sfera-new/docs/infrastructure/MONITORING_SETUP.md

Настройка мониторинга и логирования SFERA

🎯 Обзор

Комплексная система мониторинга и логирования для платформы SFERA, включающая метрики производительности, логирование ошибок, алертинг и визуализацию данных для обеспечения надежности и производительности в production окружении.

📊 Архитектура мониторинга

Компоненты системы

graph TB
    A[SFERA App] --> B[Winston Logger]
    A --> C[Prometheus Metrics]
    A --> D[OpenTelemetry]

    B --> E[Log Files]
    B --> F[ELK Stack]

    C --> G[Grafana Dashboard]
    D --> H[Jaeger Tracing]

    I[Alertmanager] --> J[Slack/Email]
    G --> I

🚨 Логирование

1. Структурированное логирование с Winston

Установка зависимостей

npm install winston winston-daily-rotate-file
npm install --save-dev @types/winston

Конфигурация логгера

Создание src/lib/logger.ts:

import winston from 'winston'
import DailyRotateFile from 'winston-daily-rotate-file'

// Определение уровней логирования
const levels = {
  error: 0,
  warn: 1,
  info: 2,
  http: 3,
  verbose: 4,
  debug: 5,
  silly: 6,
}

// Цвета для консольного вывода
const colors = {
  error: 'red',
  warn: 'yellow',
  info: 'green',
  http: 'magenta',
  verbose: 'white',
  debug: 'cyan',
  silly: 'grey',
}

winston.addColors(colors)

// Формат для production
const productionFormat = winston.format.combine(
  winston.format.timestamp({ format: 'YYYY-MM-DD HH:mm:ss:ms' }),
  winston.format.errors({ stack: true }),
  winston.format.json(),
)

// Формат для разработки
const developmentFormat = winston.format.combine(
  winston.format.timestamp({ format: 'YYYY-MM-DD HH:mm:ss:ms' }),
  winston.format.colorize({ all: true }),
  winston.format.printf(
    (info) => `${info.timestamp} ${info.level}: ${info.message}${info.stack ? '\n' + info.stack : ''}`,
  ),
)

// Транспорты для production
const productionTransports: winston.transport[] = [
  // Консольный вывод
  new winston.transports.Console({
    level: 'info',
    format: productionFormat,
  }),

  // Ротация логов по дням - общие логи
  new DailyRotateFile({
    filename: 'logs/application-%DATE%.log',
    datePattern: 'YYYY-MM-DD',
    zippedArchive: true,
    maxSize: '20m',
    maxFiles: '14d',
    level: 'info',
    format: productionFormat,
  }),

  // Отдельный файл для ошибок
  new DailyRotateFile({
    filename: 'logs/error-%DATE%.log',
    datePattern: 'YYYY-MM-DD',
    zippedArchive: true,
    maxSize: '20m',
    maxFiles: '30d',
    level: 'error',
    format: productionFormat,
  }),

  // HTTP запросы
  new DailyRotateFile({
    filename: 'logs/http-%DATE%.log',
    datePattern: 'YYYY-MM-DD',
    zippedArchive: true,
    maxSize: '20m',
    maxFiles: '7d',
    level: 'http',
    format: productionFormat,
  }),
]

// Транспорты для разработки
const developmentTransports: winston.transport[] = [
  new winston.transports.Console({
    level: 'debug',
    format: developmentFormat,
  }),
]

// Создание логгера
export const logger = winston.createLogger({
  level: process.env.NODE_ENV === 'production' ? 'info' : 'debug',
  levels,
  format: process.env.NODE_ENV === 'production' ? productionFormat : developmentFormat,
  transports: process.env.NODE_ENV === 'production' ? productionTransports : developmentTransports,
  exitOnError: false,
})

