Multi-Site Health Monitor
Overview
The Multi-Site Health Monitor skill automates continuous monitoring of 10-100+ websites with configurable health checks, intelligent alert routing, and automatic incident escalation. This production-grade monitoring solution integrates with Slack, PagerDuty, Datadog, Google Sheets, and WordPress to provide real-time visibility into your digital infrastructure.
Why This Matters
- - Prevent Revenue Loss: Detect downtime in seconds, not hours
- Reduce Alert Fatigue: Smart thresholds and deduplication prevent notification overload
- Automate Incident Response: Auto-restart failed services, escalate to on-call teams
- Multi-Channel Alerts: Route critical issues to PagerDuty, warnings to Slack, metrics to Datadog
- Historical Analysis: Track uptime trends, identify patterns, generate compliance reports
Key Integrations
- - Slack: Real-time alerts, incident channels, status dashboards
- PagerDuty: Automatic incident creation, on-call escalation, incident tracking
- Datadog: Metric ingestion, custom dashboards, anomaly detection
- Google Sheets: Automated reporting, SLA tracking, audit logs
- WordPress: Monitor plugin health, theme updates, core vulnerabilities
- AWS/Azure: Auto-restart EC2 instances, trigger Lambda functions, scale infrastructure
Quick Start
Try these example prompts immediately:
Example 1: Monitor 5 Critical Sites with Slack Alerts
CODEBLOCK0
Example 2: Auto-Restart Failed Services
CODEBLOCK1
Example 3: WordPress Multi-Site Monitoring
CODEBLOCK2
Example 4: Performance Threshold Monitoring
Monitor https://api.example.com/metrics every 10 minutes.
Alert if:
- Response time > 2000ms (warning) or > 5000ms (critical)
- Error rate > 1% (warning) or > 5% (critical)
- CPU usage > 70% (warning) or > 90% (critical)
- Memory usage > 80% (warning) or > 95% (critical)
Send metrics to Datadog with tags: env:prod, service:api, team:backend
Capabilities
1. Multi-Protocol Health Checks
Monitor endpoints via:
- - HTTP/HTTPS: GET, POST, HEAD requests with custom headers
- TCP: Port connectivity checks (e.g., database ports 3306, 5432)
- DNS: Domain resolution, DNS propagation verification
- SSL/TLS: Certificate validity, expiration warnings, chain verification
- Ping/ICMP: Basic connectivity for infrastructure nodes
Example: Monitor API health with custom authentication
CODEBLOCK4
2. Intelligent Alert Routing
- - Severity-Based Routing: Critical → PagerDuty + Slack + SMS, Warning → Slack only, Info → Sheets log
- Deduplication: Suppress duplicate alerts within 5-minute window
- Escalation Rules: Auto-escalate if critical issue unresolved for 30+ minutes
- Custom Thresholds: Define per-endpoint sensitivity (e.g., API endpoint stricter than blog)
- Quiet Hours: Suppress non-critical alerts during maintenance windows
3. Automatic Incident Response
- - Webhook Triggers: POST to custom endpoints (restart services, scale infrastructure)
- AWS Integration: Auto-restart EC2 instances, trigger Lambda functions
- Service Restart: Execute shell commands on remote servers via SSH
- Rollback Triggers: Revert deployments if health checks fail
- Notification Actions: Create tickets in Jira, GitHub Issues, or Linear
4. Performance Metrics & Trending
- - Response Time Tracking: Detect slowdowns before they become critical
- Uptime Calculation: Real-time SLA tracking (99.9%, 99.95%, 99.99%)
- Error Rate Monitoring: Track HTTP 4xx, 5xx, timeout errors
- Datadog Integration: Send custom metrics for dashboards and alerts
- Historical Reporting: Generate monthly uptime reports, SLA compliance docs
5. WordPress-Specific Monitoring
- - Core Updates: Alert when WordPress core updates available
- Plugin Vulnerabilities: Check against WordPress vulnerability database
