Skip to main content

Delete a Data Flow

Use the flows API to delete data flows and their associated resources. You can delete a flow from any point (data source, data set, or data sink) and choose whether to delete only downstream resources or the entire flow.

Delete Flow Endpoints​

The flows API provides multiple endpoints for deleting flows, allowing you to delete from any resource in the flow:

Delete Flow: Request
# Delete from flow node
DELETE /flows/{flow-node-id}

# Delete from data source
DELETE /data_sources/{data-source-id}/flow

# Delete from data set
DELETE /data_sets/{data-set-id}/flow

# Delete from data sink
DELETE /data_sinks/{data-sink-id}/flow

Delete Options​

When deleting a flow, you have two main approaches to choose from. The default behavior is safer and more targeted, while the full deletion option removes the entire flow structure.

Delete Downstream Only (Default)​

By default, deleting a flow removes only the specified resource and all downstream resources:

DELETE /data_sets/{data-set-id}/flow

This will delete:

  • The specified data set
  • All downstream data sets
  • All downstream data sinks
  • Associated transforms and data maps

Delete Entire Flow​

To delete the entire flow including upstream resources, include the ?all=1 or ?full_tree=1 query parameter:

DELETE /data_sets/{data-set-id}/flow?all=1
DELETE /flows/{flow-node-id}?full_tree=1

This will delete:

  • The entire flow from the origin data source
  • All data sets in the flow
  • All data sinks in the flow
  • Associated transforms and data maps

Prerequisites​

Before deleting a flow, ensure that:

  1. All Resources are Paused: Any ACTIVE resources in the data flow will cause the request to fail
  2. Proper Permissions: You have the necessary permissions to delete the resources
  3. No Dependencies: Other flows or resources don't depend on the flow being deleted

Delete Response​

The delete operation provides clear feedback about whether the deletion was successful or if there are issues that need to be resolved first.

A successful request to delete a data flow returns 200 OK with no response body.

Failed Deletion (Active Resources)​

If any resources in the flow are still active, the request will fail with a 405 Method Not Allowed error:

Delete Flow: Error Response
{
  "data_sources": [5023],
  "data_sets": [5059, 5061, 5062],
  "data_sinks": [5029, 5030],
  "message": "Active flow resources must be paused before flow deletion!"
}

Delete Flow Examples​

The following examples demonstrate different deletion scenarios and how to use the various endpoints effectively.

Example 1: Delete from Data Source​

Delete all flows originating from a specific data source:

DELETE /data_sources/5023/flow

Example 2: Delete from Data Set​

Delete a data set and all downstream resources:

DELETE /data_sets/5061/flow

Example 3: Delete Entire Flow​

Delete the entire flow from a downstream resource:

DELETE /data_sets/5061/flow?all=1

Example 4: Delete from Flow Node​

Delete a flow starting from a specific flow node:

DELETE /flows/10001

Cascading Deletion​

When you delete a flow, the following resources are automatically deleted:

  • Data Sets: All downstream data sets in the flow
  • Data Sinks: All data sinks associated with deleted data sets
  • Transforms: All transforms applied to deleted data sets
  • Data Maps: All data maps associated with deleted data sets
  • Flow Nodes: All flow nodes in the deleted portion of the flow

Recovery Considerations​

Important: Deleted flows cannot be recovered through the API. Consider the following before deletion:

  1. Export the Flow: Use the export functionality to create a backup
  2. Verify Dependencies: Ensure no other resources depend on the flow
  3. Documentation: Keep records of the flow configuration for potential recreation

Best Practices​

  1. Always Pause First: Pause flows before attempting deletion
  2. Use Selective Deletion: Start with downstream-only deletion unless you need to remove the entire flow
  3. Export Before Deletion: Create backups of important flows
  4. Test in Non-Production: Test deletion workflows in development environments first
  5. Monitor Dependencies: Ensure deletion won't break other workflows

Legacy Endpoints​

Note: The legacy /data_flows endpoints are still available for backward compatibility but are deprecated:

  • DELETE /data_flows/data_source/{data_source_id}
  • DELETE /data_flows/{data_set_id}
  • DELETE /data_flows/data_sink/{data_sink_id}

It's recommended to use the new /flows endpoints for all new development.