This n8n workflow demonstrates a sophisticated approach to handling unreliable API responses by implementing custom retry and delay logic. Designed to overcome the default limitations of retries (max 5) and delay periods (max 5 seconds), this workflow allows for configurable, extended retries with adjustable delays, including features like exponential backoff.

The process begins with a manual trigger, allowing users to initiate the workflow intentionally. It then proceeds to set initial parameters such as ‘max_tries’ and ‘delay_seconds’, which define the total retry limit and initial delay interval. The core of the workflow involves making an HTTP request to a specified URL, with error handling that continues on failure, enabling retries.

If the HTTP request fails, the workflow updates the delay and maximum attempts dynamically, then checks if the retry count has exceeded the limit. If so, it stops execution and raises an error indicating service unavailability. Otherwise, it waits for the specified delay before retrying, creating a loop that provides granular control over retry behavior, which is especially useful for managing API rate limits or temporary outages.

This workflow is ideal in scenarios where robust error handling and controlled retries are critical, such as integrating with flaky external APIs, avoiding overloading services, or implementing custom backoff strategies in automation pipelines.