Git钩子脚本如何检查告警策略

wen 实用脚本 2026-07-15 1

本文目录导读：

Git钩子脚本如何检查告警策略

准备工作
检查告警策略的示例脚本
特定运维场景的检查脚本
集成测试框架
最佳实践建议
注意事项

Git 钩子脚本可以通过在代码提交或合并前运行自定义检查来监控告警策略相关的配置,以下是几种常见的实现方式：

准备工作

创建钩子脚本

# 进入项目仓库
cd your-repo
# 创建 pre-commit 钩子（或其他钩子）
vim .git/hooks/pre-commit
chmod +x .git/hooks/pre-commit

检查告警策略的示例脚本

#!/bin/bash
set -e
# 告警策略检查脚本
echo "🔍 正在检查告警策略配置..."
# 定义检查规则
ERROR_COUNT=0
WARNING_COUNT=0
# 1. 检查 YAML/JSON 格式的告警策略文件
check_alerting_files() {
    local files=$(git diff --cached --name-only --diff-filter=ACM | grep -E '\.(yaml|yml|json)$' || true)
    for file in $files; do
        # 检查文件语法
        if [[ $file == *.yaml ]] || [[ $file == *.yml ]]; then
            if ! python3 -c "import yaml; yaml.safe_load(open('$file'))" 2>/dev/null; then
                echo "❌ YAML语法错误: $file"
                ((ERROR_COUNT++))
            fi
        elif [[ $file == *.json ]]; then
            if ! python3 -c "import json; json.load(open('$file'))" 2>/dev/null; then
                echo "❌ JSON语法错误: $file"
                ((ERROR_COUNT++))
            fi
        fi
    done
}
# 2. 检查告警阈值配置
check_alert_thresholds() {
    local alert_files=$(git diff --cached --name-only --diff-filter=ACM | grep -E '(alert|alarm|threshold)' || true)
    for file in $alert_files; do
        if [[ -f "$file" ]]; then
            # 检查阈值是否在合理范围内
            local threshold=$(grep -oP '"threshold":\s*\K[0-9.]+' "$file" || true)
            if [[ -n "$threshold" ]]; then
                if (( $(echo "$threshold > 10000" | bc -l) )); then
                    echo "⚠️  警告: $file 中的告警阈值过高: $threshold"
                    ((WARNING_COUNT++))
                fi
            fi
            # 检查是否配置了通知渠道
            if ! grep -q '"notify\|notification\|webhook' "$file" 2>/dev/null; then
                echo "⚠️  警告: $file 缺少通知渠道配置"
                ((WARNING_COUNT++))
            fi
        fi
    done
}
# 3. 检查是否引用了不存在的服务或指标
check_service_references() {
    local changed_files=$(git diff --cached --name-only --diff-filter=ACM)
    for file in $changed_files; do
        if grep -qP '(service|metric)_(name|id):' "$file" 2>/dev/null; then
            local refs=$(grep -oP '(service|metric)_(name|id):\s*\K\w+' "$file")
            for ref in $refs; do
                # 检查引用的服务或指标是否存在于配置文件中
                if ! grep -q "$ref" "config/services.yaml" 2>/dev/null && \
                   ! grep -q "$ref" "config/metrics.yaml" 2>/dev/null; then
                    if [[ "$file" != "config/services.yaml" ]] && \
                       [[ "$file" != "config/metrics.yaml" ]]; then
                        echo "⚠️  警告: $file 引用了不存在的服务/指标: $ref"
                        ((WARNING_COUNT++))
                    fi
                fi
            done
        fi
    done
}
# 4. 检查重复的告警规则
check_duplicate_rules() {
    local files=$(git diff --cached --name-only --diff-filter=ACM | grep -E '(alert|rule)' || true)
    for file in $files; do
        if [[ -f "$file" ]]; then
            # 检查是否有重复的告警名称
            local rule_names=$(grep -oP '(alert|rule)[_ ]?name:\s*\K\w+' "$file" 2>/dev/null)
            local duplicates=$(echo "$rule_names" | sort | uniq -d)
            if [[ -n "$duplicates" ]]; then
                echo "❌ 发现重复的告警规则名称:"
                echo "$duplicates" | while read -r dup; do
                    echo "   - $file 中的 '$dup'"
                    ((ERROR_COUNT++))
                done
            fi
        fi
    done
}
# 5. 执行所有检查
check_alerting_files
check_alert_thresholds
check_service_references
check_duplicate_rules
# 输出检查结果
echo ""
if [[ $ERROR_COUNT -gt 0 ]] || [[ $WARNING_COUNT -gt 0 ]]; then
    echo "📊 检查结果:"
    echo "   - 错误: $ERROR_COUNT"
    echo "   - 警告: $WARNING_COUNT"
    if [[ $ERROR_COUNT -gt 0 ]]; then
        echo "❌ 存在严重错误，提交被拒绝"
        exit 1
    fi
fi
echo "✅ 告警策略检查通过"
exit 0

特定运维场景的检查脚本

Prometheus Alertmanager 配置检查

#!/bin/bash
# pre-commit 钩子 - 检查 Prometheus 告警规则
echo "检查 Prometheus 告警规则..."
# 使用 promtool 验证
for file in $(git diff --cached --name-only | grep 'rules/.*\.yml'); do
    if ! promtool check rules "$file" 2>/dev/null; then
        echo "❌ Prometheus 规则语法错误: $file"
        exit 1
    fi
    # 检查告警表达式语法
    if ! promtool test rules "$file" 2>/dev/null; then
        echo "❌ Prometheus 规则测试失败: $file"
        exit 1
    fi
done

云平台告警策略检查

#!/bin/bash
# 检查云平台告警策略
check_cloud_alerts() {
    local files=$(git diff --cached --name-only | grep -E '(alarm|alert|monitor)')
    for file in $files; do
        echo "检查: $file"
        # 检查必要的字段
        required_fields=("metric" "threshold" "period" "evaluation_count")
        for field in "${required_fields[@]}"; do
            if ! grep -q "\"$field\":" "$file" 2>/dev/null && \
               ! grep -q "$field:" "$file" 2>/dev/null; then
                echo "⚠️  缺少必需字段: $field"
            fi
        done
        # 检查 Actions 配置
        if grep -q '"actions"' "$file" 2>/dev/null; then
            if ! grep -q '"notify"' "$file" 2>/dev/null; then
                echo "⚠️  告警未配置通知方式"
            fi
        fi
    done
}

集成测试框架

#!/bin/bash
# pre-commit 钩子 - 运行告警策略测试
# 运行专门的测试脚本
if [[ -f "tests/test_alerts.py" ]]; then
    python3 -m pytest tests/test_alerts.py -v --tb=short
    if [[ $? -ne 0 ]]; then
        echo "❌ 告警策略测试失败"
        exit 1
    fi
fi
# 运行告警模拟测试
if command -v alert_simulator &> /dev/null; then
    for file in $(git diff --cached --name-only | grep -E 'alert|alarm|rule'); do
        if ! alert_simulator simulate "$file"; then
            echo "❌ 告警模拟测试失败: $file"
            exit 1
        fi
    done
fi

最佳实践建议

配置管理

# 检查配置版本兼容性
check_version_compatibility() {
    local files=$(git diff --cached --name-only | grep 'alert')
    for file in $files; do
        local api_version=$(grep -oP 'apiVersion:\s*\K\w+' "$file" 2>/dev/null)
        if [[ -n "$api_version" ]] && [[ "$api_version" != "v1" ]] && [[ "$api_version" != "v2" ]]; then
            echo "⚠️  不支持的 API 版本: $api_version"
        fi
    done
}