This tutorial covers the steps for monitoring dataflows for streaming sources using the Sources workspace.
This tutorial requires a working understanding of the following components of Adobe Experience Platform:
In the Platform UI, select Sources from the left navigation bar to access the Sources workspace. The Catalog screen displays a variety of sources for which you can create an account with.
To view existing dataflows for streaming sources, select Dataflows from the top header.
The Dataflows page contains a list of all existing dataflows in your organization, including information about their source data, account name, and dataflow run status.
Select the name of the dataflow you want to view.
The following table contains more information on dataflow run statuses:
The Dataflow Activity page displays specific information on your streaming dataflow. The top banner contains the cumulative number of records ingested and records failed for all of your streaming dataflow runs in your selected date range.
By default, the data displayed contains ingestion rates from the last seven days. Select Last 7 days to adjust the time frame of records displayed.
A calendar pop-up window appears, providing you options for alternative ingestion time frames. You can configure the dataflow run time frame to view flow runs from the previous seven days or the last 30 days. Alternatively, you configure the interactive calendar to set a custom time frame of your choice. When finished, select Apply.
The lower half of the page displays information on the number of records received, ingested, and failed, per flow run. Each flow run is recorded within an hourly window.
Each individual dataflow run shows the following details:
successwith a warning. Note: Support for ingesting records with warnings is only available to streaming sources.
The Dataflow run overview page contains additional information on your dataflow, such as its corresponding dataflow run ID, target dataset, and organization ID.
A flow run with errors also contains the Dataflow run errors panel, which displays the particular error that led to the failure of the run, as well as the total count of records that failed.
Records with warnings displays a list of mapper transformation warnings that occurred during your flow run. Rows that are partially ingested are considered successful and are appended with warnings if any mapper transformation errors are found.
By default, all mapper transformation errors are considered as warnings, except if they are any of the following:
To view error diagnostics, select Preview error diagnostics.
The Error diagnostics preview window allows you to preview up to 100 errors and/or warnings regarding your dataflow run. From here, you can also download the ingestion failure manifest for more information, using the Data Access API.
By following this tutorial, you have successfully used the Sources workspace to monitor your streaming dataflows and identify the errors that led to any failed dataflows. See the following documents for more information: