OpenAIRetryServiceAdapter retries non-retryable errors

[phelps-sg]

Currently OpenAIRetryServiceAdapter retries all errors except OpenAIScalaTokenCountExceededException. However, there are many other errors returned by the API that should not be retried, e.g. 401 and 402 (https://platform.openai.com/docs/guides/error-codes/api-errors).

I think in general it is only 429 errors that should be retried. Relevant OpenAI docs quoted below. See also #9.

429 - Rate limit reached for requests	Cause: You are sending requests too quickly. Solution: Pace your requests. Read the Rate limit guide.
429 - You exceeded your current quota, please check your plan and billing details	Cause: You have hit your maximum monthly spend (hard limit) which you can view in the account billing section. Solution: Apply for a quota increase.
429 - The engine is currently overloaded, please try again later	Cause: Our servers are experiencing high traffic. Solution: Please retry your requests after a brief wait.

429 - Rate limit reached for requests Cause: You are sending requests too quickly.
Solution: Pace your requests. Read the Rate limit guide.
429 - You exceeded your current quota, please check your plan and billing details Cause: You have hit your maximum monthly spend (hard limit) which you can view in the account billing section.
Solution: Apply for a quota increase.
429 - The engine is currently overloaded, please try again later Cause: Our servers are experiencing high traffic.
Solution: Please retry your requests after a brief wait.

[peterbanda]

Hey @phelps-sg,

Very nice summary. Thanks for the feedback! Will check the error codes and apply retry only on the relevant ones (will need new exceptions).

Cheers

[phelps-sg]

Many thanks! Have submitted an initial unit test for OpenAIRetryServiceAdapter in PR #18 in case it is useful.