Salesforce Data 360 targets

• •Due to a third-party limitation, Salesforce Data 360 Connector cannot perform read and write operations on the same Data Lake Object (DLO). To work around this, you must manually create a recovery table in the target and specify the recovery table name in the writerRecoveryTableNameAlias custom property before you run the task.
Perform the following tasks to create and configure the recovery table:
1. 1Create a schema file named INFORMATICA_CDC_RECOVERY.yml with the following content:
openapi: 3.0.3
components:
schemas:
INFORMATICA_CDC_RECOVERY:
type: object
properties:
TASK_ID:
type: string
TYPE:
type: string
SCHEMA_NAME:
type: string
TABLE_NAME:
type: string
CYCLE_NUMBER:
type: number
SEQUENCE:
type: string
LAST_CYCLE_UPDATE_TIME:
type: string
2. 2Log in to Salesforce and open the Salesforce Data Cloud app.
3. 3Navigate to DataCloud Setup and select the Ingestion API associated with your task.
4. 4Upload the INFORMATICA_CDC_RECOVERY.yml schema to create the recovery table.
5. 5To create a data stream for the recovery table, perform the following tasks:
• ▪ Open the Salesforce Data Cloud app launcher and go to the Data Stream window.
• ▪ Create a new data stream.
• ▪ Select Ingestion API as the data stream source, and proceed.
• ▪ Select the INFORMATICA_CDC_RECOVERY table for the Ingestion API.
6. 6To configure the primary key for the data stream, perform the following tasks:
• ▪ In the New Data Stream window, create a new formula field with the following transformation formula:
CONCAT(sourceField['TASK_ID'], "_", sourceField['TYPE'], "_",
sourceField['SCHEMA_NAME'], "_", sourceField['TABLE_NAME'])
• ▪ Set the formula return type as text.
• ▪ Select this newly created formula field as the primary key.
7. 7Select the data space configured for your task and deploy the data stream.
8. 8After deployment, open the data stream to copy the Object API Name of the recovery table and keep it handy.
9. 9When you configure a database ingestion and replication task, enter the copied object API name in the writerRecoveryTableNameAlias custom property on the Destination page before you run the task.
• •Salesforce Data 360 reserves certain column names such as cdp_sys_SourceVersion, DataSource, DataSourceObject, and KQ_Id for internal use. If your source object contains any of these column names, rename them before you deploy the task to avoid failure.
• •In incremental load and combined load jobs, delete operations are ignored and skipped.
• •For a task that loads data to a Salesforce Data 360 target, the source object must have one primary key. If the source object has no primary key or more than one primary key, the deployment fails.
• •Audit Mode and Soft Deletes apply modes are not available for Salesforce Data 360 targets. These apply modes require creating objects without a primary key. Salesforce Data 360 does not permit this, because all data lake objects require a primary key.