Webhooks / Retries
Retries
If your endpoint is down or slow, Swum keeps the event and tries again. You don't lose events to a short outage.
Schedule
A try fails when your endpoint doesn't answer with a 2xx within 10 seconds. Swum then tries again on this schedule, up to eight tries. Each delay begins after the previous failed try; the final delay is 24 hours, not a 24-hour total retry window.
| Try | After |
|---|---|
| 1 | Immediately |
| 2 | 1 minute after try 1 |
| 3 | 5 minutes after try 2 |
| 4 | 30 minutes after try 3 |
| 5 | 2 hours after try 4 |
| 6 | 5 hours after try 5 |
| 7 | 10 hours after try 6 |
| 8 | 24 hours after try 7 |
Every try carries the same event id and a fresh signature time. Store the ids you have handled and skip repeats.
When an endpoint keeps failing
The endpoint shows as Failing in Developers & Integrations. After 3 days of failures, Swum turns it off and emails the school owner. Turn it back on once it's fixed; new events flow again, and you can resend missed ones from the Delivery Log.
Delivery Log
Each endpoint keeps a 30-day log of what was sent: the event, the status code your endpoint gave, how long it took, and the start of its answer. Retry sends any event again, with the same id.
Send a test event
Send Test Event posts a sample event to your endpoint, signed with your real secret, so you can check your signature check end to end.
updated_since (see Pagination).