations. If a failure occurs in a specific record, the error behavior that you defined in the Error behavior field (above) is triggered for that record only. Only this record is rolled back in all its Destinations. If the error behavior defined above is Continue (default), the error will be handled according to the defined settings and the Data Mapper will continue its operation with the next record. If the error behavior defined above is Exit, the step execution will be terminated.
-
Auto Commit: This instructs the database to commit each SQL statement automatically, provided there are no prior open transactions on the same context (Magic xpi does not support nested transactions). In cases where a transaction is already open, the Data Mapper joins this transaction and ignores the DB Transaction Level value. If you want to connect to a remote database and perform a record level transaction, selecting Auto Commit will significantly enhance performance. This means that there will be a marked decrease in the time taken to complete the transaction.
|
|
-
A database transaction defined at the step level is ignored if there is a transaction defined at the flow level. However, in DB2/400 databases, if Auto Commit is selected in an Insert operation or the WITH NC keyword is added to the SQL statement in Delete and Update operations (regardless of transaction type), the step level is not ignored.
-
Call flows from the Data Mapper will be part of the same transaction. All databases that take part in the process including databases used only in the called flow, have to be defined in advance. For Rollback and Commit, you should expect unpredictable behavior if not all databases are defined.
-
If a transaction cannot be opened, an error message will be displayed.
-
The DB Transaction Level cannot be defined as Record when a DB/400 Source contains BLOB data type fields. You should use a Mapper level transaction instead.
|
|
|
DB Transaction
|
You can define which databases this transaction is performed on. Any database resource that was not defined as part of the transaction will perform the auto commit procedure, regardless of the selected transaction level in the step.
To do this, click to open the Database Resources List. Select a database from the list, and then click OK.
|
