Clarify throttling messages to specify actual wait time

[waldekmastykarz]

Currently, when a throttling event occurs, the message shown is:

retry #1 in 10.0s (retry-after: 42s), 429, Too Many Requests

It's unclear to users whether the script will wait 10 seconds or 42 seconds due to the "retry-after" value. This ambiguity can cause confusion regarding the actual delay between retries.

Suggestion:

• Update the throttling message to clearly indicate which value is being used for the wait before the next retry (e.g., "Waiting 42s before retry Bump word-wrap from 1.2.3 to 1.2.5 #1 as instructed by retry-after header").
• Consider removing or explaining the other value if it's not used for the actual wait, or provide a clear explanation of both values if they are relevant.

Clearer messaging will help users better understand the script's behavior during throttling situations.

[github-actions]

Thank you for raising this issue. The report clearly describes the confusion caused by the current throttling message and provides concrete examples of the ambiguity. The suggestion to clarify which value is being used (and to provide context about the "retry-after" header) is helpful.

To help address the issue, could you clarify:

• Are there cases where the script uses a calculated wait time instead of the "retry-after" value from the header? If so, could you provide an example scenario?
• Is the concern mainly about messaging, or is there uncertainty about whether the correct wait logic is being applied as well?

Providing this additional context will help ensure that the root cause of the ambiguity is addressed and that the message accurately reflects the script's behavior.

AI-generated content by issue-reviewer may be incorrect. Use reactions to eval.

[waldekmastykarz]

The concern is mainly about messaging

[pelikhan]

i should start using devproxy for these. could you help configure the project to add 1 retry every time?

[pelikhan]

Added more logging to message, also DEBUG=genaiscript:fetch:retry logger

[github-actions]

Summary:

The issue focuses on unclear throttling messages during script execution, specifically confusion over which value (a hardcoded wait time vs. the "retry-after" header from the server) is actually used to determine the delay between retries. The request is to clarify messaging so users understand the actual wait behavior during throttling (HTTP 429 responses).

Comments clarify:

• The main concern is with the clarity of user-facing messages, not the throttling logic.
• Suggestions include more explicit messaging about which delay is in effect, potentially leveraging logging or debug tools.
• There is also discussion about configuring retries for testing and improvements to logging for better traceability.

AI-generated content by issue-bot may be incorrect. Use reactions to eval.

[waldekmastykarz]

i should start using devproxy for these. could you help configure the project to add 1 retry every time?

We've got a step-by-step description of the configuration here: https://learn.microsoft.com/microsoft-cloud/dev/dev-proxy/how-to/test-that-my-application-handles-throttling-properly. Please, let me know if anything's unclear