Retry Strategy: Implement Configurable Retry Logic with Exponential Backoff

Question

## Round 1 - Coding

## Problem

Implement a retry decorator/wrapper that retries a failing function call up to `max_retries` times with configurable backoff strategies. Support fixed delay, linear backoff, and exponential backoff with optional jitter.

```python
import time
from typing import Callable, Type

class RetryStrategy:
    def __init__(self, max_retries: int, backoff: str = "exponential",
                 base_delay: float = 1.0, jitter: bool = False,
                 retryable_exceptions: tuple = (Exception,)):
        ...
    def execute(self, func: Callable, *args, **kwargs):
        ...
```

## Example

```
def flaky_api_call():
    import random
    if random.random() < 0.7:
        raise ConnectionError("timeout")
    return "success"

strategy = RetryStrategy(max_retries=3, backoff="exponential", base_delay=1.0)
result = strategy.execute(flaky_api_call)
# Attempt 1: fails -> wait 1s
# Attempt 2: fails -> wait 2s
# Attempt 3: fails -> wait 4s
# Attempt 4: succeeds -> returns "success"
# If all 4 fail -> raises the last exception
```

## Follow-ups

1. Why add jitter to exponential backoff? What problem does it solve in distributed systems?
2. How do you distinguish retryable errors (network timeout) from non-retryable ones (HTTP 400 bad request)?
3. How would you implement a circuit breaker on top of this retry strategy?
4. How would you make this work with `async`/`await` for async functions?

Retry Strategy: Implement Configurable Retry Logic with Exponential Backoff

Question Details

Full Details

Topics

More from MongoDB