Monitoring and Observability

Monitoring and Observability

Alita Robot provides comprehensive monitoring capabilities including health checks, Prometheus metrics, and resource monitoring.

Health Endpoint

The health endpoint provides real-time status of the bot and its dependencies.

Endpoint

GET /health

Response Format

{
  "status": "healthy",
  "checks": {
    "database": true,
    "redis": true
  },
  "version": "1.0.0",
  "uptime": "24h30m15s"
}

Response Fields

Field	Type	Description
status	string	Overall health: healthy or unhealthy
checks.database	boolean	PostgreSQL connection status
checks.redis	boolean	Redis connection status
version	string	Bot version
uptime	string	Time since bot started

HTTP Status Codes

Code	Meaning
200	All systems healthy
503	One or more checks failed

Health Check Logic

The health check performs:

1. Database check: Pings PostgreSQL with a 2-second timeout
2. Redis check: Sets and gets a test key with a 2-second timeout

Both checks must pass for healthy status.

Usage Examples

# Simple health check
curl http://localhost:8080/health

# Docker health check (built-in)
/alita_robot --health

# Kubernetes liveness probe
livenessProbe:
  httpGet:
    path: /health
    port: 8080
  initialDelaySeconds: 30
  periodSeconds: 10

Prometheus Metrics

The metrics endpoint exposes Prometheus-compatible metrics for monitoring.

Endpoint

GET /metrics

Prometheus Scrape Configuration

Add to your prometheus.yml:

scrape_configs:
  - job_name: 'alita-robot'
    static_configs:
      - targets: ['alita:8080']
    scrape_interval: 15s
    scrape_timeout: 10s
    metrics_path: /metrics

Docker Compose with Prometheus

services:
  prometheus:
    image: prom/prometheus:latest
    volumes:
      - ./prometheus.yml:/etc/prometheus/prometheus.yml
      - prometheus_data:/prometheus
    ports:
      - "9090:9090"
    depends_on:
      - alita

volumes:
  prometheus_data:

Grafana Dashboard

For visualization, add Grafana:

services:
  grafana:
    image: grafana/grafana:latest
    volumes:
      - grafana_data:/var/lib/grafana
    ports:
      - "3000:3000"
    environment:
      - GF_SECURITY_ADMIN_PASSWORD=admin
    depends_on:
      - prometheus

volumes:
  grafana_data:

Resource Monitoring

Alita Robot includes automatic resource monitoring to prevent resource exhaustion.

Configuration

# Maximum goroutines before triggering cleanup
# Default: 1000, Recommended: 1000-5000
RESOURCE_MAX_GOROUTINES=1000

# Maximum memory usage in MB before triggering cleanup
# Default: 500, Recommended: 500-2000
RESOURCE_MAX_MEMORY_MB=500

# Memory threshold for triggering garbage collection
# Default: 400, Recommended: 80% of RESOURCE_MAX_MEMORY_MB
RESOURCE_GC_THRESHOLD_MB=400

Auto-Remediation

When thresholds are exceeded, the system automatically:

1. Triggers garbage collection
2. Logs warnings about resource usage
3. Takes corrective action if configured

Activity Monitoring

Track group activity automatically:

# Days of inactivity before marking chat as inactive
# Default: 30, Range: 1-365
INACTIVITY_THRESHOLD_DAYS=30

# Hours between automatic activity checks
# Default: 1, Range: 1-24
ACTIVITY_CHECK_INTERVAL=1

# Enable automatic cleanup of inactive chats
# Default: true
ENABLE_AUTO_CLEANUP=true

Activity Metrics

The system tracks:

• DAG: Daily Active Groups
• WAG: Weekly Active Groups
• MAG: Monthly Active Groups

Groups are automatically:

• Marked inactive after the threshold period
• Reactivated when they become active again

Performance Monitoring

Enable Performance Tracking

# Enable automatic performance monitoring
ENABLE_PERFORMANCE_MONITORING=true

# Enable background statistics collection
ENABLE_BACKGROUND_STATS=true

Collected Metrics

• Message processing time
• Database query duration
• Cache hit/miss rates
• API response times
• Goroutine count
• Memory usage

Debug Mode

For detailed logging during development:

DEBUG=true

Debug mode:

• Increases log verbosity
• Disables background monitoring
• Shows detailed error stack traces

Log Analysis

Log Fields

Structured log entries include:

Field	Description
update_id	Telegram update ID
error_type	Error type (e.g., *TelegramError)
file	Source file where error occurred
line	Line number
function	Function name

Example Log Entry

{
  "level": "error",
  "msg": "Handler error occurred: user blocked bot",
  "update_id": 123456789,
  "error_type": "*gotgbot.TelegramError",
  "file": "alita/modules/admin.go",
  "line": 45,
  "function": "handleAdminCommand",
  "time": "2024-03-15T10:30:00Z"
}

Log Levels

Level	Description
DEBUG	Verbose debugging (DEBUG=true only)
INFO	Normal operation events
WARN	Expected issues (e.g., user blocked bot)
ERROR	Unexpected errors
FATAL	Critical errors that stop the bot
PANIC	Unrecoverable errors

Alerting

Prometheus Alerting Rules

Create alerts.yml:

groups:
  - name: alita-alerts
    rules:
      - alert: AlitaUnhealthy
        expr: up{job="alita-robot"} == 0
        for: 1m
        labels:
          severity: critical
        annotations:
          summary: "Alita Robot is down"

      - alert: HighMemoryUsage
        expr: process_resident_memory_bytes > 500000000
        for: 5m
        labels:
          severity: warning
        annotations:
          summary: "High memory usage detected"

      - alert: DatabaseConnectionFailed
        expr: alita_health_database == 0
        for: 1m
        labels:
          severity: critical
        annotations:
          summary: "Database connection failed"