Retry idempotent GET requests in Python

A Python function that retries an idempotent GET request a fixed number of times with a delay between attempts, raising a RuntimeError only after all retries fail.

Easy Python 3.9+ Aug 9, 2026 Microservices patterns 14 views 0 copies

retry idempotent urllib http

Python code

23 lines

Python 3.9+

import time
import urllib.error
import urllib.request
from http.client import HTTPException

def fetch_with_retry(url, max_retries=3, delay=1.0):
    for attempt in range(1, max_retries + 1):
        try:
            with urllib.request.urlopen(url, timeout=5) as response:
                return response.read().decode()
        except (urllib.error.URLError, HTTPException, TimeoutError) as e:
            print(f"Attempt {attempt} failed for {url}: {e}")

            if attempt < max_retries:
                time.sleep(delay)
    raise RuntimeError(f"Failed after {max_retries} attempts: {url}")

if __name__ == "__main__":
    try:
        result = fetch_with_retry("http://127.0.0.1:9999/mock", max_retries=3, delay=0.5)
        print(f"Success: {result}")
    except RuntimeError as err:
        print(err)

Output

stdout

Attempt 1 failed for http://127.0.0.1:9999/mock: <urlopen error [Errno 111] Connection refused>
Attempt 2 failed for http://127.0.0.1:9999/mock: <urlopen error [Errno 111] Connection refused>
Attempt 3 failed for http://127.0.0.1:9999/mock: <urlopen error [Errno 111] Connection refused>
Failed after 3 attempts: http://127.0.0.1:9999/mock

How it works

The urllib.request.urlopen call performs the GET request inside a context manager, automatically closing the response. The try/except catches network-level failures like URLError, HTTPException, and TimeoutError, which are the typical exceptions for transient network issues. The loop iterates from 1 to max_retries, printing the attempt number on each failure and sleeping delay seconds between attempts. If all attempts fail, the function raises a RuntimeError so the caller can handle the final failure explicitly. Because GET requests are idempotent, retrying them is safe, which makes this a common pattern in microservice clients.

Common mistakes

Retrying non-idempotent requests (POST, PUT) without idempotency keys, which can cause duplicate side effects.
Not catching both `urllib.error.URLError` and `HTTPException`, missing timeouts.
Sleeping before the first attempt or after the last one, wasting time.
Forgetting to close the response object when not using a context manager.

Variations

Use `requests.Session` with `HTTPAdapter` to get built-in retry support.
Add exponential backoff by multiplying `delay` by a factor on each retry.

Real-world use cases

A microservice calling a downstream HTTP API that occasionally fails due to transient network errors.
A health check endpoint that retries a few times before declaring a service unhealthy.
A background worker fetching configuration from a remote endpoint with short-lived connectivity issues.

Retry idempotent GET requests in Python

Python code

Output

How it works

Common mistakes

Variations

Real-world use cases

More from Microservices patterns

Tutorials

Quizzes

Python code

Output

How it works

Common mistakes

Variations

Real-world use cases

More from Microservices patterns

Keep learning

Tutorials

Quizzes