// Middleware для Express/Next.js
export const loggerMiddleware = (req: any, res: any, next: any) => {
  const start = Date.now()

  res.on('finish', () => {
    const duration = Date.now() - start
    logger.http('HTTP Request', {
      method: req.method,
      url: req.url,
      status: res.statusCode,
      duration: `${duration}ms`,
      userAgent: req.get('User-Agent'),
      ip: req.ip,
    })
  })

  next()
}

// Утилиты для логирования
export const logError = (error: Error, context?: object) => {
  logger.error('Application Error', {
    message: error.message,
    stack: error.stack,
    ...context,
  })
}

export const logInfo = (message: string, meta?: object) => {
  logger.info(message, meta)
}

export const logWarn = (message: string, meta?: object) => {
  logger.warn(message, meta)
}

export const logDebug = (message: string, meta?: object) => {
  logger.debug(message, meta)
}

2. Интеграция с Next.js API

API Routes логирование

// src/app/api/graphql/route.ts
import { logger } from '@/lib/logger'

export async function POST(request: Request) {
  const startTime = Date.now()

  try {
    logger.info('GraphQL Request Started')

    // Основная логика GraphQL
    const result = await handleGraphQLRequest(request)

    logger.info('GraphQL Request Completed', {
      duration: Date.now() - startTime,
      success: true,
    })

    return result
  } catch (error) {
    logger.error('GraphQL Request Failed', {
      duration: Date.now() - startTime,
      error: error.message,
      stack: error.stack,
    })

    throw error
  }
}

GraphQL Resolvers логирование

// src/graphql/resolvers.ts
import { logger } from '@/lib/logger'

export const resolvers = {
  Query: {
    getUser: async (parent: any, args: any, context: any) => {
      const { userId } = args

      logger.info('Getting user', { userId, requestId: context.requestId })

      try {
        const user = await prisma.user.findUnique({
          where: { id: userId },
        })

        logger.info('User retrieved successfully', { userId })
        return user
      } catch (error) {
        logger.error('Failed to get user', {
          userId,
          error: error.message,
        })
        throw error
      }
    },
  },
}

📈 Метрики и мониторинг

1. Prometheus метрики

Установка зависимостей

npm install prom-client
npm install --save-dev @types/prom-client

Настройка метрик

Создание src/lib/metrics.ts:

import promClient from 'prom-client'

// Создание реестра метрик
export const register = new promClient.Registry()

// Добавление стандартных метрик
promClient.collectDefaultMetrics({
  register,
  prefix: 'sfera_',
})

// HTTP запросы
export const httpRequestsTotal = new promClient.Counter({
  name: 'sfera_http_requests_total',
  help: 'Total number of HTTP requests',
  labelNames: ['method', 'route', 'status'],
  registers: [register],
})

export const httpRequestDuration = new promClient.Histogram({
  name: 'sfera_http_request_duration_seconds',
  help: 'Duration of HTTP requests in seconds',
  labelNames: ['method', 'route', 'status'],
  buckets: [0.005, 0.01, 0.025, 0.05, 0.1, 0.25, 0.5, 1, 2.5, 5, 10],
  registers: [register],
})

// GraphQL метрики
export const graphqlOperationsTotal = new promClient.Counter({
  name: 'sfera_graphql_operations_total',
  help: 'Total number of GraphQL operations',
  labelNames: ['operation_name', 'operation_type', 'success'],
  registers: [register],
})

export const graphqlOperationDuration = new promClient.Histogram({
  name: 'sfera_graphql_operation_duration_seconds',
  help: 'Duration of GraphQL operations in seconds',
  labelNames: ['operation_name', 'operation_type'],
  buckets: [0.001, 0.005, 0.01, 0.025, 0.05, 0.1, 0.25, 0.5, 1, 2.5, 5],
  registers: [register],
})

// База данных
export const databaseConnectionsActive = new promClient.Gauge({
  name: 'sfera_database_connections_active',
  help: 'Number of active database connections',
  registers: [register],
})

export const databaseQueryDuration = new promClient.Histogram({
  name: 'sfera_database_query_duration_seconds',
  help: 'Duration of database queries in seconds',
  labelNames: ['query_type'],
  buckets: [0.001, 0.005, 0.01, 0.025, 0.05, 0.1, 0.25, 0.5, 1],
  registers: [register],
})

