1

curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.40.3/install.sh | bash

1
2

echo 'export NVM_DIR="$([ -z "${XDG_CONFIG_HOME-}" ] && printf %s "${HOME}/.nvm" || printf %s "${XDG_CONFIG_HOME}/nvm")"' >> ~/.bashrc
echo '[ -s "$NVM_DIR/nvm.sh" ] && \. "$NVM_DIR/nvm.sh"' >> ~/.bashrc

1

source ~/.bashrc

1

nvm --version

1

nvm install 20

1

npm install -g npm

1

npm install -g @anthropic-ai/claude-code

1

claude --version

1

1.0.96 (Claude Code)

1

npm install -g @musistudio/claude-code-router

1
2

export API_BASE_URL=https://xxx.com/v1/chat/completions
export API_KEY=sk-xxx

1

mkdir -p ~/.claude-code-router

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34

cat > ~/.claude-code-router/config.json <<EOF
{
  "LOG": true,
  "LOG_LEVEL": "trace",
  "HOST": "127.0.0.1",
  "PORT": 3456,
  "API_TIMEOUT_MS": "600000",
  "Providers": [
    {
      "name": "oneapi",
      "api_base_url": "${API_BASE_URL}",
      "api_key": "${API_KEY}",
      "models": [
        "deepseek-v3",
        "deepseek-r1"
      ],
      "transformer": {
        "use": [
          [
            "maxtoken",
            {
              "max_tokens": 16384
            }
          ]
        ]
      }
    }
  ],
  "Router": {
    "default": "oneapi,deepseek-v3",
    "think": "oneapi,deepseek-r1"
  }
}
EOF

1

ccr restart

1

ccr code

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19

   Enter to confirm · Esc to exit
╭───────────────────────────────────────────────────╮
│ ✻ Welcome to Claude Code!                         │
│                                                   │
│   /help for help, /status for your current setup  │
│                                                   │
│   cwd: /root                                      │
│                                                   │
│   ─────────────────────────────────────────────── │
│                                                   │
│   Overrides (via env):                            │
│                                                   │
│   • API timeout: 600000ms                         │
│   • API Base URL: http://127.0.0.1:3456           │
╰───────────────────────────────────────────────────╯

╭────────────────────────────────────────────────────────────────────────────────────────╮
│ >帮我检查当前节点上 GPU 卡的状态                                                           │
╰────────────────────────────────────────────────────────────────────────────────────────╯

1

ccr code -p "怎么样给k8s集群续证书"

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40

    Service not running, starting service...
    给Kubernetes集群续证书主要涉及以下几个步骤：

    1. 检查当前证书有效期：
    ```bash
    kubeadm alpha certs check-expiration
    ````

    2. 续订所有证书（控制平面）：

    ```bash
    kubeadm alpha certs renew all
    ```

    3. 重启控制平面组件：

    ```bash
    docker restart $(docker ps | grep kube-apiserver | awk '{print $1}')
    docker restart $(docker ps | grep kube-controller-manager | awk '{print $1}')
    docker restart $(docker ps | grep kube-scheduler | awk '{print $1}')
    ```

    4. 更新 kubeconfig 文件：

    ```bash
    kubeadm alpha kubeconfig user --org system:masters --client-name kubernetes-admin
    ```

    5. 验证新证书：

    ```bash
    openssl x509 -noout -text -in /etc/kubernetes/pki/apiserver.crt | grep Not
    ```

    注意：

    - 这些操作需要在 master 节点上执行
    - 如果是高可用集群，需要在所有 master 节点上执行
    - 续订操作不会影响正在运行的 pod 和服务
    claude code router service has been successfully stopped.