Issue Station
Sign In
Tools hereToolbox

openclaw - 💡(How to fix) Fix [Bug]: Native Codex runtime shows mixed openai-codex model state and idle-timeouts in Telegram direct sessions [1 pull requests]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…

With the recommended native Codex configuration (model.primary = "openai/gpt-5.5", agentRuntime.id = "codex"), some Telegram direct sessions appear to hang or become non-responsive and later show a mixed state with Model: openai-codex/gpt-5.5 and Runtime: OpenAI Codex, suggesting a possible fallback, auto-pin, or routing-state issue.

Root Cause

With the recommended native Codex configuration (model.primary = "openai/gpt-5.5", agentRuntime.id = "codex"), some Telegram direct sessions appear to hang or become non-responsive and later show a mixed state with Model: openai-codex/gpt-5.5 and Runtime: OpenAI Codex, suggesting a possible fallback, auto-pin, or routing-state issue.

Fix Action

Fixed

RAW_BUFFERClick to expand / collapse

Bug type

Behavior bug (incorrect output/state without crash)

Beta release blocker

No

Summary

With the recommended native Codex configuration (model.primary = "openai/gpt-5.5", agentRuntime.id = "codex"), some Telegram direct sessions appear to hang or become non-responsive and later show a mixed state with Model: openai-codex/gpt-5.5 and Runtime: OpenAI Codex, suggesting a possible fallback, auto-pin, or routing-state issue.

Steps to reproduce

  1. Configure a Telegram direct-chat agent with model.primary = "openai/gpt-5.5" and agentRuntime.id = "codex".
  2. Reset an existing Telegram direct session that was previously using the stable PI route with openai-codex/gpt-5.5.
  3. Send a normal message to the agent and wait for the native Codex runtime to handle the turn.
  4. Observe whether the session becomes slow, hangs, or stops responding.
  5. Check the session status and logs.
  6. In affected sessions, status may show Model: openai-codex/gpt-5.5 with Runtime: OpenAI Codex, and logs may include idle-timeout or selected-model-unavailable messages.

Expected behavior

Based on prior observed fresh native Codex canary sessions, a Telegram direct session configured with model.primary = "openai/gpt-5.5" and agentRuntime.id = "codex" should show Model: openai/gpt-5.5, Runtime: OpenAI Codex, and logs indicating agentHarnessId: codex.

Actual behavior

In affected Telegram direct sessions after migration to the native Codex runtime, the agent becomes slow or non-responsive, while status shows the mixed state Model: openai-codex/gpt-5.5 with Runtime: OpenAI Codex; observed logs include codex app-server turn idle timed out waiting for completion and selected model unavailable, and observed routing behavior includes openai/gpt-5.5 -> openai-codex/gpt-5.5.

OpenClaw version

OpenClaw 2026.5.7 (eeef486)

Operating system

macOS 26.4.1, build 25E253

Install method

Global npm package installed under the Homebrew prefix; binary at /opt/homebrew/bin/openclaw, symlinked to /opt/homebrew/lib/node_modules/openclaw/openclaw.mjs.

Model

Target model under test: openai/gpt-5.5, but Observed affected/effective model state: openai-codex/gpt-5.5

Provider / routing chain

Prior stable route: openai-codex/gpt-5.5 with Runtime: OpenClaw Pi Default. Target native Codex route: model.primary = "openai/gpt-5.5" with agentRuntime.id = "codex", expected to show Runtime: OpenAI Codex. Affected sessions show mixed state: Model: openai-codex/gpt-5.5 with Runtime: OpenAI Codex.

Additional provider/model setup details

No response

Logs, screenshots, and evidence

Impact and severity

No response

Additional information

No response

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

FAQ

Expected behavior

Based on prior observed fresh native Codex canary sessions, a Telegram direct session configured with model.primary = "openai/gpt-5.5" and agentRuntime.id = "codex" should show Model: openai/gpt-5.5, Runtime: OpenAI Codex, and logs indicating agentHarnessId: codex.

#batch processing#GPU compatibility#latency issue#model loading#dependency error

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING