MoonshotAI / MoonshotAI/kimi-cli

在云端服务器部署的kimiclaw 无回复 CLI 持续 429 engine_overloaded 超过 48 小时,已导出诊断文件

Open
#2,209 4 comments 3 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug
Dominant language
Python
Stars
11.4k
Forks
1.3k
Avg merge
9h 47m
Merged PRs (30d)
2

Description

What version of Kimi Code CLI is running?

v1.41.0(已从 v1.24.0 升级,升级后问题依旧)

Which open platform/subscription were you using?

Kimi 官方平台(通过 /login 登录)

Which model were you using?

Kimi-k2.6(升级前为 kimi-for-coding / k2.5,两者均 429)

What platform is your computer?

Linux x86_64(远程服务器,非本地)

What issue are you seeing?

从 2026-05-08 起,kimi-cli 所有对话请求均返回 429 engine_overloaded,持续超过 48 小时。

错误信息:
LLM provider error: Error code: 429 - {'error': {'message': 'The engine is currently overloaded, please try again later', 'type': 'engine_overloaded_error'}}

关键观察:

  • 飞书端可正常对话,网页版正常,仅 CLI 通道完全不可用
    -飞书端我让他重启gateway,然后他就敲键盘8个小时了。。。看起来是完全卡住了,但是定时任务还能跑
  • 本地进程正常(lsof -i :18789 确认端口监听正常)
  • 额度未耗尽,账号状态正常
  • 已尝试重启服务、升级 CLI(v1.24→v1.41)、切换模型(k2.5→k2.6),均无效
  • Session ID:3f1a45bc-0727-41fb-86e2-9e9c780b7dd8
What steps can reproduce the bug?
  1. 远程 Linux 服务器上执行 kimi-cli
  2. 进入交互界面,模型显示为 Kimi-k2.6
  3. 输入任意消息(如 nihao)并回车
  4. 立即返回 429 engine_overloaded
  5. 等待 10-30 分钟后重试,结果相同
  6. 重启服务、升级 CLI、切换模型后重试,结果相同
What is the expected behavior?

CLI 应正常返回模型回复,而非持续 429。飞书端和网页版目前均可正常使用。

Additional information
  • 已执行 kimi export 导出诊断文件(20KB ZIP),包含完整日志与配置
  • 诊断文件涉及敏感信息(API Key、路径等),未在此上传,请提供邮箱或私发方式
  • 该问题疑似服务端对 CLI 通道/该会话的持续限流,非临时过载

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the kimi-cli entry point and reproduce the 429 engine_overloaded response on a remote Linux x86_64 server using the steps in the issue. Review the exported diagnostic archive carefully because it contains sensitive API keys and paths, and compare the CLI behavior with the working Feishu and web channels. Done means CLI requests receive model responses instead of persistent 429 errors.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
cli
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
38/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.