Durable Execution
How do I choose between retry policies and compensation in a durable agent workflow?
Choose retries when the failed step can be safely re-executed without side effects, such as calling an idempotent API. Choose compensation when a step completed but later steps fail, requiring rollback of prior actions, like refunding a payment. Catalyst lets you define retry policies per activity and compensation activities per workflow. For non-idempotent steps, always prefer compensation over infinite retries.
Was this article helpful?
Your feedback helps improve Diagrid's FAQ experience.
Keep reading
More Diagrid FAQ articles
- Durable Execution
What is durable execution in AI agent workflows?
Durable execution means an AI agent workflow can keep its progress even when a process crashes, a tool call fails, or the system restarts.
- Durable Execution
Why do production AI agents need durable workflows?
Production AI agents need durable workflows because real agent tasks rarely finish in a single clean request.
- Durable Execution
Is checkpointing enough for production AI agents?
Checkpointing helps, but it is usually not enough by itself for production AI agents. It explains the production reliability impact for AI agent workflows.