Auto-Retries
Automatically retry failed step executions.
Last updated
Automatically retry failed step executions.
Last updated
When auto-retry is enabled on a step, the step automatically retries if the step execution fails. This is useful for ensuring your workflows run successfully when encountering intermittent errors, such as API rate limits.
Auto-retry is available on the following step types:
Integrations (e.g. Salesforce, Slack, Jira)
If a step with auto-retry enabled fails, Paragon will perform several retries, waiting after each failed execution before performing the next attempt. All attempts will only count as one task used.
The timing of retry attempts is as follows:
1st retry: around 10 seconds after initial attempt
2nd retry: around 20 seconds after 1st retry
3rd retry: around 30 seconds after 2nd retry
4th retry: around 50 seconds after 3rd retry
5th retry: around 80 seconds after 4th retry
To enable auto-retry for a step, click on the step, scroll to the bottom of the sidebar, and toggle auto-retry.
Note: Enabling auto-retry only retries the specific step, not your entire workflow.
Deployed Workflows: Paragon attempts to run your step up to five times with an exponential back-off. Retried steps appear as "Running" in Task History.
Test Workflow or Test Step: Paragon does not retry step executions.
While auto-retry is enabled, no error emails for the workflow will be sent during the auto-retry window. If Paragon fails to retry the step after the last attempt, we will send you an email indicating the error, and workflow execution in your Task History.
