Improve error logs when Codex fails to create a task due to timeout
Feature request
When Codex fails to create a task, it would be very helpful to provide more detailed logs and error messages to help users identify the root cause.
I recently encountered a task creation timeout issue. After investigation, I found that one of my MCP servers was responding very slowly. Codex was waiting for the MCP server response, which eventually caused the task creation to fail with a timeout.
However, the main problem was that there were no clear logs or error messages indicating that the MCP server response was the cause. Without this information, troubleshooting took a lot of unnecessary time.
Suggestion
Could Codex provide more explicit diagnostics when task creation fails, for example:
- Which step is currently blocking (MCP initialization, tool discovery, network request, etc.)
- Which MCP server or component is causing the delay
- Timeout reason and relevant timing information
- A suggestion to check or disable problematic MCP servers
Better error reporting would significantly improve the debugging experience, especially for users with custom MCP configurations.
This is just a suggestion. Thanks for building Codex!
2 Comments
Potential duplicates detected. Please review them and close your issue if it is a duplicate.
Powered by Codex Action
This issue is related to #31359, but focuses on improving diagnostics rather than only handling unreachable MCP servers.
#31359 highlights that an unavailable MCP server can block task creation and cause a timeout. My suggestion here is that, regardless of the specific MCP failure mode (unreachable, slow response, initialization hang, etc.), Codex should provide clearer diagnostic logs so users can quickly identify what is blocking task creation.
Better observability would help users troubleshoot similar issues without spending significant time guessing the root cause.