scripting

使用 Acquia CLI 编写脚本与自动化

适用于：

• 在 Shell 脚本中自动运行 acli 命令
• 将 acli 集成到 CI/CD 流水线中
• 以非交互方式运行 acli

在脚本、CI/CD 流水线和自动化工作流中使用 acli，以编程方式管理 Acquia Cloud 资源。

---

非交互式命令

所有 acli 命令都可以通过提供选项以非交互方式运行。这对于脚本和 CI/CD 至关重要。

基本模式

# Interactive - prompts for missing information
acli ide:create

# Non-interactive - provides all info upfront
acli ide:create \
  --application=abc123 \
  --label="Automated IDE"

常用选项

# No prompts - fails if info is missing
acli <command> --no-interaction

# Don't wait for async operations
acli <command> --no-wait

# Skip confirmations
acli <command> --yes

---

简单自动化示例

示例 1：按计划部署

#!/bin/bash
# deploy.sh - Deploy to staging nightly

BRANCH="develop"
ENVIRONMENT="staging"
ENV_ID="<environment-id>"

acli api:environments:code-switch $ENV_ID $BRANCH

# Verify Drupal status
acli remote:drush status

MEO（V3）订阅： 将 api:environments:code-switch 行替换为 acli api:v3:environments:create-deployment $ENV_ID true $BRANCH（参见 MEO Deployments）。同样，acli remote:drush cr 也有由 API 驱动的对应命令 acli api:v3:environments:clear-caches <environmentId> <domains> （参见 MEO Environments）。

通过 cron 运行：

# crontab -e
# Deploy every night at 2am
0 2 * * * /path/to/deploy.sh >> /tmp/deploy.log 2>&1

示例 2：清除缓存工作流

#!/bin/bash
# cache-clear.sh - Clear caches across environments

set -e  # Exit on any error

acli remote:drush cr
echo "Caches cleared!"

---

CI/CD 集成

GitHub Actions

在推送到 main 时部署：

# .github/workflows/deploy.yml
name: Deploy to Acquia

on:
  push:
    branches: [main]

jobs:
  deploy:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v3

      - name: Install acli
        run: |
          curl -fsSL https://github.com/acquia/cli/releases/latest/download/acli \
            -o /usr/local/bin/acli
          chmod +x /usr/local/bin/acli

      - name: Authenticate
        env:
          ACQUIA_KEY: ${{ secrets.ACQUIA_KEY }}
          ACQUIA_SECRET: ${{ secrets.ACQUIA_SECRET }}
        run: |
          mkdir -p ~/.acquia
          cat > ~/.acquia/credentials.json <<EOF
          {
            "acquia-cloud-api": {
              "key": "$ACQUIA_KEY",
              "secret": "$ACQUIA_SECRET"
            }
          }
          EOF

      - name: Deploy to production
        run: |
          acli api:environments:code-switch ${{ vars.ACQUIA_PROD_ENV_ID }} main

GitLab CI/CD

# .gitlab-ci.yml
deploy_production:
  stage: deploy
  image: ubuntu:latest
  before_script:
    - curl -fsSL https://github.com/acquia/cli/releases/latest/download/acli \
        -o /usr/local/bin/acli
    - chmod +x /usr/local/bin/acli
    - mkdir -p ~/.acquia
    - |
      cat > ~/.acquia/credentials.json <<EOF
      {
        "acquia-cloud-api": {
          "key": "$ACQUIA_KEY",
          "secret": "$ACQUIA_SECRET"
        }
      }
      EOF
  script:
    - acli api:environments:code-switch $ACQUIA_PROD_ENV_ID $CI_COMMIT_BRANCH
  only:
    - main

---

错误处理

检查退出代码

#!/bin/bash

acli ide:create --application=abc123

if [ $? -eq 0 ]; then
    echo "IDE created successfully"
else
    echo "IDE creation failed"
    exit 1
fi

妥善处理错误

#!/bin/bash
set -e  # Exit on first error

trap 'echo "Error occurred on line $LINENO"' ERR

acli api:environments:code-switch $ENV_ID main
echo "Deploy successful!"

重试逻辑

#!/bin/bash

retry_count=0
max_retries=3

while [ $retry_count -lt $max_retries ]; do
    if acli remote:drush updatedb; then
        echo "Database updates completed"
        exit 0
    fi
    retry_count=$((retry_count + 1))
    if [ $retry_count -lt $max_retries ]; then
        echo "Attempt $retry_count failed. Retrying..."
        sleep 10
    fi
done

echo "Failed after $max_retries attempts"
exit 1

---

日志记录与监控

记录输出

#!/bin/bash

LOGFILE="/var/log/acli-deploy.log"

{
    echo "=== Deploy started at $(date) ==="
    acli api:environments:code-switch $ENV_ID main
    echo "=== Deploy finished at $(date) ==="
} >> $LOGFILE 2>&1

发送通知

#!/bin/bash

SLACK_WEBHOOK="$SLACK_WEBHOOK_URL"

send_slack() {
    curl -X POST $SLACK_WEBHOOK \
        -H 'Content-Type: application/json' \
        -d "{
            \"text\": \"Acquia deployment: $1\",
            \"channel\": \"#deployments\"
        }"
}

if acli api:environments:code-switch $ENV_ID main; then
    send_slack "✓ Production deployment successful"
else
    send_slack "✗ Production deployment failed"
    exit 1
fi

---

进阶：API 密钥身份验证

在服务器上进行自动化时，请使用 API 密钥，而不是浏览器身份验证。

生成 API 凭据

在 Acquia Cloud UI 中：

1. 前往 Settings → API Tokens
2. 生成新令牌
3. 复制 Key 和 Secret

配合 acli 使用

# Method 1: Environment variables
export ACQUIA_KEY="your-key-here"
export ACQUIA_SECRET="your-secret-here"
acli api:applications:list

# Method 2: Credentials file
mkdir -p ~/.acquia
cat > ~/.acquia/credentials.json <<EOF
{
  "acquia-cloud-api": {
    "key": "your-key-here",
    "secret": "your-secret-here"
  }
}
EOF

acli api:applications:list

安全管理凭据

# Option 1: AWS Secrets Manager
aws secretsmanager get-secret-value --secret-id acquia-cli-key

# Option 2: HashiCorp Vault
vault kv get secret/acquia/cli-key

# Option 3: GitHub Secrets (for Actions)
# Set ACQUIA_KEY and ACQUIA_SECRET as repository secrets

---

最佳实践

1. 使用 --no-interaction — 在脚本中始终传入此选项，以防提示阻塞进程。
2. 使用 set -e — 遇到任何错误时立即退出，以免失败后仍悄然继续运行。
3. 安全存储凭据 — 使用环境变量或密钥管理器，切勿硬编码。
4. 使用 app:task-wait — 执行异步操作后，等待其完成再进行下一步。

---

故障排除

CI 中身份验证失败

设置 ACLI_NO_INTERACTION 环境变量并显式提供凭据：

export ACLI_NO_INTERACTION=1
export ACLI_KEY=your-api-key
export ACLI_SECRET=your-api-secret
acli api:applications:list

命令挂起并等待输入

在脚本中始终传入 --no-interaction：

acli ide:create --application=abc123 --label="CI IDE" --no-interaction

退出代码未正确传递

在脚本顶部使用 set -e：

#!/bin/bash
set -e
acli auth:login --no-interaction
acli api:environments:code-switch $ENV_ID main