- Theme Security: Monitor for outdated or vulnerable themes
- Database Health: Monitor wp_options, table integrity, query performance
- User Activity: Track suspicious login attempts, new admin accounts
- Backup Verification: Confirm backups complete successfully
6. Compliance & Audit Logging
- - Google Sheets Integration: Automatic logging of all checks, alerts, actions
- Audit Trail: Who triggered what, when, and what happened
- SLA Reports: Monthly/quarterly compliance reports (99.9% uptime proof)
- Change Tracking: Document all configuration changes with timestamps
- Export Formats: CSV, JSON, PDF for compliance submissions
Configuration
Required Environment Variables
CODEBLOCK5
Configuration File Format (YAML)
CODEBLOCK6
Setup Instructions
- 1. Create monitoring config: Save YAML above as INLINECODE0
- Set environment variables: Source
.env file with all required API keys - Initialize Google Sheets: Create spreadsheet, share with service account email
- Test endpoints: Run
multi-site-health-monitor --validate to verify all URLs respond - Deploy: Run as systemd service or Docker container for continuous monitoring
Example Outputs
Slack Alert (Critical)
CODEBLOCK7
Google Sheets Log Entry
CODEBLOCK8
PagerDuty Incident
CODEBLOCK9
Datadog Metrics Sent
multi_site_monitor.health_check.response_time:245ms (tags: service:api, env:prod)
multi_site_monitor.health_check.status:200 (tags: service:api, env:prod)
multi_site_monitor.health_check.availability:99.87 (tags: service:api, env:prod)
multi_site_monitor.auto_restart.attempts:1 (tags: service:api, env:prod)
Tips & Best Practices
1. Optimal Check Intervals
- - Critical APIs: 60-120 seconds (detects issues in 2-4 minutes)
- Standard Services: 300 seconds (5 minutes, good balance)
- Non-Critical Endpoints: 600-900 seconds (10-15 minutes, reduces noise)
- Batch Jobs: 1800+ seconds (30+ minutes, less frequent monitoring)
2. Threshold Tuning
- - Start Conservative: Begin with loose thresholds, tighten over 2 weeks
- Account for Variance: Set response time thresholds 2-3x slower than baseline
- Error Rate: 0.1-1% warning, 5%+ critical (adjust per service SLA)
- Test Thresholds: Deliberately fail endpoints to verify alert routing works
3. Reducing Alert Fatigue
- - Deduplication: Suppress identical alerts within 5-minute window
- Smart Escalation: Only escalate if issue persists >30 minutes
- Quiet Hours: Disable non-critical alerts 2am-6am (adjust per timezone)
- Severity Mapping: Not everything is critical; use warning/info for minor issues
4. WordPress-Specific Best Practices
- - Check Core Updates Weekly: Set interval to 604,800 seconds (7 days)
- Monitor Plugin Health Daily: Check for vulnerabilities, outdated plugins
- Database Backups: Verify backup completion status in health endpoint
- Staging Environment: Monitor staging sites separately to catch issues before production
- Custom Health Endpoints: Create
/wp-json/custom/health returning comprehensive data
5. Cost Optimization
- - Batch Checks: Group 5-10 checks into single HTTP request where possible
- Datadog Sampling: Send detailed metrics every 5 minutes, summary every hour
- Google Sheets: Batch writes (max 100 rows per request) to reduce API calls
- PagerDuty: Use deduplication to avoid triggering duplicate incidents
6. Security Hardening
- - API Key Rotation: Rotate all API keys monthly
- VPC Monitoring: Monitor internal endpoints from private subnets only
- IP Whitelisting: Restrict health check endpoints
多站点健康监控
概述
多站点健康监控技能可自动对10-100+个网站进行持续监控，支持可配置的健康检查、智能告警路由和自动事件升级。这套生产级监控解决方案集成了Slack、PagerDuty、Datadog、Google Sheets和WordPress，为您的数字基础设施提供实时可见性。
为何重要
- - 防止收入损失：在数秒内（而非数小时）检测到停机
- 减少告警疲劳：智能阈值和去重机制防止通知过载
- 自动化事件响应：自动重启失败服务，升级至值班团队
- 多渠道告警：将关键问题路由至PagerDuty，警告路由至Slack，指标路由至Datadog
- 历史分析：跟踪正常运行时间趋势，识别模式，生成合规报告
关键集成
