Error Handling
Error handling is critical for building resilient workflows. Orcher provides multiple strategies for handling failures — from automatic retries with configurable backoff policies, to the saga pattern — where you pair each step with a compensating action to undo completed work on failure — to circuit breakers that prevent cascading failures. You can also implement graceful degradation with fallback data sources, set timeouts to bound long-running operations, and route persistent failures to dead letter queues for manual inspection. This guide covers each strategy with complete examples.
Retry & Compensation
Automatic retry with configurable backoff, plus saga compensation patterns you compose
Error Codes
The code vocabulary shared across SDKs, and what to branch on instead of message text
Resilience Patterns
Circuit breakers, graceful degradation, and timeout handling
Dead Letter Queue
Route persistent failures to a DLQ for manual inspection and replay
Best Practices & Troubleshooting
Guidelines for retry policies, idempotent compensation, logging, timeouts, and common issues