// Бизнес метрики
export const usersOnline = new promClient.Gauge({
  name: 'sfera_users_online',
  help: 'Number of users currently online',
  registers: [register],
})

export const ordersTotal = new promClient.Counter({
  name: 'sfera_orders_total',
  help: 'Total number of orders created',
  labelNames: ['organization_type', 'status'],
  registers: [register],
})

export const messagesTotal = new promClient.Counter({
  name: 'sfera_messages_total',
  help: 'Total number of messages sent',
  labelNames: ['message_type'],
  registers: [register],
})

// Redis/кэш метрики
export const cacheHitsTotal = new promClient.Counter({
  name: 'sfera_cache_hits_total',
  help: 'Total number of cache hits',
  labelNames: ['cache_key_pattern'],
  registers: [register],
})

export const cacheMissesTotal = new promClient.Counter({
  name: 'sfera_cache_misses_total',
  help: 'Total number of cache misses',
  labelNames: ['cache_key_pattern'],
  registers: [register],
})

// Middleware для сбора HTTP метрик
export const metricsMiddleware = (req: any, res: any, next: any) => {
  const start = Date.now()

  res.on('finish', () => {
    const duration = (Date.now() - start) / 1000
    const route = req.route?.path || req.path

    httpRequestsTotal.labels(req.method, route, res.statusCode.toString()).inc()

    httpRequestDuration.labels(req.method, route, res.statusCode.toString()).observe(duration)
  })

  next()
}

API endpoint для метрик

// src/app/api/metrics/route.ts
import { NextResponse } from 'next/server'
import { register } from '@/lib/metrics'

export async function GET() {
  try {
    const metrics = await register.metrics()

    return new NextResponse(metrics, {
      headers: {
        'Content-Type': register.contentType,
      },
    })
  } catch (error) {
    return NextResponse.json({ error: 'Failed to generate metrics' }, { status: 500 })
  }
}

2. OpenTelemetry трассировка

Установка зависимостей

npm install @opentelemetry/api @opentelemetry/sdk-node
npm install @opentelemetry/instrumentation-http
npm install @opentelemetry/instrumentation-graphql
npm install @opentelemetry/exporter-jaeger

Конфигурация трассировки

Создание src/lib/tracing.ts:

import { NodeSDK } from '@opentelemetry/sdk-node'
import { HttpInstrumentation } from '@opentelemetry/instrumentation-http'
import { GraphQLInstrumentation } from '@opentelemetry/instrumentation-graphql'
import { JaegerExporter } from '@opentelemetry/exporter-jaeger'
import { Resource } from '@opentelemetry/resources'
import { SemanticResourceAttributes } from '@opentelemetry/semantic-conventions'

// Настройка экспортера для Jaeger
const jaegerExporter = new JaegerExporter({
  endpoint: process.env.JAEGER_ENDPOINT || 'http://localhost:14268/api/traces',
})

// Настройка SDK
const sdk = new NodeSDK({
  resource: new Resource({
    [SemanticResourceAttributes.SERVICE_NAME]: 'sfera-app',
    [SemanticResourceAttributes.SERVICE_VERSION]: '1.0.0',
  }),
  traceExporter: jaegerExporter,
  instrumentations: [
    new HttpInstrumentation({
      applyCustomAttributesOnSpan: (span, request, response) => {
        span.setAttributes({
          'http.request.body.size': request.headers['content-length'] || 0,
          'http.response.body.size': response.getHeader('content-length') || 0,
        })
      },
    }),
    new GraphQLInstrumentation({
      mergeItems: true,
      allowValues: true,
    }),
  ],
})