- - Slack：实时告警、事件频道、状态仪表板
- PagerDuty：自动创建事件、值班升级、事件跟踪
- Datadog：指标采集、自定义仪表板、异常检测
- Google Sheets：自动报告、SLA跟踪、审计日志
- WordPress：监控插件健康、主题更新、核心漏洞
- AWS/Azure：自动重启EC2实例、触发Lambda函数、扩展基础设施
快速入门
立即尝试以下示例提示：
示例1：监控5个关键站点并发送Slack告警
每5分钟监控以下站点，如有失败则通过Slack告警：
- - https://api.example.com/health
- https://app.example.com/status
- https://cdn.example.com/ping
- https://wordpress.example.com/wp-json/health
- https://db.example.com/check
告警规则：
- - 严重（页面宕机）：Slack #incidents + PagerDuty
- 警告（响应慢 >3秒）：Slack #alerts
- 信息（证书过期 <30天）：Google Sheets日志
示例2：自动重启失败服务
每2分钟监控 https://payment-service.example.com/health。
如果连续失败3次：
- 1. POST请求至 https://restart-api.example.com/restart-payment-service
- 向PagerDuty发送事件Payment Service Down
- 通知Slack #critical-incidents
- 记录至Google Sheets，包含时间戳、错误详情、重启状态
响应超时：10秒
预期响应：HTTP 200，返回{status:healthy}
示例3：WordPress多站点监控
监控以下WordPress站点的健康与安全：
- - https://site1.example.com/wp-json/wp/v2/health-check
- https://site2.example.com/wp-json/wp/v2/health-check
- https://site3.example.com/wp-json/wp/v2/health-check
检查项：
- - 核心更新可用（超过1周未更新则警告）
- 插件漏洞（发现任何漏洞则严重）
- 数据库连接（断开则严重）
- SSL证书过期（少于30天则警告）
告警目标：
- - 严重：PagerDuty + Slack #wordpress-critical
- 警告：Slack #wordpress-alerts
- 信息：Google Sheets #monitoring-log
示例4：性能阈值监控
每10分钟监控 https://api.example.com/metrics。
在以下情况下告警：
- - 响应时间 > 2000ms（警告）或 > 5000ms（严重）
- 错误率 > 1%（警告）或 > 5%（严重）
- CPU使用率 > 70%（警告）或 > 90%（严重）
- 内存使用率 > 80%（警告）或 > 95%（严重）
向Datadog发送指标，标签：env:prod, service:api, team:backend
功能
1. 多协议健康检查
通过以下方式监控端点：
- - HTTP/HTTPS：GET、POST、HEAD请求，支持自定义头
- TCP：端口连通性检查（如数据库端口3306、5432）
- DNS：域名解析、DNS传播验证
- SSL/TLS：证书有效性、过期警告、链验证
- Ping/ICMP：基础设施节点的基本连通性
示例：使用自定义认证监控API健康
端点：https://api.example.com/health
方法：POST
头：
Authorization: Bearer YOURAPIKEY
User-Agent: MultiSiteMonitor/1.0.0
预期状态：200
预期响应体：{status:healthy,version:2.1.0}
超时：10秒
2. 智能告警路由
- - 基于严重级别的路由：严重 → PagerDuty + Slack + SMS，警告 → 仅Slack，信息 → Sheets日志
- 去重：在5分钟窗口内抑制重复告警
- 升级规则：如果严重问题30分钟以上未解决，自动升级
- 自定义阈值：为每个端点定义灵敏度（如API端点比博客更严格）
- 静默时段：在维护窗口期间抑制非关键告警
3. 自动事件响应
- - Webhook触发器：POST至自定义端点（重启服务、扩展基础设施）
- AWS集成：自动重启EC2实例、触发Lambda函数
- 服务重启：通过SSH在远程服务器上执行Shell命令
- 回滚触发器：如果健康检查失败，回滚部署
- 通知操作：在Jira、GitHub Issues或Linear中创建工单
4. 性能指标与趋势分析
- - 响应时间跟踪：在问题变得严重之前检测到性能下降
- 正常运行时间计算：实时SLA跟踪（99.9%、99.95%、99.99%）
- 错误率监控：跟踪HTTP 4xx、5xx、超时错误
- Datadog集成：为仪表板和告警发送自定义指标
- 历史报告：生成月度正常运行时间报告、SLA合规文档
5. WordPress专用监控
- - 核心更新：当WordPress核心更新可用时告警
- 插件漏洞：对照WordPress漏洞数据库检查
- 主题安全：监控过时或存在漏洞的主题
- 数据库健康：监控wp_options、表完整性、查询性能
- 用户活动：跟踪可疑登录尝试、新管理员账户
- 备份验证：确认备份成功完成
6. 合规与审计日志
- - Google Sheets集成：自动记录所有检查、告警、操作
- 审计追踪：谁在何时触发了什么操作，结果如何
- SLA报告：月度/季度合规报告（99.9%正常运行时间证明）
- 变更跟踪：记录所有配置变更及时间戳
- 导出格式：CSV、JSON、PDF，用于合规提交
配置
必需的环境变量
bash
Slack通知
export SLACK
WEBHOOKURL=https://hooks.slack.com/services/YOUR/WEBHOOK/URL
export SLACK_CHANNEL=#incidents # 或 #alerts、#monitoring等
PagerDuty事件创建
export PAGERDUTY
APIKEY=YOUR
PAGERDUTYAPI_KEY
export PAGERDUTY
SERVICEID=YOUR
SERVICEID
Datadog指标采集
export DATADOG
APIKEY=YOUR
DATADOGAPI_KEY
export DATADOG
APPKEY=YOUR
DATADOGAPP_KEY
export DATADOG_SITE=datadoghq.com # 或 datadoghq.eu
Google Sheets日志
export GOOGLE
SHEETSID=YOUR
SPREADSHEETID
export GOOGLE
SERVICEACCOUNT_JSON=/path/to/service-account.json
AWS自动重启（可选）
export AWS
ACCESSKEY
ID=YOURAWS_KEY
export AWS
SECRETACCESS
KEY=YOURAWS_SECRET
export AWS_REGION=us-east-1
SSH远程服务重启（可选）
export SSH
PRIVATEKEY=/path/to/private/key
export SSH_USER=deploy
配置文件格式（YAML）
yaml
monitors.yaml
monitors:
- name: 生产API
url: https://api.example.com/health
interval: 300 # 秒
timeout: 10
method: GET
expected_status: 200
expected
bodycontains: healthy
alert_rules:
critical:
- slack_channel: #critical-incidents
- pagerduty_severity: critical
warning:
- slack_channel: #alerts
auto_restart:
enabled: true
command: systemctl restart api-service
max_retries: 3
retry_delay: 60
