Bug Description

When changing only the Options of a Trigger node (e.g. adding forceReconnect to an Email Trigger (IMAP) node) and publishing the workflow, the running trigger process is not restarted. The new options are persisted to the database, but the active trigger continues running with the old configuration.

A trigger restart only occurs when a structural change is made (e.g. moving a node/note on the canvas).

Additional context

• The workflow export confirms "forceReconnect": 55 is correctly stored in the database after the first Publish
• The updatedAt timestamp on the workflow entity updates correctly
• Container logs show zero log entries during the Publish event, indicating n8n did not deactivate/reactivate the trigger
• After the second Publish (with structural change), the trigger restarts correctly and forceReconnect works as expected
• N8N_LOG_LEVEL=debug was used to verify the presence/absence of reconnect log messages

This is likely related to the v2.0 migration from Active/Inactive toggle to Publish/Unpublish — the publish logic may only detect structural changes (node additions/removals, connection changes) as reasons to restart triggers, missing options-only changes.

To Reproduce

1. Create and publish a workflow with an Email Trigger (IMAP) node (typeVersion 2), no forceReconnect option set
2. Confirm the workflow is active and the IMAP trigger is connected
3. Open the IMAP trigger node → Options → add Force Reconnect Every Minutes = 55
4. Save and Publish the workflow
5. Set N8N_LOG_LEVEL=debug and restart the container (required because the reconnect message is logged at debug level)
6. Wait 55+ minutes and check container logs: no Forcing reconnect to IMAP server message appears → the trigger was not restarted with the new options
7. Now make a structural change (e.g. move a Sticky Note) and Publish again
8. Wait 55+ minutes and check container logs: Forcing reconnect to IMAP server appears → the trigger was restarted
9. Remove N8N_LOG_LEVEL=debug and restart the container to restore default logging

Expected behavior

Publishing a workflow should always restart active trigger nodes, regardless of whether the change is structural (node position, connections) or configuration-only (node options/parameters).

Debug Info

Debug info

core

• n8nVersion: 2.18.1
• platform: docker (self-hosted)
• nodeJsVersion: 24.14.1
• nodeEnv: production
• database: sqlite
• executionMode: regular
• concurrency: -1
• license: enterprise (production)
• consumerId: fb4613cc-eb98-48aa-bfc4-d1649f71047d

storage

• success: all
• error: all
• progress: false
• manual: true
• binaryMode: filesystem

pruning

• enabled: true
• maxAge: 336 hours
• maxCount: 10000 executions

Generated at: 2026-05-07T11:24:10.390Z

Operating System

Ubuntu 24.04 LTS

n8n Version

2.18.1

Node.js Version

24.14.1

Database

SQLite (default)

Execution mode

main (default)

Hosting

self hosted