The worst thing about making calls to external service is that those calls sometimes fail because of network errors or because the service is temporarily unavailable. It makes sense to retry the function call a few times and hope that the problem disappears in those situations.

The retry package gives us an easy method to implement retries in Python. All we need is a decorator added to the function we want to retry.

First, we have to install the package. If we use pip, the installation looks like this: pip install retry.

After that, we must find the definition of the function we want to retry, for example, this one:

1
2
def a_function_that_fails_for_no_reason():
    ...

Now, we add the decorator to the function. To define a decorator, we need the names of the errors that should cause a retry. All other errors will be propagated to the caller without retrying the function.

Parsing machine learning logs with Ahana, a managed Presto service, and Cube, a headless BI solution

Check out my article published on the Cube.dev blog!

Additionally, we need the number of retries and the delay between the subsequent function calls.

1
2
3
4
5
from retry import retry

@retry((HTTPError, DNSError), tries = 3, delay = 2, backoff = 2)
def a_function_that_fails_for_no_reason():
    ...