Health & Monitoring

Check API health, service status, and diagnostics.

Health Endpoint


GET /api/health

Returns the overall health status of the API and its dependencies.

Response


{
  "status": "healthy",
  "timestamp": "2026-02-15T12:00:00Z",
  "version": "0.2.0",
  "checks": {
    "database": {
      "status": "pass",
      "latencyMs": 12
    },
    "solana": {
      "status": "pass",
      "latencyMs": 89,
      "message": "Slot: 245789012"
    },
    "redis": {
      "status": "pass",
      "latencyMs": 5
    }
  },
  "uptime": 86400,
  "circuitBreakers": {}
}

Status Values

Status	HTTP Code	Description
`healthy`	200	All systems operational
`degraded`	200	Optional services unavailable
`unhealthy`	503	Critical services down

Check Results

Each service check returns:

Field	Type	Description
`status`	`"pass"` \| `"fail"`	Health status
`latencyMs`	number	Response time in ms
`message`	string?	Additional info

Simple Uptime Check

For load balancers and uptime monitors:


HEAD /api/health

Returns 200 OK if the API is running (no body).

Circuit Breakers

When webhook endpoints fail repeatedly, circuit breakers prevent cascade failures.

The health response includes circuit breaker status:


{
  "circuitBreakers": {
    "webhook.example.com": {
      "state": "open",
      "failures": 3
    },
    "api.partner.io": {
      "state": "closed",
      "failures": 0
    }
  }
}

Circuit Breaker States

State	Description
`closed`	Normal operation
`open`	Requests blocked (3+ failures)
`half-open`	Testing if service recovered

Circuits reset after 5 minutes of being open.

Request Tracing

All API requests include a correlation ID for tracing:

Request Header (optional)


X-Request-ID: your-custom-id

Response Header (always present)


X-Request-ID: 550e8400-e29b-41d4-a716-446655440000

Use this ID when reporting issues or debugging webhook deliveries.

Rate Limit Headers

All responses include rate limit information:


X-RateLimit-Limit: 60
X-RateLimit-Remaining: 42
X-RateLimit-Reset: 1700000060

When rate limited (HTTP 429):


Retry-After: 60
X-RateLimit-Remaining: 0

Monitoring Best Practices

Health Check Intervals

Environment	Interval	Timeout
Production	30s	10s
Staging	60s	15s
Development	5m	30s

Alerting Thresholds

Metric	Warning	Critical
Health status	`degraded`	`unhealthy`
Response time	> 2s	> 5s
Error rate	> 1%	> 5%
Circuit breakers open	1+	3+

Example: Prometheus Scrape Config


scrape_configs:
  - job_name: 'etch-api'
    metrics_path: /api/health
    static_configs:
      - targets: ['etch.film.fun']
    relabel_configs:
      - source_labels: [__address__]
        target_label: instance

Dead Letter Queue

Failed launch jobs are saved to a dead letter queue for manual review.

Jobs end up in the DLQ when:

All retry attempts exhausted
Non-recoverable errors (e.g., invalid config)
Solana transaction failures after retries

Contact support to review and retry DLQ items.