Skip to main content
InfraAudit delivers webhook events with at-least-once semantics and retries failed deliveries using exponential backoff. Understanding the retry behavior helps you build a reliable receiver.

​
Delivery guarantees

  • At-least-once: InfraAudit may deliver the same event more than once — for example, if your endpoint acknowledges after a network timeout on InfraAudit’s side. Use the delivery_id field in each payload to deduplicate.
  • No ordering guarantee: Events may arrive out of order if retries from one delivery overlap with a later successful delivery.

​
What counts as success

A delivery succeeds if your endpoint returns any 2xx HTTP status code within 10 seconds. A delivery fails if:
  • The endpoint returns a 4xx or 5xx status
  • No response is received within 10 seconds
  • The connection is refused or the DNS lookup fails
InfraAudit only checks the status code — the response body is ignored.

​
Retry schedule

After all 3 retries are exhausted (roughly 2.5 hours from the initial attempt), the delivery is marked failed and no further retries occur.

​
Viewing delivery history

From the web UI, go to Settings → Webhooks → [webhook name] → Deliveries. Each delivery record shows:
  • Event type and delivery ID
  • Attempt number and HTTP status returned
  • Response time
  • Request and response headers for debugging
Via the API:

​
Handling failures

If deliveries are consistently failing:
1

Check the delivery history

Look at the HTTP status code. A 401 means your endpoint is rejecting the signature. A 5xx means your application is erroring.
2

Fix your receiver

Resolve the issue within the 2.5-hour retry window to avoid losing events.
3

Recover missed events from the API

For events that exhausted all retries, query the corresponding InfraAudit API endpoint directly — for example, GET /api/v1/drifts/{id} for a missed drift.detected event.
Manual retry of failed deliveries is not yet available in the UI or API. Use the InfraAudit API to fetch missed data directly.

​
Disabling a webhook during maintenance

To suppress deliveries while your endpoint is down, disable the webhook:
Re-enable it after maintenance. Events that fired while the webhook was disabled are not replayed — query the API for any data you need from that window.