Previous
Error handling and retries
Last Updated
>
Retry failed steps, choose a fallback, and avoid duplicate runs.
Things fail: an API times out, a token expires, a service is down. Error handling decides what Shard does next.
Retries
Each action can retry automatically. Choose how many times to retry and how long to wait between attempts. Shard waits longer after each failure, so a struggling service isn't hit repeatedly.
When retries run out
Choose what happens if an action still fails:
Stop the run and mark it as failed.
Continue with the next step.
Run a fallback, like sending an alert to your team.
Avoiding duplicate runs
If a trigger fires twice for the same event, Shard can recognize it and run the automation only once.
Tip: add a fallback alert to any automation that matters. A failed run nobody sees is worse than one that pages you.
