n8n - 💡(How to fix) Fix Schedule Trigger stopped firing — workflow Published & active, manual executions succeed, no automated fires for 2+ hours

Hi n8n team,

We have a production workflow whose Schedule Trigger has stopped invoking despite being correctly configured, Published, and active. Manual executions of the same workflow succeed cleanly end-to-end, so the workflow logic itself is healthy. The issue is purely on the scheduler-registration side.

Tenant / instance: sfcai1.app.n8n.cloud Workflow ID: KNktCTVaB1tXiDh5 Workflow name: "Equitas — Deal Stage Progression Watcher" Latest versionId (at time of writing): 30861dd0-af09-4943-b8da-5efd18595499 Trigger node: "Schedule Trigger" (n8n-nodes-base.scheduleTrigger) Configured interval: every 5 minutes (rule.interval = [{field:"minutes", minutesInterval:5}])

Timeline (all timestamps UTC, 2026-05-28):

• 14:30 → 17:55 Workflow was firing on cadence every 5 minutes as expected. Last successful auto-fire (mode=trigger) was execution id 80886 at 17:55:58.043Z.
• 16:35 One execution (id 80787) failed inside a downstream Code node. That was the only error in the window.
• After 17:55:58, no further trigger-mode executions have occurred. Now is 20:15Z+ — over 2 hours and 19 minutes with zero auto-fires, despite scheduler being "active".

Diagnostic actions we performed:

1. PUT /api/v1/workflows/KNktCTVaB1tXiDh5 with two corrective node updates (active=true preserved). PUT returned 200 OK.
2. Confirmed workflow shows active=true via GET /api/v1/workflows/KNktCTVaB1tXiDh5.
3. UI Unpublish → Publish cycle (twice). Both times the dialog "Workflow published, your schedule trigger will now trigger executions on the schedule you have defined" appeared. No fires followed.
4. Deleted the existing Schedule Trigger node via the UI, added a fresh Schedule Trigger node, wired it to 5 downstream Search nodes (via API for precision), set rule to every-5-minutes.
5. After the fresh trigger node was installed, we did another full UI Unpublish + Publish cycle. The first Publish attempt was rejected ("Workflow could not be published. Your most recent changes may be lost, because someone else just updated this workflow"); after a hard browser reload we successfully Published.
6. Manual executions via the editor's "Execute workflow" button succeed every time and the workflow runs end-to-end cleanly (e.g., execution ids 80982, 80989, 80995, 80997, 81016 — all status=success, mode=manual).
7. No further mode=trigger executions have been recorded since 17:55:58Z.

What we believe is happening: The scheduler cluster's registration for this specific workflow appears to be stuck in a state where it does not pick up the Publish state changes. UI/API-side everything reports correctly registered and active; backend scheduler is not invoking the trigger. This appears to be a per-workflow registration wedge rather than a tenant-wide issue (we have other workflows on this tenant that are firing normally, e.g. "Equitas — Broker Intake Webhook" handles incoming webhooks reliably — though that's webhook-mode, not schedule-mode).

What we'd like:

1. Please force a scheduler re-registration for workflow KNktCTVaB1tXiDh5 on the backend.
2. If a backend restart of our tenant's scheduler instance is needed, please let us know a maintenance window so we can plan around it.
3. If there's anything we should look for in our workflow definition that would prevent registration, please flag it. The trigger node config is the canonical { rule: { interval: [{ field: "minutes", minutesInterval: 5 }] } }.

Thank you. Happy to provide any further diagnostic data — execution ids, full workflow JSON, recent versionId history, etc.

Best, Kyri Stylianou [your email tied to the n8n Cloud account]