Prometheus Metrics

The API exposes Prometheus metrics at /metrics for monitoring and observability.

curl http://localhost:3000/metrics

Available Metrics

HTTP

Metric	Type	Labels	Description
http_request_duration_seconds	Histogram	method, route, status_code	Request duration
http_requests_total	Counter	method, route, status_code	Total requests

Signup

Metric	Type	Labels	Description
signup_requests_total	Counter	endpoint, status	Signup requests
signup_duration_seconds	Histogram	endpoint	Signup processing time

Google Sheets

Metric	Type	Labels	Description
sheets_requests_total	Counter	operation, status	Sheets API requests
sheets_request_duration_seconds	Histogram	operation	Sheets API latency

Turnstile

Metric	Type	Labels	Description
turnstile_requests_total	Counter	status	Verification requests
turnstile_validation_duration_seconds	Histogram	-	Verification latency

Default Node.js process metrics (node_cpu_*, node_memory_*, process_*) are also collected.

Configuration

Metrics are enabled by default. Disable with:

ENABLE_METRICS=false

Prometheus Scrape Config

scrape_configs:
  - job_name: 'signup-api'
    scrape_interval: 15s
    static_configs:
      - targets: ['localhost:3000']
        labels:
          service: 'signup-api'

Docker Compose with Prometheus

services:
  signup-api:
    image: your-registry/signup-api:latest
    ports:
      - "3000:3000"

  prometheus:
    image: prom/prometheus:latest
    ports:
      - "9090:9090"
    volumes:
      - ./prometheus.yml:/etc/prometheus/prometheus.yml

Useful Queries

# Error rate
rate(signup_requests_total{status="error"}[5m])

# P95 latency
histogram_quantile(0.95, rate(signup_duration_seconds_bucket[5m]))

# Success rate
rate(signup_requests_total{status="success"}[5m]) / rate(signup_requests_total[5m])

# Sheets API latency
rate(sheets_request_duration_seconds_sum[5m]) / rate(sheets_request_duration_seconds_count[5m])

# Turnstile success rate
rate(turnstile_requests_total{status="success"}[5m]) / rate(turnstile_requests_total[5m])

# HTTP 5xx rate
rate(http_requests_total{status_code=~"5.."}[5m])

Grafana Dashboard Panels

1. Request Rate: rate(signup_requests_total[1m])
2. Error Rate: rate(signup_requests_total{status="error"}[5m])
3. P95 Latency: histogram_quantile(0.95, rate(signup_duration_seconds_bucket[5m]))
4. Sheets API Status: rate(sheets_requests_total[1m])
5. Turnstile Verification: rate(turnstile_requests_total[1m])

Alert Rules

groups:
  - name: signup_api
    interval: 30s
    rules:
      - alert: HighErrorRate
        expr: rate(signup_requests_total{status="error"}[5m]) > 0.1
        for: 5m
        labels:
          severity: warning
        annotations:
          summary: High error rate on signup API

      - alert: HighLatency
        expr: histogram_quantile(0.95, rate(signup_duration_seconds_bucket[5m])) > 5
        for: 5m
        labels:
          severity: warning
        annotations:
          summary: P95 latency above 5 seconds

      - alert: SheetsAPIFailure
        expr: rate(sheets_requests_total{status="error"}[5m]) > 0.05
        for: 5m
        labels:
          severity: critical
        annotations:
          summary: Google Sheets API failures detected