Monitoring Webhook Deliveries and Request Logs

Overview

The API console gives you two audit surfaces for everything your integrations do: Deliveries records every webhook attempt StretchDev made, and Request logs records every API call made against StretchDev. Together they answer the two questions you'll always ask when an integration misbehaves: did my webhook go out? and what did my API key just do?

Read the delivery log

1. Open the API console and go to Deliveries.
2. Each row is one delivery attempt and shows:
- Webhook — which subscription it came from, and its target URL.
- Event type — for example repo.push or a dev.ping test.
- Status — the delivery outcome (for example delivered or queued).
- Response status — the HTTP code your endpoint returned.
- Attempts — how many times StretchDev tried.
- Duration — how long the attempt took, in milliseconds.
- Delivered at — when it completed.
3. Filter by a specific webhook to isolate one integration's history.

Use the delivery log to confirm a webhook fired, see exactly what your endpoint returned, and spot patterns (for example, a run of non-2xx responses after an endpoint change).

Read the request log

1. Go to Request logs.
2. Each row is one API request and shows the method and path, the status code, the duration in milliseconds, the caller's IP and user agent, the scopes the request used, and the API key that made it (by name and prefix).
3. Open a single log entry to see its full request and response detail.
4. Filter by status to focus — for example, show only 4xx/5xx requests to find failures.

Request logs are your audit trail: which key did what, when, from where, and whether it succeeded.

Example

A teammate reports the nightly report script "isn't working." You open Request logs, filter to status 401, and see a cluster of 401 responses from the key named Nightly report starting last night. The key's prefix matches one you rotated yesterday — the script is still using the old secret. You hand over the new secret, and the next run logs 200s. Separately, you check Deliveries and confirm the CI webhook is still returning 200 in ~80 ms, so that integration is healthy.

Tips

• When an integration breaks, check Request logs first (was the call even made, and what status came back?) then Deliveries for webhook-driven flows.
• Filter request logs by status code to jump straight to failures.
• Match the key prefix in a log to the key in your API-keys list to attribute traffic without exposing secrets.
• Watch duration for creeping latency on your own endpoints (deliveries) and on API calls.

Troubleshooting

• A delivery shows a non-2xx response status. Your receiver rejected or errored; inspect the entry's detail and your endpoint logs, fix the endpoint, and re-send a Test delivery.
• Repeated 401/403 in request logs. The API key is invalid, paused, revoked, expired, or missing a required scope. Check the key's status and scopes, or rotate it.
• A delivery is stuck as queued. The target may be unreachable; confirm the URL and that the webhook is active, then test again.
• "Request log not found." The log ID isn't in your organization; confirm the workspace.