Resilience Patterns

Resilience Patterns

Chronos provides robust mechanisms to handle task failures and network hiccups.

Execution Policy

Define how failures are handled using ExecutionPolicy.

Stop on Failure (Default)

The task enters a FAILED state and will not run again.

ExecutionPolicy.stopOnFailure();

Continue on Failure

The task logs the error (via listeners) and continues to the next scheduled interval.

ExecutionPolicy.continueOnFailure();

Retry with Backoff

Automatically retry a failed task before giving up.

ExecutionPolicy.retry(
    RetryPolicy.exponentialBackoff(
        Duration.ofSeconds(1),  // Initial delay
        Duration.ofMinutes(5),  // Max delay
        2.0,                    // Multiplier
        3                       // Max attempts
    )
);

Jitter

Add randomness to your schedules to prevent "thundering herd" issues when many tasks are scheduled for the same time.

// Add +/- 0 to 500ms
SchedulerSpec.builder()
    .defaultJitter(Jitter.uniform(0.5))
    .build();

Or Gaussian jitter:

// Gaussian distribution (mean 0, stdDev 100ms)
Jitter.gaussian(0.1);