Skip to main content

Fix a failing webhook

When your system stops getting Jaah's events, the endpoint's delivery log says what your receiver answered. This guide finds the failing endpoint, reads why, checks the fix with a test, and sends the missed events again. What a delivery is and how it moves from Queued to Delivered or Failed is in How webhooks work.

You need to be an Owner or Admin of your organization, on a plan that includes webhooks.

  1. Open Settings, select Integrations, then Webhooks.
  2. Read the Last delivery column: the latest attempt's result and how long ago it was, such as Failed · 2h. Hover it for your receiver's status code. A hollow mark before the name means the endpoint is Disabled and sends nothing. Select the failing endpoint's name.
  3. Select the Deliveries tab. Each row is one delivery, newest first: its Event, Status, your receiver's last Code, its Latency and how many Tries it took.
  4. Hover a Failed or Retrying status to read why the last attempt failed, and when the next one is due. The Code points at the cause:
    • —: Jaah got no answer in time, or could not reach the URL. Check that the URL is public and your receiver answers within the time allowed.
    • 401 or 403: your receiver refused the request, often because its signature check fails. Check it holds the current secret, as in Verify the signature.
    • 3xx: Jaah does not follow redirects. Enter the final URL on Settings and select Save.
    • 5xx: your receiver failed while handling the event. Check its own logs.
  5. Fix your receiver, or the endpoint's URL on Settings. If the endpoint shows Disabled, tick Enabled there and select Save.
  6. Select Send test. Jaah sends a ping event at once and opens Deliveries, where its row shows Delivered once your receiver answers with success. A test is tried once, never retried.
  7. On each delivery that failed while your receiver was down, select Redeliver. Jaah sends it again at once, with a fresh round of retries, and the row shows the new result. Select Load more to reach older rows; the log keeps 30 days.

Events that happened while the endpoint was Disabled were never sent, so there is nothing to redeliver for them.