Codex app repeatedly fails with remote environment offline and model capacity errors

What version of the Codex App are you using (From “About Codex” dialog)?

26.908.40834

What subscription do you have?

ChatGPT Pro20x

What platform is your computer?

Darwin 25.6.0 arm64 arm

What issue are you seeing?

All Codex chats and models fail repeatedly.

The affected models include GPT-6 Astra and GPT-5.6 Terra. The app reports:
“Selected model is at capacity. Please try a different model.”

It also reports:
“Error running remote compact task: Codex ran out of room in the model's context window.”

After reinstalling the app and logging in again, the problem remains. The account usage page shows that usage is allowed and the account is not rate-limited.

Local proxy transport is working, but the Codex desktop logs repeatedly show:

• Remote environment is offline
• Opening handshake has timed out
• Codex app-server is not available

This affects all chats, including new chats.

What steps can reproduce the bug?

Feedback ID: 01a0a56e-efa9-7c22-8a86-7caac0d6c0a8

What is the expected behavior?

A new chat should accept a short prompt and run normally using the selected model.

Additional information

Feedback ID: 01a0a56e-efa9-7c22-8a86-7caac0d6c0a8

The issue appears to be account-side remote execution routing or a remote environment failure rather than a local installation problem. Please reset or repair the remote execution environment for this account.

Start with the Codex desktop logs and Feedback ID 01a0a56e-efa9-7c22-8a86-7caac0d6c0a8, focusing on the remote environment offline, handshake timeout, and app-server unavailable messages. Done means identifying and resolving the account-side remote execution failure so a new chat can run a short prompt with the selected model.