// Инициализация трассировки
if (process.env.NODE_ENV === 'production') {
  sdk.start()
  console.log('Tracing started successfully')
}

export { sdk }

📱 Dashboard и визуализация

1. Grafana Dashboard конфигурация

Docker Compose для мониторинга стека

Создание docker-compose.monitoring.yml:

version: '3.8'

services:
  prometheus:
    image: prom/prometheus:latest
    container_name: sfera-prometheus
    ports:
      - '9090:9090'
    volumes:
      - ./monitoring/prometheus.yml:/etc/prometheus/prometheus.yml
      - prometheus_data:/prometheus
    command:
      - '--config.file=/etc/prometheus/prometheus.yml'
      - '--storage.tsdb.path=/prometheus'
      - '--web.console.libraries=/etc/prometheus/console_libraries'
      - '--web.console.templates=/etc/prometheus/consoles'
      - '--web.enable-lifecycle'
    restart: unless-stopped

  grafana:
    image: grafana/grafana:latest
    container_name: sfera-grafana
    ports:
      - '3001:3000'
    volumes:
      - grafana_data:/var/lib/grafana
      - ./monitoring/grafana/provisioning:/etc/grafana/provisioning
      - ./monitoring/grafana/dashboards:/etc/grafana/dashboards
    environment:
      - GF_SECURITY_ADMIN_PASSWORD=admin123
      - GF_INSTALL_PLUGINS=grafana-piechart-panel
    restart: unless-stopped

  jaeger:
    image: jaegertracing/all-in-one:latest
    container_name: sfera-jaeger
    ports:
      - '16686:16686'
      - '14268:14268'
    environment:
      - COLLECTOR_OTLP_ENABLED=true
    restart: unless-stopped

  alertmanager:
    image: prom/alertmanager:latest
    container_name: sfera-alertmanager
    ports:
      - '9093:9093'
    volumes:
      - ./monitoring/alertmanager.yml:/etc/alertmanager/alertmanager.yml
    restart: unless-stopped

volumes:
  prometheus_data:
  grafana_data:

Prometheus конфигурация

Создание monitoring/prometheus.yml:

global:
  scrape_interval: 15s
  evaluation_interval: 15s

rule_files:
  - 'rules/*.yml'

alerting:
  alertmanagers:
    - static_configs:
        - targets:
            - alertmanager:9093

scrape_configs:
  - job_name: 'sfera-app'
    static_configs:
      - targets: ['host.docker.internal:3000']
    metrics_path: '/api/metrics'
    scrape_interval: 30s

  - job_name: 'prometheus'
    static_configs:
      - targets: ['localhost:9090']

  - job_name: 'node-exporter'
    static_configs:
      - targets: ['node-exporter:9100']

Grafana Dashboard JSON

Создание monitoring/grafana/dashboards/sfera-dashboard.json:

{
  "dashboard": {
    "id": null,
    "title": "SFERA Application Dashboard",
    "tags": ["sfera"],
    "timezone": "browser",
    "panels": [
      {
        "id": 1,
        "title": "HTTP Request Rate",
        "type": "graph",
        "targets": [
          {
            "expr": "rate(sfera_http_requests_total[5m])",
            "legendFormat": "{{method}} {{route}}"
          }
        ],
        "yAxes": [
          {
            "label": "Requests/sec",
            "min": 0
          }
        ]
      },
      {
        "id": 2,
        "title": "Response Time",
        "type": "graph",
        "targets": [
          {
            "expr": "histogram_quantile(0.95, rate(sfera_http_request_duration_seconds_bucket[5m]))",
            "legendFormat": "95th percentile"
          },
          {
            "expr": "histogram_quantile(0.50, rate(sfera_http_request_duration_seconds_bucket[5m]))",
            "legendFormat": "50th percentile"
          }
        ]
      },
      {
        "id": 3,
        "title": "GraphQL Operations",
        "type": "graph",
        "targets": [
          {
            "expr": "rate(sfera_graphql_operations_total[5m])",
            "legendFormat": "{{operation_name}} ({{operation_type}})"
          }
        ]
      },
      {
        "id": 4,
        "title": "Database Connections",
        "type": "singlestat",
        "targets": [
          {
            "expr": "sfera_database_connections_active",
            "legendFormat": "Active Connections"
          }
        ]
      },
      {
        "id": 5,
        "title": "Error Rate",
        "type": "graph",
        "targets": [
          {
            "expr": "rate(sfera_http_requests_total{status=~\"5..\"}[5m])",
            "legendFormat": "5xx Errors"
          }
        ]
      },
      {
        "id": 6,
        "title": "Orders Created",
        "type": "graph",
        "targets": [
          {
            "expr": "rate(sfera_orders_total[5m])",
            "legendFormat": "{{organization_type}}"
          }
        ]
      }
    ],
    "time": {
      "from": "now-1h",
      "to": "now"
    },
    "refresh": "30s"
  }
}

