error_handling

Implement a circuit breaker by tracking failure counts and state (CLOSED, OPEN, HALF_OPEN) around external calls. After a failure threshold, requests are rejected until a timeout passes, then trial requests test recovery before closing the circuit again.

Use exceptions for unexpected, exceptional conditions and Result types for expected failures like validation errors. Result types make error paths explicit in function signatures, while exceptions suit truly exceptional situations that disrupt normal control flow.

Wrap the call in a retry decorator that catches specified exceptions and sleeps for backoff_factor raised to the attempt number between tries. Limit max attempts and only retry transient errors like network timeouts, not validation failures.

Yes, async error handling uses try-catch around awaited calls, with specific handling per error type such as returning defaults for NotFound or retrying network errors. Unhandled promise rejections must be caught explicitly to avoid silent failures.

Catching generic Exception hides programming bugs like null references and makes debugging harder. Catch specific exception types, log unexpected errors with full context, and re-raise them wrapped in meaningful application errors.