--- title: Avoiding duplicate workflows description: Update sets manage the published state of all versions of a workflow prior to committing the workflow version on a local instance.Follow the steps in this page to commit a workflow in an update set.It is not possible to have multiple published versions as a result of update set commits. However, this does not eliminate risk, and care should be taken when migrating update sets.Update Set B is migrated and committed to the production instance. locale: en-US canonical_url: https://www.servicenow.com/docs/r/australia/build-workflows/legacy-workflow/c\_AvoidingDuplicateWorkflows.html release: australia product: Legacy Workflow classification: legacy-workflow topic_type: concept last_updated: "2026-03-12" reading_time_minutes: 2 breadcrumb: [Input variable movement, Workflow administration, Classic Workflow, Build workflows] --- # Avoiding duplicate workflows Update sets manage the published state of all versions of a workflow prior to committing the workflow version on a local instance. The last version of a workflow committed as an Insert or Update using an update set becomes the currently published version, regardless of the publishing sequence for the workflow versions. **Parent Topic:**[Input variable movement](https://raw.githubusercontent.com/ServiceNow/ServiceNowDocs/australia/markdown/australia/build-workflows/legacy-workflow/c_InputVariableMovement.md) ## Commit a workflow in an update set Follow the steps in this page to commit a workflow in an update set. ### Procedure 1. Workflow A - Version 1 is created and published in Update Set A. 2. Update Set A is completed and migrated to a local instance. 3. When the update set is committed, the system sets all prior versions of Workflow A to published = false. In the first migration, there are no prior versions. 4. Workflow A - Version 1 becomes the only published version of the workflow. ## Update set migration example It is not possible to have multiple published versions as a result of update set commits. However, this does not eliminate risk, and care should be taken when migrating update sets. Consider this example: 1. Workflow A - Version 1 is migrated and committed to the production instance. 2. Update Set B is created. 3. Update Set C is created. 4. Workflow A - Version 2 is published in Update Set B. A customer update record is added to Update Set B with the Version 2 payload. A customer update record is added to Update Set B with the Version 1 workflow left unpublished. 5. Update Set B is completed. 6. Workflow A - Version 3 is published in Update Set C. A customer update record is added to Update Set C with the Version 3 payload. A customer update record is added to Update Set C with the Version 2 workflow left unpublished. 7. Update Set C is completed. 8. Update Set C is migrated and committed to the production instance. Workflow A - Version 1 is set to unpublished. Workflow A - Version 2 update is skipped since Update Set B, which contains Version 2, was never migrated. Workflow A - Version 3 is committed and becomes the only published version of the workflow. ## Update set migration risk Update Set B is migrated and committed to the production instance. 1. Workflow A - Version 3 is set to unpublished. 2. Workflow A - Version 1 remains unpublished. 3. Workflow A - Version 2 is committed and becomes the only published version of the workflow. The workflow has gone back a version, perhaps unintentionally. The regressed version becomes the currently published version.