2. Alerting правила

Prometheus правила алертинга

Создание monitoring/rules/alerts.yml:

groups:
  - name: sfera.alerts
    rules:
      # Высокий уровень ошибок
      - alert: HighErrorRate
        expr: rate(sfera_http_requests_total{status=~"5.."}[5m]) > 0.1
        for: 2m
        labels:
          severity: critical
        annotations:
          summary: 'High error rate detected'
          description: 'Error rate is {{ $value }} requests/sec'

      # Медленные ответы
      - alert: HighResponseTime
        expr: histogram_quantile(0.95, rate(sfera_http_request_duration_seconds_bucket[5m])) > 1
        for: 5m
        labels:
          severity: warning
        annotations:
          summary: 'High response time detected'
          description: '95th percentile response time is {{ $value }}s'

      # Падение приложения
      - alert: ApplicationDown
        expr: up{job="sfera-app"} == 0
        for: 1m
        labels:
          severity: critical
        annotations:
          summary: 'Application is down'
          description: 'SFERA application is not responding'

      # Много активных подключений к БД
      - alert: HighDatabaseConnections
        expr: sfera_database_connections_active > 50
        for: 5m
        labels:
          severity: warning
        annotations:
          summary: 'High number of database connections'
          description: '{{ $value }} active database connections'

      # Мало дискового пространства
      - alert: DiskSpaceLow
        expr: (node_filesystem_avail_bytes / node_filesystem_size_bytes) * 100 < 10
        for: 5m
        labels:
          severity: critical
        annotations:
          summary: 'Disk space is low'
          description: 'Only {{ $value }}% disk space remaining'

      # Высокое использование памяти
      - alert: HighMemoryUsage
        expr: (1 - (node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes)) * 100 > 90
        for: 5m
        labels:
          severity: warning
        annotations:
          summary: 'High memory usage'
          description: 'Memory usage is {{ $value }}%'

Alertmanager конфигурация

Создание monitoring/alertmanager.yml:

global:
  smtp_smarthost: 'localhost:587'
  smtp_from: 'alerts@sfera.com'

route:
  group_by: ['alertname']
  group_wait: 10s
  group_interval: 10s
  repeat_interval: 1h
  receiver: 'web.hook'

receivers:
  - name: 'web.hook'
    slack_configs:
      - api_url: 'YOUR_SLACK_WEBHOOK_URL'
        channel: '#alerts'
        title: 'SFERA Alert'
        text: '{{ range .Alerts }}{{ .Annotations.summary }}: {{ .Annotations.description }}{{ end }}'

    email_configs:
      - to: 'admin@sfera.com'
        subject: 'SFERA Alert: {{ .GroupLabels.alertname }}'
        body: |
          {{ range .Alerts }}
          Alert: {{ .Annotations.summary }}
          Description: {{ .Annotations.description }}
          {{ end }}

inhibit_rules:
  - source_match:
      severity: 'critical'
    target_match:
      severity: 'warning'
    equal: ['alertname', 'dev', 'instance']

