Bug type

Regression (worked before, now fails)

Beta release blocker

No

Summary

Sessions in full timeout spam (unusable) when Active Memory ON

2026.4.14
webchat became “one message then stuck”
openai-codex showed refresh_token_reused / timeout behavior
disabling Active Memory made the session usable again

Steps to reproduce

Openclaw 2026.4.14

Provider: Openai-Codex (OAuth)

Active Memory ON

Expected behavior

No errors, regular webchat ui flow.

Actual behavior

08:25:12+00:00 warn agent/embedded {"subsystem":"agent/embedded"} {"event":"embedded_run_agent_end","tags":["error_handling","lifecycle","agent_end","assistant_error"],"runId":"43077d1c-64ea-4035-80bc-869fbb39a58a","isError":true,"error":"terminated","failoverReason":null,"model":"gpt-5.4","provider":"openai-codex","rawErrorPreview":"terminated","rawErrorHash":"sha256:e8c95a2a7271","providerRuntimeFailureKind":"unknown"} embedded run agent end

OpenClaw version

2026.4.14

Operating system

macOS

Install method

npm

Model

openai-codex/gpt-5.4

Provider / routing chain

direct

Additional provider/model setup details

No response

Logs, screenshots, and evidence

Impact and severity

No response

Additional information

No response