Search code examples
pythonerror-handlingopenai-api

OpenAI API: How to catch all 5xx errors in Python?

I want to catch all 5xx errors (e.g., 500) that OpenAI API sends so that I can retry before giving up and reporting an exception.

Right now I'm basically doing the following:

try:
    response = openai.ChatCompletion.create(req)
except InvalidRequestError as e:
    reportError
except ServiceUnavailableError as e:
    retry
except Exception as e:
    response = f"Exception: {e}"
    raise Exception(response)

Some 5xx errors are getting caught as unknown errors (last case) which I want to catch so that I can retry them as I do in the case of the ServiceUnavailableError. But I don't know how to go about catching all the 5xx errors for retry. The docs just talk about how to catch the specifically named errors.

Solution

  • All 5xx errors belong to the ServiceUnavailableError. Take a look at the official OpenAI documentation:

    TYPE OVERVIEW
    APIError Cause: Issue on our side.
    Solution: Retry your request after a brief wait and contact us if the issue persists.
    Timeout Cause: Request timed out.
    Solution: Retry your request after a brief wait and contact us if the issue persists.
    RateLimitError Cause: You have hit your assigned rate limit.
    Solution: Pace your requests. Read more in our Rate limit guide.
    APIConnectionError Cause: Issue connecting to our services.
    Solution: Check your network settings, proxy configuration, SSL certificates, or firewall rules.
    InvalidRequestError Cause: Your request was malformed or missing some required parameters, such as a token or an input.
    Solution: The error message should advise you on the specific error made. Check the documentation for the specific API method you are calling and make sure you are sending valid and complete parameters. You may also need to check the encoding, format, or size of your request data.
    AuthenticationError Cause: Your API key or token was invalid, expired, or revoked.
    Solution: Check your API key or token and make sure it is correct and active. You may need to generate a new one from your account dashboard.
    ServiceUnavailableError Cause: Issue on our servers.
    Solution: Retry your request after a brief wait and contact us if the issue persists. Check the status page.

    Handle the ServiceUnavailableError as follows:

    try:
  # Make your OpenAI API request here
  response = openai.Completion.create(prompt="Hello world",
                                      model="text-davinci-003")

except openai.error.ServiceUnavailableError as e:
  # Handle 5xx errors here
  print(f"OpenAI API request error: {e}")
  pass