github / github/app

Temporary rate limits stop active task execution instead of retrying

Open
#3,113 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
No language data
Stars
2.1k
Forks
153
PR merge metrics
No merged PRs in 30d

Description

### Description

Copilot App sometimes stops an active task after displaying a temporary rate-limit error:

> You’ve hit your rate limit. Please wait for your limit to reset in under a minute or switch to auto model to continue.

### Steps to reproduce

1. Start a longer-running task using Sol with Extra High effort.
2. If a temporary rate limit occurs, observe the session behavior.

### Expected behavior

When the rate limit is temporary, Copilot App should wait and retry the task using the same model and effort level rather than stopping execution.

The retry behavior should be considerate of upstream providers, allowing sufficient time between attempts and limiting retries so that recovery attempts do not add unnecessary load.

If the task cannot be resumed after a reasonable period, the app should stop and clearly notify the user.

### Actual behavior

Execution stops after the rate-limit error and does not automatically resume, despite the message stating that the limit should reset in under a minute.

### Impact

Long-running or delegated tasks can terminate because of a short-lived rate limit. The user must notice the interruption and manually continue or restart the task.

Contributor guide

Open the contributing guide

Research direction

No files or tests are named. Start by tracing active task execution through temporary rate-limit errors and identify the existing retry and user-notification paths; done means temporary limits retry with the same model and effort, retries are bounded and spaced, and unrecoverable tasks stop with a clear notice.

Written by the indexing model from the issue text.

Assessment

Domain
desktop
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Active
Clarity
Mostly clear
Newbie friendliness
52/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.