[agent/embedded] embedded run agent end: runId=d63c9b38-d041-4ac9-ab0f-edebfe6bc999 isError=true error=LLM request timed out.
我本地有7890端口的代理。
4 个赞
让终端走7890端口代理？
例如这样：
export http_proxy="http://127.0.0.1:7890"
export https_proxy="http://127.0.0.1:7890"
2 个赞
windows系统，我根据豆包给的方案，这样设置，就是还是超时。（有测试过，终端里这样配置能获取模型）
PS C:\Users\Windows> $env:HTTP_PROXY=“http://127.0.0.1:7890”
PS C:\Users\Windows> $env:HTTPS_PROXY=“http://127.0.0.1:7890”
PS C:\Users\Windows> openclaw gateway
OpenClaw 2026.3.8 (3caab92) — I’m like tmux: confusing at first, then suddenly you can’t live without me.
22:14:52 [canvas] host mounted at http://127.0.0.1:18789/__openclaw__/canvas/ (root C:\Users\Windows.openclaw\canvas)
22:14:52 [heartbeat] started
22:14:52 [health-monitor] started (interval: 300s, startup-grace: 60s, channel-connect-grace: 120s)
22:14:52 [gateway] agent model: google/gemini-3-flash-preview
22:14:52 [gateway] listening on ws://127.0.0.1:18789, ws://[::1]:18789 (PID 49360)
22:14:52 [gateway] log file: C:\Users\Windows\AppData\Local\Temp\openclaw\openclaw-2026-03-10.log
22:14:52 [browser/server] Browser control listening on http://127.0.0.1:18791/ (auth=token)
22:14:59 [ws] webchat connected conn=eefec111-a32d-4f15-a81e-b327e722e66a remote=127.0.0.1 client=openclaw-control-ui webchat v2026.3.8
22:15:01 [tools] tools.profile (coding) allowlist contains unknown entries (apply_patch, image). These entries won’t match any tool unless the plugin is enabled.
22:15:03 [ws] webchat connected conn=e590e7f8-4317-4433-b35e-491e3f05599f remote=127.0.0.1 client=openclaw-control-ui webchat v2026.3.8
22:15:12 [agent/embedded] embedded run agent end: runId=01f7d2a8-022f-4400-914a-ab2db076c656 isError=true error=LLM request timed out.
22:15:25 [agent/embedded] embedded run agent end: runId=01f7d2a8-022f-4400-914a-ab2db076c656 isError=true error=LLM request timed out.
那可能配置没有生效到openclaw，我部署用的是docker，直接在服务器上开的tun模式
实在不行就走tun模式吧 
或者让它自查，先配上kimi或者其他国内的模型，让他测试调用谷歌的模型，然后告诉他代理在7890，让他直接配好
你这个是http http代理也可以试下socks5代理。我电脑不在身边明天截图给你
开全局模式试一下。
这应该是配置问题吧 （我平时调用gpt claude应该都得走代理 但实际上我没有手动给他配置代理 clashverge开的规则模式）
会不会是url配置问题？站内有一篇openclaw疑难汇总
直接tun模式一了百了