---
category: [Administration & Integrations, Administration & Integrations/Custom Integrations & Apps, Platform and Product Extensions, Platform and Product Extensions/Workday Extend, Platform and Product Extensions/Integration]
keyword: [Extend, Orchestration Builder, Orchestrate, Orchestration]
nav: wcp_docs
parent_url: /wcp_docs/
title: >- 
  Concept: Orchestration Processes
layout: subsection
---

An orchestration doesn't always execute as a whole. Instead, Orchestrate may divide each orchestration into smaller execution units called processes, which follow each other seamlessly. Every orchestration has at least 1 process, created by Orchestrate at launch.

Each process has these characteristics:

-   Can execute for up to 60 minutes in an app that's been promoted to production or 45 minutes in an app not yet promoted to production.
-   Has its own unique ID to enable tracking.
-   Can automatically save its state when complete.
-   Passes control to the next process in the sequence.


You can think of processes as relay runners passing a baton. Each runner completes their segment, then hands off to the next runner to continue the race.

Orchestrate creates a new process when it encounters a step that triggers a potentially long-running operation. This is known as checkpointing. The components that create checkpoints are:

-   <b>Send Workday RaaS Request</b>.
-   <b>Trigger PDF Generation</b>.
-   <b>Trigger Business Process</b>.
-   <b>Trigger Integration</b>.
-   <b>Send HTTP Request</b> (when used with polling).


This is the sequence of events that occurs when Orchestrate reaches one of these checkpoints:

-   The current process saves its data and state.
-   The long-running operation executes.
-   When the long-operation completes, a new process begins.
-   The new process loads the saved data and state and executes.
-   The 60/45 minute timeout resets.


<b>Design Considerations</b>

Always bear in mind that the maximum time allowed for any single process to complete is <b>60 minutes</b> on a production tenant or <b>45 minutes</b> on a nonproduction tenant, while the entire orchestration is subject to its own separate timeout: For more information, see: [Reference: Orchestration Runtime Limits](/wcp_docs/GUID-412c9e62-1ba2-4975-a36b-b794d2b559be.html).

Avoid timeouts by using checkpointing steps strategically for operations that might take a long time. Remember that each checkpoint step preserves data while providing a new processing window.

<b>Monitoring and Troubleshooting</b>

Use the Orchestration Activity dashboard to view details about your process chain and identify issues, paying particularly attention to process execution time.

When monitoring your orchestrations, the key pieces of data are:

-   <b>Run ID</b>: The unique instance ID for an orchestration run.
-   <b>Process IDs</b>: The unique ID for an individual process. Use this to identify specific processes that are causing issues.
-   <b>Common Request ID</b>: An ID that that links all the processes in an orchestration.
-   <b>Process Status</b>: Shows which process is currently running or completed.


### Related Information

[Concept: Orchestration Activity](/wcp_docs/GUID-fe14647a-603a-43a1-a33a-c920d5c90934.html)

[Reference: Orchestration Summary Report Details](/wcp_docs/GUID-046dfceb-f16a-4ccc-9357-9fbc1f4ca67b.html)