Repository navigation
Codex server_is_overloaded stream errors are not retried #25730
Description
Description
OpenAI Codex subscription streams can return transient overload errors in this shape:
{"type":"error","sequence_number":2,"error":{"type":"service_unavailable_error","code":"server_is_overloaded","message":"Our servers are currently overloaded. Please try again later.","param":null}}OpenCode should treat this as a retryable provider overload. Today this nested stream error shape can miss retry classification because the overload code/type are nested under error.
Related broader reports: #16214 and #21893.
Plugins
None
OpenCode version
Latest dev
Steps to reproduce
- Use an OpenAI Codex subscription model.
- Hit a transient streamed overload response with
error.code: server_is_overloaded. - Observe that the error can terminate instead of entering the existing retry path.
Screenshot and/or share link
No response
Operating System
No response
Terminal
No response
Metadata
Metadata
Assignees
Labels
No labels
Activity
github-actions commentedon May 4, 2026
This issue might be a duplicate of existing issues. Please check:
session/retry.tsandprovider/error.tsare involved. Theserver_is_overloaded/service_unavailable_errorshape from Codex may be a specific sub-case of what that issue tracks.If this
server_is_overloadederror shape is not yet handled by any fix or PR against #21893, please note that distinction in the existing issue or confirm this is a separate gap.spark4862 commentedon Jun 2, 2026
Related fix in #30323 — includes plus the broader OpenAI/Codex stream envelope set in one classifier. Happy to coordinate if #25728 is preferred for overload-only.
spark4862 commentedon Jun 2, 2026
Related fix in #30323 — includes server_is_overloaded plus the broader OpenAI/Codex stream envelope set in one classifier. Happy to coordinate if #25728 is preferred for overload-only.
github-actions commentedon Aug 2, 2026
To stay organized issues are automatically closed after 60 days of no activity. If the issue is still relevant please open a new one.