Staging Process Workflow (Master Data Services)
During each step of the staging process, fields in the staging and batch tables are updated. The following workflow explains how and when the most important fields in each table are updated.
Step 1: Populate the staging tables
When you initially populate the three staging tables (mdm.tblStgMember, mdm.tblStgMemberAttribute, and mdm.tblStgRelationship), the Batch_ID for each record should be NULL. The Status_ID field for each record defaults to 0, which indicates that the member is ready to be processed.
The batch table (mdm.tblStgBatch) does not yet contain a record for the batch.
After you have imported data into the staging tables, open the Master Data Manager user interface (UI) to confirm the number of records available to process. In the Integration Management functional area in the Unbatched Staging Records pane, choose a model. The number of records available to stage is displayed. This is the total number of staging records that contain:
The model in the ModelName field.
The logged-in user’s name or no user name in the UserName field.
Step 2: Initiate the staging process
To initiate the staging process in the UI, select the version to import data into and click Process staged data.
When the staging process starts, a row is added to the batch table. An ID is assigned in the ID field and the Status_ID field is updated to 1 to indicate that the batch is queued for processing.
Each row in the staging tables is updated with the batch ID from the batch table. The Status_ID remains 0 to indicate that processing has not started.
Staging batches that are queued to run are processed in sequence, and processing begins at an interval determined by a setting in Master Data Services Configuration Manager. The staging tables are processed one after another. mdm.tblStgMember is processed first, followed by mdm.tblStgMemberAttribute, and then mdm.tblStgRelationship. You cannot initiate the processing of one table at a time.
Step 3: Batch completes
As each row in the staging table is imported into the appropriate Master Data Services database tables, the Status_ID field for the row is updated with 1 or 2 to indicate success or failure, respectively. Each row’s ErrorCode field is also updated. These error codes are displayed along with descriptions on the Staging Batch Errors page in Master Data Manager. For more information, see Staging Error Messages (Master Data Services).
When all members in a batch have been processed, the batch table Status_ID field is updated to 2 and the LastRunEndDTM field displays the date and time that the process completed.
Records that were successfully loaded are now available in Explorer.
ERR210000 is not an error; it is displayed on successfully loaded records.
The following illustration shows a summary of the workflow.
Members remain in the staging tables until you clear them. For more information, see Clearing Batches Workflow (Master Data Services) or How to: Clear a Batch from the Staging Queue (Master Data Services).
- Business rules are not automatically applied to your newly-imported data. You must validate the model version against business rules. For more information, see How to: Validate a Version (Master Data Services).