RNode Server Metrics Examples

Quick examples of how to use and extend RNode Server metrics.

🚀 Basic Usage

import { createApp } from 'rnode-server';

const app = createApp({ 
  metrics: true  // Enable Prometheus metrics
});

app.listen(4546, () => {
  console.log('Server running with metrics on /metrics');
});

📊 Available Metrics

HTTP Metrics

• http_requests_total - Total request count
• http_requests_duration_seconds - Request duration
• rnode_server_http_status_total - Status code counts

System Metrics

• rnode_server_process_cpu_usage_percent - CPU usage
• rnode_server_process_memory_kb - Memory usage
• rnode_server_uptime_seconds - Server uptime

Business Metrics

• rnode_server_cache_hits_total - Cache hits
• rnode_server_cache_misses_total - Cache misses
• rnode_server_slow_requests_total - Slow requests

🔧 Custom Metrics

import { record_cache_hit, record_cache_miss } from 'rnode-server';

// Cache middleware
app.use((req, res, next) => {
  const cacheKey = req.url;
  
  if (cache.has(cacheKey)) {
    record_cache_hit();
    return res.json(cache.get(cacheKey));
  }
  
  record_cache_miss();
  next();
});

📈 Monitoring Queries

Request Rate

rate(http_requests_total[5m])

Error Rate

rate(http_requests_total{status=~"5.."}[5m])

95th Percentile Response Time

histogram_quantile(0.95, rate(http_requests_duration_seconds_bucket[5m]))

Cache Hit Ratio

rate(rnode_server_cache_hits_total[5m]) / (rate(rnode_server_cache_hits_total[5m]) + rate(rnode_server_cache_misses_total[5m]))

🎯 Performance Alerts

# High error rate
expr: rate(http_requests_total{status=~"5.."}[5m]) > 0.1

# Slow requests
expr: rate(rnode_server_slow_requests_total[5m]) > 0.05

# High CPU usage
expr: rnode_server_process_cpu_usage_percent > 80

# High memory usage
expr: rnode_server_process_memory_kb > 1048576  # 1GB

📱 Quick Test

# Start server with metrics
npm run dev

# Check metrics endpoint
curl http://localhost:4546/metrics

# Generate some load
ab -n 1000 -c 10 http://localhost:4546/hello

# Check metrics again
curl http://localhost:4546/metrics | grep http_requests_total