🔧 Практические примеры использования

1. Логирование в компонентах

// src/components/orders/order-processing.tsx
import { logger } from '@/lib/logger'
import { ordersTotal } from '@/lib/metrics'

export function OrderProcessor({ orderId }: { orderId: string }) {
  const processOrder = async () => {
    logger.info('Starting order processing', { orderId })

    try {
      const result = await processOrderLogic(orderId)

      // Инкремент метрики
      ordersTotal.labels('SELLER', 'completed').inc()

      logger.info('Order processed successfully', {
        orderId,
        processingTime: result.processingTime
      })

      return result
    } catch (error) {
      logger.error('Order processing failed', {
        orderId,
        error: error.message,
        stack: error.stack
      })

      ordersTotal.labels('SELLER', 'failed').inc()
      throw error
    }
  }

  return (
    <button onClick={processOrder}>
      Process Order
    </button>
  )
}

2. Мониторинг GraphQL запросов

// src/lib/graphql-monitoring.ts
import { graphqlOperationsTotal, graphqlOperationDuration } from '@/lib/metrics'
import { logger } from '@/lib/logger'

export const graphqlMiddleware = {
  requestDidStart() {
    return {
      didResolveOperation(requestContext: any) {
        const { operationName, operation } = requestContext.request
        logger.info('GraphQL operation started', {
          operationName,
          operationType: operation.operation,
        })
      },

      willSendResponse(requestContext: any) {
        const { operationName, operation } = requestContext.request
        const { errors } = requestContext.response
        const success = !errors || errors.length === 0

        graphqlOperationsTotal.labels(operationName || 'unknown', operation.operation, success.toString()).inc()

        if (errors) {
          logger.error('GraphQL operation failed', {
            operationName,
            errors: errors.map((e) => e.message),
          })
        }
      },
    }
  },
}

3. Мониторинг бизнес-метрик

// src/hooks/useRealtime.ts
import { usersOnline, messagesTotal } from '@/lib/metrics'
import { logger } from '@/lib/logger'

export const useRealtime = ({ onEvent }: { onEvent: (event: any) => void }) => {
  useEffect(() => {
    const socket = io()

    socket.on('connect', () => {
      usersOnline.inc()
      logger.info('User connected to realtime', { userId: socket.id })
    })

    socket.on('disconnect', () => {
      usersOnline.dec()
      logger.info('User disconnected from realtime', { userId: socket.id })
    })

    socket.on('message:new', (message) => {
      messagesTotal.labels(message.type).inc()
      logger.info('New message received', {
        messageId: message.id,
        conversationId: message.conversationId,
      })
      onEvent({ type: 'message:new', data: message })
    })

    return () => socket.disconnect()
  }, [onEvent])
}

🚀 Запуск мониторинга

1. Локальная среда

# Запуск стека мониторинга
docker-compose -f docker-compose.monitoring.yml up -d

# Доступ к сервисам
# Prometheus: http://localhost:9090
# Grafana: http://localhost:3001 (admin/admin123)
# Jaeger: http://localhost:16686
# Alertmanager: http://localhost:9093

2. Production среда

# Создание необходимых директорий
mkdir -p monitoring/{grafana/provisioning,rules}
mkdir -p logs

# Установка прав доступа
chmod -R 755 monitoring/
chmod -R 777 logs/

# Запуск с production конфигурацией
docker-compose -f docker-compose.yml -f docker-compose.monitoring.yml up -d

🎯 Заключение

Система мониторинга и логирования SFERA обеспечивает:

1. Полную видимость: Метрики, логи, трассировка
2. Проактивный мониторинг: Алерты и уведомления
3. Производительность: Мониторинг производительности в реальном времени
4. Отладка: Детализированное логирование и трассировка
5. Бизнес-аналитика: Метрики по заказам, пользователям, сообщениям

Эта система гарантирует надежность и высокую производительность платформы SFERA в production окружении.