Failed records for Reverse ETL and audience syncs

RudderStack has made failed records for Reverse ETL and audience syncs generally available across all its workspaces, regardless of the plan chosen. Previously, a failed Reverse ETL sync would only indicate the number of records that failed, accompanied by a sample error based on the status code. However, the new feature now provides a comprehensive breakdown of failures for each individual record.

Upon opening a completed sync, users can access the "Failed records" panel, which displays a detailed analysis of failures categorized by error codes. This information can be further expanded by turning on the "Retain and show failures" option for the connection. The panel will then present a list of every failed record along with the precise error message recorded by RudderStack. Users can also open any specific record to view its source row, which is read directly from their warehouse.

With this enhanced visibility into failed records, the complete set of data is seamlessly integrated into the warehouse as a queryable and joinable table. This empowers users to analyze and manipulate the data as needed, just like any other table within the warehouse.

For audience syncs, the process is similar. Users can open the sync and enable error details in the "Failed records" panel. The same setting applies across the board. It is important to note that capturing error details is a per-connection opt-in and will apply from the next sync onwards. It may take up to a minute for the error details to be available in the data plane. Please note that existing runs will not retroactively fill in the failure details.