Skip to main content

Retry Engine

The Retry Engine in Solvix ensures that transient failures do not break your application.

It automatically retries failed requests based on configurable strategies, making your system more resilient.

Why Retry Matters​

In real-world systems, failures are common:

  • Network instability
  • Temporary server errors
  • Rate limits
  • Timeouts

Retrying intelligently can significantly improve success rates.

Basic usage​

const client = createClient({
  retry: {
    retries: 3,
  },
});

Retry configuration​

retry: {
  retries: number,
  baseDelay?: number,
  maxDelay?: number,
  shouldRetry?: (error) => boolean
}

Options explained​

retries​

Number of retry attempts.

retries: 3;

baseDelay​

Initial delay (ms) before retry.

baseDelay: 300;

maxDelay​

Maximum delay cap.

maxDelay: 2000;

shouldRetry​

Custom retry logic.

shouldRetry: (error) => {
  return error.status >= 500;
};

Exponential backoff​

Solvix uses exponential backoff:

delay = baseDelay * 2^attempt

This prevents overwhelming servers.

Example​

const client = createClient({
  retry: {
    retries: 3,
    baseDelay: 200,
  },
});

Retry flow​

  1. Request fails
  2. Check retry condition
  3. Wait (backoff)
  4. Retry request
  5. Repeat until success or limit reached

Retry + Error integration​

try {
  await client.get("/api");
} catch (error) {
  console.log(error.attempts);
}

Best practices​

  • Avoid high retry counts
  • Use retry only for transient errors
  • Combine with circuit breaker
  • Use custom logic when needed

Summary​

The Retry Engine improves reliability by automatically handling temporary failures.