Pipelines

Pipelines in UDMG define a file transfer path between two Endpoints: a source and a destination.

Pipelines provide the following features:

• Flexibility and Reusability: Through Configuration Structure, Pipelines let you reuse standardized Endpoints across multiple transfer scenarios by applying Pipeline-specific configuration, so you do not need to create new Endpoints from scratch every time.
• Protocol Versatility: Supporting both server and client scenarios means UDMG can adapt to your partners' existing infrastructure rather than forcing them to change their setup. Whether you need to receive files, push them out, or pull them from remote systems, you can create a Pipeline Type that fits.
• Scalable Organization: Through the Account Authentication model, the hierarchical structure (Pipelines → Account Groups → Accounts) makes it easy to manage permissions and access at scale. You can set up Pipelines once and apply them to multiple Accounts, or create partner-specific variations as needed.
• Workflow Integration: Through Workflows, Pipelines support automated processing steps during transfer execution, allowing Tasks to run at specific transfer lifecycle events.
• Enforced Data Integrity: Through Pipeline Validation, UDMG automatically validates Pipeline configurations to prevent file path conflicts that could cause data loss or overwrites.

Configuration Structure

Each Pipeline is defined by a source, a destination, and Pipeline-specific configuration that determines how the transfer behaves between them.

A Pipeline consists of the following components:

• Source Endpoint: the Endpoint from which files are retrieved or received. It is selected from the list of available Endpoints already created in UDMG.
• Destination Endpoint: the Endpoint to which files are delivered. The available options depend on the selected Source Endpoint and the supported Pipeline Types.
• Pipeline Configuration: the required and optional settings that apply to the Pipeline itself, such as file and folder permissions or a virtual path. This configuration allows the same Endpoint to be reused across multiple Pipeline scenarios without redefining the base Endpoint each time.

Pipeline Types

UDMG supports multiple Pipeline types designed to accommodate different file transfer scenarios. These types are grouped according to the role UDMG plays in the transfer.

UDMG as Server

When UDMG acts as a Server, an External Partner connects to a Local Server Endpoint using a supported client or a web browser, depending on the protocol.

Depending on the Pipeline type and permission, partners may be able to upload files or upload, download, and manage folders. In all UDMG as Server Pipelines, the destination is a Local Filesystem, so all files handled through the local server Endpoint are written on the UDMG Server.

info

For more details and the full list of supported Pipelines, see UDMG as Server.

UDMG as Client

When UDMG acts as a Client, it initiates file transfers to or from remote endpoints using supported protocols.

Client-initiated transfers are executed through the Transfers Scheduled API, which allows you to programmatically trigger file transfers.

Depending on the Pipeline type, UDMG may upload files to a remote system (push scenario) or download files from a remote system (pull scenario). In both cases, the Local Filesystem is always involved, either as the source of files being sent or the destination for files being received.

info

For more details and the full list of supported Pipelines, see UDMG as Client.

UDMG as Gateway

When UDMG acts as a Gateway, it facilitates file transfers between external partners and a Remote Cloud Storage Endpoint through a local server Endpoint.

In this model, UDMG serves as an intermediary between partner systems and cloud storage, allowing partners to interact with familiar protocols without requiring direct access to the cloud provider.

Depending on permissions, partners can upload, download, and manage files and folders in the remote cloud storage through the server Endpoint

info

For more details and the full list of supported Pipelines, see UDMG as Gateway.

Workflows

Each Pipeline includes a Workflow that defines automated processing steps (Tasks) executed during a transfer.

The Workflow is organized around transfer lifecycle events (triggers). Tasks can be assigned to these triggers to perform additional operations when the corresponding event occurs.

At a high level, configuring automated processing with a Workflow involves two steps:

1. Create and configure a Task that defines the action to perform.
2. Assign that Task to one or more Task Triggers in a Pipeline's Workflow so the Task runs when the corresponding event occurs.

This allows Tasks to be defined once and then reused across multiple Pipelines and Task triggers as needed.

info

For Tasks of type Publish Event, additional steps are required. For more information, see Universal Event Integration.

Assigning a Task to a Pipeline

To assign a Task to a Pipeline, follow these steps:

1. From the sidebar, select Configuration > Pipelines.
2. Click the Name of the Pipeline where you want to assign a Task.
3. Click the Workflow tab.
4. Click Edit.
5. The Workflow editor displays five columns:
   • Available Tasks
   • On Staged Tasks
   • On Sent Tasks
   • On Received Tasks
   • On Error Tasks
6. Drag and drop Task cards as needed:
   • To assign a Task to the Pipeline Workflow, move a card from Available Tasks to one of the trigger columns.
   • To unassign a Task from the Pipeline, click the X icon on the Task card.
   • Click Cancel to discard all changes.
7. Click Update.

info

Not all Task Triggers are available for every Pipeline type. For more information, see Trigger Availability & Behavior by Pipeline Type.

Pipeline Validation

UDMG validates Pipeline configurations to prevent path conflicts and preserve configuration integrity. If a conflict is identified, then the Configuration Item is blocked from being created or associated, or a detailed error message is displayed.

Validation Trigger	Validation Check
When a Pipeline is created or updated	The combination of Source Endpoint, Destination Endpoint, Endpoint-specific Configuration, and Virtual Path must be unique.
When a Pipeline is assigned to an Account Group	The combination of Source Endpoint and Virtual Path must be unique across all Pipelines assigned to the same Account Group.
When an Account is associated with an Account Group	The combination of Source Endpoint and Virtual Path must be unique across all Pipelines associated with an Account through its Account Groups.

tip

The most common validation conflicts occur when multiple Pipelines expose the same Source Endpoint and Virtual Path combination to the same Account Group or Account.