Make webhook queue MCP tools: stuck items and Sume redelivery

When a webhook queue item looks stuck, check Make's queue and Sume's delivery status separately. Sume can re-POST a job's terminal event with Redeliver.

4 min readSume
All posts

If a webhook queue item looks stuck in a Make scenario that waits on Sume, inspect two places: Make's queue, and the delivery status on the Sume job. If Sume's delivery shows exhausted or failed, call POST /v1/jobs/{job_id}/webhook/redeliver and Sume re-sends the real terminal event.

Make's release notes dated September 25, 2026 mention "MCP tools for webhook queues" for queue management; the snapshot read 2026-10-01 is a one-line entry, so this post does not describe those tools. The Sume side is from Webhooks.

Which side holds the item?

Sume sends the callback; Make receives it. An item missing from Make's queue may never have arrived, and an item present in Make may be waiting on your scenario. Sume's job object and job events show the delivery side, including the webhook.delivery event.

Where to look, from the Sume docs read 2026-10-01.
QuestionWhere to checkWhat you see
Did Sume try to deliver?Job object or job eventswebhook.delivery event and delivery status
Is the job finished?status_urlJob status completed, failed or canceled
Did Make accept the call?Make's own queueUse Make's tools; not covered here

What delivery statuses does Sume report?

The documented vocabulary is pending, delivering, delivered, retrying, failed and exhausted. Automatic delivery makes up to 10 attempts. Ten refused attempts leave a failed delivery and a job that still reached its real terminal state, so the work is not lost.

How do I resend the event?

Use Redeliver, either on the delivery row or with POST /v1/jobs/{job_id}/webhook/redeliver (scope jobs:write). Sume re-POSTs the job's real terminal event with a fresh timestamp and signature. It still works after the automatic attempts are exhausted and does not consume one of the 10. It does not change the destination URL; a new URL is a new job.

Do not use Send test for this. It posts a dummy webhook.test payload with no job id, so it proves the URL is reachable but replays nothing.

What if the callback never arrives?

Keep status_url polling available. The docs call delivery an optimization, never the only recovery path. Treat job_id as the idempotency key in your scenario so a redelivered event does not run the downstream steps twice. For the retry-limit side, see Make webhook queue full 400 and Sume's ten attempts.

Sources

Related posts

More in Integrations

All Integrations posts

Written by Sume