[REQUEST] Retry Mechanism with Backoff
Is your feature request related to a problem? Please describe. Workflows frequently encounter transient failures when interacting with external services, leading to unnecessary total failures instead of graceful recovery.
Describe the solution you'd like A configurable retry mechanism that provides:
- Exponential backoff settings
- Maximum retry attempts configuration
- Customizable retry conditions
- Jitter options to prevent thundering herd problems
- Timeout controls
- Ability to apply to any block in the workflow
Describe alternatives you've considered Manual implementation of retry logic in each block, but this is error-prone and creates significant duplication.
Additional context This would dramatically improve workflow reliability when dealing with rate-limited APIs, network instability, and other temporary failures.
Can you assign me to this one?
@aadamgough let's hold off on this one for now - there are some other tools I think that are higher priority