Loading

Data 360: Guide to Deploying Different Data Streams Using Change Sets and Key Considerations

Publiceringsdatum: Mar 16, 2026
Beskrivning

Data Streams serve as the foundation for ingesting and unifying data from multiple systems within Data 360. When moving configurations between environments (for example, from sandbox to production), Change Sets provide a secure and structured way to deploy Data Streams and their associated metadata. This article covers how to package and deploy Data Streams from various connectors using Change Sets, along with key considerations to ensure consistency and successful activation in the target org.

Lösning
General Deployment Steps for Any Connector:

1. Add the Data Stream to a Data Kit
- This automatically includes the associated mappings, DMOs, and connector metadata in the package.
2. Create an Outbound Change Set
- Add the Data Kit and all related dependencies.
3. Deploy to the Target Org
- In the target org, go to the Inbound Change Set and deploy it.
4. Validate Deployment
5. Once deployed, confirm that the following components appear correctly in the target org UI:
- Data Stream
- Connector (specific to the data source)
- Mappings
- Any custom DMOs included
6. Activate the Connector (If Required)
- Some connectors deploy in an Inactive state and must be manually activated before use.
- Ensure that the connector configuration matches the original org setup (for example, database or source credentials).

Key Considerations with the Names of the Entities:
Ensure that the names of the DLO and the Schema (Data 360 Entities) remain consistent across orgs.
- If you have manually changed the name of the entities (e.g., DLO), it is suggested to create a new data kit in the source org and redeploy the entities with system-generated names to ensure a smooth transaction.
- Refer Change of API Names of Data Cloud Entities with DevOps Data Kit Deployment for more information.

Connector-Specific Considerations
While the deployment flow remains consistent across connectors, each connector type may have unique post-deployment actions or limitations. Below are some current examples:

Considerations when Deploying File Upload Data Streams:

1. Connector is auto-created: When a File Upload Data Stream is added, the corresponding connector is created automatically as part of the configuration.
2. Post-deployment activation required: After deploying the Data Kit to the target org, make sure to activate the connector first, and then activate the respective data streams.
 
Considerations when Deploying CRM Data Streams:

1. Deploy core metadata first
- Ensure that the standard and custom objects, along with any updated fields, are deployed to the target org before deploying the data streams.
2. Verify permissions post-deployment
- After deployment, confirm that the required object- and field-level permissions are granted in the "Data Cloud Salesforce Connector" permission set in the target org.
3. Deploy data streams after metadata
- Once metadata and permissions are in place, proceed with deploying the related data streams.
4. CRM connections for external-source orgs
- If a data stream uses a source org other than the home org, the relevant CRM connection is included in the Data Kit when you add the Data stream. After deployment to the target org, the connection will appear in an inactive state and will need to be re-authenticated.
5. Maintain consistent CRM connector alias names
- Keep CRM connector alias names consistent across orgs to avoid metadata conflicts and alignment issues.

Considerations when Deploying Ingestion API Data Streams:

1. Ensure that the API names of the DLO and the Schema remain consistent across orgs.
2. It is not recommended to manually create the Ingestion API connection in the target org.
3. Avoid deployment of multiple data kits containing the same data stream.
 
Considerations when Deploying External Connector Data Stream (e.g., Snowflake, Databricks):

1. Always include the Connector metadata in the Change Set.
2. The Connector cannot be manually modified or recreated in the target org.
- Prior to activating, the Connector can be manually modified in the target org; however, it is advised not to manually modify or recreate the deployed connection in the target org as it may fail the deployment.
3. Once the connection is deployed to the target org, its connection details should not be modified to a different Snowflake account.
4. If the connection is updated in the target org and later redeployed from the source org, the deployment will overwrite the connection details with the values from the source org. This is expected behavior.
5. The Connector must reference the same database as in the source org. Changing the database is currently not supported.
6. Once deployed, manually activate the Connector in the target org.
 
Best Practices:

1. Keep API names and schema definitions identical across orgs to avoid mapping issues.
2. Validate data source credentials after deployment.
3. Avoid overlapping deployments of the same Data Stream.
4. If errors occur during activation, review the Data Kit logs and connector configuration before re-deploying.
Ytterligare resurser

Packaging in Data Cloud

Knowledge-artikelnummer

005228027

 
Laddar
Salesforce Help | Article